Citation
New horizons for media framing analysis and military spouse employment

Material Information

Title:
New horizons for media framing analysis and military spouse employment
Creator:
Culer, Kallie Jae ( author )
Language:
English
Physical Description:
1 electronic file (65 pages) : ;

Subjects

Subjects / Keywords:
Mass media -- Social aspects ( lcsh )
Military spouses -- Employment ( lcsh )
Genre:
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Review:
Media framing analysis has developed in theory and application as a subset of political communication theory for the past several decades. Recent efforts among its scholars have sought to identify and create generalizable frames that can be applied across numerous social and political issues for the purposes of further examining the impact media coverage has on public opinion and policy development. This study utilizes previously established frames to examine media coverage of military spouse employment over the past twenty years, in order to better identify what role media coverage has played in the development of employment policy and support programs for military spouses.
Bibliography:
Includes bibliographical references.
System Details:
System requirements: Adobe Reader.
Statement of Responsibility:
by Kallie Jae Culver.

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
953488656 ( OCLC )
ocn953488656
Classification:
LD1193.P86 2016m C95 ( lcc )

Downloads

This item has the following downloads:


Full Text
NEW HORIZONS FOR MEDIA FRAMING ANALYSIS AND MILITARY SPOUSE
EMPLOYMENT
by
KALLIE JAE CULVER
B.A., University of Arkansas at Little Rock, 2011
A thesis submitted to the
Faculty of the Graduate School of the
University of Colorado in partial fulfillment
of the requirements for the degree of
Masters of Public Administration
Public Administration
2016


2016
KALLIE JAE CULVER
ALL RIGHTS RESERVED
11


This thesis for the Master of Public Administration degree by
Kallie Jae Culver
has been approved for the
Public Administration Program
By
Christopher Weible, Chair
Malcolm Goggins
Rebecca Glazier
Date: 12 April 2016
in


Culver, Kallie Jae (MPA, Public Administration)
New Horizons for Media Framing Analysis and Military Spouse Employment
Thesis directed by Associate Professor Christopher Wieble
ABSTRACT
Media framing analysis has developed in theory and application as a subset of political
communication theory for the past several decades. Recent efforts among its scholars have
sought to identify and create generalizable frames that can be applied across numerous social
and political issues for the purposes of further examining the impact media coverage has on
public opinion and policy development. This study utilizes previously established frames to
examine media coverage of military spouse employment over the past twenty years, in order
to better identify what role media coverage has played in the development of employment
policy and support programs for military spouses.
The form and content of this abstract are approved. I recommend its publication.
Approved: Christopher Wieble
IV


TABLE OF CONTENTS
CHAPTER
I. MILITARY SPOUSE EMPLOYMENT AND MEDIA FRAMING ANALYSIS...................1
Introduction...............................................................1
Background and Significance..............................................4
Research Questions.......................................................7
II. FRAMING THEORY AND MILITARY SPOUSE EMPLOYMENT POLICY
DEVELOPMENT................................................................8
Review of Literature.......................................................8
Media Framing Theory Development.........................................8
Research and Policy Development on Military Spouse Employment...........13
III. RESEARCH DESIGN AM) METHODS......................................... 18
Materials and Methods Utilized............................................18
Research Design.........................................................18
Research Population and Sample..........................................18
Materials and Preparation...............................................20
Coding Reliability......................................................23
Research Hypotheses.....................................................24
Measurements and Calculations...........................................25
IV. MEDIA COVERAGE RESULTS................................................27
Findings and Implications.................................................27
Differences in Media Coverage Across Time...............................27
v


Nature of Frame Differences Across Media Coverage................31
Differences in Media Coverage Between Groups.......................34
Nature of Frame Differences in Media Coverage Between Groups.....35
Frame Usage Differences in Media Coverage Between Groups.........39
Conclusion...........................................................45
REFERENCES...........................................................50
APPENDIX
A. NEW HORIZONS FOR MEDIA FRAMING ANALYSIS AND MILITARY SPOUSE
EMPLOYMENT: CODEBOOK.................................................54
CODING CATEGORIES:.................................................54
CODING PROCESS:....................................................54
RECORDING CODING CATEGORIES:.......................................58
vi


LIST OF TABLES
TABLE
1 - Increase of Media Coverage 1996-2015.................................28
2 - Yearly Increase of Media Coverage 1996-2015.........................29
3 Nature of Frame Usage 1996-2015......................................32
4 Nature of Frame Usage Counts.........................................37
5 - Frame Comparisons Across Newspaper Groups...........................41
6 - Frame Usage Rate Between Media Groups...............................42
i


CHAPTER I
MILITARY SPOUSE EMPLOYMENT AND MEDIA FRAMING ANALYSIS
Air Force Times, July 24, 1995 Were hearing that were officers spouses
and we do not need to work because our husbands make enough money, said
Audrey Moore, wife of a Marine stationed at Twenty-nine Palms, California.
Therefore, we should be staying at home and having children. Theyre not
coming out directly, but senior officers are indirectly putting pressure on
junior officers about [their] wives." (lowers, A Rocky Road, 1995)
Associated Press, June 29, 2011 In an effort to address military spouses
employment challenges, the Defense Department today launched a program to
expand career opportunities for military spouses worldwide, and to recognize
the skills and talents they bring to the employment table. (Sanchez, 2011)
Introduction
In todays struggling economy, there is a particular subset of individuals struggling
against great odds to maintain gainful employmentthese individuals are the spouses of US
Military Members. In contrast to the general public, military spouses today struggle with a
notoriously difficult unemployment rate reported to be 26% or higher and a wage gap of 25%
(Military Spouse Employment Partnership, 2014; Bushatz, 2014). In todays bit-driven 24-
hour news cycle, why would there ever be media coverage of such a select group of the
American population dealing with such unique employment challenges? Wouldnt that
coverage largely only exist within military media publications? Not when national figures
take on the cause of addressing those very unique challengesnamely, the nations First
Ladies, Michelle Obama and Jill Biden.
So the question then arises, why military spouse employment? What is it about these
challenges military spouses face that interested such figures in taking up their cause? Does
the employment of a military spouse play a vital role in maintaining financial security and
family wellness among military families? What is public perception of military spouse
1


employment, both within the military and US Government communities, as well within the
public at large? If it is an issue of national policy to bring awareness and program support to,
has it always been this way? Was it ever different? The first news excerpt displayed above
gives a glimpse into negative perceptions of military spouse employment that were long held
among the military community itself. The stark difference in the news excerpts presented at
the beginning of this thesis suggests that coverage is very different today from what it was
twenty years ago. This thesis aims to reveal exactly how this coverage has changedin both
amount and qualityover time.
These changes matter because media coverage influences public opinion and public
policy. Scholars across the fields of communication, psychology, and political science began
several decades ago to build a cross-field body of research on the impact media has in
setting, influencing, and impacting public policy and public perception of many social issues.
One important aspect of this research is media framing analysis, which researches
specifically how the way the media frames an issue might impact public policy and public
perception of that issue. In other words, does the weight given to an issue, or emphasis
placed on it, or tone used, or word choices the media utilizes to portray any given issue have
any impact on public perception and future policy development surrounding that issue? A
multitude of studies developed around measuring this phenomenon argues that it does
(Gamson & Modigliani, 1989; Iyengar, 1990; Druckman, 2001; and Boydstun, Gross,
Resnik, & Smith, 2013).
Given its cross-field nature however, many of these studies on media framing
analysis are uniquely diverse in both their definition and their methods. Thus, todays experts
in these fields are pressed with a need to create a standardized definition for media frames, as
2


well as consistent and applicable methods for any researcher within any one of these fields to
conduct media framing analysis on any given issue. Recent research by scholars Boydstun et.
al. however appears to be the beginnings of addressing this very need. By creating both a
systematic approach to identifying and coding media frames, as well as creating two models
for systematic computer analysis of the data, future research now has a potential guide map
to follow (Boydstun, Gross, Resnik, & Smith, 2013).1 With this addition of bringing
centralized definition and method instruction to the body of existing academic research on
framing analysis, an opportunity now exists for those interested to test the applicability of
their frames and analysis methods on issues where there has been a noticeable change in
policy development and public perception.
As revealed in the beginning, one such issue that merits further research is examining
how media coverage of military spouse employment has changed over the last twenty years.
Today the news cycles when it comes to the US Military, more often than not targets
coverage of troop movement, troop reductions, drone strikes, pay cuts, defense spending and
budget battles, retirement plan changes, or scandals within the Veterans Administration.
With the backdrop of an ongoing, politically controversial and laborious war in the Middle
East that has kept the country engaged in conflict far longer than any other American war -
most in the general public stand conveniently distanced from the daily woes of the military
life and frustrated at the seemingly never-ending conflict America has found itself in. Despite
these quandaries and frustrations, a unique window of opportunity has risen under the current
presidential administration, led by first ladies Michelle Obama and Jill Biden. This window
of opportunity has been a joint effort between public and private sectors, led by the White
House to highlight the importance of employment and transition assistance for veterans,
Retrieved at: http ://www.cs.emu, edu/nasmith/temp/frames-2013.pdf
3


transitioning military members, and military spouses (Hall, 2011; Sanner, 2011; Superville,
2013; and Military Spouse Employment Partnership, 2014). This national level of effort has
not gone unnoticed among military communities or traditional military media outlets. Given
the support growing from both the private sector and state governments for these efforts, the
employment plights of military members as well as their spouses have now also gained the
medias attention across national and regional media platforms.
Background and Significance
From its earliest beginnings, the U. S. Military community has relied heavily on the
volunteer support of its active-duty members counterpartsin a predominantly male
military force usually this meant their wivesto provide large amounts of unpaid social and
community support. In many ways it still does today, although the image of the military
spouse now includes both genders given increasing diversity among both those who serve
and those who join the supporting ranks by marriage. The traditional manner of volunteer
family-led support for the military community and its social fabric has long been referred to
by many as the two-person nature of a military career meaning that for any service member
who stays in the service for any length of time and takes on any leadership or staff role
(officer or enlisted) will also bring with those positions an amount of support expected from
their spouse as well. This need for spouse support and leadership has historically played a
direct role in influencing US military family policy, US Department of Defense (DoD)
family policy, and public perception on the subject of military spouse employment.
News archives from the mid-1980s tell of an investigation the Air Force began, after
some key command spouses went public claiming they were pressured to quit gainful
employment in order to provide full-time volunteer, unpaid community support for their
active-duty spouses command position. The consequence of not doing so? the very real
4


threat that it would affect command decisions about their husbands position, next promotion,
or next assignment (Black, 1987 and Fisher, 1987). Even with a military wide policy
directive forbidding this type of command influence established shortly after, this systemic
command influence continued to impact public perception of the role a military spouse
should occupy for years to come. However it was not the only obstacle military spouses
faced in terms of employment. To this day, female and male spouses attached to a military
member face significant difficulties in obtaining and maintaining gainful employment
because of the numerous moves, deployments, and family commitments associated with the
military lifestyle (Segal, 1986, pp. 22-30 and Harrell, 2001).
Much has changed over the past three decades in military policy, public perception,
and public policy support for a military spouses ability and personal need to balance
multiple roles and their right to be gainfully employed. The most visible developments as of
late, again, are the legislative and policy development efforts again led by first ladies,
Michelle Obama and Jill Biden, for the White House Joining Forces Initiative. With this
national leadership and supportnow advocacy organizations, governmental organizations,
state governments, and the private sector have been brought together to address the problem
of military spouse employment (The White House, 2013). Far from an in-house investigation
to determine attitudes of command influence on the issue, this level of policy development
and public support indicates change many spouses in the 1980s would have probably deemed
inconceivable.
In selecting a public policy issue to utilize as a subject of research for media framing
analysis, military spouse employment policy was chosen for its recent development on the
national policy level, as well as its personal significance to my own experiences as a military
5


spouse. I, too, am a military spouse determined to excel in my education and to find a way to
develop a meaningful career, while also actively supporting my husbands military career,
even with knowing the very real challenges and limitations military spouses face in
maintaining gainful employment. Through previous efforts to research military spouse policy
development, one of the key changes I observed were increased research efforts supporting
policy change, as well as increased media coverage of the employment challenges military
spouses face and efforts being made to address those challenges. Thus, it became apparent
that with enough individual champions, systemic research, and increasing media coverage
over the past thirty years policy changes have occurred (Semetko & Valkenburg, 2000;
Matthes & Kohring, 2008; and Dimitrova, Kaid, Williams, & and Trammell, 2005). This
realization alone showed me that media-framing analysis could now provide the necessary
grounded theory and methods for measuring how media coverage of military spouse
employment has changed over time and how it has framed the issue of military spouse
employment to both military audiences and the larger public.
Ultimately, both of these topics have a rich history of academic research, with expert
scholars supplying significant work to both framing theory and military spouse policy.
Bringing these bodies of research together will benefit each through providing insightful
connections and timely application. Consequently, this research project proposes to utilize
media framing analysis to identify, track, and measure media portrayal of the military spouse
and military spouse employment over the last thirty years. Thus my research questions were
designed to aid in discovering what changes have occurred and what differences exist
amongst media coverage of military spouse employment.
6


Research Questions
Research Question 1: How has coverage of military spouse employment across military and
nonmilitary sources changed over time, both in amount and content?
Research Question 2: How does coverage of military spouse employment differ across
military and nonmilitary sources?
These questions will be used to guide the data analysis process, so that testing the
hypotheses laid out in the methods section will be conducted in a way that adds meaningful
and valid research findings to the body of research already conducted on both media framing
analysis and military spouse employment. In order to properly situate how this research
project would add to existing research in these fields, the next section will briefly review
seminal academic literature and theoryfirst, for media framing analysis and secondly, for
military spouse employment.
7


CHAPTER II
FRAMING THEORY AND MILITARY SPOUSE EMPLOYMENT POLICY
DEVELOPMENT
Review of Literature
This project examines media coverage of military spouse employment over the past
twenty years from 1996 to 2015. In order to do that, systematic methods must be outlined and
established. For that to happen, it is important to first review how other scholars have
accomplished similar studies of both media coverage and of military spouse employment.
This will not only provide foundational theory and credence to the methods chosen for this
project, but it will also help situate this project and what value it hopes to add to both media
framing research and developing research on military spouse employment.
Media Framing Theory Development
Framing Theory has developed along a multitude of academic disciplines spanning
across sociology, psychology, political science, and communication, to name a few.
Beginning in the late 1970s, foundational theorists began studying framing effects and
theorized what influence they had both on the individual decision level and on a larger scale
through media communication (Kahneman & Tversky, 1979; Kahneman & Tversky, 1984;
Gamson & Modigliani, 1989; Iyengar, 1990; Entman, 1993; and Nelson, Oxley, & and
Clawson, 1997). Given this theorys multi-disciplinary nature, finding consensus on a
definition of what a framing effect is, what frame types can be applied systematically across
issues and fields of study, and what model of framing analysis provides the most applicable
methods has been ongoing for the past thirty years.
Formative work by Gamson & Modigliani in the 1980s, developed a definition of
framing within the media as, a central organizing idea, or frame, for making sense of
8


relevant events, suggesting what is at issue (1989, p. 3). This definition was later built on
and expanded by Entman, who said, Framing essentially involves selection and salience. To
frame is to select some aspects of a perceived reality and make them more salient in a
communicating text... (Entman, 1993, p. 52). These two definitions of framing suggest that
how media chooses to categorize and structure their reports and thoughts on events has a
measureable impact on its audience, just as the actual message of the text content does.
Nelson, et. al.s work on the Psychology of Framing Effects also helped to establish
definition by distinguishing between framing effects used by the media and direct messages
of persuasion and belief changethe former being an indirect use of structure and emphasis,
and the latter being a form of propaganda.
The turn of the century brought with it further development in media framing theory.
Druckman in his study on framing effects on citizen competence reveals another key
definition needed to truly understand two major trends of framing research the public level
and the individual level. According to Druckman, some disciplines have focused on showing
framing effects on an individuals thoughts and perceptions, while other disciplines have
attempted to measure framing effects on public and media communication (2001, p. 228).
Understanding how these vary in nature, also gives light to greater clarity of defining exactly
what framing is and whom it affects. So with the understanding of what framing effects are
and how they occur, the next question to be addressed is how have framing effects been
analyzed and measured?
Framing theory has largely developed by focusing on measuring framing effects
along two levels, as previously introducedthe individual thought level and the public
discourse level. At the individual level multiple disciplines often overlap in studies on
9


decision-making that attempt to determine if framing an individual choice in one way or
another might affect an individuals decision between the two, whether in survey research,
political polling, consumer choice, etc.. Legacy work here began with Kahneman &
Tverskys studies on prospect theory (also known as risky choice) and the development of
loss versus gain valence frames. These frames specifically examine how an individual
decision is affected when a choice is framed as a loss or a gain (Kahneman & Tversky, 1979
and 1984). In 1998, Levin and others contributed significant work to framing theory, by
creating a typology of three frame types commonly studied and compiling an extensive
review of existing studies on each frame type. These frame types were risky choice framing
effects, attribute framing effects, and goal framing effects (Levin, Schneider, & and Gaeth,
1998). Even more recently, political science scholars Druckman (2004) and Boydstun &
Glazier (2013) have further developed risky choice framing theory by respectively,
considering the role emotions play in responses to gain/loss frames and how the media
utilized these same frames to portray the war on terror. Knowing these scholars individual
contributions to the continued development of framing theory helps to establish not only that
framing effects are real, but also importantly reveals how they can transcend just one topic
and apply to a host of different social issues. Studying how framing effects occur across
social issues whether at the individual level or public discourse level will be essential for
better understanding what role they have played in media coverage of military spouse
employment.
Perhaps of all the scholars studying framing effects at the public discourse level,
Iyengars founding work on thematic versus episodic framing is especially key, as thematic
and episodic frames were found to be the most applicable to media coverage of military
10


spouse employment when selecting a general frame type to measure for. Thematic frames are
when the media discuss a particular issue in general or in the abstract, whereas episodic
frames refer to when the media frame an issue according to a specific example of it.
Iyengars work studied the framing effects within public media and its resulting impact on
public opinion concerning poverty and government welfare. He found through multiple
studies that framing poverty according to episodic framesnamely according to one poor
persons experiencestimulated more of an individual connection and expectation of self-
reliance, whereas the use of thematic framing to portray poverty as a collective social malady
elicited more of a desire for a collective response and an expectation for the government to
address the problem. Thus, his findings revealed that a mere framing of the issue played a
significant role in influencing public opinion on social welfare policy (Iyengar, 1990, pp. 21-
22, 28-29, 35-36).
Capella and Jamieson also contributed important work on media framing effects by
revealing its impact on public trust, levels of public cynicism, and public trust in the media
itself (Cappella & Hall Jamieson, 1996). De Vreese was another scholar that with the help of
colleagues conducted several studies on media framing effects, particularly revealing how the
media utilizes frames such as conflict, human interest, responsibility, and economic
consequences (Valkenburg, Semetko, & and De Vreese, 1999 and De Vreese, 2004). In the
end, regardless of diversity in approach, which level of framing is studied, or the number of
frames examinedmultiple studies have shown framing effects to be real at both the
individual decision level and throughout media discourse at large.
Framing theory does not come without its critiques and apparent limitations,
however. Druckman in particular questions its relevance and reveals that its impact largely
11


depends on the context and the individuals involved. He also goes on to enquire whether
framing theory is methodologically sound both in terms of validity and in terms of bias as he
wonders whether there might be a strong focus on only positive framing research results
(Druckman, 2004, pp. 684-85). Scheufele also presents a strong critique of framing analysis
and its methodology after compiling a classification and extensive review of existing
literature and studies across disciplines. In the end, two of his main criticisms are the need for
a replicable, longitudinal field study design, and the need for further research to incorporate
both the macro and micro (public discourse and decision-making) levels of framing analysis,
as he too often found one ignoring strengths of the other (Scheufele, 2004, p. 422).
Out of all recent research, Boydstun et. al.s recent development of a unified coding
scheme and systematic framing analysis models seems to present a strong response to these
critiques. Only time for further research and application, however, will determine if they are.
Along with Iyengars seminal work on episodic versus thematic framing, Boydstun et. als
research was chosen as the key source in theory behind the specific framing methods to be
utilized for this project (Boydstun, Gross, Resnik, & Smith, 2013). The value in utilizing
their research is that their framing structure and generalizeable frames were created to have
broad application across social issues, which in turn provides systematic means for
examining media coverage of military spouse employment. Iyengars work on the nature of
frame utilized by media sourcesthematic versus episodicsuggests to be the most helpful
for tracking the policy impact of media coverage on military spouse employment across time,
as well as for examining differences in thematic versus episodic framing usage between
military media and nonmilitary media. Given that thematic framing is reported to have the
greater policy impact, it is expected that military media will have the most invested interest
12


in utilizing it. This expectation also comes into play when examing media coverage for
specific frame usage differences, according to Boydstun et. al.s list of specific frames. Here
again, if military media sources have an invested interest in finding, portraying, and
encouraging policy solutions to military spouses employment challenges, it arguably follows
that of the specific frame types, economic frames and policy prescription and evaluation
frames will be their most popular choices given these frames practical emphasis. In the end,
however, these are mere suppositions and the only way to truly know which framing methods
are utilized across time and how the two media groups differ in their usage of both levels of
framing will be by examining the data and analyzing it.
Research and Policy Development on Military Spouse Employment
In conducting a literature review for this research project, it is also important to
briefly review important academic scholarship that has been done on military spouses and
military spouse employment. It is also important to note key policy changes that have
occurred throughout the last 20 years that arguably result from the increased awareness and
policy development that academic research and increased media coverage have created.
Since the mid 1980s, scholars at RAND, a research organization with large funding support
from the US Government, and others interested in US military related research have been
conducting studies on the effects military spouses have on morale, enlistment, and retention
rates (RAND Coorporationj.This research began largely after the policy change implemented
across military branches through the DOD Directive 1400.33 created in 1988. This directive
specifically prohibited military commanders from considering spouses employment status or
their volunteer work in making promotions, assignment decisions, and performance reports
(Department of Defense, 1988). Before this directive, a spouses volunteer involvement and
employment status could be commented on in official performance reports and often played a
13


factor in command decisions on promotion and new base assignments (Harrell, 2001, p. 56).
The existence of improper command perception of spouse employment and unfair command
influence, as well as the shifting numbers of women into the workforce at that time
nationwide, necessitated that US military-family policymakers reconsider what support was
offered and what expectations were placed on spouses in the future (Segal, 1986 and
Schwartz, Wood, & Griffith, 1991). Research soon revealed that in addition to military
community expectations and commitments, military spouses continually faced the following
challenges in finding and keeping employment: frequent moves, licensure portability,
childcare considerations, and deployments.
With the turn of the century, a new interest in research and demand for policy
development occurred, as the military began to realize the very real impact the spouse and
family had on a military members retention and reenlistment. Research by social and
military policy experts largely recommended solutions that would greatly improve spouse
employment chances, having been convinced that an employed spouse and thus a satisfied
spouse directly impacts the military members decision to stay in the military, and vice versa.
Some recommendations were shortening moving cycles, developing programs giving
incentives to local businesses to hire military veterans and military spouses, and finally
developing policies and legislation at the State and Federal level to address licensure and
certification portability issues many spouses face with moving across state lines frequently
(Harrell, 2001; Harrell, Lim, Castaneda, & Golinelli, 2004; Castaneda & Harrell, 2008).
Some of the most key policy developments for military spouse employment in the last
decade has, first, been the development of the military spouse federal hiring preference,
created under President Bush in 2008 with Executive Order 13473 and then expanded again
14


in definition and application under President Obama in 2009. As it stands currently, this
special hiring preference administrated by the U. S. Office of Personnel Management gives
military spouses the ability to develop a portable civil service career alongside their active
duty military member (The US Department of Defense, 2012 and The U.S. Office of
Personnel Management, 2012). In addition to this special hiring preference, the Military
Spouse Employment Partnership forged under the First Ladies Joining Forces Campaign as
an expansion of an Army specific military spouse employment program has brought
additional career progression opportunities to military spouses in the past five years within
the private sector. This program created in 2011 is now administered across all branches of
the military and has created a partnership with numerous private companies promising to hire
and provide portable careers for veterans and military spouses (Military Spouse Employment
Partnership, 2014).
Most recently in terms of academic research, The RAND Corporation has
spearheaded several research projects looking into the roles of military spouses and the
challenges they currently face with employment. In 2009 RAND Project Air Force released a
report detailing survey results conducted on 802 Air Force military spouses. (Miller,
Meadows, Hanser, & Taylor, 2011) The Institute for Veterans and Military Families in
coordination with the Military Officers Association of America also recently conducted a
survey of Military Spouses in 2013, and released its report and findings in February of 2014.
(Maury & Stone, 2014) Another military advocacy organization, Blue Star Families also
conducted a survey in cooperation with The Institute for Veterans and Military Families on
military lifestyle issues just this year, which also added key insights for further research into
the challenges and efforts being made for military spouse employment. (Blue Star Families,
15


2015) Ultimately these research findings on issues military spouses face with finding and
maintaining employment are valuable for the White House Joining Forces Initiative, the
Department of Defense, for military families, and for employers and businesses across the
nation looking to partner with veterans and their families.
For purposes here, reviewing both the development of framing theory and research
surrounding the policy development of military spouse employment efforts, provides key
insight into how the first can assist with better tracking and measuring the second. In order to
better understand why and how media coverage of military spouse employment has changed
and increased over the past twenty years, one must first have a means for measuring that
coverage. Framing theory and the role it plays in media analysis will supply the necessary
means. Noting increasing academic research efforts and policy developments for military
spouse employment also plays an important role for this project, in that they are often the
subject of said media coverage, and thus provide specific events to target and measure
coverage before and after within the set time period of the past twenty years. Through
researching both media framing and military spouse employment thus far, there are two key
expectations that create the foundation for the research hypotheses discussed in the next
section, which will provide specific and measureable means to answer the general research
questions discussed earlier. These expectations are first that media coverage will show
changes across all sources in amount and content through both levels of framingnamely,
the nature of frame utilized and the specific types of frames discussed. The second follows
more specifically, in that differences and changes between the groups of media sources are
also expected. Thus, with these outlined expectations and having established both a relevant
academic history of both media framing and military spouse employment, the next section
16


will outline the specific methods used to track, measure, and and analyze media coverage of
military spouse employment.
17


CHAPTER III
RESEARCH DESIGN AND METHODS
Materials and Methods Utilized
Research Design
The proposed research design draws primarily from theory and structural designs
illustrated in the works of Boydstun, Gross, Resnik, & Smith, 2013; Dimitrova, Kaid,
Williams, & and Trammell, 2005; Matthes & Kohring, 2008; Iyengar, 1990; and Semetko &
Valkenburg, 2000. A deductive structure was first chosen, as it required the variables and
frames selected for measurement to be clearly defined prior to any research. This structure
directly targets limiting researcher bias. This was important to establish given the fact that
while being a student researcher, I am also a military spouse inquiring into a topic that
directly connects to my current life. Other framing analysis studies have taken on a more
inductive nature, where researchers discover frames as they go, but with this comes the
danger of selection bias making these studies harder to validate and replicate (Macnamara,
2005, p. 9).
Research Population and Sample
The target population being studied in this research project is media coverage of US
military spouse employment over the past 20 years. The research has utilized content analysis
to examine printed media sources discussing any issues related to military spouse
employment. Thus the unit of analysis was printed media articles accessed electronically,
gathered from the following sample sources in two distinct categories: military media sources
and nonmilitary media sources. In the military category, sources utilized were specific
military and government newspapers, namely The Air Force Times, The Navy Times, The
Army Times, The Federal Times, and DOD documents in the form of government press
18


releases and official government blog posts. In the nonmilitary media category, sources were
selected first from the following national newspapers: The Associated Press, The Washington
Post, The Washington Times, The New York Times, The Wall Street Journal, and USA Today.
After retrieving all data available from these sources, local newspapers were also accessed
based on their close vicinity to military bases and subsequent coverage of military issues
namely the Virginian Pilot located in Norfolk, Virginia; The El Paso Times located in El
Paso, Texas; the Florida Times-Union located in Jacksonville, Florida; The Deseret News
located in Salt Lake City, Utah; and The Fayetteville Observer located in Fayetteville, North
Carolina.
All of these sources were discovered, selected, and retrieved from the following
databases: Lexus Nexus, Americas News, Pro Quest, and Academic Search Premier Plus.
These sources were chosen based on the following criteria: 1) connection to target
population, 2) national, local, and military reputable sources to provide differing
perspectives, and 3) accessibility (Babbie, 2013, pp. 332-335). The national papers were
selected as well to account for potential bias in coverage, in that both liberal and conservative
media sources by reputation were chosen. Several military branch media sources, as well as
two from the Federal government were also chosen to ensure a fair sampling of perspective
and coverage across the military branches and DOD. In terms of sampling, articles were
gathered from a 20-year span, starting in 1996 up to present day. A total of 235 articles on
military spouse employment were collected from military news sources, and a total of 117
nonmilitary sources were gathered. During the coding process, a final check for relevant
material within each article was made and the final total of articles came to 314, which broke
down into 204 military articles and 110 nonmilitary articles. For sample selections, this
19


research project utilized systematic random sampling to select the articles from each
newspaper. Given the total number of articles from each newspaper varied selection was
conducted as follows: every 10th unit from groups of articles larger than 40, every 5th for
groups of sources ranging in number from 15-40, and one unit was randomly sampled from
groups of sources that amounted to 10 or less articles.
Materials and Preparation
Research preparation for this project began with collecting the necessary articles and
creating a codebook. This was completed prior to coding any data, so that once the actual
sample materials were drawn from the entire collection they would each be sampled in the
same systematic manner. The codebook and list of coding categories are included as
appendices and provide clear operational variables that were recorded and measured. The
codebook also lists the coding processes and parameters in detail for any future research that
might want to utilize them. In brief, coding categories included: the source id, the newspaper
group, the article date, the relevancy of the article, the primary frame, the primary tone, the
nature of the frame, and finally whether the article was printed in more than one newspaper.
The primary text frames and tones were preselected from Boydstun et. al. list of
generalizable frames and frame tones prior to research. The frames chosen from their listing
initially were as follows: Economic, Quality of Life, Cultural Identity, Political, Public
Opinion, and Policy Prescription and Evaluation. These frames were chosen based on their
direct applicability and connection to the issue of military spouse employment. In
determining how relevant they were, Boydstuns detailed examples and defined uses for each
frame were considered in connection to military spouse employment to see how applicable
they were. If there was no direct connection to be made than that frame was not selected
20


(Boydstun, Gross, Resnik, & Smith, 2013, pp. 4-5 and Boydstun & Gross, Policy Frames
Codebook, 2014).
The choice to look at the nature or type of frames being used, namely thematic or
episodic frames, stems again from Iyengars seminal work on media framing (1990).
Stepping back to look at the overall nature of the frames used, according to his research,
points to valuable data on how the public will largely respond to coverage of a given issue, as
discussed in the literature review of framing theory. Here the use of thematic frames refers to
media coverage of military spouse employment in general, which according to Iyengar will
largely elicit a response of an expected collective responsibility for solutions; whereas
episodic refers to coverage of military spouse employment in terms of a specific spouse or
group of spouses, which according to Iyengars work the use of elicited a response of self-
reliance and individual responsibility for solutions (Iyengar, 1990). Thus, operationally I
looked to see whether an article covered military spouse employment from the overarching
perspective of military spouses at large, or if it discussed the perspective of specifically
named spouses and their issues with employment. This process of selecting frames, tone, and
the nature of frames created by other researchers, again, was part of the deductive design of
the project, to prevent researcher selection bias.
Once I began conducting research and collecting data however, it soon became
apparent that utilizing a purely deductive structure, where I would only use previously
established frames, frame types, and tone values made it difficult to clearly identify what
impact and value those predetermined variables had in relation to my specific data in a way
that would enable other potential researchers to code and replicate my processes consistently.
Boydstun et. al. frames in particular were so broad in nature within my coding processes that
21


it was difficult to not find overlap or a presence of all frames in every article. Thus their
existence revealed very little about their specific impact or role long term in media coverage
trends concerning military spouse employment. So in response to these difficulties, I needed
to find a way to make the frames more targeted in how they were identified across each
article. This in turn would also show the value in the usage of different frames and difference
of frame and tone usage across multiple media sources. Thus I decided to reduce the number
of Boydstun et. al. primary frames I would use to the four most applicable frames of
Economic, Quality of Life, Cultural Identity, and Policy Prescription and Evaluation.
Next I chose to create and use specific sub-frames for each of these primary frames
that produced issue specific variables consistently found through the data. Those subframes
created are as follows: Economic broke down into licensure portability, unemployment,
employer interest, and income/financial contributor; Quality of Life broke down into
retention, support programs, separation, and frequent moves; Cultural Identity broke down
into volunteerism, entrepreneurship, and career flexibility; and finally the Policy Prescription
and Evaluation frame was deemed specific enough to not need subframes.2 I also
encountered difficulty during this process, in utilizing Boydstun et. al. tone values, due to
concerns of subjectivity and an inability to replicate implicit tone values across coders. So in
response to this, I first limited the tone values to be identified to explicitly pro, explicitly
negative, balanced (a mixture of explicitly pro and anti), or no tone at all. In making these
adjustments to both the frames and tone values, I conducted a practice code of a sample of
my data, as well as worked with training a second coder through several sample sets so as to
ensure my coding methods and selection of subframes and tone values applied consistently
Detailed definitions and examples of how to identify each subframe throughout the data are supplied in the codebook appendix.
22


and were valid. It was during this process that both my second coder and I identified that
measuring for tone in a consistent and reliable manner proved to be impossible, given that it
was far too subjective in nature to consistently code. The reason for this we largely
determined to be that the subject nature itself was neutral in tone more often than not and
thus rarely presented clearly positive or clearly negative examples of tone. In discovering
this, I decided to remove tone as a measurable variable from the project.
Coding Reliability
In most content analysis studies, two coders will blindly code a set of articles selected
to determine coding reliability. Once coding is completed a final reliability test then
determines the percentage of agreement between coders in order to ensure intercoder
reliability (Macnamara, 2005, pp. 10- 11, 20). As materials were coded, all coding results
were entered into a data table using Excel in the form of a sample tally sheet. (Babbie, 2013,
p. 338) This tally sheet recorded all important code categories. This tally sheet later formed
the dataset that was then used for statistical analysis to determine whether there were any
significant differences in the groups of sources, as well as differences across time. Given the
total dataset included 314 articles, the sample sets the 2nd coder worked with were selected in
groups of 20-30 articles. At the end of the coding process, there were two sets of intercoder
reliability tests runpercentage of agreement and Cohens kappa. In looking at the coding
reliability for the nature of frame variable, the percentage of agreement between coders
obtained was 100% and the resulting Cohens kappa value was 1.00. In looking at the coding
reliability for the individual frames utilized by media groups, each test was conducted for
both the collective primary frame variable and the individual subframe variables. The
collective primary frame variable results and their levels of reliability however were directly
linked to how reliable the individual subframe values were, as they were a summation of
23


each of those individual subframes presence or absence. Across the individual subframes the
percentage of agreement figures ranged from 90-100% and the Cohens kappa values ranged
from .77 to 1.00. Ultimately though, the primary frame reliability figures sufficiently
revealed that the level of coder reliability needed was reached with a percentage of
agreement value of 95% and a Cohens kappa value of .94.
Research Hypotheses
This research project, as stated earlier, sought to discover whether there were any
differences between military and nonmilitary media coverage of the issue, or whether there
were any differences in the frames or tones utilized between the groups of sources. In light of
this, the following research hypotheses were created to specifically address research
questions 1 and 2. Hypotheses 1(a) and 1(b) directly answer the first research question, and
Hypotheses 2(a), 2(b), and 2(c) all seek to answer the second research question.
Hypothesis 1(a): Media coverage of military spouse employment will increase across all
sources from 1996-2015. A noted spike in coverage is expected following President Obamas
election in 2008.
Hypothesis 1(b): Coverage of military spouse employment across all sources will use both
thematic and episodic frames.
Hypothesis 2(a): Military sources will have more coverage of military spouse employment
than nonmilitary sources.
Hypothesis 2(b): Military sources will use more thematic frames than nonmilitary sources.
Hypothesis 2(c): There will be differences in frames utilized by military media sources and
nonmilitary media sources. It is expected that nonmilitary sources will utilize more
Economic and Policy Prescription and Evaluation frames than military sources.
Utilizing this research design and examining the data according to these parameters made
statistical analysis possible to test these hypotheses. In doing so, it was possible to discover
what role military and nonmilitary media sources played in covering the issue of military
24


spouse employment, how coverage has increased over the last twenty years, and what
differences in their coverage existed in terms of frame usage.
Measurements and Calculations
There are a number of measurements and calculations available to analyze the data
this research project gathered, but there are two specific measurement purposes that it sought
to accomplish. First, the hypotheses and research questions formulated specifically required
descriptive statistical measurements that will describe the occurrence of media coverage on
the chosen issue over the set time period. Descriptive statistics revealed which of the
preselected frames and tones each newspaper and each media group utilized the most
frequently. The second measurement purpose these parameters required was to determine
what interventions or differences were at play between the two media groups and whether
they were statistically significant. This allowed for statistical analysis of the difference in the
types of frames and tone each media group utilizes.
For descriptive purposes, some common measurements and calculations are measures
of central tendency, such as the mean and the mode, and a frequency distribution. Calculating
the mean and mode reveals important data, such as the most frequently utilized frame or
tone. Calculating a frequency distribution of the data gathered in graphical form helps to
reveal the progress of media coverage over time across each source individually, as well as
collectively across each group (California State University, Northridge).
It is always important to do more than just describe what the data was revealing. This
is where the second purpose of determining statistically significant differences between the
groups of data gathered becomes key. There are several tests available to measure differences
between two groups, but the effectiveness of each largely depends on the type of variables
being examined. In this research project, the nature of frames, primary frames, and subframes
25


examined and coded for are considered nominal variables, in that they are each distinctively
different and not ordinal, interval, or ratio in nature. Thus given their nominal nature, the
Chi-Square independent test, represents the best form of statistical measurement for
measuring the difference between the groups. This test largely describes and reveals the
difference between observed frequencies and expected frequencies between two groups,
similar to the Cohens Kappa reliability tests run on the sample set coded by both myself and
the second coder (California State University, Northridge).
In more technical terms, the hypotheses and research questions also include in their
breadth the basic questions that the chi-square test always seeks to answerthat of the null
hypothesis and the alternative hypothesis. The former implies that there is no relationship
between the two variables studied, and the latter would imply that there is. Using the Chi-
square independent test for analysis requires the level of significance to be outlined, followed
by calculating the degrees of freedom, expected frequencies, a test statistic, and finally a p-
value from the sample data. Once the data is analyzed and the tests run, the results will reveal
what, if any, significant differences exist between the variables (Babbie, 2013, pp. 475-477).
This form of statistical analysis was also predicated in a similar form of media framing
content analysis by Dimitrova et. al. in 2005. The only way to determine this is through
statistical analysis. Thus the difference between military and nonmilitary media coverage in
terms of their frame and nature of frame usage need to be examined for statistical
significance. Thus, the next section will discuss what results, examining and analyzing the
data revealed and what differences, if any, occurred across time and between the two media
groups.
26


CHAPTER IV
MEDIA COVERAGE RESULTS
Findings and Implications
From the beginning this project has sought to discover, track, and identify changing
media coverage of military spouse employment over the past twenty years. The research
questions were specifically designed to approach the data for answers on not only how the
coverage has changed over time in both amount and content, but also how that coverage
differs between the two key groups of media sources studiedmilitary and nonmilitary
printed media. These research questions led to hypotheses that outlined measureable
expectations and outcomes for this project that could either be found and confirmed or at
least provide differing measureable results that could be clearly explained. The data itself has
revealed both expected and surprising results. Either way, audiences interested in the study of
media framing or military spouse employment or both can glean valuable new insights to the
role the media has played in covering military spouse employment over the past two decades,
as well as how military spouse employment itself has changed and increased since the early
1990s.
Differences in Media Coverage Across Time
The first research question sought to discover how media coverage of military spouse
employment has changed over time in both amount and content. Hypothesis 1(a) predicted
that the amount of media coverage would increase across all sources over the set 20-year
period, but that the greatest spike in coverage would be seen after 2008 when President
Obama was elected. The data revealed both these expectations to be largely true.
27


Table 1 Increase of Media Coverage 1996-2015
Increase of Media Coverage on Military Spouse Employment
1996 to 2015 (in 5-year increments)
Nonmilitary Newspapers Military Newspapers
2001-2005 2011-2015 2001-2005 2011-2015
Nonmilitaiy Total Number of Articles = 110
Military Total Number of Articles = 204
28


Table 2 Yearly Increase of Media Coverage 1996-2015
Yearly Increase of Media Coverage on Military Spouse Employment
1996-2015
Nonmilitary Newspapers Military Newspapers
so-................................... .............................
Year
Nonmilitary Total Number of Articles = 110
Military Total Number of Articles = 204
H- t ro ro ro to ro ro to ro h- h- ro ro ro ro ro ro to ro
VO VO O O O O o VO VO o o O O O o o O
VO VO o o O o o t H- t VO VO o o O O O * H-
00 o ro -u Ov 00 o ro 0> 00 o ro 2k O 00 o ro A
29


Coverage along both groups of sources increased slowly over the first decade as shown in
Table 1. The greatest increase in coverage was seen after President Obamas election in
2008, however it was most notably seen in the years of 2010-2012 as seen in Table 2. This
spike suggests that the creation of President Obamas Joining Forces White House Initiative
in 2011, spearheaded by First Ladies Michelle Obama and Jill Biden played a pivotal role in
raising awareness and drawing media coverage to the issue of military spouse employment
(Sanner, 2011 and Sanchez, 2011).
Military Spouse employment itself was and continues to be a primary issue targeted
by the Joining Forces Initiative. As mentioned earlier, an important part of the Joining Forces
Initiative that made spouse employment such a targeted issue was the expansion of the Army
Spouse Employment Program into an all-branch wide DoD program into MSEP, the Military
Spouse Employment Partnership that brought companies across the private and nonprofit
sectors on board as military spouse hiring partners (Jowers, 2011). Examples discussing
Joining Forces and its focus on Military Spouse Employment abound amongst the coverage
across both groups of sources. The Associated Press reported following Joining Forces
launch in April of 2011, First Lady Michelle Obama called on companies.. .to recruit and
hire members of the military and their spouses, saying they offer unique skills and qualities
(Sanner, 2011). USA Today reported on the launch of Joining Forces by sharing, Military
families will be Michelle Obamas second signature issue... She and Jill Biden, the vice
presidents wife, say they intend to build a program that can outlast their husbands
tenure.. .Focused on helping families cope with issues related to employment, education, and
health, Joining Forces will be managed by a three-member board that includes retired
general Stanley McChrystal (Hall, 2011). Military sources also reported on the initiative
30


saying, Joining Forces engages the nation on military community issues, and highlights
areas where support is needed.. .that includes military spouses (lowers, 2011). Ultimately,
this increased coverage within the past five years has led to raised awareness, especially
among military spouses, of what efforts on both a national policy level and branch specific
program level have occurred since the First Ladies took up the torch.
Nature of Frame Differences Across Media Coverage
The coverage in terms of content also shows some changes across time throughout
both groups of media sources. In the first decade stories primarily focused on the challenges
military spouses faced in finding, gaining, and maintaining employment. In the second
decade media coverage focused more on finding policy solutions at the national and state
levels for addressing those still occurring and well-known challenges, as well as raising
awareness to existing military and community-based programs that have been created to
specifically help military spouses with employment. These trends can be seen on a larger
scale by looking at the nature of frames chosen across both groups of media, through the
examining their use of thematic versus episodic framing or a mixture of the two. Hypothesis
1(b) projected that that both groups of sources would utilize both thematic and episodic
frames.
By and large both groups increasingly utilized thematic frames the most, or a decided
mixture of the two, which still supports the original hypothesis that both groups would utilize
both types. However it was surprising to see how little episodic frames were used alone.
Very few between both groups of sources would ever use purely episodic framing throughout
an article. The use of thematic frames allowed both military and nonmilitary media sources
to continually portray the issue of military spouse employment as a common issue for a large
number of military spouses. One example from 1996 in a regional, nonmilitary newspaper
31


printed the following, The military has undergone many changes since it became an all-
volunteer force, says Carolyn Becraft, deputy assistant secretary of Defense for personnel
support, family and education. It has gone from a largely unmarried force under conscription
to a predominately married force today. According to Becraft, more than 60 percent of
military spouses have jobs (The Deseret News, 1996). The article then continued on in a
thematic nature to portray the growth of military family services such as base childcare, as a
direct response to the growing dual-career nature of the married military households and the
challenges that inherently presents.
The utilization of both thematic and episodic frames, allowed the media to address
common employment struggles for spouses with frequent examples of how individual
military spouses uniquely responded. One article from The Florida-Times Union shares of
the larger employment challenges military spouses face such as professional licensure
portability, high unemployment rates, and little to no unemployment insurance after
relocating due to military orders. In the midst of these larger issues though the article also
highlights a particular spouse Amy Wolford and the rare success she had in finding
employment on her military base thanks to the military spouse employment assistant program
specialist on base (Piatt, 2004). Of the very few articles presented solely from an episodic
frame, they commonly follow a particular spouses career journey alongside his/her military
spouses active military career and how they managed to make a dual career work or what
sacrifices the civilian spouse was often required to make (Moon, 2001 and Tuzzolino, 2008).
Ultimately though, over time the usage of thematic frames would grow increasingly greater
across both groups of media.
32


Table 3 Nature of Frame Usage 1996-2015
Nature of Frame Usage
1996-2015 (in 5-year increments)
1996-2000 2001-2005 2006-2010 2011-2015
70-'
33


Differences in Media Coverage Between Groups
The second research question looks to see what differences exist within coverage
between the two media groupsmilitary and nonmilitary. Hypothesis 2(a) first addresses this
question in suggesting that overall military media sources will have more coverage of
military spouse employment than nonmilitary media sources will. In terms of total amount of
coverage in the past two decades, military media sources carried nearly double the amount of
articles collected. As mentioned earlier, military articles totaled to 204 versus the nonmilitary
articles, which amounted to a total of 110 articles. Given coverage differs largely in amount
and frequency over the years, in that military media coverage of this issue is nearly double
that of nonmilitary media coverage of military spouse employment, one might question how
article counts and coverage between the two groups are directly comparable. To address this,
first, it is important to note that it only makes sense for military media coverage to be greater
overall given it is a subject that directly impacts their audience. As discussed previously,
studies on military spouse employment have revealed it to play a key role in determining and
impacting military retention. This factor alone makes following its development and
progression worthwhile for military media sources. As coverage has increased, military
spouses themselves have long relied on military media sources directly or indirectly shared
through base officials and services to continually update them to new programs, services, and
policy efforts on their behalf.
Even more recently, media sources, military service providers, and military interest
groups are continually seeking how to reach younger and newer generations of military
spouses with the most current information on what programs and services exist for them. As
one recent article in The Navy Times discussed, two of their newest efforts to get the word
out on military spouse employment efforts is both through generational targeted efforts via
34


social media online and through partnering with military interest groups to assist with
spreading information to their members and chapters (lowers, 2014). In the end, whether it
is for military retention interests or for informing interested audience membersnamely
military spouses themselvesmilitary media sources have played a key role in the amount of
media coverage military spouse employment has gained in the past twenty years. Knowing
this also helps for accounting for why media coverage differs so greatly in amount across the
two groups.
Last but not least, the argument for still comparing coverage between these two media
groups lies in the need to assess the differing audiences reached through their coverage. To
only measure media framing trends within military media sources, would provide a very
skewed picture, as military media sources typically only reach military audiences and they do
so at a different frequency than nonmilitary media does. National and regional nonmilitary
media sources however widen the scope of what audiences are reached and at what frequency
they are reached. Examining and comparing both groups for trends within their media
coverage, arguably gives a better understanding and holistic picture of how framing effects
are occurring.
Nature of Frame Differences in Media Coverage Between Groups
The next aspect of coverage examined for differences between the two media groups
was measured through looking at their different framing usage at the general level by
examining what nature of frames were being utilizedthematic, episodic, or both. Utilizing
thematic or both types of frames were increasingly common to both sets of media sources,
and there were no statistically significant differences amongst their usage of them. However,
it is still interesting to note how they did differ in their usage between the two groups over
time. Table 3 shows the differences between the two groups over time graphically, but given
35


the differences numerically are small, Table 4 reveals the differences in actual figures. As
revealed in Table 4, nonmilitary newspapers prioritized both thematic and episodic types of
frames more than solely thematic from 1996 to 2010 and it was only in the last five years that
their usage of solely thematic frames outgrew their usage of both. In the military media
coverage however, it was only in the first five-year period from 1996 to 2001 that their usage
of both types outweighed their usage of solely thematic frames. Here again the data reveals
that their dependence on thematic frames grew increasingly over the time period, with its
greatest spike in the past five years from 2010 onward.
The military medias use of more thematic frames lines up with Hypothesis 2(b),
which predicted that the military media would use more thematic frames than nonmilitary
media sources would. This is true even if you calculate the percentage rate each used, where
military media utilized purely thematic frames 61% of the time, versus nonmilitary media
sources utilizing them 48% throughout the time period. Earlier in the research process, it was
expected to find more usage of episodic frameshowever as shown in Tables 3 and 4 their
usage was quite infrequent between both groups. In terms of percentage, nonmilitary media
used solely episodic frames less than 1%, while military sources used them just less than 3%
of the time.3 The expectation to find more usage of purely episodic framing stems from
Iyengars founding work on framing effects, as discussed earlier. His fundamental work
revealed that how a policy or issue is framed, whether episodically or thematically, has
considerable impact on how the public responds to it. Episodic framing elicited an
expectation of individual responsibility whereas thematic framing of an issue as a collective
problem elicited an expectation for collective social responsibility (Iyengar, 1990).
3
Percentages were calculated by taking the number of specifically framed articles within each group and dividing them by
the total number of articles for each group. Exact figures found in Table 4.
36


Table 4 Nature of Frame Usage Counts
Newspaper Type Year Group Total
1996- 2000 2001- 2005 2006-2010 2011-2015
Thematic 4 6 12 31 53
Nature of
Nonmilitary Frame Both 7 9 14 26 56
Newspaper^ Episodic 0 1 0 0 1
Total 11 16 26 57 110
Thematic 9 20 27 69 125
Military Nature of Both Frame 12 14 19 28 73
Newspapers Episodic 1 0 2 3 6
Total 22 34 48 100 204
Thematic 13 26 39 100 178
Nature of
Total Frame Both 19 23 33 54 129
Episodic 1 1 2 3 7
Total 33 50 74 157 314
37


Iyengar was not alone in suggesting the very real impact framing effects within the
media and public discourse carry, but his work provides a key foundation for understanding
the role thematic framing has largely played in the coverage of military spouse employment.
Given both military and nonmilitary media sources have continually increased their use of
thematic framing, the data suggests according to framing theory that addressing the
challenges of military spouse employment is not a problem military spouses can individually
fix on their own. Given many of the challenges they face in employment are a direct result of
their collective service and connection to the military itself, media coverage implied more
and more that a collective response was/is necessary.
Perhaps this has been nowhere more apparent than in coverage of the Joining Forces
Initiative, which itself corresponds with the largest increase in the reliance on thematic
framing in the past five years across both media groups. Statements like the following were
made by Michelle Obama, "We should all be working together on this. These are pretty solid
Americans out here that are making these sacrifices quietly for all the rest of us" (Hall,
2011). Whether you're in finance or technology or the food industry, every single one of you
can ask yourselves that same question: 'What more can we do? Think outside the box, take
real risks and work together to make big, bold commitments to hire our veterans and military
spouses... (Superville, 2013). Again and again coverage thematically shows that the First
Ladies, the President, and numerous other advocates refer to helping military families and
addressing the issue of military spouse employment as a collective responsibility demanding
a collective response at all levels, national, state, and local. Given so many examples of
thematic framing increasing and knowing what response thematic framing purports to illicit
in terms of public response, the data suggests that media coverage itself was a key factor for
38


the increase in policy efforts, program funding, and legislative efforts made in the last decade
on behalf of military spouses and their employment.
Frame Usage Differences in Media Coverage Between Groups
The last and perhaps most extensive portion of the data examined for differences
between the two media groups occurred in looking at the differences in specific frame usage.
The frame choices, were again selected from Boydstun et. al. list of generalizable frames
according to which of them best applied to the issue of Military Spouse Employment. The
frame choices selected were an Economic Frame, a Quality of Life Frame, a Cultural Identity
Frame, and a Policy and Prescription Evaluation Frame. How these frames were specifically
identified and applied through the use of subframes for each frame category was established
during the coding process, as outlined in the previous methods section. When conducting the
final data analysis, the codes for each subframe were tallied and recorded in two additional
ways across the data sheetfirst they were compiled to give a sum of the number of
subframes used for each frame category and secondly a tally was taken to merely note the
presence or absence of any or all subframes with the identifier of 0 or 1, to be able to
accurately compare frame usage across articles and media groups so that some articles did
not skew the data with a heavier use or lighter use of subframes. These final tally results were
then transferred into an SPSS dataset for further statistical analysis.
In response to the second research question of looking for coverage differences
between the media groups, Hypothesis 2(c) sought to specifically identify how the groups
would differ in their specific frame usage. It proposed that nonmilitary media sources would
utilize more economic and policy and prescription frames that military media sources would.
The results were initially surprising in that specific frame usage did not differ between the
two groups as significantly as expected. In fact very little statistically significant differences
39


in frame usage were found. The significant differences that did occur actually occurred
across both media groups over time, rather than between them. This is not to say that there
frame usage was identical, or that there were no differences at all. What differences did occur
between the two groups, even if not statistically significant are still important to note. As
Table 5 reveals, comparing the means of frame usage across both media groups shows that
their overall frame usage was largely similar. Interestingly enough and different from what
Hypothesis 2(c) suggested, both media groups primarily relied on the use of economic frames
and quality of life frames, more so than the other two. However, the use of Policy frames did
become stronger over time, whereas the use of Cultural Identity frames remained the least
used between both groups throughout the entire time period examined.
Looking at the mean percentage rates of frame usage, as Table 6 shows, also reveals
how the frame usage between the groups was largely similar over the years. Small
differences did occur, the most notable of which are the use of the quality of life frames and
the policy frames between the two groups. The original hypothesis again, suggested that
nonmilitary media would utilize economic and policy frames more than military media
would. In reality both groups prioritized economic frame use throughout the entire period.
Interestingly enough though, both groups began with using more quality of life frames and
although economic frames began to show stronger usage over time across both groups, each
group still continued to use quality of life frames in fairly proportional amounts throughout
the time period.
40


Table 5 Frame Comparisons Across Newspaper Groups
Frame Comparisons Across Newspaper Groups in 5 Year Increments
Nonmilitary Newspapers Military Newspapers
I Economic
Quality of Life
Cultural Identity
Policy
41


Table 6 Frame Usage Rate Between Media Groups
Economic
Frame Usage Rate Between Media Groups
Quality of Life
Cultural Identity
Policy
1996-2000
2001-2005
42


The greatest difference in frame usage found between the groups and over time was
in their use of policy frames. Nonmilitary papers started utilizing policy frames more in the
first decade, and it wasnt till after 2006 that military papers started using it more frequently,
going from a 3% rate of use in 2001-2005 to a 26% rate of use in the 2006-2010 year range.
Reasons for this rapid growth in policy frame use across both media groups in the 2006-2010
time frames arguably stem from significant policy efforts made for military spouses during
those years under both President Bush and President Obama, upon his election. As discussed
in the Literature and Theory Development section, it was during this time that the Federal
Military Spouse hiring preference was created, both Presidents began advocating for hiring
support for military families, and by 2010 the coordination of the Joining Forces White
House initiative had begun. This is further reflected in the statistical analysis conducted on
the data collected. The primary statistical test for measuring the differences in the use of
frames between the media groups was the chi-square test, as outlined previously in the
methods section. Running the chi square tests on frame usage between military and
nonmilitary papers revealed very few statistically significant differences in their frame usage
overall. In fact, the only significant difference found was in the time period from 2001 to
2005, where their usage confirmed the null hypothesis of there being no significant
difference in military versus nonmilitary usage of policy frames to be incorrect. There was a
significant difference in their usage of policy frames during these years only, given there was
a significant value of .027 with the degrees of freedom being l.4
Given both military media sources and nonmilitary sources primarily used similar
amounts of each frame type over the last twenty years, the question becomes: what do these
4 In order to account for the low cell count due to a small sample size, the Fisher Exact Test
results within the Chi-square test were the final significant values used and recorded.
43


frame types emphasize in terms of content and how do they connect with the respective
media audiences? Looking at the larger, overall nature of frames already gave us the
impression that both military and nonmilitary media sources often frame military spouse
employment as a collective issue prompting a government-led, collective response according
to framing theory (Iyengar, 1990). Specifically though, which frames were utilized the most
to portray content on military spouse employment? As the tables above have already
revealed, the two most common frames utilized by both groups over time were Economic
frames and Quality of Life frames. According to Boydsun et. al. 2013 study and codebook,
Economic frames emphasize, the costs, benefits, or monetary/financial implications of the
issue (to an individual, family, community or to the economy as a whole). Quality of Life
frames however, instead focus on the effects of a policy on individuals wealth, mobility,
access to resources, happiness, social structures, ease of day-to-day routines, quality of
community life, etc. (Boydstun, Gross, Resnik, & Smith, 2013, pp. 4-5) Emphasizing both
of these aspects, by framing content covering military spouse employment according to each
of these perspectives seems logical. For the military audience it continually frames the issue
and any challenges or proposed solutions according to both important economic
considerations military families must continually assess, while also providing insight into
how those same issues and solutions derive from or impact their welfare and daily life. For
the nonmilitary audience, given much of military spouse employment is already framed as a
collective responsibility for the general public to care about addressing, framing the specifics
of the issue as having both economic and quality of life considerations arguably makes the
connection both practical and personal.
Last but not least, Policy Prescription and Evaluation frames are defined as,
44


particular policies proposed for addressing an identified problem, and figuring out if certain
policies will work, or if existing policies are effective (Boydstun, Gross, Resnik, & Smith,
2013, p. 4). This growth in the use of policy frames over time between both media groups,
knowing framing theory and how it suggests an impact on policy development suggests that
media coverage played a role in the development of military spouse economic policy. This
would also be true from the aspect of the use of thematic framing increasing across both
media groups as well. Looking at both the growth and use of specific frames, as well as the
nature of frames used over time both present compelling evidence that as coverage increased,
an awareness of military spouse employment challenges grew, which thus created motivation
to reexamine existing military policy and programs and to consider new policy solutions.
Conclusion
Ultimately, this project sought to identify and measure framing differences by first
asking targeted research questions about media framing and coverage of military spouse
employment and secondly, outlining specific hypotheses for answering them. The research
questions again aimed to find what framing differences occurred within the media coverage
across time and between the groups of media sources. The hypotheses defined how to track
those differences and framing changes through noting specific factors such as important dates
and events within the set time period, changes in the amount of coverage, the varied use of
thematic versus episodic framing, and the varied use of the chosen specific frames
Economic, Quality of Life, Cultural Identity, and Policy Prescription and Evaluation. In the
end, after conducting a significant amount of research into both media framing content
analysis and military spouse employment policy development, coverage over time and
coverage comparisons between military media sources and nonmilitary media sources clearly
45


showed some framing changes and differences.
This projects data largely revealed an increased amount of media coverage in recent
years with a subsequent, noticeable increase in program support and policy efforts for
military spouse employment endeavors. Thus, framing theory as it stands currently would
stipulate that how the media has portrayed military spouses and how it has portrayed policy
efforts made for them played a part, potentially a significant part, in the development process
and public support for it over time (Iyengar, 1990, p. 20). Iyengars seminal work on the
impact in thematic framing supports this when looking at the growth of public policy efforts
and public support for military spouse employment efforts at large. In addition to this, other
framing theorists previously examined, such as DeVreese, Capella and Jamieson, and
Boydstun et. al. all contributed significant studies analyzing and measuring the impact that
the use of specific frames such as conflict and economic frames, opportunity versus risk
frames, or a categorized list of commonly used specific frames have -both within media
coverage and public response to coverage of any given issue. Bodystun et. al. generalizable,
but specific frames in particular provided the necessary structure that this studied needed to
examine what type of frames the media commonly used to portray the issue of military
spouse employment. This studys results showed the usage of an economic frame was by far
the most common, whereas the usage of the policy frame showed the most change and
increase over time across both media groups (De Vreese, 2004; Cappella & Hall Jamieson,
1996; and Boydstun, Gross, Resnik, & Smith, 2013). In the end, the only way to confirm
whether media coverage and its usage of framing had any impact on military spouse
employment policy was to measure for framing effects and to examine media coverage of the
issue. This project, while revealing small differences in framing effects between the two
46


media groups, still strongly presented that the growth in thematic framing and the specific
frames utilized tracked in correlation to policy development on military spouse employment.
Ultimately, framing theorys advocates have long been studying to find a way to
bring more validity and reliability to its methodology. Over a similar time period, military
spouses have been studied and recognized for not only their contributions to the military
community, but also for their need and right to gainful employment. For researchers wishing
to better understand the forces behind recent policy development and national initiatives for
military spouse employment, media framing content analysis provided clear methods for
better identifying what role media coverage played alongside policy development for military
spouse employment. For researchers looking to expand framing theorys validity and
reliability, Boydstun et. al. recent generalizable framing methods need to be replicated to
further support their systematic worth. Embarking on this research project sought to not only
add valuable efforts in supporting and testing their methods, but it also helped to further
establish a clearer understanding of how military spouse employment policy is developing
and how media coverage has contributed to that process.
Framing theory largely thrives today as a multi-disciplinary field of study. This is
arguably one of its greatest strengths, as each peer-reviewed and systematic addition
purposes to increase theory validity and reliability. As one scholar praises it,
Framings value, however, does not hinge on its potential as a unified research
domain but... as a provocative model that bridges parts of the field that need to be in
touch with each other: quantitative and qualitative, empirical and interpretive,
psychological and sociological, and academic and professional. If the most interesting
happens at the edges of disciplinesand in the center of policy debatesthen
framing certainly has the potential to bring disciplinary perspectives together in
interesting ways (Reese, 2007, p. 148).
In noting this theorys distinct ability to cross research disciplines, this research projects
topic and focus were also partly inspired by the number of studies found utilizing framing
47


analysis to measure framing effects on gender, a specific population, and a social
movementnamely the womens movement. Some of these studies such as Terkildsen &
Schnell, 1997; Akhavan-Majid & Ramprasad, 1998; Lind & Salo, 2002; and Barnett, 2005
all reveal in diverse ways the impact that media had in framing issues directly feeding public
perception and policy development within the various waves of the womens movement.
These studies in particular and the multi-disciplinary nature of framing analysis itself,
debatably present that utilizing framing analysis to analyze public perception and policy
development for the specific population of military spouses and military spouse employment
policy would enable similar success. Thus as an addition to framing theory scholarship, this
study hoped to contribute not only valuable efforts in testing Boydstun et. al. methods and
generalizable frames, but to also build upon framing theorists efforts to study framing effects
around a specific issue that largely impacts a specific population.
This study itself also proved to be multi-disciplinary in nature, not only in how it
drew upon a multitude of framing theory sources, but also because of how it additionally
purposed to add to the growing field of academic research studying military spouse
employment. For the issue of military spouse employment alone, this project was able to not
only track the growth of policy and program support for military spouse employment over
the past twenty years, but it also revealed who key players were, the timing and impact of
various events, and a clearer understanding of what employment conditions for military
spouses today still prove to be challenging. If nothing else, this study hopes to pave the way
for continued, future research on military spouse employment. Questions like, what role does
gender now play in employment statistics and programmatic efforts given the growing
number of male military spouses or how is generational diversity impacting how military
48


spouses learn of resources and support efforts will be key for future research to address.
Lastly, by drawing these two fields of study together, this project hopes to provide a unique
and valuable bridge between the two that will enable future research new ideas and verifiable
methods, whether that be in studying further media framing impacts or the development of
military spouse employment.
49


REFERENCES
Akhavan-Majid, R., & Ramprasad, J. (1998). Framing and Ideology: A Comparative Anlysis
of US and Chinese Newspaper Coverage of the Fourth United Nations Conference on
Women and the NGO Forum. Mass Communication and Society, 1 (3/4), 131-152.
Babbie, E. (2013). The Practice of Social Research (13th Edition ed.). Belmont, CA:
Wadsworth Cengage Learning.
Barnett, B. (2005). Feminists Shaping News: A Framing Analysis of News Releases from the
National Organization for Women. Journal of Public Relations Research, 17 (4),
341-362.
Black, N. (1987, September 15). AIRFORCE STUDYING WHETHER WIVES BEING
PRESSURED ON CAREERS. Associated Press.
Blue Star Families. (2015). Blue Star Families 2015 Annual Military Family Lifestyle Survey.
Blue Star Families, Syracuse University, & The Institute of Veterans and Military
Families.
Boydstun, A. E., & Glazier, R. A. (2013). A Two-Tiered method for Identifying Trends in
Media Framing of Policy Issues: The Case of the War on Terror. Policy Studies
Journal, 41 (4), 707-736.
Boydstun, A. E., & Gross, J. H. (2014). Policy Frames Codebook. *Not Published.
Boydstun, A. E., Gross, J. H., Resnik, P., & Smith, N. A. (2013, September 16). Identifying
Media Frames and Frame Dynamics Within and Across Policy Issues. 1-13.
Bushatz, A. (2014, February 12). Survey Finds Most Military Spouses Underemployed.
Retrieved December 3, 2015, from Military.Com News:
http ://www. military. com/daily-news/2014/02/12/survey-finds-most-military-spouses-
underemployed.html?comp=7000023317843&rank=2#disqus_thread
California State University, Northridge. (n.d.). When to Use a Particular Statistical Test
PDF. Retrieved April 25, 2015, from California State University, Northridge:
http://www.csun.edU/~amarenco/Fcs%20682/When%20to%20use%20what%20test.p
df
Cappella, J. N., & Hall Jamieson, K. (1996). News Frames, Political Cynicism, and Media
Cynicism. The Annals of the American Academy of Political and Social Science, 546,
71-84.
50


Castaneda, L. W., & Harrell, H. M. (2008). Military Spouse Employment: A Grounded
Theory Approach to Experiences and Perceptions. Armed Forces & Society, 34 (3),
389-412.
De Vreese, C. H. (2004). The Effects of Frames in Political Television News on Issue
Interpretation and Frame Salience. Journalism &Mass Communication Quarterly, 81
(1), 36-52.
Department of Defense. (1988, February 10). Department of Defense Directive: Number
1400.33. USA.
Dimitrova, D. V., Kaid, L. L., Williams, A. P., & and Trammell, K. D. (2005). War on the
Web: The Immediate News Framing of Gulf War II. The Harvard International
Journal of Press/Politics, 10 (1), 22-44.
Druckman, J. N. (2004). Political Preference Formation: Competition, Deliberation, and the
(Irrelevance of Framing Effects. The American Political Science Review 98 (4),
671-686.
Druckman, J. N. (2001). The Implications of Framing Effects for Citizen Competence.
Political Behavior, 23 (3), 225-256.
Entman, R. M. (1993). Framing: Toward Clarification of a Fractured Paradigm. Journal of
Communication, 43 (4), 51-58.
Fisher, J. (1987, November 23). Officers' Wives Feel Heat to Stay in the Kitchen. Retrieved
March 9, 2015, from America's
News:http://infoweb.newsbank.com/resources/doc/nb/news
/0EB4EBFF92920DlE?p=NewsBank
Gamson, W. A., & Modigliani, A. (1989). Media Discourse and Public Opinion on Nuclear
Power: A Constructionist Approach. American Journal of Sociology, 95 (1), 1-37.
Hall, M. (2011, April 13). Michelle Obama, Jill Biden are 'Joining Force s'-Project Aims to
Lend Support to Military Families. USA Today.
Harrell, M. C. (2001). Officers' Spouses: Have the White Gloves Been Mothballed? Armed
Forces & Society, 28 (1), 55-75.
Harrell, M. C., Lim, N., Castaneda, L. W., & Golinelli, D. (2004). Working Around the
Military: Challenges to Military Spouse Employment and Education. RAND National
Defense Research Institute. Santa Monica: RAND.
Iyengar, S. (1990). Framing Responsibility for Political Issues: The Case of Poverty.
Political Behavior, 72 (1), 19-40.
51


lowers, K. (1995, July 24). A Rocky Road. Air Force Times, 55 (51), pp. 12-13.
lowers, K. (2014, March 31). Crank It Up: Service Providers Want to Get You a Message.
The Navy Times.
lowers, K. (2011, July 21). Mission: Family: Portable-jobs Program May Give Spouses a
Long Awaited Leg Up. Army Times.
Kahneman, D., & Tversky, A. (1984). Choices, Values, and Frames. American Psychologist,
39 (4), 341-350.
Kahneman, D., & Tversky, A. (1979). Prospect Theory: An Analysis of Decision under Risk.
Econometrica, 47 (2), 263-292.
Levin, I. P., Schneider, S. L., & and Gaeth, G. J. (1998). All Frames Are Not Created Equal:
A Typology and Critical Analysis of Framing Effects. Organizational Behavior and
Human Decision Processes, 76(2), 149-188.
Lind, R. A., & Salo, C. (2002, March). The Framing of Feminists and Feminism in News and
Public Affairs Programs in U.S. Electronic Media. Journal of Communication ,211-
228.
Macnamara, J. (2005). Media Content Analysis: Its uses; benefits and best practice
methodology. Asia Pacific Public Relations Journal, 6( 1), 1-34.
Matthes, J., & Kohring, M. (2008). The Content Analysis of Media Frames: Toward
Improving Reliability and Validity. Journal of Communication 58, 258-279.
Military Spouse Employment Partnership. (2014). MSEP: About Us. Retrieved December 3,
2015, from MSEP Jobs: https://msepjobs.militaryonesource.mil/msep/content/about-
us
Moon, S. S. (2001, February 24). She's Been There. The Fayetteville Observer.
Nelson, T. E., Oxley, Z. M., & and Clawson, R. A. (1997). Toward a Pyschology of Framing
Effects. Political Behavior, 19 (3), 221-246.
Piatt, G. (2004, February 24). State Pushes Effort to Put Military Families at Ease, Duval to
join program to help Spouses find Jobs and their Kids move easily into classroom.
The Florida-Times Union.
RAND Coorporation. (n.d.). Married to the Military: The Employment and Earnings of
Military Wives Compared with Those of Civilian Wives Retrieved April 21, 2014,
from RAND Web site: http://www.rand.org/natsec_area/products/married.html
52


Reese, S. D. (2007). The Framing Project: A Bridging Model for Media Research Revisited.
Journal of Communication 57, 148-154.
Sanchez, E. (2011, June 29). DoD News: DOD Launches Military Spouse Employment
Partnership. Retrieved March 9, 2015, from U.S. Department of Defense:
http://www.defense.gov/news/newsarticle.aspx?id=64509
Sanner, A. (2011, April 14). First Lady: Military members, Spouses Need Jobs. The
Associated Press.
Scheufele, B. (2004). Framing-effects Approach: A Theoretical and Methodological Critique.
Communications, 29, 401-428.
Schwartz, J. B., Wood, L. L., & Griffith, J. D. (1991). The Impact of Military Life on Spouse
Labor Force Outcomes. Armed Forces & Society, 17 (3), 385-407.
Segal, M. W. (1986). The Military and the Family as Greedy Institutions. Armed Forces &
Society, 13 (1), 9-38.
Semetko, H. A., & Valkenburg, P. M. (2000). Framing European Politics: A Content
Analysis of Press and Television News. Journal of Communication, 50 (2), 93-109.
Superville, D. (2013, March 13). Michelle Obama urges CEOs to hire more veterans. The
Associated Press.
Terkildsen, N., & Schnell, F. (1997). How Media Frames Move Public Opinion: An Analysis
of the Women's Movement. Political Research Quarterly, 50 (4), 879-900.
The Deseret News. (1996, May 6). Online Document: Many Families in the Military Today:
How They Fare. The Deseret News.
The U.S. Office of Personnel Management. (2012, August 12). Rules and Regulations:
Noncompetitive Appointment of Certain Military Spouses The Federal Register, 74
(154).
The US Department of Defense. (2012, March 19). NUMBER 1400.25, Volume 315 USA.
The White House. (2013). Joining Forces: States Step Up to Help Military Spouses Get Back
to Work. Retrieved April 21, 2014, from White House Web Site:
http://www.whitehouse.gov/joiningforces/military-spouses-back-to-work
Tuzzolino, T. (2008, October 27). Job Plan Would Help Military Spouses Face Unique
Challenges. The Navy Times.
Valkenburg, P. M., Semetko, H. A., & and De Vreese, C. H. (1999). The Effects of News
Frames on Readers Thoughts and Recall. Communication Research, 26 (5), 550-569.
53


APPENDIX A
NEW HORIZONS FOR MEDIA FRAMING ANALYSIS AND MILITARY SPOUSE
EMPLOYMENT: CODEBOOK
Last Updated: 22 December 2015
Kallie J. Culver
Graduate Student at University of ColoradoDenver
kallie.culver@ucdenver.edu
This codebook lays out the coding categories and parameters utilized to review all
data collected for this project. This codebook and its list of coding categories will provide
clear operational variables to record and measure (Babbie, 2013, pp. 335-339).
CODING CATEGORIES:
Coding categories include the following: the article ID, the Newspaper ID, the group
under which the newspaper falls (military or nonmilitary), the date of the article, the articles
relevance, the nature of the frame (episodic or thematic), the primary frame evaluation
(economic, quality of life, cultural identity, and policy prescription and evaluation) which is
recorded based on the presence or absence of various sub-frames identified within each
primary frame, and whether the article was printed in multiple newspapers.5
CODING PROCESS:
Once data has been gathered and is filed in an accessible manner, coding should be
conducted according to the following processes.
STEP 1:
A data table in either Excel or SPSS must first be created and contain columns for
tallying all coding categories (Babbie, 2013, p. 338). This tally sheet will record all the code
categories, and thus will provide numerical data for later analysis in SPSS.
STEP 2:
Coding should begin with recording the first four coding categories: newspaper ID,
5 The primary frames and tones were first selected from Boydstun et. al.s list of frames and tones. Their codebook can be
referenced for detailed definitions and examples of their frames and tones. The sub frames and tone categories were further adjusted to
specifically apply to military spouse employment. Iyengars work can be referenced for further background definition in determining the
nature of the frame.
54


newspaper group, date of the article, and the relevancy of the article. To determine the
relevancy of the article, the coder should skim the headline and the article to determine if the
article is addressing the topic of military spouse employment. If yes, code as relevant, if not
code as irrelevant and set aside. Examples of an irrelevant article would be an article that
references military, spouse, and employment as separate and entirely unrelated nouns or an
article that discusses military employment or veterans employment issues without any
reference to military spouses. In addition to determining whether there is relevant
coverage of military spouse employment, it is also important to remember that some
articles content will be entirely relevant and other articles will only address the issue in
a portion of the content. With this it is important for the coder to apply all further
coding processes to the relevant portions of the article. Each coding process should be
applied in direct application and connection to content covering the issue of military
spouse employment.
STEP 3:
Once the initial coding is complete, the coder then moves on to code the actual
content to determine the primary frame, tone, and nature of the frame. Here the order changes
though, as the coder must code for the various sub-frame categories first. Code these sub-
frames according to binary code, noting their presence in the article as 1 and their absence in
the article as 0. The sub-frames are listed below with examples of each:
1. Economic Sub Frames
Portability: any reference to military spouse career portability, whether
that be difficulties with maintaining career licenses across state lines or
specific jobs or career fields or companies that allow for portability.
Unemployment: any reference to unemployment rates, difficulties, or
benefits.
Employer Interest/Hiring: any reference to employers interested in
military spouses such as employers involved in the military spouse
employment partnership (MSEP) or the federal government or comments
on hiring practices concerning military spouses.
Income/Financial Contributor: any reference to military spouses income or
55


their ability or inability to be a financial contributor.
2. Quality of Life Sub Frames
Retention: any reference to military retention (keeping personnel in the
military) issues in relation to military spouse employment.
Separation: any reference to military members leaving the military due to
military spouse employment issues.
Support Programs: any reference to programs created to support military
spouses, whether that is a government or military or state or local
community program. Common examples mentioned are the Military
Spouse Employment Program, Military Spouse Employment Partnership,
programs run by base family centers, programs run by organizations such
as Military Officers Association of America and Blue Star Families, and
many others. Note: these programs are often mentioned in conjunction
with legislative and policy efforts. Code for support programs, when it is
discussing program implementation of those efforts.
Frequent Moves: any reference to the military lifestyle of frequent moves
and its impact on military spouse employment.
3. Cultural Identity Sub Frames:
Volunteerism: any reference to military spouse volunteer work or their
choice to volunteer in the absence of gainful employment.
Entrepreneurship: any reference to military spouses starting their own
business or being an entrepreneur.
Career Flexibility: any reference to military spouses being willing to take
any job or take pay cuts or be flexible in terms of their career, as well as
any comments on having diverse resumes or work experience due to their
military lifestyle. Remember that merely being able to move around with a
job should be coded under portability.
4. Policy Prescription and Evaluation
Policies or Fegislation: Any reference to policies or legislation impacting
military spouse employment, such as the Joining Forces Initiative, Federal
Military Spouse Hiring Preference Rule (noncompetitive hiring authority)
56


as authorized by the Federal Office of Personnel Management, Military
Spouse Employment Directive, state legislation passed on licensure
portability, etc. Another common legislative effort mentioned in
conjunction with Military Spouse Employment are Base Realignment and
Closure (BRAC) hearings and decisions.
Now, for determining the primary frame assess which sub-frames were coded as present, and
select the correct option available: whether that be one frame, a mixture of frames, all of the
frames, none of the frames, or finally select other if none of the options apply. Here again
you are coding the primary frames based on merely their presence or absence, so even if
there are more sub-frames utilized in one primary frame over another, you would still select
an option that would note both primary frames as present.
Next you want to determine the nature or types of the frames being utilized. Are they
episodic or thematic or a mixture of both? Here you will code for the presence of episodic. If
the article is episodic in nature, then code a 1, if it is thematic code a 0, and if it is both
episodic and thematic then code the articles as .5. An episodic frame would be an article
framed according to the perspective of a specific person or specific set of persons. A thematic
framing of the article would instead be generalized, speaking about groups or people in
general. Some articles will use both types of frames throughout. Examples of an episodic
frame would be a human-interest story on a specific military spouse, whereas a thematic
frame would discuss military spouses at large.
Last but not least, after reviewing the content of the article compare it with the data
set and note whether the article was printed more than once in a different newspaper. Often
times this will be discovered earlier in the coding process as you are reading the article and
recognize familiar content. Be sure to compare dates, news sources, and all content. If the
article is printed in a different news service, then code 1 for repeat. If the article is similar but
contains different portions of text and is printed on a different date/and or news service code
0 for repeat and treat as two separate articles. If the article is the same in terms of content and
news source, but printed on separate but close dates in time, then code the second as a double
in the newspaper ID column and empty all other codes for it. Also, only code for repeated
articles that reach an entirely different news audience than the first. In most cases, this will
only occur across military news sources as they attempt to reach every branch of the military.
57


RECORDING CODING CATEGORIES:
Coding categories should be recorded as follows:
A. Newspaper ID
1. Associated Press 9. Navy Times
2. Washington Post 10. Army Times
3. W ashington T imes 11. DOD Documents
4. New York T imes 12. Federal Times
5. Wall Street Journal 13. USA Today
6. El Paso T imes 14. The Fayetteville Observer
7. The Virginian Pilot 15. The Florida Times-Union
8. Air Force Times 16. The Deseret News
B. Newspaper Group ID 0. Non-Military 1. Military
C. Date of Article
D. Relevance 0. Not Relevant 1. Relevant
E. Primary Frame Utilized
1. Economic Frame 9. Q&P
2. Quality of Life Frame 10. C & P
3. Cultural Identity Frame 11. E, Q, &C
4. Prescription and Policy 12. E, Q, &P
Evaluation Frame 13. E, C, &P
5. E&C 14. Q, C, & P
6. E&Q 15. All
7. E&P 16. None
8. Q&C 17. Other
F. Nature of Frame Utilized 0. Thematic 1. Episodic .5 Both
G. Repetition. 0. No Repeat 1. Repeat Article
58


Full Text

PAGE 1

NEW HORIZONS FOR MED IA FRAMING ANALYSIS AND MILITARY SPOUSE EMPLOYMENT by KALLIE JAE CULVER B.A., University of Arkansas at Little Rock, 2011 A thesis submitted to the Faculty of the Graduate School of the University of Colorado in partial fulfillment of the requirements for the degree of Master s of Public Administration Public Administration 2016

PAGE 2

ii 2016 KALLIE JAE CULVER ALL RIGHTS RESERVED

PAGE 3

iii This thesis for the Master of Public Administration degree by Kallie Jae Culver has been approved for the Public Administration Program B y Christopher Weible Chair Malcolm Goggins Rebecca Glazier Date: 12 April 2016

PAGE 4

iv Culver, Kallie Jae (MPA, Public Administration ) New Horizons for Media Framing Analysis and Military Spouse Employment Thesis directed by Associate Professor Christopher Wieble ABSTRACT Media framing analysis has developed in theory and application as a subset of political communication theory for the past several decades. Recent efforts among its scholars have sought to identify and create generalizable frames that can be applied across numerous social and political issues for the purposes of further examining the impact media coverage has on public opinion and policy development This study utilizes previously established frames to examine media coverage of military spouse employment ove r the past twenty years, in order to better identify what role media coverage has played in the development of employment policy and support programs for military spouses The form and content of this abstract are approved. I recommend its publication. Ap proved: Christopher Wieble

PAGE 5

v TABLE OF CONTENTS CHAPTER I. MILITARY SPOUSE EMPLOYMENT AND MEDIA FRAMING ANALYSIS ............... 1 Introduction ................................ ................................ ................................ ............................... 1 Background and Significance ................................ ................................ ............................... 4 Research Questions ................................ ................................ ................................ ............... 7 II. FRAMING THEORY AND MILITARY SPOUSE EMPLOYMENT POLICY DEVELOPMENT ................................ ................................ ................................ ..................... 8 Review of Literature ................................ ................................ ................................ ................. 8 Media Framing Theory Development ................................ ................................ ................... 8 Research and Policy Development on Military Spouse Employment ................................ 13 III. RESEARCH DESIGN AND METHODS ................................ ................................ ........ 18 Materials and Methods Utilized ................................ ................................ .............................. 18 Research Design ................................ ................................ ................................ .................. 18 Research Population and Sample ................................ ................................ ........................ 18 Materials and Preparation ................................ ................................ ................................ ... 20 Coding Reliabil ity ................................ ................................ ................................ ............... 23 Research Hypotheses ................................ ................................ ................................ .......... 24 Measurements and Calculations ................................ ................................ ......................... 25 IV. MEDIA COVERAGE RESULTS ................................ ................................ .................... 27 Findings and Implications ................................ ................................ ................................ ....... 27 Differences in Media Coverage Across Time ................................ ................................ ..... 27

PAGE 6

vi Nature of Frame Differences Across Media Coverage ................................ ................... 31 Differences in Media Coverage Between Groups ................................ .............................. 34 Nature of Frame Differences in Media Coverage Between Groups ............................... 35 Frame Usage Differences in Media Coverage Between Groups ................................ .... 39 Conclusion ................................ ................................ ................................ .............................. 45 REFERENCES ................................ ................................ ................................ ....................... 50 APPENDIX A. NEW HORIZONS FOR MEDIA FRAMING ANALYSIS AND MILITARY SPOUSE EMPLOYMENT: CODEBOOK ................................ ................................ ............................ 54 C ODING CATEGORIES: ................................ ................................ ................................ .. 54 CODING PROCESS: ................................ ................................ ................................ .......... 54 RECORDING CODING CATEGORIES: ................................ ................................ .......... 58

PAGE 7

i LIST OF TABLES TABLE 1 Increase of Media Coverage 1996 2015 ................................ ................................ ............ 28 2 Yearly Increase of Media Coverage 1996 2015 ................................ ................................ 29 3 Nature of F rame Usage 1996 2015 ................................ ................................ ................... 32 4 Nature of Frame Usage Counts ................................ ................................ ......................... 37 5 Frame Comparisons Across Newspaper Groups ................................ .............................. 41 6 Frame Usage Rate Between Media Groups ................................ ................................ ...... 42

PAGE 8

1 CHAPTER I MILITARY SPOUSE EMPLOYMENT AND MEDIA FRAMING ANALYSIS Air Force Times, July 24, 1995 "We're hearing that we're officers' spouses and we do not need to work because our husbands make enough money," said Audrey Moore, wife of a Marine stationed at Twen ty nine Palms, California. "Therefore, we should be staying at home and having children. They're not coming out directly, but senior officers are indirectly putting pressure on junior officers about [their] wives." (Jowers, A Rocky Road, 1995) Associated Press June 29, 2011 "In an effort to address military spouses' employment challenges, the Defense Department today launched a program to expand career opportunities for military spouses worldwide, and to recognize the skills and talents they bring to the employment table." (Sanchez, 2011) Introduction In today's struggling economy, there is a particular subset of individuals struggling against great odds to maintain gainful employment these individuals are the spouses of US Military Members. In contrast to the general public, military spouses today stru ggle with a notoriously difficult unemployment rate reported to be 26% or higher and a wage gap of 25% (Military Spouse Employment Partnership, 2014; Bushatz, 2014) In today's bit driven 24 hour news cycle, why would there ever be media coverage of such a select group of the American population dealing with such unique employment challenges? Wouldn't that coverage largely only exist within military media publications? Not when national figures take on the cause of addressing those very unique challenges na mely, the nation's First Ladies, Michelle Obama and Jill Biden. So the question then arises, why military spouse employment? What is it about these challenges military spouses face that interested such figures in taking up their cause? Does the employment of a military spouse play a vital role in maintaining financial security and family wellness among military families? What is public perception of military spouse

PAGE 9

2 employment, both within the military and US Government communities, as well within the public at large? If it is an issue of national policy to bring awarene ss and program support to, has it always been this way? Was it ever different? The first news excerpt displayed above gives a glimpse into negative perceptions of military spouse employment that were long held among the military community itself. The stark difference in the news excerpts presented at the beginning of this thesis suggests that coverage is very different today from what it was twenty years ago. This thesis aims to reveal exactly how this coverage has changed in both amount and quality over ti me. These changes matter because media coverage influences public opinion and public policy. Scholars across the fields of communication, psychology, and political science began several decades ago to build a cross field body of research on the impact med ia has in setting, influencing, and impacting public policy and public perception of many social issues. One important aspect of this research is media framing analysis, which researches specifically how the way the media frames an issue might impact publi c policy and public perception of that issue. In other words, does the weight given to an issue, or emphasis placed on it, or tone used, or word choices the media utilizes to portray any given issue have any impact on public perception and future policy de velopment surrounding that issue? A multitude of studies developed around measuring this phenomenon argues that it does (Gamson & Modigliani, 1989; Iyengar, 1990; Druckman, 2001; and Boydstun, Gross, Resnik, & Smith, 2013). Given its cross field nature how ever, many of these studies on media framing analysis are uniquely diverse in both their definition and their methods. Thus, today's experts in these fields are pressed with a need to create a standardized definition for media frames, as

PAGE 10

3 well as consistent and applicable methods for any researcher within any one of these fields to conduct media framing analysis on any given issue. Recent research by scholars Boydstun et. al. however appears to be the beginnings of addressing this very need. By creating both a systematic approach to identifying and coding media frames, as well as creating two models for systematic computer analysis of the data, future research now has a potential guide map to follow (Boydstun, Gross, Resnik, & Smith, 2 013) 1 With this addition of bringing centralized definition and method instruction to the body of existing academic research on framing analysis, an opportunity now exists for those interested to test the applicability of their frames and analysis me thods on issues where there has been a noticeable change in policy development and public perception. As revealed in the beginning, one such issue that merits further research is examining how media coverage of military spouse employment has changed over the last twenty years. Today the news cycles when it comes to the US Military, more often than not targets coverage of troop movement, troop reductions, drone strikes, pay cuts, defense spending and budget battles, retirement plan changes, or scandals with in the V eteran s A dministration. With the backdrop of an ongoing, politically controversial and laborious war in the Middle East that has kept the country engaged in conflict far longer than any other American war most in the general public stand conveni ently distanced from the daily woes of the military life and frustrated at the seemingly never ending conflict America has found itself in. Despite these quandaries and frustrations, a unique window of opportunity has risen under the current presidential a dministration, led by first ladies Michelle Obama and Jill Biden. This window of opportunity has been a joint effort between public and private sectors, led by the White House to highlight the importance of employment and transition assistance for veterans 1 Retrieved at: http://www.cs.cmu.edu/~nasmith/temp/frames 2013.pdf

PAGE 11

4 transitioning military members, and military spouses (Hall, 2011; Sanner, 2011; Superville, 2013; and Military Spouse Employment Partnership, 2014). This national level of effort has not gone unnoticed among military communities or traditional military m edia outlets. Given the support growing from both the private sector and state governments for these efforts, the employment plights of military members as well as their spouses have now also gained the media's attention across national and regional media platforms. Background and Significance From its earliest beginnings, the U. S. Military community has relied heavily on the volunteer support of its active duty member's counterparts in a predominantly male military force usually this meant their wives to provide large amounts of unpaid social and community support. In many ways it still does today, although the image of the military spouse now includes both genders given increasing diversity among both those who serve and those who join the supporting ran ks by marriage. The traditional manner of volunteer family led support for the military community and its social fabric has long been referred to by many as the two person nature of a military career meaning that for any service member who stays in the s ervice for any length of time and takes on any leadership or staff role (officer or enlisted) will also bring with those positions an amount of support expected from their spouse as well. This need for spouse support and leadership has historically played a direct role in influencing US military family policy, US Department of Defense (DoD) family policy, and public perception on the subject of military spouse employment. News archives from the mid 1980s tell of an investigation the Air Force began, after some key command spouses went public claiming they were pressured to quit gainful employment in order to provide full time volunteer, unpaid community support for their active duty spouse's command position. The consequence of not doing so? the very real

PAGE 12

5 threat that it would affect command decisions about their husband's position, next promotion, or next assignment (Black, 1987 and Fisher, 1987). Even with a military wide policy directive forbidding this type of command influence established shortly after this systemic command influence continued to impact public perception of the role a military spouse should occupy for years to come. However it was not the only obstacle military spouses faced in terms of employment. To this day, female and male spouses attached to a military member face significant difficulties in obtaining and maintaining gainful employment because of the numerous moves, deployments, and family commitments associated with the military lifestyle (Segal, 1986, pp. 22 30 and Harrell, 2001) M uch has changed over the past three decades in military policy, public perception, and public policy support for a military spouse's ability and personal need to balance multiple roles and their right to be gainfully employed. The most visible developme nts as of late, again, are the legislative and policy development efforts again led by first ladies, Michelle Obama and Jill Biden, for the White House Joining Forces Initiative. With this national leadership and support now advocacy organizations, governm ental organizations, state governments, and the private sector have been brought together to address the problem of military spouse employment (The White House, 2013) Far from an in house investigation to determine attitudes of command influence on the issue, this level of policy development and public support indicates change many spouses in the 1980s would have probably deemed inconceivable. In selecting a public po licy issue to utilize as a subject of research for media framing analysis, military spouse employment policy was chosen for its recent development on the national policy level, as well as its personal significance to my own experiences as a military

PAGE 13

6 spouse I too am a military spouse determined to excel in my education and to find a way to develop a meaningful career, while also actively supporting my husband's military career, even with knowing the very real challenges and limitations military spouses fa ce in maintaining gainful employment. Through previous efforts to research military spouse policy development, one of the key changes I observed were increased research efforts supporting policy change, as well as increased media coverage of the employment challenges military spouses face and efforts being made to address those challenges. Thus it became apparent that with enough individual champions, systemic research, and increasing media coverage over the past thirty years policy changes have occurred ( Semetko & Valkenburg, 2000; Matthes & Kohring, 2008; and Dimitrova, Kaid, Williams, & and Trammell, 2005). This realization alone showed me that media framing analysis could now provide the necessary grounded theory and methods for measuring how media cove rage of military spouse employment has changed over time and how it has framed the issue of military spouse employment to both military audiences and the larger public. Ultimately, both of these topics have a rich history of academic research, with expert scholars supplying significant work to both framing theory and military spouse policy. Bringing these bodies of research together will benefit each through providing insightful connections and timely application. Consequently, this research project propos es to utilize media framing analysis to identify, track, and measure media portrayal of the military spouse and military spouse employment over the last thirty years. Thus my research questions were designed to aid in discovering what changes have occurred and what differences exist amongst media coverage of military spouse employment.

PAGE 14

7 Research Questions Research Question 1: How has coverage of military spouse employment across military and nonmilitary sources changed over time, both in amount and content? Research Question 2: How does coverage of military spouse employment differ across military and nonmilitary sources? These questions will be used to guide the data analysis process, so that testing the hypotheses laid out in the methods section will be c onducted in a way that adds meaningful and valid research findings to the body of research already conducted on both media framing analysis and military spouse employment. In order to properly situate how this research project would add to existing researc h in these fields, the next section will briefly review seminal academic literature and theory first, for media framing analysis and secondly, for military spouse employment.

PAGE 15

8 CHAPTER II FRAMING THEORY AND MILITARY SPOUSE EMPLOYMENT POLICY DEVELOPMENT R eview of Literature This project examine s media coverage of military spouse employment over the past twenty years from 1996 to 2015. In order to do that, systematic methods must be outlined and established. For that to happen, it is important to first review how other scholars have accomplished s imilar studies of both media coverage and of military spouse employment. This will not only provide foundational theory and credence to the methods chosen for this project, but it will also help situate this project and what value it hopes to add to both m edia framing research and developing research on military spouse employment. Media Framing Theory Development Framing Theory has developed along a multitude of academic disciplines spanning across sociology, psychology, political science, and communication to name a few. Beginning in the late 1970s, foundational theorists began studying framing effects and theorized what influence they had both on the individual decision level and on a larger scale through media communication (Kahneman & Tversky, 1979; Kah neman & Tversky, 1984; Gamson & Modigliani, 1989; Iyengar, 1990; Entman, 1993 ; and Nelson, Oxley, & and Clawson, 1997) Given this theory's multi disciplinary nature, finding consensus on a definition of what a framing effect is, what frame types can be ap plied systematically across issues and fields of study, and what model of framing analysis provides the most applicable methods has been ongoing for the past thirty years. Formative work by Gamson & Modigliani in the 1980s, developed a definition of fram ing within the media as, "a central organizing idea, or frame for making sense of

PAGE 16

9 relevant events, suggesting what is at issue" (1989, p. 3) This definition was later built on and expanded by Entman, who said, "Framin g essentially involves selection and salience. To frame is to select some aspects of a perceived reality and make them more salient in a communicating text..." (Entman, 1993, p. 52) These two definitions of framing suggest that how media chooses to categorize and structure their reports and thoughts on events has a measureable impact on its audience, just as the actual message of the text content does. Nelson, et. al. 's work on "the Psychology of Framing Effects" also helped to establish definition by distinguishing between framing effects used by the media and direct messages of persuasion and belief change the former being an indirect use of structure and emphasis, and the latter be ing a form of propaganda. The turn of the century brought with it further development in media framing theory. Druckman in his study on framing effects on citizen competence reveals another key definition needed to truly understand two major trends of fra ming research the public level and the individual level. According to Druckman, some disciplines have focused on showing framing effects on an individual's thoughts and perceptions, while other disciplines have attempted to measure framing effects on pub lic and media communication (2001, p. 228) Understanding how these vary in nature, also gives light to greater clarity of defining exactly what framing is and whom it affects. So with the understanding of what fr aming effects are and how they occur, the next question to be addressed is how have framing effects been analyzed and measured? Framing theory has largely developed by focusing on measuring framing effects along two levels, as previously introduced the in dividual thought level and the public discourse level. At the individual level multiple disciplines often overlap in studies on

PAGE 17

10 decision making that attempt to determine if framing an individual choice in one way or another might affect an individual's dec ision between the two, whether in survey research, political polling, consumer choice, etc.. Legacy work here began with Kahneman & Tversky's studies on prospect theory (also known as risky choice) and the development of loss versus gain valence frames. Th ese frames specifically examine how an individual decision is affected when a choice is framed as a loss or a gain (Kahneman & Tversky, 1979 and 1984). In 1998, Levin and others contributed significant work to framing theory, by creating a typology of thre e frame types commonly studied and compiling an extensive review of existing studies on each frame type. These frame types were risky choice framing effects, attribute framing effects, and goal framing effects (Levin, Schneider, & a nd Gaeth, 1998) Even more recently, political science scholars Druckman (2004) and Boydstun & Glazier (2013) have further developed risky choice framing theory by respectively, considering the role emotions play in responses to gain/loss frames and how the media utilized these same frames to portray the war on terror. Knowing these scholars individual contributions to the continued development of framing theory helps to establish not only that framing effects are real, but also importantly reveals how they can transcend just one topic and apply to a host of different social issues. Studying how framing effects occur across social issues whether at the individual level or public di scourse level will be essential for better understanding what role they have played in media coverage of military spouse employment. Perhaps of all the scholars studying framing effects at the public discourse level, Iyengar's founding work on thematic ve rsus episodic framing is especially key, as thematic and episodic frames were found to be the most applicable to media coverage of military

PAGE 18

11 spouse employment when selecting a general frame type to measure for. Thematic frames are when the media discuss a p articular issue in general or in the abstract, whereas episodic frames refer to when the media frame an issue according to a specific example of it. Iyengar's work studied the framing effects within public media and its resulting impact on public opinion c oncerning poverty and government welfare. He found through multiple studies that framing poverty according to episodic frames namely according to one poor person's experience stimulated more of an individual connection and expectation of self reliance, whe reas the use of thematic framing to portray poverty as a collective social malady elicited more of a desire for a collective response and an expectation for the government to address the problem. Thus his findings revealed that a mere framing of the issue played a significant role in influencing public opinion on social welfare policy (Iyengar, 1990, pp. 21 22, 28 29, 35 36) Capella and Jamieson also contributed important work on media framing effects by revealing its impact on public trust, levels of public cynicism, and public trust in the media itself (Cappella & Hall Jamieson, 1996) D e Vreese was another scholar that with the help of colleagues conducted several studies on media framing effects, particularly revealing how the media utilizes frames such as conflict, human interest, responsibility, and economic consequences (Valkenburg, Semetko, & and De Vreese, 1999 and De Vreese, 2004). In the end, regardless of diversit y in approach, which level of framing is studied, or the number of frames examined multiple studies have shown framing effects to be real at both the individual decision level and throughout media discourse at large. Framing theory does not come without its critiques and apparent limitations however. Druckman in particular questions its relevance and reveals that its impact largely

PAGE 19

12 depends on the context and the individuals involved. He also goes on to enquire whether framing theory is methodologically s ound both in terms of validity and in terms of bias as he wonders whether there might be a strong focus on only positive framing research results (Druckman, 2004, pp. 684 85) Scheufele also presents a strong critique of framing analysis and its methodology after compiling a classification and extensive review of existing literature and studies across disciplines. In the end, two of his main criticisms are the need for a replic able, longitudinal field study design, and the need for further research to incorporate both the macro and micro (public discourse and decision making) levels of framing analysis, as he too often found one ignoring strengths of the other (Scheufele, 2004, p. 422) Out of all recent research, Boydstun et. al. 's recent development of a unified coding scheme and systematic framing analysis models seems to present a strong response to these critiques. Only time for further re search and application, however, will determine if they are. Along with Iyengar's seminal work on episodic versus thematic framing, Boydstun et. al's research was chosen as the key source in theory behind the specific framing methods to be utilized for thi s project (Boydstun, Gross, Resnik, & Smith, 2013). The value in utilizing their research is that their framing structure and generalizeable frames were created to have broad application across social issues, which in turn provides systematic means for exa mining media coverage of military spouse employment. Iyengar's work on the nature of frame utilized by media sources thematic versus episodic suggests to be the most helpful for tracking the policy impact of media coverage on military spouse employment acr oss time as well as for examining differences in thematic versus episodic framing usage between military media and nonmilitary media. Given that thematic framing is reported to have the greater policy impact, it is expected that military media will have t he most invested interest

PAGE 20

13 in utilizing it. This expectation also comes into play when examing media coverage for specific frame usage differences, according to Boydstun et. al.'s list of specific frames. Here again, if military media sources have an invest ed interest in finding, portraying, and encouraging policy solutions to military spouses employment challenges, it arguably follows that of the specific frame types, economic frames and policy prescription and evaluation frames will be their most popular c hoices given these frames practical emphasis In the end, however, these are mere suppositions and the only way to truly know which framing methods are utilized across time and how the two media groups differ in their usage of both levels of framing will b e by examining the data and analyzing it. Research and Policy Development on Military Spouse Employment In conducting a literature review for this research project, it is also important to briefly review important academic scholarship that has been done on military spouses and military spouse employment. It is also important to note key policy changes that have occurred throughout the last 20 years that arguably result from the increased awareness and policy development that academic research and increas ed media coverage have created. Since the mid 1980s, scholars at RAND, a research organization with large funding support from the US Government, and others interested in US military related research have been conducting studies on the effects military spo uses have on morale, enlistment, and retention rates (RAND Coorporation) This research began largely after the policy change implemented across military branches through the DOD Directive 1400.33 created in 1988. This directive specifically prohibited military commanders from considering spouses' employment status or their volunteer work in making promotions, assignment decisions, and performance reports (Department of Defense, 1988) Before this dir ective, a spouse's volunteer involvement and employment status could be commented on in official performance reports and often played a

PAGE 21

14 factor in command decisions on promotion and new base assignments (Harrell, 2001, p. 56 ) The existence of improper command perception of spouse employment and unfair command influence, as well as the shifting numbers of women into the workforce at that time nationwide, necessitated that US military family policymakers reconsider what support was offered and what expectations were placed on spouses in the futur e (Segal, 1986 and Schwartz, Wood, & Griffith, 1991). Research soon revealed that in addition to military community expectations and commitments, military spouses continually faced the f ollowing challenges in finding and keeping employment: frequent moves, licensure portability, childcare considerations, and deployments. With the turn of the century, a new interest in research and demand for policy development occurred, as the military b egan to realize the very real impact the spouse and family had on a military member's retention and reenlistment. Research by social and military policy experts largely recommended solutions that would greatly improve spouse employment chances, having been convinced that an employed spouse and thus a satisfied spouse directly impacts the military member's decision to stay in the military, and vice versa. Some recommendations were shortening moving cycles, developing programs giving incentives to local busin esses to hire military veterans and military spouses, and finally developing policies and legislation at the State and Federal level to address licensure and certification portability issues many spouses face with moving across state lines frequently (Harr ell, 2001; Harrell, Lim, Castaneda, & Golinelli, 2004; Castaneda & Harrell, 2008). Some of the most key policy developments for military spouse employment in the last decade has, first, been the development of the military spouse federal hiring preferenc e, created under President Bush in 2008 with Executive Order 13473 and then expanded again

PAGE 22

15 in definition and application under President Obama in 2009. As it stands currently, this special hiring preference administrated by the U. S. Office of Personnel Ma nagement gives military spouses the ability to develop a portable civil service career alongside their active duty military member (The US Department of Defense, 2012 and The U.S. Office of Personnel Management, 2012). In addition to this special hiring pr eference, the Military Spouse Employment Partnership forged under the First Ladies' Joining Forces Campaign as an expansion of an Army specific military spouse employment program has brought additional career progression opportunities to military spouses i n the past five years within the private sector. This program created in 2011 is now administered across all branches of the military and has created a partnership with numerous private companies promising to hire and provide portable careers for veterans and military spouses (Military Spouse Employment Partnership, 2014) Most recently in terms of academic research, The RAND Corporation has spearheaded several research projects looking into the roles of military spouses and th e challenges they currently face with employment. In 2009 RAND Project Air Force released a report detailing survey results conducted on 802 Air Force military spouses. (Miller, Meadows, Hanser, & Taylor, 2011) The Institute for Veterans and Military Famil ies in coordination with the Military Officers' Association of America also recently conducted a survey of Military Spouses in 2013, and released its report and findings in February of 2014. (Maury & Stone, 2014) Another military advocacy organization, Blu e Star Families also conducted a survey in cooperation with The Institute for Veterans and Military Families on military lifestyle issues just this year, which also added key insights for further research into the challenges and efforts being made for mili tary spouse employment. (Blue Star Families,

PAGE 23

16 2015) Ultimately these research findings on issues military spouses face with finding and maintaining employment are valuable for the White House Joining Forces Initiative, the Depar tment of Defense, for military families, and for employers and businesses across the nation looking to partner with veterans and their families. For purposes here, reviewing both the development of framing theory and research surrounding the policy develo pment of military spouse employment efforts, provides key insight into how the first can assist with better tracking and measuring the second. In order to better understand why and how media coverage of military spouse employment has changed and increased over the past twenty years, one must first have a means for measuring that coverage. Framing theory and the role it plays in media analysis will supply the necessary means. Noting increasing academic research efforts and policy developments for military sp ouse employment also plays an important role for this project, in that they are often the subject of said media coverage, and thus provide specific events to target and measure coverage before and after within the set time period of the past twenty years. Through researching both media framing and military spouse employment thus far, there are two key expectations that create the foundation for the research hypotheses discussed in the next section, which will provide specific and measureable means to answe r the general research questions discussed earlier. These expectations are first that media coverage will show changes across all sources in amount and content through both levels of framing namely, the nature of frame utilized and the specific types of fr ames discussed. The second follows more specifically, in that differences and changes between the groups of media sources are also expected. Thus, with these outlined expectations and h aving established both a relevant academic history of both media framin g and military spouse employment, the next section

PAGE 24

17 will outline the specific methods used to track, measure, and and analyze media coverage of military spouse employment.

PAGE 25

18 CHAPTER III RESEARCH DESIGN AND METHODS Materials and Methods Utilized Research Design The proposed research design draws primarily from theory and structural designs illustrated in the works of Boydstun, Gross, Resnik, & Smith, 2013 ; Dimitrova, Kaid, Williams, & and Trammell, 2005 ; Matthes & Kohring, 2008; Iyengar, 1990 ; and Semetko & Valkenburg, 2000 A deductive structure was first chosen, as it required the variables and frames selected for measurement to be clearly defined prior to any research. This structure directly targets limiting researcher bias. This was important to establ ish given the fact that while being a student researcher, I am also a military spouse inquiring into a topic that directly connects to my current life. Other framing analysis studies have taken on a more inductive nature, where researchers discover frames as they go, but with this comes the danger of selection bias making these studies harder to validate and replicate (Macnamara, 2005, p. 9) Research Population and Sample The target population being studied in this researc h project is media coverage of US military spouse employment over the past 20 years. The research has utilized content analysis to examine printed media sources discussing any issues related to military spouse employment. Thus the unit of analysis was prin ted media articles accessed electronically, gathered from the following sample sources in two distinct categories: military media sources and nonmilitary media sources. In the military category, sources utilized were specific military and government newspa pers, namely The Air Force Times, The Navy Times, The Army Times, The Federal Times, and DOD documents in the form of government press

PAGE 26

19 releases and official government blog posts. In the nonmilitary media category, sources were selected first from the foll owing national newspapers: The Associated Press, The Washington Post, The Washington Times, The New York Times, The Wall Street Journal, and USA Today After retrieving all data available from these sources, local newspapers were also accessed based on the ir close vicinity to military bases and subsequent coverage of military issues namely the Virginian Pilot located in Norfolk, Virginia; The El Paso Times located in El Paso, Texas; the Florida Times Union located in Jacksonville, Florida; The Deseret News located in Salt Lake City, Utah; and The Fayetteville Observer located in Fayetteville, North Carolina. All of these sources were discovered, selected, and retrieved from the following databases: Lexus Nexus, America's News, Pro Quest, and Academic Searc h Premier Plus. These sources were chosen based on the following criteria: 1) connection to target population, 2) national, local, and military reputable sources to provide differing perspectives, and 3) accessibility (Babbie, 2013, pp. 332 335) The natio nal papers were selected as well to account for potential bias in coverage, in that both liberal and conservative media sources by reputation were chosen. Several military branch media sources, as well as two from the Federal government were also chosen to ensure a fair sampling of perspective and coverage across the military branches and DOD. In terms of sampling, articles were gathered from a 20 year span, starting in 1996 up to present day. A total of 235 articles on military spouse employment were colle cted from military news sources, and a total of 117 nonmilitary sources were gathered. During the coding process, a final check for relevant material within each article was made and the final total of articles came to 314, which broke down into 204 milita ry articles and 110 nonmilitary articles. For sample selections, this

PAGE 27

20 research project utilized systematic random sampling to select the articles from each newspaper. Given the total number of articles from each newspaper varied selection was conducted as follows: every 10 th unit from groups of articles larger than 40, every 5 th for groups of sources ranging in number from 15 40, and one unit was randomly sampled from groups of sources that amounted to 10 or less articles. Materials and Preparation Researc h preparation for this project began with collecting the necessary articles and creating a codebook. This was completed prior to coding any data, so that once the actual sample materials were drawn from the entire collection they would each be sampled in t he same systematic manner. The codebook and list of coding categories are included as appendices and provide clear operational variables that were record ed and measure d The codebook also lists the coding processes and parameters in detail for any future research that might want to utilize them. In brief, coding categories included: the source id, the newspaper group, the article date, the relevancy of the article, the p rimary frame, the primary tone, the nature of the frame, and finally whether the article was printed in more than one newspaper. The primary text frames and tones were preselected from Boydstun et. al. list of generalizable frames and frame tones prior to research. The frames chosen from their listing initially were as follows: E conomic, Q uality of L ife, C ultural I dentity, P olitical, P ublic O pinion, and P olicy P rescription and E valuation. These frames were chosen based on their direct applicability and con nection to the issue of military spouse employment. In determining how relevant they were, Boydstun's detailed examples and defined uses for each frame were considered in connection to military spouse employment to see how applicable they were. If there wa s no direct connection to be made than that frame was not selected

PAGE 28

21 (Boydstun, Gross, Resnik, & Smith, 2013, pp. 4 5 and Boydstun & Gross, Policy Frames Codebook, 2014). The choice to look at the nature or type of frames being used, namely thematic or epis odic frames stems again from Iyengar's seminal work on media framing (1990) Stepping back to look at the overall nature of the frames used, according to his research, points to valuable data on how the public will larg ely respond to coverage of a given issue, as discussed in the literature review of framing theory. Here the use of thematic frames refers to media coverage of military spouse employment in general, which according to Iyengar will largely elicit a response of an expected collective responsibility for solutions; whereas episodic refers to coverage of military spouse employment in terms of a specific spouse or group of spouses, which according to Iyengar's work the use of elicited a response of self reliance a nd individual responsibility for solutions (Iyengar, 1990) Thus, operationally I looked to see whether an article covered military spouse employment from the overarching perspective of military spouses at large, or if it discu ssed the perspective of specifically named spouses and their issues with employment. This process of selecting frames, tone, and the nature of frames created by other researchers, again, was part of the deductive design of the project, to prevent researche r selection bias. Once I began conducting research and collecting data however, it soon became apparent that utilizing a purely deductive structure, where I would only use previously established frames, frame types, and tone values made it difficult to cl early identify what impact and value those predetermined variables had in relation to my specific data in a way that would enable other potential researchers to code and replicate my processes consistently. Boydstun et. al. frames in particular were so bro ad in nature within my coding processes that

PAGE 29

22 it was difficult to not find overlap or a presence of all frames in every article. Thus their existence revealed very little about their specific impact or role long term in media coverage trends concerning mili tary spouse employment. So in response to these difficulties, I needed to find a way to make the frames more targeted in how they were identified across each article. This in turn would also show the value in the usage of different frames and difference of frame and tone usage across multiple media sources. Thus I decided to reduce the number of Boydstun et. al. primary frames I would use to the four most applicable frames of E conomic, Q uality of L ife, C ultural I dentity, and P olicy P rescription and E valuati on. Next I chose to create and use specific sub frames for each of these primary frames that produced issue specific variables consistently found through the data. Those subframes created are as follows: E conomic broke down into licensure portability, une mployment, employer interest, and income/financial contributor; Q uality of L ife broke down into retention, support programs, separation, and frequent moves; C ultural I dentity broke down into volunteerism, entrepreneurship, and career flexibility; and finally the P olicy P rescription and E valuation frame was deemed specific enough to not need subframes. 2 I also encountered difficulty during this process, in utilizing Boy dstun et. al. tone values, due to concerns of subjectivity and an inability to replicate implicit tone values across coders. So in response to this, I first limited the tone values to be identified to explicitly pro, explicitly negative, balanced (a mixtur e of explicitly pro and anti), or no tone at all. In making these adjustments to both the frames and tone values, I conducted a practice code of a sample of my data, as well as worked with training a second coder through several sample sets so as to ensure my coding methods and selection of subframes and tone values applied consistently 2 Detailed definitions and examples of how to identify each subframe throughout the data are supplied in the codebook appendix.

PAGE 30

23 and were valid. It was during this process that both my second coder and I identified that measuring for tone in a consistent and reliable manner proved to be impossible, g iven that it was far too subjective in nature to consistently code. The reason for this we largely determined to be that the subject nature itself was neutral in tone more often than not and thus rarely presented clearly positive or clearly negative exampl es of tone. In discovering this, I decided to remove tone as a measurable variable from the project. Coding Reliability In most content analysis studies, two coders will blindly code a set of articles selected to determine coding reliability. Once coding is completed a final reliability test then determines the percentage of agreement between coders in order to ensure "intercoder reliability" (Macnamara, 2005, pp. 10 11, 20) As materials were coded, all coding results were entered into a data table using Excel in the form of a sample tally sheet. (Babbie, 2013, p. 338) This tally sheet recorded all important code categories. This tally sheet later formed the dat a set that was then used for statistical analysis to determine whether there were any significan t differences in the groups of sources, as well as differences across time. Given the total dataset included 314 articles, the sample sets the 2 nd coder worked with were selected in groups of 20 30 articles. At the end of the coding process, there were two sets of intercoder reliability tests run percentage of agreement and C ohen's kappa. In looking at the coding reliability for the nature of frame variable, the percentage of agreement between coders obtained was 100% and the resulting C ohen's kappa value w as 1.00. In looking at the coding reliability for the individual frames utilized by media groups, each test was conducted for both the collective primary frame variable and the individual subframe variables. The collective primary frame variable results an d their levels of reliability however were directly linked to how reliable the individual subframe values were, as they were a summation of

PAGE 31

24 each of those individual subframes presence or absence. Across the individual subframes the percentage of agreement figures ranged from 90 100% and the C ohen's kappa values ranged from .77 to 1.00. Ultimately though, the primary frame reliability figures sufficiently revealed that the level of coder reliability needed was reached with a percentage of agreement value of 95% and a C ohen's kappa value of .94. Research Hypotheses This research project, as stated earlier, sought to discover whether there were any differences between military and nonmilitary media coverage of the issue, or whether there were any differences in the frames or tones utilized between the groups of sources. In light of this, the following research hypotheses were created to specifically address research questions 1 and 2. Hypotheses 1(a) and 1(b) directly answer the first research question, and Hy potheses 2(a), 2(b), and 2(c) all seek to answer the second research question. Hypothesis 1(a): Media coverage of military spouse employment will increase across all sources from 1996 2015. A noted spike in coverage is expected following President Obama' s election in 2008. Hypothesis 1(b): Coverage of military spouse employment across all sources will use both thematic and episodic frames. Hypothesis 2(a): Military sources will have more coverage of military spouse employment than nonmilitary sources Hypothesis 2(b): Military sources will use more thematic frames than nonmilitary sources. Hypothesis 2(c): There will be differences in frames utilized by military media sources and nonmilitary media sources. It is expected that nonmilitary sources will utilize more E conomic and P olicy P rescription and Evaluation frames than military sources. Utilizing this research design and examining the data according to these parameters made statistical analysis possible to test these hypotheses. In doing so, it was possible to discover what role military and nonmilitary media sources played in covering th e issue of military

PAGE 32

25 spouse employment, how coverage has increased over the last twenty years, and what differences in their coverage existed in terms of frame usage. Measurements and Calculations There are a number of measurements and calculations availab le to analyze the data this research project gathered, but there are two specific measurement purposes that it sought to accomplish. First, the hypotheses and research questions formulated specifically required descriptive statistical measurements that wil l describe the occurrence of media coverage on the chosen issue over the set time period. Descriptive statistics revealed which of the preselected frames and tones each newspaper and each media group utilized the most frequently. The second measurement pur pose these parameters required was to determine what interventions or differences were at play between the two media groups and whether they were statistically significant. This allowed for statistical analysis of the difference in the types of frames and tone each media group utilizes. For descriptive purposes, some common measurements and calculations are measures of central tendency, such as the mean and the mode, and a frequency distribution. Calculating the mean and mode reveals important data, such a s the most frequently utilized frame or tone. Calculating a frequency distribution of the data gathered in graphical form helps to reveal the progress of media coverage over time across each source individually, as well as collectively across each group (California State University, Northridge) It is always important to do more than just describe what the data was revealing. This is where the second purpose of determining statistically significant differences between the groups of data gathered becomes key. There are several tests available to measure differences between two groups, but the effectiveness of each largely depends on the type of variables being examined. In this research project, the nature of frames, primary frame s, and subframes

PAGE 33

26 examined and coded for are considered nominal variables, in that they are each distinctively different and not ordinal, interval, or ratio in nature. Thus given their nominal nature, the Chi Square independent test, represents the best fo rm of statistical measurement for measuring the difference between the groups. This test largely describes and reveals the difference between observed frequencies and expected frequencies between two groups, similar to the Cohen's Kappa reliability tests r un on the sample set coded by both myself and the second coder (California State University, Northridge) In more technical terms, the hypotheses and research questions also include in their breadth the basic questions that the chi square test always seeks to answer that of the null hypothesis and the alternative hypothesis. The former implies that there is no relationship between the two variables studied, and the latter would imply that there is. Using the Chi square independe nt test for analysis requires the level of significance to be outlined, followed by calculating the degrees of freedom, expected frequencies, a test statistic, and finally a p value from the sample data. Once the data is analyzed and the tests run, the res ults will reveal what, if any, significant differences exist between the variables (Babbie, 2013, pp. 475 477) This form of statistical analysis was also predicated in a similar form of media framing content analysi s by Dimitrova et. al. in 2005. The only way to determine this is through statistical analysis. Thus the difference between military and nonmilitary media coverage in terms of their frame and nature of frame usage need to be examined for statistical signif icance. Thus, the next section will discuss what results, examining and analyzing the data revealed and what differences, if any, occurred across time and between the two media groups.

PAGE 34

27 CHAPTER IV MEDIA COVERAGE RESULTS Findings and Implications From the beginning this project has sought to discover, track, and identify changing media coverage of military spouse employment over the past twenty years. The research questions were specifically designed to approach the data for answers on not only how the coverage has changed over time in both amount and content, but also how that coverage differs between the two key groups of media sources studied military and nonmilitary printed media. These research questions led to hypotheses that outlined measurea ble expectations and outcomes for this project that could either be found and confirmed or at least provide differing measureable results that could be clearly explained. The data itself has revealed both expected and surprising results. Either way, audien ces interested in the study of media framing or military spouse employment or both can glean valuable new insights to the role the media has played in covering military spouse employment over the past two decades, as well as how military spouse employment itself has changed and increased since the early 1990s. Differences in Media Coverage Across Time The first research question sought to discover how media coverage of military spouse employment has changed over time in both amount and content. Hypothesis 1(a) predicted that the amount of media coverage would increase across all sources over the set 20 year period, but that the greatest spike in coverage would be seen after 2008 when President Obama was elected. The data revealed both these expectations to be largely true.

PAGE 35

28 Table 1 Increase of Media Coverage 1996 2015

PAGE 36

29 Table 2 Yearly Increase of Media Coverage 1996 2015

PAGE 37

30 Coverage along both groups of sources increased slowly over the first decade as shown in Table 1. The greatest increase in coverage was seen after President Obama's election in 2008, however it was most notably seen in the years of 2010 2012 as seen in Tab le 2. This spike suggests that the creation of President Obama's Joining Forces White House Initiative in 2011, spearheaded by First Ladies Michelle Obama and Jill Biden played a pivotal role in raising awareness and drawing media coverage to the issue of military spouse employment (Sanner, 2011 and Sanchez, 2011). Military Spouse employment itself was and continues to be a primary issue targeted by the Joining Forces Initiative. As mentioned earlier, an important part of the Joining Forces Initiative that made spouse employment such a targeted issue was the expansio n of the Army Spouse Employment Program into an all branch wide DoD program into MSEP, the Military Spouse Employment Partnership that brought companies across the private and nonprofit sectors on board as military spouse hiring partners (Jowers, 2011) Examples discussing Joining Forces and its focus on Military Spouse Employment abound amongst the coverage across both groups of sources. The Associated Press reported following Joining Forces' launch in April of 2011, "First Lady Michelle Obama called on companiesto recruit and hire members of the military and their spouses, saying they offer unique skills and qualities" (Sanner, 2011) USA Today reported on the launch of Joining Forces by sharin g, "Military families will be Michelle Obama's second signature issueShe and Jill Biden, the vice president's wife, say they intend to build a program that can outlast their husbands' tenureFocused on helping families cope with issues related to employme nt, education, and health, Joining Forces' will be managed by a three member board that includes retired general Stanley McChrystal" (Hall, 2011) Military sources also reported on the initiative

PAGE 38

31 saying, "Joining Forces engage s the nation on military community issues, and highlights areas where support is neededthat includes military spouses" (Jowers, 2011) Ultimately, this increased coverage within the past five years has led to raised awarene ss, especially among military spouses, of what efforts on both a national policy level and branch specific program level have occurred since the First Ladies took up the torch. Nature of Frame Differences Across Media Coverage The coverage in terms of con tent also shows some changes across time throughout both groups of media sources. In the first decade stories primarily focused on the challenges military spouses faced in finding, gaining, and maintaining employment. In the second decade media coverage fo cused more on finding policy solutions at the national and state levels for addressing those still occurring and well known challenges, as well as raising awareness to existing military and community based programs that have been created to specifically he lp military spouses with employment. These trends can be seen on a larger scale by looking at the nature of frames chosen across both groups of media, through the examining their use of thematic versus episodic framing or a mixture of the two. Hypothesis 1 (b) projected that that both groups of sources would utilize both thematic and episodic frames. By and large both groups increasingly utilized thematic frames the most, or a decided mixture of the two, which still supports the original hypothesis that both groups would utilize both types. However it was surprising to see how little episodic frames were used alone. Very few between both groups of sources would ever use purely episodic framing throughout an article. The use of thematic frames allowed both mil itary and nonmilitary media sources to continually portray the issue of military spouse employment as a common issue for a large number of military spouses. One example from 1996 in a regional, nonmilitary newspaper

PAGE 39

32 printed the following, "The military ha s undergone many changes since it became an all volunteer force,' says Carolyn Becraft, deputy assistant secretary of Defense for personnel support, family and education. It has gone from a largely unmarried force under conscription to a predominately mar ried force today.' According to Becraft, more than 60 percent of military spouses have jobs" (The Deseret News, 1996) The article then continued on in a thematic nature to portray the growth of military family services such as base childcare, as a direct response to the growing dual career nature of the married military households and the challenges that inherently prese nts. The utilization of both thematic and episodic frames, allowed the media to address common employment struggles for spouses with frequent examples of how individual military spouses uniquely responded. One article from The Florida Times Union shares o f the larger employment challenges military spouses face such as professional licensure portability, high unemployment rates, and little to no unemployment insurance after relocating due to military orders. In the midst of these larger issues though the ar ticle also highlights a particular spouse Amy Wolford and the rare success she had in finding employment on her military base thanks to the military spouse employment assistant program specialist on base (Piatt, 2004) Of the v ery few articles presented solely from an episodic frame, they commonly follow a particular spouse's career journey alongside his/her military spouse's active military career and how they managed to make a dual career work or what sacrifices the civilian s pouse was often required to make (Moon, 2001 and Tuzzolino, 2008). Ultimately though, over time the usage of thematic frames would grow increasingly greater across both groups of media.

PAGE 40

33 Table 3 Nature of Frame Usage 1996 2015

PAGE 41

34 Differences in Media Coverage Between Groups The second research question looks to see what differences exist within coverage between the two media groups military and nonmilitary. Hypothesis 2(a) first addresses this question in suggest ing that overall military media sources will have more coverage of military spouse employment than nonmilitary media sources will. In terms of total amount of coverage in the past two decades, military media sources carried nearly double the amount of article s collected. As mentioned earlier, m ilitary articles totaled to 204 versus the nonmilitary articles, which amounted to a total of 110 articles. Given coverage differs largely in amount and frequency over the years, in that military media coverage of this i ssue is nearly double that of nonmilitary media coverage of military spouse employment, one might question how article counts and coverage between the two groups are directly comparable. To address this, first it is important to note that it only makes se nse for military media coverage to be greater overall given it is a subject that directly impacts their audience. As discussed previously, studies on military spouse employment have revealed it to play a key role in determining and impacting military reten tion. This factor alone makes following its development and progression worthwhile for military media sources. As coverage has increased, military spouses themselves have long relied on military media sources directly or indirectly shared through base offi cials and services to continually update them to new programs, services, and policy efforts on their behalf. Even more recently, media sources, military service providers, and military interest groups are continually seeking how to reach younger and newer generations of military spouses with the most current information on what programs and services exist for them As one recent article in The Navy Times discussed, two of their newest efforts to "get the word out" on military spouse employment efforts is b oth through generational targeted efforts via

PAGE 42

35 social media online and through partnering with military interest groups to assist with spreading information to their members and chapters (Jowers, 2014) In the end, whether it is for military retention interests or for informing interested audience members namely military spouses themselves military media sources have played a key role in the amount of media coverage military spouse employment has gained in the past twenty ye ars. Knowing this also helps for accounting for why media coverage differs so greatly in amount across the two groups. Last but not least, the argument for still comparing coverage between these two media groups lies in the need to assess the differing aud iences reached through their coverage. To only measure media framing trends within military media sources, would provide a very skewed picture, as military media sources typically only reach military audiences and they do so at a different frequency than n onmilitary media does. National and regional nonmilitary media sources however widen the scope of what audiences are reached and at what frequency they are reached. Examining and comparing both groups for trends within their media coverage, arguably gives a better understanding and holistic picture of how framing effects are occurring. Nature of Frame Differences in Media Coverage Between Groups The next aspect of coverage examined for differences between the two media groups was measured through looking at their different framing usage at the general level by examining what nature of frames were being utilized thematic, episodic, or both. Utilizing thematic or both types of frames were increasingly common to both sets of media sources, and there were no sta tistically significant differences amongst their usage of them. However, it is still interesting to note how they did differ in their usage between the two groups over time. Table 3 shows the differences between the two groups over time graphically, but gi ven

PAGE 43

36 the differences numerically are small, Table 4 reveals the differences in actual figures. As revealed in Table 4, nonmilitary newspapers prioritized both thematic and episodic types of frames more than solely thematic from 1996 to 2010 and it was only in the last five years that their usage of solely thematic frames outgrew their usage of both. In the military media coverage however, it was only in the first five year period from 1996 to 2001 that their usage of both types outweighed their usage of sole ly thematic frames. Here again the data reveals that their dependence on thematic frames grew increasingly over the time period, with its greatest spike in the past five years from 2010 onward. The military media's use of more thematic frames lines up wit h Hypothesis 2(b), which predicted that the military media would use more thematic frames than nonmilitary media sources would. This is true even if you calculate the percentage rate each used, where military media utilized purely thematic frames 61% of th e time, versus nonmilitary media sources utilizing them 48% throughout the time period. Earlier in the research process, it was expected to find more usage of episodic frames however as shown in Tables 3 and 4 their usage was quite infrequent between both groups. In terms of percentage, nonmilitary media used solely episodic frames less than 1%, while military sources used them just less than 3% of the time. 3 The expectation to find more usage of purely episodic framing stems from Iyengar's founding work on framing effects, as discussed earlier. His fundamental work revealed that how a policy or issue is framed, whether episodically or thematically, has consi derable impact on how the public responds to it. Episodic framing elicited an expectation of individual responsibility whereas thematic framing of an issue as a collective problem elicited an expectation for collective social responsibility (Iyengar, 1990) 3 Percentages were calculated by taking the number of specifically framed articles within each group and dividing them by the total number of articles for each group. Exact figures found in Tabl e 4.

PAGE 44

37 Table 4 Nature of Frame Usage Counts

PAGE 45

38 Iyengar was not alone in suggesting the very real impact framing effects within the media and public discourse carry, but his work provides a key foundation for understanding the role thematic framing has largely played in the coverage of military spouse e mployment. Given both military and nonmilitary media sources have continually increased their use of thematic framing, the data suggests according to framing theory that addressing the challenges of military spouse employment is not a problem military spou ses can individually fix on their own. Given many of the challenges they face in employment are a direct result of their collective service and connection to the military itself, media coverage implied more and more that a collective response was/is necess ary. Perhaps this has been nowhere more apparent than in coverage of the Joining Forces Initiative, which itself corresponds with the largest increase in the reliance on thematic framing in the past five years across both media groups. Statements like the following were made by Michelle Obama, "We should all be working together on this. These are pretty solid Americans out here that are making these sacrifices quietly for all the rest of us" (Hall, 2011) "Whether you're in finance or technology or the food industry, every single one of you can ask yourselves that same question: `What more can we do? Think outside the box, take real risks and work together to make big, bold commitments to hire our veterans and military spouses..." (Superville, 2013) Again and again coverage thematically shows that the First Ladies, the President, and numerous other advocates refer to helping military families and addressing the issue of militar y spouse employment as a collective responsibility demanding a collective response at all levels, national, state, and local. Given so many examples of thematic framing increasing and knowing what response thematic framing purports to illicit in terms of p ublic response, the data suggests that media coverage itself was a key factor for

PAGE 46

39 the increase in policy efforts, program funding, and legislative efforts made in the last decade on behalf of military spouses and their employment. Frame Usage Differences in Media Coverage Between Groups The last and perhaps most extensive portion of the data examined for differences between the two media groups occurred in looking at the differences in specific frame usage. The frame choices, were again selected from Boyd stun et. al. list of generalizable frames according to which of them best applied to the issue of Military Spouse Employment. The frame choices selected were an Economic Frame, a Quality of Life Frame, a Cultural Identity Frame, and a Policy and Prescripti on Evaluation Frame. How these frames were specifically identified and applied through the use of subframes for each frame category was established during the coding process, as outlined in the previous methods section. When conducting the final data analy sis, the codes for each subframe were tallied and recorded in two additional ways across the data sheet first they were compiled to give a sum of the number of subframes used for each frame category and secondly a tally was taken to merely note the presenc e or absence of any or all subframes with the identifier of 0 or 1, to be able to accurately compare frame usage across articles and media groups so that some articles did not skew the data with a heavier use or lighter use of subframes. These final tally results were then transferred into an SPSS dataset for further statistical analysis. In response to the second research question of looking for coverage differences between the media groups, Hypothesis 2(c) sought to specifically identify how the groups w ould differ in their specific frame usage. It proposed that nonmilitary media sources would utilize more economic and policy and prescription frames that military media sources would. The results were initially surprising in that specific frame usage did n ot differ between the two groups as significantly as expected. In fact very little statistically significant differences

PAGE 47

40 in frame usage were found. The significant differences that did occur actually occurred across both media groups over time, rather than between them. This is not to say that there frame usage was identical, or that there were no differences at all. What differences did occur between the two groups, even if not statistically significant are still important to note. As Table 5 reveals, comp aring the means of frame usage across both media groups shows that their overall frame usage was largely similar. Interestingly enough and different from what Hypothesis 2(c) suggested, both media groups primarily relied on the use of economic frames and q uality of life frames, more so than the other two. However, the use of P olicy frames did become stronger over time, whereas the use of Cultural Identity frames remained the least used between both groups throughout the entire time period examined. Looking at the mean percentage rates of frame usage, as Table 6 shows, also reveals how the frame usage between the groups was largely similar over the years. Small differences did occur, the most notable of which are the use of the quality of life frames and the policy frames between the two groups. The original hypothesis again, suggested that nonmilitary media would utilize economic and policy frames more than military media would. In reality both groups prioritized economic frame use throughout the entire peri od. Interestingly enough though, both groups began with using more quality of life frames and although economic frames began to show stronger usage over time across both groups, each group still continued to use quality of life frames in fairly proportiona l amounts throughout the time period.

PAGE 48

41 Table 5 Frame Comparisons Across Newspaper Groups

PAGE 49

42 Table 6 Frame Usage Rate Between Media Groups

PAGE 50

43 The greatest difference in frame usage found between the groups and over time was in their use of policy frames. Nonmilitary papers started utilizing policy frames more in the first decade, and it wasn't till after 2006 that military papers started using i t more frequently, going from a 3% rate of use in 2001 2005 to a 26% rate of use in the 2006 2010 year range. Reasons for this rapid growth in policy frame use across both media groups in the 2006 2010 time frames arguably stem from significant policy effo rts made for military spouses during those years under both President Bush and President Obama, upon his election. As discussed in the Literature and Theory Development section, it was during this time that the Federal Military Spouse hiring preference was created, both Presidents began advocating for hiring support for military families, and by 2010 the coordination of the Joining Forces White House initiative had begun. This is further reflected in the statistical analysis conducted on the data collected. The primary statistical test for measuring the differences in the use of frames between the media groups was the chi square test, as outlined previously in the methods section. Running the chi square tests on frame usage between military and nonmilitary p apers revealed very few statistically significant differences in their frame usage overall. In fact, the only significant difference found was in the time period from 2001 to 2005, where their usage confirmed the null hypothesis of there being no significa nt difference in military versus nonmilitary usage of policy frames to be incorrect. There was a significant difference in their usage of policy frames during these years only, given there was a significant value of .027 with the degrees of freedom being 1 4 Given both military media sources and nonmilitary sources primarily used similar amounts of each frame type over the last twenty years, the question becomes : what do these 4 In order to account for the low cell count due to a small sample size, the Fisher Exact Test results within the Chi square test were the final significant values used and recorded.

PAGE 51

44 frame types emphasize in terms of content and how do they connect with the respe ctive media audiences ? Looking at the larger, overall nature of frames already gave us the impression that both military and nonmilitary media sources often frame military spouse employment as a collective issue prompting a government led, collective respo nse according to framing theory (Iyengar, 1990) Specifically though, which frames were utilized the most to portray content on military spouse employment? As the tables above have already revealed, the two most common frames utilized by both groups over time were Economic frames and Quality of Life fra mes. According to Boydsun et. al. 2013 study and codebook, Economic frames emphasize, "the costs, benefits, or monetary/financial implications of the issue (to an individual, family, community or to the economy as a whole)." Quality of Life frames however, instead focus on "the effects of a policy on individuals' wealth, mobility, access to resources, happiness, social structures, ease of day to day routines, quality of community life, etc." (Boydstun, Gross, Resnik, & Smith, 2 013, pp. 4 5) Emphasizing both of these aspects, by framing content covering military spouse employment according to each of these perspectives seems logical. For the military audience it continually frames the issue and any challenges or proposed sol utions according to both important economic considerations military families must continually assess, while also providing insight into how those same issues and solutions derive from or impact their welfare and daily life. For the nonmilitary audience, gi ven much of military spouse employment is already framed as a collective responsibility for the general public to care about addressing, framing the specifics of the issue as having both economic and quality of life considerations arguably makes the connec tion both practical and personal. Last but not least, Policy Prescription and Evaluation frames are defined as,

PAGE 52

45 "particular policies proposed for addressing an identified problem, and figuring out if certain policies will work, or if existing policies are effective" (Boydstun, Gross, Resnik, & Smith, 2013, p. 4) This growth in the use of policy frames over time between both media groups, knowing framing theory and how it suggests an impact on policy development suggests th at media coverage played a role in the development of military spouse economic policy. This would also be true from the aspect of the use of thematic framing increasing across both media groups as well. Looking at both the growth and use of specific frames as well as the nature of frames used over time both present compelling evidence that as coverage increased, an awareness of military sp ouse employment challenges grew, which thus created motivation to reexamine existing military policy and programs and t o consider new policy solutions. Conclusion Ultimately, t his project sought to identify and measure framing differences by first asking targeted research questions about media framing and coverage of military spouse employment and secondly, outlining spec ific hypotheses for answering them. The research questions again aimed to find what framing differences occurred within the media cover age across time and between the groups of media sources. The hypotheses defined how to track those differences and framin g changes through noting specific factors such as important dates and events within the set time period, changes in the amount of coverage, the varied use of thematic versus episodic framing, and the varied use of the chosen specific frames Economic, Quali ty of Life, Cultural Identity, and Policy Prescription and Evaluation. In the end, a fter conducting a significant amount of research into both media framing content analysis and military spouse employment policy development, coverage over time and coverage comparisons between military media sources and nonmilitary media sources clearly

PAGE 53

46 showed some framing changes and differences. This project's data largely revealed an increased amount of media coverage in recent years with a subsequent, noticeable increas e in program support and policy efforts for military spouse employment endeavors Thus, framing theory as it stands currently would stipulate that how the media has portrayed military spouses and how it has portrayed policy efforts made for them played a part, potentially a significant part, in the development process and public support for it over time (Iyengar, 1990, p. 20) Iyengar's seminal work on the impact in thematic framing supports this when looking at the growth of public policy efforts and publ ic support for military spouse employment efforts at large. In addition to this, other framing theorists previously examined, such as DeVreese, Capella and Jamieson, and Boydstun et. al. all contributed significant studies analyzing and measuring the impac t that the use of specific frames such as conflict and economic frames, opportunity versus risk frames, or a categorized list of commonly used specific frames have both within media coverage and public response to coverage of any given issue. Bodystun et. al. generalizable, but specific frames in particular provided the necessary structure that this studied needed to examine what type of frames the media commonly used to portray the issue of military spouse employment. This study's results showed the usage of an economic frame was by far the most common, whereas the usage of the policy frame showed the most change and increase over time across both media groups (De Vreese, 2004; Cappella & Hall Jamieson, 1996; and Boydstun, Gross, Resnik, & Smith, 2013). I n the end, t he only way to confirm whether media coverage and its usage of framing had any impact on military spouse employment policy was to measure for framing effects and to examine media coverage of the issue. This project, while revealing small differ ences in framing effects between the two

PAGE 54

47 media groups, still strongly presented that the growth in thematic framing and the specific frames utilized tracked in correlation to policy development on military spouse employment. Ultimately, framing theory's ad vocates have long been studying to find a way to bring more validity and reliability to its methodology. Over a similar time period, military spouses have been studied and recognized for not only their contributions to the military community, but also for their need and right to gainful employment. For researchers wishing to better understand the forces behind recent policy development and national initiatives for military spouse employment, media framing content analysis provided clear methods for better i dentifying what role media coverage played alongside policy development for military spouse employment. For researchers looking to expand framing theory's validity and reliability, Boydstun et. al. recent generalizable framing methods need to be replicated to further support their systematic worth. Embarking on this research project sought to not only add valuable efforts in supporting and testing their methods, but it also helped to further establish a clearer understanding of how military spouse employmen t policy is developing and how media coverage has contributed to that process. Framing theory largely thrives today as a multi disciplinary field of study This is arguably one of its greatest strengths, as each peer reviewed and systematic addition purpos es to increase theory validity and reliability. As one scholar praises it, Framing's value, however, does not hinge on its potential as a unified research domain but as a provocative model that bridges parts of the field that need to be in touch with each other: quantitative and qualitative, empirical and interpretive, psycholog ical and sociological, and academic and professional. If the most interesting happens at the edges of disciplines and in the center of policy debates then framing certainly has the potential to bring disciplinary perspectives together in interesting ways (Reese, 2007, p. 148) In noting this theory's distinct ability to cross research disciplines, this research project's topic and focus were also partly inspired by the number of studies found utilizing framing

PAGE 55

48 analysis to measure framing effects on gender, a specific population, and a social movement namely the women's movement. Some of these studies such as Terkildsen & Schnell, 1997 ; Akhavan Majid & Ramprasad, 1998; Lind & Salo, 2002; and Barnett, 2005 all reveal in diver se ways the impact that media had in framing issues directly feeding public perception and policy development within the various waves of the women's movement. These studies in particular and the multi disciplinary nature of framing analysis itself, debata bly present that utilizing framing analysis to analyze public perception and policy development for the specific population of military spouses and military spouse employment policy would enable similar success. Thus as an addition to framing theory schola rship, this study hoped to contribute not only valuable efforts in testing Boydstun et. al. methods and generalizable frames, but to also build upon framing theorists efforts to study framing effects around a specific issue that largely impacts a specific population. This study itself also proved to be multi disciplinary in nature, not only in how it drew upon a multitude of framing theory sources, but also because of how it additionally purposed to add to the growing field of academic research studying military spouse employment. For the issue of military spouse employment alone, this proj ect was able to not only track the growth of policy and program support for military spouse employmen t over the past twenty years, but it also revealed who key players were, the timing and impact of various events, and a clearer understanding of what employment conditions for military spouses today still prove to be challenging. If nothing else, this stud y hopes to pave the way for continued, future research on military spouse employment. Questions like, what role does gender now play in employment statistics and programmatic efforts given the growing number of male military spouses or how is generational diversity impacting how military

PAGE 56

49 spouses learn of resources and support efforts will be key for future research to address. Lastly, by drawing these two fields of study together, this project hopes to provide a unique and valuable bridge between the two th at will enable future research new ideas and verifiable methods, whether that be in studying further media framing impacts or the development of military spouse employment.

PAGE 57

50 REFERENCES Akhavan Majid, R., & Ramprasad, J. (1998). Framing and Ideology: A Comparative Anlysis of US and Chinese Newspaper Coverage of the Fourth United Nations Conference on Women and the NGO Forum. Mass Communication and Society, 1 (3/4), 131 152. Babbie, E. (2013). The Practice of Social Research (13th Edition ed .). Belmont, CA: Wadsworth Cengage Learning. Barnett, B. (2005). Feminists Shaping News: A Framing Analysis of News Releases from the National Organization for Women. Journal of Public Relations Research 17 (4), 341 362. Black, N. (1987, September 15). AIR FORCE STUDYING WHETHER WIVES BEING PRESSURED ON CAREERS. Associated Press Blue Star Families. (2015). Blue Star Families 2015 Annual Military Family Lifestyle Survey. Blue Star Families, Syracuse University, & The Inst itute of Veterans and Military Families. Boydstun, A. E., & Glazier, R. A. (2013). A Two Tiered method for Identifying Trends in Media Framing of Policy Issues: The Case of the War on Terror. Policy Studies Journal 41 (4), 707 736. Boydstun, A. E., & Gross, J. H. (2014). Policy Frames Codebook. *Not Published. Boydstun, A. E., Gross, J. H., Resnik, P., & Smith, N. A. (2013, September 16). Identifying Media Frames and Frame Dynamics Within and Across Policy Issues. 1 13. Bushatz, A. (2014, February 1 2). Survey Finds Most Military Spouses Underemployed Retrieved December 3, 2015, from Military.Com News: http://www.military.com/daily news/2014/02/12/survey finds most military spouses underemployed.html?comp=7000023317843&rank=2#disqus_thread Califo rnia State University, Northridge. (n.d.). When to Use a Particular Statistical Test PDF. Retrieved April 25, 2015, from California State University, Northridge: http://www.csun.edu/~amarenco/Fcs%20682/When%20to%20use%20what%20test.p df Cappella, J. N., & Hall Jamieson, K. (1996). News Frames, Political Cynicism, and Media Cynicism. The Annals of the American Academy of Political and Social Science, 546 71 84.

PAGE 58

51 Castaneda, L. W., & Harrell, H. M. (2008). Military Spouse Employment: A Grounded Theory Approac h to Experiences and Perceptions. Armed Forces & Society, 34 (3), 389 412. De Vreese, C. H. (2004). The Effects of Frames in Political Television News on Issue Interpretation and Frame Salience. Journalism & Mass Communication Quarterly, 81 (1), 36 52. D epartment of Defense. (1988, February 10). Department of Defense Directive: Number 1400.33. USA. Dimitrova, D. V., Kaid, L. L., Williams, A. P., & and Trammell, K. D. (2005). War on the Web: The Immediate News Framing of Gulf War II. The Harvard Internati onal Journal of Press/Politics 10 (1), 22 44. Druckman, J. N. (2004). Political Preference Formation: Competition, Deliberation, and the (Ir)relevance of Framing Effects. The American Political Science Review 98 (4), 671 686. Druckman, J. N. (2001). The Implications of Framing Effects for Citizen Competence. Political Behavior, 23 (3), 225 256. Entman, R. M. (1993). Framing: Toward Clarification of a Fractured Paradigm. Journal of Communication, 43 (4), 51 58. Fisher, J. (198 7, November 23). Officers' Wives Feel Heat to Stay in the Kitchen. Retrieved March 9, 2015, from America's News:http://infoweb.newsbank.com/resources/doc/nb/news /0EB4EBFF92920D1E?p=NewsBank Gamson, W. A., & Modigliani, A. (1989). Media Discourse and Public Opinion on Nuclear Power: A Constructionist Approach. American Journal of Sociology, 95 (1), 1 37. Hall, M. (2011, April 13). Michelle Obama, Jill Biden are 'Joining Forces' Project Aims to Lend Support t o Military Families. USA Today Harrell, M. C. (2001). Officers' Spouses: Have the White Gloves Been Mothballed? Armed Forces & Society, 28 (1), 55 75. Harrell, M. C., Lim, N., Castaneda, L. W., & Golinelli, D. (2004). Working Around the Military: Chal lenges to Military Spouse Employment and Education. RAND National Defense Research Institute. Santa Monica: RAND. Iyengar, S. (1990). Framing Responsibility for Political Issues: The Case of Poverty. Political Behavior, 12 (1), 19 40.

PAGE 59

52 Jowers, K. (1995, J uly 24). A Rocky Road. Air Force Times 55 (51), pp. 12 13. Jowers, K. (2014, March 31). Crank It Up: Service Providers Want to Get You a Message. The Navy Times Jowers, K. (2011, July 21). Mission: Family: Portable jobs Program May Give Spouses a Long Awaited Leg Up. Army Times Kahneman, D., & Tversky, A. (1984). Choices, Values, and Frames. American Psychologist 39 (4), 341 350. Kahneman, D., & Tversky, A. (1979). Prospect Theory: An Analysis of Decision under Risk. Econometrica 47 (2), 263 292. Levin, I. P., Schneider, S. L., & and Gaeth, G. J. (1998). All Frames Are Not Created Equal: A Typology and Critical Analysis of Framing Effects. Organizational Behavior and Human Decision Processes 76 (2), 149 188. Lind, R. A., & Salo, C. (2002, March). The Framing of Feminists and Feminism in News and Public Affairs Programs in U.S. Electronic Media. Journal of Communication 211 228. Macnamara, J. (2005). Media Content Analysis: Its uses; benefits and best practice methodology. Asia Pacific Public Relations Journal 6 (1), 1 34. Matthes, J., & Kohring, M. (2008). The Content Analysis of Media Frames: Toward Improving Reliability and Validity. Journal of Communication 58 258 279. Military Spouse Employment Partnership. (2014). MSEP: About Us Retrieved December 3, 2015, from MSEP Jobs: https://msepjobs.militaryonesource.mil/msep/content/about us Moon, S. S. (2001, February 24). She's Been There. The Fayetteville Observer Nelson, T. E., Oxley, Z. M., & and Clawson, R. A. (1997 ). Toward a Pyschology of Framing Effects. Political Behavior 19 (3), 221 246. Piatt, G. (2004, February 24). State Pushes Effort to Put Military Families at Ease, Duval to join program to help Spouses find Jobs and their Kids move easily into classroom The Florida Times Union RAND Coorporation. (n.d.). Married to the Military: The Employment and Earnings of Military Wives Compared with Those of Civilian Wives Retrieved April 21, 2014, from RAND Web site: http://www.rand.org/natsec_area/products/mar ried.html

PAGE 60

53 Reese, S. D. (2007). The Framing Project: A Bridging Model for Media Research Revisited. Journal of Communication 57 148 154. Sanchez, E. (2011, June 29). DoD News: DOD Launches Military Spouse Employment Partnership. Retrieved March 9, 2015 from U.S. Department of Defense: http://www.defense.gov/news/newsarticle.aspx?id=64509 Sanner, A. (2011, April 14). First Lady: Military members, Spouses Need Jobs. The Associated Press Scheufele, B. (2004). Framing effects Approach: A Theoretical and Methodological Critique. Communications 29 401 428. Schwartz, J. B., Wood, L. L., & Griffith, J. D. (1991). The Impact of Military Life on Spouse Labor Force Outcomes. Armed Forces & Soc iety 17 (3), 385 407. Segal, M. W. (1986). The Military and the Family as Greedy Institutions. Armed Forces & Society 13 (1), 9 38. Semetko, H. A., & Valkenburg, P. M. (2000). Framing European Politics: A Content Analysis of Press and Television News Journal of Communication, 50 (2), 93 109. Superville, D. (2013, March 13). Michelle Obama urges CEOs to hire more veterans. The Associated Press Terkildsen, N., & Schnell, F. (1997). How Media Frames Move Public Opinion: An Analysis of the Women's Movement. Political Research Quarterly 50 (4), 879 900. The Deseret News. (1996, May 6). Online Document: Many Families in the Military Today: How They Fare. The Deser et News The U.S. Office of Personnel Management. (2012, August 12). Rules and Regulations: Noncompetitive Appointment of Certain Military Spouses The Federal Register, 74 (154). The US Department of Defense. (2012, March 19). NUMBER 1400.25, Volume 31 5 USA. The White House. (2013). Joining Forces: States Step Up to Help Military Spouses Get Back to Work Retrieved April 21, 2014, from White House Web Site: http://www.whitehouse.gov/joiningforces/military spouses back to work Tuzzolino, T. (2008, Oc tober 27). Job Plan Would Help Military Spouses Face Unique Challenges. The Navy Times Valkenburg, P. M., Semetko, H. A., & and De Vreese, C. H. (1999). The Effects of News Frames on Readers Thoughts and Recall. Communication Research 26 (5), 550 569.

PAGE 61

54 APPENDIX A NEW HORIZONS FOR MEDIA FRAMING ANALYSIS AND MILITARY SPOUSE EMPLOYMENT: CODEBOOK Last Updated: 22 December 2015 Kallie J. Culver Graduate Student at University of Colorado Denver kallie.culver@ucdenver.edu This codebook lays out the coding categories and parameters utilized to review all data collected for this project. This codebook and its list of coding categories will provide clear operational variables to record and measure (Babbie, 2013, pp. 335 339) CODING CATEGORIES: Coding categories include the following: the article ID, the Newspaper ID, the group under which the newspaper falls (military or nonmilitary), the date of the article, the article's relevance, the nature of the frame (episodic or thematic), the primary frame evaluation (economic, quality of life, cultural identity, and policy prescription and evaluation) which is recorded based on the presence or absence of various sub frames identified within each primary frame, and whether the article was printed in multiple newspapers. 5 CODING PROCESS: Once data has been gathered and is filed in an accessible manner, coding should be conducted according to the following processes. STEP 1: A data table in either Excel or SPSS must first be created and contain columns for tallying all coding categories (Babbie, 2013, p. 338) This tally sheet will record all the code categories, and thus will provide numerical data for later ana lysis in SPSS. STEP 2: Coding should begin with recording the first four coding categories: newspaper ID, 5 The primary frames and tones were first selected from Boydstun e t. al.'s list of frames and tones. Their codebook can be referenced for detailed definitions and examples of their frames and tones. The sub frames and tone categories were further adjusted to specifically apply to military spouse employment. Iyengar's wo rk can be referenced for further background definition in determining the nature of the frame.

PAGE 62

55 newspaper group, date of the article, and the relevancy of the article. To determine the relevancy of the article, the coder should skim the headline and the article to determine if the article is addressing the topic of military spouse employment. If yes, code as relevant, if not code as irrelevant and set aside. Examples of an irrelevant article would be an article that references military, spouse, an d employment as separate and entirely unrelated nouns or an article that discusses military employment or veterans employment issues without any reference to military spouses. In addition to determining whether there is relevant coverage of military spouse employment, it is also important to remember that some articles' content will be entirely relevant and other articles will only address the issue in a portion of the content. With this it is important for the coder to apply all further coding processes to the relevant portions of the article. Each coding process should be applied in direct application and connection to content covering the issue of military spouse employment. STEP 3: Once the initial coding is complete, the coder then moves on to code the actual content to determine the primary frame, tone, and nature of the frame. Here the order changes though, as the coder must code for the various sub frame categories first. Code these sub frames according to binary code, noting their presence in the ar ticle as 1 and their absence in the article as 0. The sub frames are listed below with examples of each: 1. Economic Sub Frames Portability: any reference to military spouse career portability, whether that be difficulties with maintaining career licenses across state lines or specific jobs or career fields or companies that allow for portability. Unemployment: any reference to unemployment rates, difficulties, or benefits. Employer Interest/Hiring: any reference to employers interested in military spouses such as employers involved in the military spouse employment partnership (MSEP) or the federal government or comments on hiring practices concerning military spouses. Income/Financial Contributor: any reference to military spouses income or

PAGE 63

56 their ability o r inability to be a financial contributor. 2. Quality of Life Sub Frames Retention: any reference to military retention (keeping personnel in the military) issues in relation to military spouse employment. Separation: any reference to military members leaving the military due to military spouse employment issues. Support Programs: any reference to programs created to support military spouses, whether that is a government or military or state or local community program. Common examples mentioned are the Militar y Spouse Employment Program, Military Spouse Employment Partnership, programs run by base family centers, programs run by organizations such as Military Officers' Association of America and Blue Star Families, and many others. Note: these programs are ofte n mentioned in conjunction with legislative and policy efforts. Code for support programs, when it is discussing program implementation of those efforts. Frequent Moves: any reference to the military lifestyle of frequent moves and its impact on military spouse employment. 3. Cultural Identity Sub Frames: Volunteerism: any reference to military spouse volunteer work or their choice to volunteer in the absence of gainful employment. Entrepreneurship: any reference to military spouses starting their own busines s or being an entrepreneur. Career Flexibility: any reference to military spouses being willing to take any job or take pay cuts or be flexible in terms of their career, as well as any comments on having diverse resumes or work experience due to their mili tary lifestyle. Remember that merely being able to move around with a job should be coded under portability. 4. Policy Prescription and Evaluation Policies or Legislation: Any reference to policies or legislation impacting military spouse employment, such as the Joining Forces Initiative, Federal Military Spouse Hiring Preference Rule (noncompetitive hiring authority)

PAGE 64

57 as authorized by the Federal Office of Personnel Management, Military Spouse Employment Directive, state legislation passed on licensure portab ility, etc. Another common legislative effort mentioned in conjunction with Military Spouse Employment are Base Realignment and Closure (BRAC) hearings and decisions. Now, for determining the primary frame assess which sub frames were coded as present, an d select the correct option available: whether that be one frame, a mixture of frames, all of the frames, none of the frames, or finally select other if none of the options apply. Here again you are coding the primary frames based on merely their presence or absence, so even if there are more sub frames utilized in one primary frame over another, you would still select an option that would note both primary frames as present. Next you want to determine the nature or types of the frames being utilized. Are t hey episodic or thematic or a mixture of both? Here you will code for the presence of episodic. If the article is episodic in nature, then code a 1, if it is thematic code a 0, and if it is both episodic and thematic then code the articles as .5. An episod ic frame would be an article framed according to the perspective of a specific person or specific set of persons. A thematic framing of the article would instead be generalized, speaking about groups or people in general. Some articles will use both types of frames throughout. Examples of an episodic frame would be a human interest story on a specific military spouse, whereas a thematic frame would discuss military spouses at large. Last but not least, after reviewing the content of the article compare it with the data set and note whether the article was printed more than once in a different newspaper. Often times this will be discovered earlier in the coding process as you are reading the article and recognize familiar content. Be sure to compare dates, n ews sources, and all content. If the article is printed in a different news service, then code 1 for repeat. If the article is similar but contains different portions of text and is printed on a different date/and or news service code 0 for repeat and trea t as two separate articles. If the article is the same in terms of content and news source, but printed on separate but close dates in time, then code the second as a double in the newspaper ID column and empty all other codes for it. Also, only code for r epeated articles that reach an entirely different news audience than the first. In most cases, this will only occur across military news sources as they attempt to reach every branch of the military.

PAGE 65

58 RECORDING CODING CATEGORIES: Coding categories should be recorded as follows: A. Newspaper ID 1. Associated Press 2. Washington Post 3. Washington Times 4. New York Times 5. Wall Street Journal 6. El Paso Times 7. The Virginian Pilot 8. Air Force Times 9. Navy Times 10. Army Times 11. DOD Documents 12. Federal Times 13. USA Today 14. The Fayettevil le Observer 15. The Florida Times Union 16. The Deseret News B. Newspaper Group ID 0. Non Military 1. Military C. Date of Article D. Relevance 0. Not Relevant 1. Relevant E. Primary Frame Utilized 1. Economic Frame 2. Quality of Life Frame 3. Cultural Identity Frame 4. Prescription and Policy Evaluation Frame 5. E & C 6. E & Q 7. E & P 8. Q & C 9. Q & P 10. C & P 11. E, Q, & C 12. E, Q, & P 13. E, C, & P 14. Q, C, & P 15. All 16. None 17. Other F. Nature of Frame Utilized 0. Thematic 1. Episodic .5 Both G. Repetition. 0. No Repeat 1. Repeat Article