Analyzing social variables' effect on adolescent emotions

Material Information

Analyzing social variables' effect on adolescent emotions
Farmer, Candace Renah
Place of Publication:
Denver, CO
University of Colorado Denver
Publication Date:
Physical Description:
xvii, 127 leaves : ; 28 cm.


Subjects / Keywords:
Emotions in adolescence ( lcsh )
Teenagers -- Social conditions ( lcsh )
Friendship in adolescence ( lcsh )
Emotions in adolescence ( fast )
Friendship in adolescence ( fast )
Teenagers -- Social conditions ( fast )
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )


Thesis (M. A.)--University of Colorado Denver, 2008. Sociology
Includes bibliographical references (leaves 121-127
General Note:
Department of Sociology
Statement of Responsibility:
by Candace Renah Farmer.

Record Information

Source Institution:
|University of Colorado Denver
Holding Location:
|Auraria Library
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
268657809 ( OCLC )


This item has the following downloads:

Full Text


ANALYZING SOCIAL VARIABLES' EFFECTS ON ADOLESCENT EMOTIONS by Candace Renah Fanner B.A., Metropolitan State College of Denver. 2002 A thesis submitted to the University of Colorado Denver in partial fulfillment of the requirement for the degree of Masters in Sociology 2008.


This thesis for tho! Mastl!rs in Sociology degree by Candace farmer has been approved by Karl Flaming Date


ADS TRACT Farmer. Candace Rcnah of Sociology J Analyzing Social Variables' Effects on Adolescent l:motions Thesis direl:tcd hy Assistant Profe:-.sor Paula Fomby This statistical analysis seeks to determine the emotional impacts of differing social fal:tors upon adolescents through the usc of an ordinary least squares regression and beta comparisons using the public use version of the first wave of the National Longitudinal Study of Health. The hypothesis states that is important and bencfil:ial to the emotional well-being of an adolescent and an impact as strong as the popularity or bullying. The study found that friendship contact sometimes has a detrimental emotional effect and sometimes a beneficial emotional effect. This highlights a possible amplifiLation effect on emotions by tiiendship contact. Other interesting findings include; evidence reinfi>rcing the necessity of differentiating friendship quality from friendship existence sitKe friendship quality always had a beneficial impact. v.hereas friendship contact did nnt. realizing that the type of emotions being analyzed change the strengthen of the impacts of the social variables being studied, meaning that different social variables cause 'arying impacts on different emotions, and there arc differences in the components that bcnelicially or detrirnenlally impact friendship contact when separated by gender.


This abstract accurately represents the content of this candidate's thesis. I recommend its publication. ; \


DEDIC AriON I dedicate this to all the wonderful people in my life who have loved mc during the timl' it tonk to write this.


ACKNO\VI.EGEMENT I want to thank Prokssor Fomby for patient. helpful and wise guidance at turn of this Thank you for all of your work. Thank you Prokssor Dwight and Profcssor Flaming for your insighh and help. This research data from Add Health, a program project designed by .1. Richard Udry, PeterS. Bearman. and Kathleen Mullan Harris. and funded by a grant PO l-HD31921 from Eunice Kennedy Shriver National Institute of Child Health and Human Development. with cooperative funding from 17 other agencies. Special acknowledgment is due Ronald R. Rindfuss and Barbra Entwisle for assistam.:e in the original design. Persons interested in obtaining. data files from Add Health should contact Add Health. Carolina Population Center. 123 W. Franklin Street. Chapel Hill. NC 27516-2524 ( No dire..:t support was received from grant PO I-HD31921 for this analysis.


TABLE OF CONTENTS Figure ....................................................................................... .ix Tables ........................................................................................ x CHAPTER I. INTRODUCTION ...................................................................... 1 Guiding Quotes .................................................................. 1 Intro Statement ................................................................... 2 2. LITERATURE REVIEW .............................................................. 3 Developmental Factors ........................................................ .3 Early Impacts ......................................................... 4 Growing to Become More Peer Oriented ......................... 5 Sociometry and Popularity ..................................................... 6 Measurements ......................................................... 6 Questioning the Connection between Happiness and Popularity ............................................ 9 Describing Popularity ............................................... 1 0 Negative and Popular. ............................................... 14 Popularity and Peer Rejection Patterns ........................... 15 Friendships ...................................................................... 16 Friendship Variety ................................................... 17 Friendships' Positive Role ......................................... 18 Negative Consequences of Friendship Contact. ................ 19 Change in Friends ................................................................... 21 The Complexity of Social Context.. .............................. 23 X


Rejection and Peer Trouble .................................................................. 24 Peer Influences on a Child's Social Development ............ 26 Problems with Reputation ......................................... 26 Specific Social Skills .............................................. 27 Victimization ........................................................ 30 Adult Consequences of Peer Rejection ......................... 31 Blaming Social Victims ........................................... 32 Interventions .................................................................... 33 Gender. ......................................................................... 36 3. METHODS ........................................................................... .38 Data ............................................................................. 38 Variables ....................................................................... 39 Brief Description of the Analysis Methods .............................. .49 4. HYPOTHESIS AND RESULTS ................................................... 50 Hypothesis ...................................................................... 50 Results .......................................................................... 5! The Dependent Variables' Coding and Interpretations ................ .51 Visual Representations of the Three Dependent and Six Social (Independent) Variables ............................................. 52 Descriptive Statistics ......................................................... 59 Interpretation of the Results ................................................. 63 The Data Tables for the First Dependent Variable ........................ 66 The Data Tables for the Second Dependent Variable .................... 80 XI


The Data Tables for the Third Dependent Variable ...................... 94 5. DISCUSSION ......................................................................... I07 Friendship Contact Impact on Emotions Using All Feelings as the Dependent Variable ........................................................... I 07 New Findings Using Happy Feelings as a Dependent Variable ....... ll5 New Findings Using Depressed Feelings as a Dependent Variable ... II? 6. CONCLUSIONS ..................................................................... 119 BIBLOGRAPHY ................................................................................... I21 Xll


LIST OF FIGURES Figure 3.1 The Distribution of the Variable Male Friendship Contact.. ................................................. .43 3.2 The Distribution of the Variable Female Friendship Contact.. .......................................... .44 3.3 The Distribution of the Variables Male and Female Friendship Contact Combined ..................................................................................... .45 Xlll


LIST OFT ABLES Table 4.1 A table showing the creation and coding of the dependent variable all feelings ................................................................................................................................. .53 4.2 A table showing the creation and coding of the dependent variable happy feelings ................................................................................................ 54 4.3 A table showing the creation and coding of the dependent variable depressed feelings ......................................................................................... 54 4.4 A table showing the creation and coding of the independent variable male friendship contact ................................................................................... .55 4.5 A table showing the creation and coding of the independent variable female friendship contact. ................................................................................ 56 4.6 A table showing the creation and coding of the independent variable friends caring ............................................................................................... 57 4.7 A table showing the creation and coding of the independent variable first friends at school. ..................................................................................... 57 4.8 A table showing the creation and coding of the independent variable social status ............................................................................................... .58 4.9 A table showing the creation and coding of the independent variable trouble with peers ......................................................................................... .58 4.10 A table showing the descriptive statistics of the variables in the study ............................................................................................................... 59 4.11 Ordinary least squares regression results predicting adolescent emotional adjustment as a function of friendship contact and gender with unstandardized and standardized coefficients using all feelings as the dependent variable ................................................................................. 66 4.12 Ordinary least squares regression results predicting adolescent emotional adjustment as a function of friendship contact, gender, race, family structure, and parental education attainment with unstandardized and standardized coefficients using all feelings as the dependent variable ....................................... 66 XIV


4.13 Ordinary least squares regression results predicting adolescent emotional adjustment as a function of friendship contact, gender, race, family structure, parental education attainment and substance use with unstandardized and standardized coefficients using all feelings as the dependent variable .................................. 68 4.14 Ordinary least squares regression results predicting adolescent emotional adjustment as a function of friendship contact, gender, race, family structure, parental education attainment, substance use. delinquency, school trouble, grades and parental closeness with unstandardized and standardized coefficients using all feelings as the dependent variable ...................................................................... 69 4.15 Ordinary least squares regression results predicting adolescent emotional adjustment as a function of friendship contact, gender, race, family structure, parental education attainment, substance use, delinquency, school trouble, grades, parental closeness, friends caring about you and first friends at school with unstandardized and standardized coefficients using all feelings as the dependent variable .................................. 71 4.16 Ordinary least squares regression results predicting adolescent emotional adjustment as a function of friendship contact, gender, race, family structure, parental education attainment, substance use, delinquency, school trouble, grades, parental closeness, friends caring about you, first friends at school and social status with unstandardized and standardized coefficients using all feelings as the dependent variable .............................................................................................. 73 4.17 Ordinary least squares regression results predicting adolescent emotional adjustment as a function of friendship contact, gender, race, family structure, parental education attainment, substance use, delinquency, school trouble, grades, parental closeness, friends caring about you, first friends at school, social status and trouble with peers with unstandardized and standardized coefficients using all feelings as the dependent variable .................................. 75 4.18 Ordinary least squares regression results predicting adolescent emotional adjustment as a function of friendship contact and gender with unstandardized and standardized coefficients using happy feelings as the dependent variable ................................................................................................... 80 4.19 Ordinary least squares regression results predicting adolescent emotional adjustment as a function of friendship contact, gender, race, family structure, and parental education attainment with unstandardized and standardized coefficients using happy feelings as the dependent variable .................................. 81 4.20 Ordinary least squares regression results predicting adolescent emotional adjustment as a function of friendship contact, gender, race, family structure, parental education attainment and substance use with unstandardized and standardized coefficients using happy feelings as the dependent variable ............................. 82 XV


4.21 Ordinary least squares regression results predicting adolescent emotional adjustment as a function of friendship contact, gender, race, family structure, parental education attainment, substance use, delinquency, school trouble, grades and parental closeness with unstandardized and standardized coefficients using happy feelings as the dependent variable ................................................................................ 83 4.22 Ordinary least squares regression results predicting adolescent emotional adjustment as a function of friendship contact, gender, race, family structure, parental education attainment, substance use, delinquency, school trouble, grades, parental closeness, friends caring about you and first friends at school with unstandardized and standardized coefficients using happy feelings as the dependent variable ........................................................................ 85 4.23 Ordinary least squares regression results predicting adolescent emotional adjustment as a function of friendship contact, gender, race, family structure, parental education attainment, substance use, delinquency, school trouble, grades, parental closeness, friends caring about you, first friends at school and social status with unstandardized and standardized coefficients using happy feelings as the dependent variable .................................................. 87 4.24 Ordinary least squares regression results predicting adolescent emotional adjustment as a function of friendship contact, gender, race, family structure, parental education attainment, substance use, delinquency, school trouble, grades, parental closeness, friends caring about you, first friends at school, social status and trouble with peers with unstandardized and standardized coefficients using happy feelings as the dependent variable .................................................. 89 4.25 Ordinary least squares regression results predicting adolescent emotional adjustment as a function of friendship contact and gender with unstandardized and standardized coefficients using depressed feelings as the dependent variable ............................................................................................. 94 4.26 Ordinary least squares regression results predicting adolescent emotional adjustment as a function of friendship contact, gender, race, family structure and parental educational attainment with unstandardized and standardized coefficients using depressed feelings as the dependent variable ........................................................... 94 4.27 Ordinary least squares regression results predicting adolescent emotional adjustment as a function of friendship contact, gender, race, family structure, parental educational attainment and substance use with unstandardized and standardized coefficients using depressed feelings as the dependent variable ............................................. 95 xvi


4.28 Ordinary least squares regression results predicting adolescent emotional adjustment as a function of friendship contact, gender, race, family structure, parental educational attainment, substance use, delinquency, school trouble, grades and parental closeness with unstandardized and standardized coefficients using depressed feelings as the dependent variable ............................................................................................. 97 4.29 Ordinary least squares regression results predicting adolescent emotional adjustment as a function of friendship contact, gender, race, family structure, parental educational attainment, substance use, delinquency, school trouble, grades, parental closeness, friends caring about you and first friends being at school with unstandardized and standardized coefficients using depressed feelings as the dependent variable ................................................................................ 99 4.30 Ordinary least squares regression results predicting adolescent emotional adjustment as a function of friendship contact, gender, race, family structure, parental educational attainment, substance use, delinquency, school trouble, grades, parental closeness, friends caring about you, first friends being at school and social status with unstandardized and standardized coefficients using depressed feelings as the dependent variable ........................................................... 102 4.31 Ordinary least squares regression results predicting adolescent emotional adjustment as a function of friendship contact, gender, race, family structure, parental educational attainment, substance use, delinquency, school trouble, grades, parental closeness, friends caring about you, first friends being at school, social status and trouble with peers with unstandardized and standardized coefficients using depressed feelings as the dependent variable ............................................ 1 03 XVll


CHAPTER I INTRODUCTION GUIDING QUOTES our American culture places a premium on sociability and all the qualities associatcu with the extrovert. This is because we arc a nation of salesmen and sociability pays. Jf however. we rcrmwed the premium value of sociability anu emphasizeu the enjoyment value. we would no longer attempt to make all children leaders: a goal few of them can reach, hut would help each child to gain tlwt level of sociability which best sati:;fied hi" unique needs. That i!>, we would formulate our requirements for sociability on the basis of the individual ability to meet them"

INTHODUCTlON STATEf\.lENT Emotional \.VCll being is impnrtant for all peopk, y11ung and old. Peer relationships and inteructions haVl.: hcen found to to happiness, lh.:pression, loneline::;s, and self estl.!em. The patterns of peer interactions that begin in follow young people through their development and sctwol careers impacting them throughout their lives. Peer relations and interactions have many long term as well short term effects. Interactions can he managed and social development can be encouraged. The purpose of this thesis is w undastand more about the social world of adolescents and children in modern America. specifically. the emotional impacts of various social factors. The research question investigates the emotional impacts of ti"iendship tiiendship quality, populairty, trouble with petrs and having your best friends at school. The research the impacts of these variables uplll1 di ffcrcnt emotions, and how they interact \\'ith each other. The was designed to sec how having a friend. measured by friendship quality and Jiiendship contact. impacts emotions, how their impact is compared to the impact of variables like being popular or having trouble with peer:-. and how all of these variables interact with each othcr contrihuting to adolescent emotions. The differences between popularity and friendships and their impacts upon youth arc fascinating and instructive. A better understanding of the interplay between emotions and social factors can help inform. diagnose and intervene. It also will help to Jill in descriptions of social worlds, emotional worlds and self concepts. The literature review will examine the research on social factors in the lives of children and to build the contc:xt in which this study will tit. 2


LITERATURE REVIEW DEVELOPMENTAL fACTORS Family is the first forcc and shapes how childrcn approach later peer relations. The Jevcl of social competen..::y a child learns through family and ot.her early life experien..::es is a strong inlluence in a child"s later devdoprnent (Erwin. 1993: Laird. Penit, Mize. Brown & Lindsey. 1994: & Paley, Conger, & Harold, 2000). It is a given in all of the literature that peer relationships strongly impact individuals emotionally. The literature review starts at the beginning, emphasizing the need for good to start in infancy. continuing into childhood and on to adulthood Laird et al.. 1994; & Paley et al.. 2000). Here it is stated that peer competency is intluenced hy those around you and can he taughtiErwin Laird et ai.,J994). This changeahility is important to keep in mind when considering interventions and the overlapping functions of ,arious S(lcial factors. fact that relations with peers impact emotions and the variahlcs that compose social variahks intcract and can he changed is INhy the of social variables, their development. changeability and interactions is described in dcpth in this literature review. [tis also a rcason for the stuuy itself. Family will always be vital to the functioning of children and adolescents hut. peers he<.ome 'stcppingstones to demonstrute independence and allow children and adolescents to without their family !Bowerman & Kinch. 1951}) and train to become competent beings as adults (Steinberg & Silverberg. J9g6l. Relationships with family and friends are associated with how the growing child understands the adjustments and changes they experience in !Gauze, Bukowski. A411an. Assec, Sippola. & 1996). It would seem thnt family and tiiends are able to impact the adolescents' life pl1sitively. This view of the positive role of family and friends informed, in part, 3


the hypothesis it points to the importance of friends and a 4uestion. Huw do the roles of and friends separately impact and o\'erlap with otht.:r -.ocial variables in their emotional Early Various studies examining the roles of parents as socializing agents support the argument that ones socializing begins at birth and continues throughout childhood (Paley et at.. 2000). It was found that peer acceptance and social behavior can he by maternal and pattrnal variables reinforcing tht.: idea that childrens characteristics are formed young (Paley et al.. 2000). Erwin summarized the literature about early socializing impacts and concluded. warm. responsive parenting early in life. and continuing into later childhood. appear to be crucial 11n the child's social orientati

Growing 111 Bewme More Peer Oriented I3owerman and Kim:h ( 195lJ) state that children get and spend more time with peers they hccom.: rnore peer oriented. The study looked at children starting in fourth grade and ending in tenth grade. ll found a loss of f:m1ily orientation was not inevitahk. hut depended on adjustment (Bowerman & Kinch. 1()59 ). P

about the variables used in the statistical analy-;is will he revinveJ thoroughly to provide mntcxt and meaning. SOCIOrvJETRY AND POPULARITY Sociometry. the study of imerpersonal relations in populations can on looking at the interactions. liking or disliking and other varieties of n:lations within a population and each of these aspects arc measured and tested differently 1Scholield & Whitely. 19g3J. section of the literature review examines the uf popularity and acceptance. to name a few variations of sociometry, which are related. hut are not the S

he hard to manage and ddin..:. Children can also be asked til list the childr..:n tlll'Y mo-;t dislike at sehOlll or with who they would least like to play. This gives a negative sociometric reading. All of these variations haH' pros and cons. Another can he of group an:eptance or generallikcahility. This can he done by students rating how much they like each in their clas!'> on a scale: thc!'>e can then he mathematically analyztd. The intricacy of what these methods truly measure complex, detailed and necessary to distinguish nuances in group and individual functioning. Schofield and Whitely ( 1983) looked at the dilkring: results that carne from two sociometric tools which were designed to measure the sam!.' thing and concluded that interpretations must acknowledge of the effects of their measurement tools upon their data. may seem obvious, but remi.nds the reader to be mindful of the tools being used. Northway ( 194t'l) examined and listec.J some of the challenges of sociometry sucl1 as: tht difference between social acceptance or social acceptability, the causes of change in group and personal status. prclcrence or prejudice and cflccts. Some S\ICiometry and popularity r..:scan.:h covers how ortcn children and adolcsc..:nts arc picked friends by their peers and so an.: considered popular or unpopular. how the choices arc made and \vhy these choices arc made. A mentioned variety of sociometry is likcability. Lik..:ability, popularity and friendship reciprocation arc not ne(essarily the same social phenomena and as Scholfield and Whitley {I 9!\3 J pointed out. simply using a diflercnt tool to get a measurement will change what is being measured. so one must be sure to correctly for the variable they arc studying. The nuances of social variables, their different functions and their diflerent impacts must be better described to understand the variables in this analysis. The small differences arc necessary to explore and detail so that interactions causes and effects and other social phenomena can be categorized for intavcntions and understanding The way friendships and other social variablrs 7


arc measured can change how they arc undcrstoild and the way variables arc measured make a large difference when it comes to accurate tools for social variable analysis. Sociumctric nominations can vary in how they arc measured. \\-'hen studi.::s constrain friendship nomination., they cut otT information about other friends and distnrlthc true social networks of the children being studied. If a study mak.cs the nomination'> upcn then there arc questions on how to if the friendship is rcciprm:atcd, unless they hunt down all nominations. and there is still the question on how to analyze nontraditional friendships with much younger or older nominees. The concern that limiting nominations limit the true friendship network of the peers studied lead George and Hartmann ( 1996) to try to lind new ways to estimate friend:;hip patterns, with these methodological concerns in mind. and to better the social networks of unpopular. popular and average children. George and Hartmann ( 1996) found that widening the possibilities of friends listed on the nominations gaw a better \ iew of frien(bhips that can be used in interventions. {hiends that arc younger. older or found in di llcring: locations than the regular school peers of the unpopular child arc now indudcd.) They found that children \Vho arc unpopular make fewer overall friends than their mon: popular counterparts. It was abo noted that unpopular children's friends tend to be unpopular as wdl. This linding highlights the possibility that these childrcn"s social progression on one hand is hdpccl by their friends because they arc friends who provide an important function, but rna) also be inhibited because since tlu.:y arc all unpopular their mutual social skills may be more limited and their social modeling not as advanced that of more socially accepted children. Being included in an unpopular group can reinfon.:e popularity and social alienation because it may be that the group of unpopular children may he seen as less socially acceptable than the child alone. Multiple factors, including reputation, popularity, as well as hchavior and activities can intluencc friendships and how these 8


friendships impact emotions and dcvdopmcnl. A simple measure of the variable representing friendship contact in this study is limited to the contact had with the first friend. male and female, listed on their l'riendship lists. This variabk docs not represent popularity or s!H.:iometric status in this analysis. hut rathn only represents the having of a friend and spending time with that friL'nd. Having the \ariahlc he simple ami represent a small piece of a respomk:nts network helped make the results of the analysis more dear. The pros and ccms of friendships and spending time with friends is reviewed later in this literature review and should he applied to \vhat the friendship contact variables truly represent in an adolescents S11cial reality. Questioning the Connection Bet\veen Happiness and Popularity Popularity is a social variahlt: that is very much in the public eye with movies and popular culture. Huw popularity really functions is re\iewcd in the literature review to give the reader a fuller account of its aspects before heing used as a variable in the analysis. Chapin ( 1950 J explored the placement of sllciometric stars. the most popular children within a population, with their peers in a new way amlloukcd for possible implications. Chapin stated that if the stars (sociometric J arc plotted threc dimensil'nally they arc found on the periphery of peer clusters. Chapin concluded position should be seen as isolating. Chapin s article questioned the assumed ideal of the well-adjusted and happy popular child. Chapin did get some criticism and some explanations of this phenomenon. l'vtorcno ( 1950) pointed out in response that there are three different types of leaders. There is an isolated leader. a powerful leader and the popular individual. These different sorts of leaders can be noted in analysis and put into perspective \vhen considering tht: functions of reputatit1IL popularity. group acceptance, roles and friendship. Criswell and ( 1951) also responded with a critique. 9


asserting that Chapin the meaning of thL' plnllcd stan,' locations. Criswell and Jcnninf!S (I 95 I J explained that just bt!cause an individual an imbalance of people preferring them. without them returning pn:fercnce, docs not de line them as emotionally cold or isulated. They claimed that in the field or sociometry beller visual model!> arc being pursued that could heller show the 'location of stars. What is interesting about these article!> is they question popularity and happiness. ahoUl what makes a child or adolescent happy or must be questioned and addressed from a nqn-adult perspective as well as a non-culture value laden philosophy. Desnibing Popularity Popularity may faces. This section reviews what popularity can he dt!lined as and how a person might become popular. This exploration (lr popularity is meant to inform the reader on the variety of forms found in the com.:cpt of popularity. so the variahk used in analysis. popularity, can h.: put in its plm.:e within the literatun. Kennedy ( 1995) had students rate thLir own popularity at the same time as they sci f r.:ported on their personal variable!>. The analyses conducted found athletic status yielded the correlation with popularity more than any other variable for all ethnic groups and gender groupings except Black females. The models used in this study explained less than half the observed variation in popularity among peers. suggesting variables may differ for peer groups within schools. This idea of a of popularity variables invaluable especially when looking at other cultures and environments. Wheeler ( 196 I) stated results showed that although differing cultures and subcultures will produce minor differences in characteristics and their importance in popularity, that most culturt!s and suhcultures havt! agreement on desired personality trait agreement. 10


Tht-;e were made fwm comparing popularity am11ng adolcsccnh fwm Western Australia and the United States of 1\mcrica and so may not he generalizable to non western hut that question was not focused upon. It should abo he noted that a more recent rc,iew of concluded that the evidence suggests acceptability of social behavior varied .kpcnding on cultural context and even the .. local'' culture of a particular peer group can inlluenccs what is acceptable and not (Osterwcil & Nagano-Nakamura, 1992). It was unsurprising that popular individuals \Vcre often found lo usually have good social adjustment. and positive pcrs0nality and lack of negative personality traits. {Davids & Parenti. 1958) and that peers judge more attractive peers more positively ( Dion & Bers.::heid, 1974) and these types of judgments could cam.c different treatment and expectations which would help .::ause unpopular behavioral patterns (Erwin, ). Any characteristic that is chosen hy peers to designate an individual as different can make them unpopular. These can be di llcrenccs like atypical behaviors, minority status. atypical looks, physical or mental handicap. 1lhcsity. a different income leveL unattractiv.:n.:ss. an accent or being from another place (Envin, I Y9Tl. The idea that popular students arc usually more adjusted and "atlractive" is not nt'IA to the literature. hut neither is the idea that not all popular individuals were inhcn:ntly 'goud ... II


AttracrilcJWI.\ ond Pop11/ariry hook aJJrcssn.l physical and impacts upnn It concluded that although childn:n arc agents in their own an attractive child elicits different and treatment from peers and adults than docs an unattracti\'l' child. These produce di llcrcnccs in of behavior and ultimately in personality and sci r concept'' 1 Erwin 1993, p. 66). From the conclusion of it seems safe to say that attractiveness usually is an advantage. That is to say allractivcness L)r unattractiveness can impact personality. This factor of physical attractiveness not found in the analysis. hut should be remembered as one that might influence emotions and social placement. Personality Traits and Popularity l\uch more than a child's looks arc imolv!.!d in the study of popularity traits. Davids and Parl.!nti ( 1958). looked at children in residential treatml.!nt and summer and found a corrdatinn. in both hctwccn social popularity and good emotional adjustment, possl.!ssion of positive personality traits and the absence of negative traits. The overall tone of dcs.:riptivc articles seemed to he that positive traits described the popular children and IH.:gativc described the unpopular. is not to say that there arc not a lot or other reasons for popularity or unpopularity that will also be addressed. ali'>O is a good place to point out that describing the characteristics or popular and unpopular children arul adolescent's doesn't prove or imply a casual relationship. It docs not sho\\' if these characteristics were caused from being treated as unpopular or popular or cau!.e their social placement in the first pbcc. lL can always he a combination of both factors (Erwin. 1993 ). It is 12


necessary to remember that and popular traits arL' just one variahlc in the larger of popularity. just pllpularity is just one of the putzle or a child's social de\L'Iopmcnt and health. Pop11/arity and Sol'ial AJiustment Popularity docs not imply kindness. McGuire and WeisL ( 1982) fuund that children \Vith chums had a stronger sense of ahruism than those wi Lhout frienus and this was not true vvhen comparing the popular to the unpopular, which had no difference in altruism. Also. in a that looked at negative sociometric d:lla a high prestige (popular group) individual was fL)Und to he highly rejected hy her peers !Smucker. 1947). In another study. which focused on reintegrating an isolate into normal peer inll!ractions it was noted, through the observations of the group. that popularity did not guarantee good mental health and that anti-sot:iallcadcrship can help a child or adolescent rise to the top of the social hierart:hy lRcgcr. 19621. Lastly, many negative and stressful behaviors were found tll be a part of thl! functioning of '"the popular group (Adler & Adler. 1995). Often manipulation and bullying ot:curn;cl within these groups and were not simply aimed outward at unpopular children or hut inward as well. These findings inflw.:nced the hypothesis because they emphasiLed the fact that ht:ing popular did not make life perfect. So. the question came up, how does having or not having a friend come into play with these other fa(."tors? When creating the variables used in analysis. popularity was created from feeling included from the peers in the school o\'erall and was not dependent on having friends or being snciall) adjusted because popularity research proved that these are separate factors. The separateness of popularity. acceptance and having friends is more fully Jetailed in the next section. 13


NegatiH: and P11pular Smucker ( 1947 J found that the most rljected indi-.,iduals were usually located in the low prestige category. but there was also a high status individual that \Cry highl} rejected by her peers. This shows the distinction between. her .. mass appear'" .. individual appeal. .. her status and her actual frienthhips and likeability. This is an example oftbe phenomena l'fnegative popular/high statused individuab. The well known fact that not all popular individuals arc nice or liked should also he remembered during analysis. Another study about negative popular found children who have a high aggression classification can be con!:'idered popular if they use their aggression to lead and gain dominance. These effectual aggressors can make and are not bullied. hut instead often pick on others (Perry. Perry. & Kennedy. 1992! and shoulc! be addressed in intervention if they arc picking on other children in an ongoing manner. The possibility and reality of socially undesirable or damaging beha\ l'Oming from popular peers needs to he emphasized to fully understand thl' social dynamics of a population. Popular children who have friends, hut act in damaging and anti social ways arc a stwng reminder to never equate friendship and popularity with social pro-adjustment or behaviors. Being popular does not necessarily mean being good or happy. The fact that being popular did not guarantee being socially adjusted. having real friends. or being nice supported the hypothesis that having real friends could be as emotionally positive as being popular. 14


Popularity and Popularity and generallikeahility are different variables. This truth'' '>cems to be accepted in popular culture and can he seen relle<.:ted in the media. Thi'> study would like to question if the mean girls are happy even if they do not have good friendships. It would also like to tJUestion if the hullied hoy feels alright (or better) if he has friend., who care'' The group behaviors of popular and unpopular children and adolescents vary hy subcultures. but patterns do 0ccur. Alder and Adler I I <)95) interviewt:d elementary school children in a longitudinal study that also incorporated parti(.ipation and observation. They found in cliques inclusion-exclusion tcchniLJues. power and manipulation. and in-group/\JUt-group differentiation. These dynamics, they wncludcd. could in part reproduce racism. sexism, and other forms of bigotry and discrimination. l'vtuldoon ( 1955 l found that the more a group articulates its popular and unpopular memhas the less it functions wdl as a cohesive unit that is friendly. illustrating how structure and cnvironmt.:nt influences pct.:r group bt.:haviors and :.Htitudes which in turn intluem:es individuals' social dewlopment. Analyting these hehaviors and dynamics can aid educators and others to reform the social environments children and adoksct.:nts arc cxpost!d to and sncialite within. These patterns are not directly studied in this analysis. but their effect, as measured hy being part or the popular crowd or nllt. is measured by the variable popularity which in this study is hy inclusion. 15


FR lENDS I-IlPS arc a social factor that must he considered in context as well as independently from acceptance. popularity and bullying. Many aspcl:h of friendships, their formation ami dissolution and their 4uality and structure, have hcen studied and have produced more areas to research. The quality of a l:hild's friendships is not measured hy sociometric tests (Hogan & Mankin, 1970). Extremely liked children may not h;ne rel:iprocal best friends; average children have their desired number of reciprocated friendships, and even. half of children of low status have at least l)ne best friend whom reciprocates (Asher & Parker. 19tJJ ). Friendship quality has its own measures. and structures. Friendship, looked at by has a large amount of research detailing its many questions and complexities. In this literature review some of these complexities v.ill be covered in an attempt Ill undnstand friendship, whik leaving room for its complil:ations. and fitting it into a model of sol:ial understanding along with bullying and pllpularity. interactions \vith these and other variabks an: invaluable for an understanding of the social worlds of adoll'scents and children and to better describe friendship functions. Burhrmcster ( 1990) concluded it bCCllllles more important to establish close fricnJships in early aJolcscence than earlier in childho0d. Having friends is related to many functions. Feeling emotionally being close w others. help and advice, ego and a relationship in which w develop social competence are a few of the fum:tions of friendship (George & Hartman. 1996). Given the of the functions of friendships it was concluded that. "al-risk" children would benefit from friendships 16


Friendship Variety Acknowledging the various types of and their dynamics is vital to an a.:curate or their functions because different types or friendships are bound to have vastly divergent impacts. emotionally and dl'\doprnentally. The myriad of friendship possibilities and outcomes should he remembered during analysis and will be partiaJiy covered in this literature review. Bonney ( 1946) reviewed numerous dealing \Vith mutual friendships and .:oncluued it was hard to lind agreement on what facwrs attracted friends to each other. Why and how children and adolescents become friends would be a good place f(,r more study. Regardless of how they an: formed. establishing close friendships becomes more important during early adolescence when compared to earlier ages (Buhnnestcr, llJ90). The tinding that friendship is needed for development seems to be a ..::onscnsus. but it is important to note that there arc more variables contributing to development other than simply having or not having friends and there i!. also the of negatin: asp.:cts of fricnuship. Hart up ( 1996) begins his article by concluding that the killing of a mother by a fourteen year old and his best friend. was an unlikely event until these two antisocial friends reached consensus about doing it .. (Hanup. I 996. p. I). Hart up ( 1996) argues that b.:ing liked"" and 'being have independent ( Iiom friendship) impacts and correlations to social compctem:e. Then: an: differences between having friends. their identity and the friendship quality in that friendship (Hart up I 9lJ6). Hartup (I 996) concludes that the evidence from the literature supports that friends arc needed to he .:og.nitive and social scaffolders for each other providing. positive support. Hnrtup ( 1996) contends that knowing the behavioral characteristics. attitudes. and the quality of the friendship is necessary to make predictions and to more fully understand friendships and their developmental significance. The and t11:g.ative aspects of friendship must he conceded to get a balancl'd view of its functions. 17


The dillerence hct\vcen the impacts of having friends and spending time with them. and friendship quality is the reasnn why friendships were analy;ed two variables. Strong evidence reinforcing and reiterating this dillnencc is outlined in thl' twn sections. Fril'ndships Positiw Role The importance of friendships for children and adolescents has many important functions like. emotional security. ego support and validation. intimacy affection. guidance and assistance. companionship and stimulation. a sense of reliable alliance. and a forum for development of social compelt!nce'' I George and Hartmann. 19%, p. 2302 l. These friendships may be able to shield the unpllpular children from universal peer rejection and help them build sol:ial skills that could aid in their integration with peers. The positive and possible "shielding"' l:apahilities of friendships will be l'xplored in this study. H,w; much do these positivl' aspects offril'ndships impal:t emoti

1}) security, (4) ami (5) conilict. There are many more scales that can he used and tailored to Ill the need of the study llr intervention. Thi:-. study measured friendship quality u-;ing llnc component. hnw n\uch the re"pondent felt their friends cared about them. This was used the questiun was available. answered using a sL:ale and sl.!crned to represent many strong wmponents of fricnd>.hip quality: affection. concern. sympathy and empathy. Negative Consequences of Friendship Contact Since friendship contact and hip quality are two of the main variables in the the negative sides of friendship (the lack of quality and trouble caused by friend contact) must ht.! linked to their impacts and t.!valuated as a ran of the variables. Windle ( 1994) found from self reported survcyo, used to identify hetvveen Friendship characteristics and behaviors, that overt hostility. cllvert hostility and kss reciprocal relationships occurring in close friendships arc associated with higher lcvds of drinking. delinquency. depression and suicidal hehaviors. Thi>. study emphasizes the irnportanl:c of peer interactions. net works and quality and their afl'l!ct on hehaviurs and health. Kupcrsrnidt, Bun.:hinal and Patterson ( 1995) L:oncludcd that for some rejected children having a best friend made them more likely than their rejected counterparts to develop antisoL:ial problems. A proposed explanation from other literature for this is that the dtaractcristics of the the rejected children were making were deviant and they were ncgatiw peer influences and encouraged delinquent activities like, drugs, alcohol, risky sex and dropping out ( Dishion, Andrews, & Crosby. I '}95: Tremblay, Masse, Vi taro & Dobkin. I '}95 l our soL:iety is structured hierarchically and actively pushes children and adults towards forming relationships with similar others'' (Erwin. 1 Q93. p. 145). Pressure Ill be similar does not simply 19


set guidelines. hut rather put prl'SSUI"C upon the individuals in the group to conform 111 group norms. The phenoml.!nnn of dcml.!ntary-s..:hool childrl.!n conf11rming stringently is due in part fnr the need Ill he similar. gain approval and ..:omparl.! favorably to onl.!s peer grllup (Erwin. The drin to he and yet an individual. can hl: in a variety of social settings. drive is strong and motivates many behaviors and actions despite negative consequences. The drive to be accepted as normal can be seen in friendships, pPpularity hids, bullying and bullying avoidance. The power of social similarity lor soml.! youth i<> so strong that many actinns. styles. and traits may he partially or totally explained by the Kandel ( ll)?g l conclutkd that her study contirmed that adllit:!sccnts groups will \h)rk to be similar to their friends when it came to frequen..:y of marijuana usc. levels of educational aspirations. political orientation. and participation in minor delinquency. If there was unbalanced llr largely differing attitudes between an adolescent and their friends' attitudes. the adolescent would either leave the friendship and seek new ones or change their own behavior or attitudes to tit. These findings emphasize the nl.!cd to have togetht.:r in healthy and prodw.:tivc groups as much as possible of the pressure to ..:on form. N!.!gativc peer inllucn..:cs and how this inllucnce power and momentum should he further. ami not just in unpopular groups. The inllucncc of pressure and ncgati\'e activities arc a detrimental of fril.!ndship that needs to be ad:nowlcdgl.!d in their own right. To analyze the impacts of friendship contact and quality without the intluencc of peer pressure, suhstan.:c abuse and delinquency arc important control \'ariablcs that an: added into the analysis Another negative aspect of friendship is the effect of co-rumination. Co-rumination is when friends dwell on their problems. speak of them. and focus on them so often that they increase depression and anxiety (Rose. Carlson, & Waller. 2007 ). This study on co-rumination found that for girls. increased depression and anxiety as well as increased quality was predicted hy 20


co-rumination. Co-rumination was also predicted In incr.:ase when anxiety and friendship qmlity increased. trade off was healthy as their friendships were strengthened. hut unhealthy they took on emotional For eo-rumination preJi..:ted higher quality, hut not the depression and anxiety. The implication being that some girls may Jevelop undetected internalizing problems hecausc: they have friendships that arc dose and seLm supportive. hut arc fostering anxiety and (Rose. Carlson. & Waller, 2007 ). This negative emotional effect of frienuships is one which must again he kept in mind when analyzing the emotional benefits and costs of different social variables since a variable was not created to measure co-rumination and its impacts. Despite the possible negative consequences of certain friendship is nonetheless vital to children and adolescents. Loneliness and isolation are not viable alternatives to poor frienuships. Other options and alternatives must he pursued. It should alSll be noted that although total lad. of companionship or friendship is ahnormal and hurts development there will he some normal and renewal of friendships. will he made and lost and depending on their patterns may simply he the normal ..:hanging and gfllwing many youth go through on their way to maturity. Changes in Friends Friendships and friendship pallerns change in a child's and adolescent's life. Change is normal. but it is alsn true that panerns in friendships can be observed and if there is an ongoing friendship pattern then it is likely In have an ongoing emotional and developmental impact. The relative stability of friendship patterns for individual adolscenls strengthens the conclusions about the 21


\ariable:-. emotional impacts because they arc not random. Parker and Seal t I t)l)()) analyzed the formation and dissolution of child friendships at summer camp and detailed that there were four differing tracks that friendships and relations to the group could takc during the temporal camp setting. More importantly. children lrom.the differing differed from ca.:h other and differed distinctly from the children that wen: chroni.:ally friendless. in their behavioral pro1iles. ThL' four groups were unique. One group showed a pattern of friendship ongoing rotation. the second group. showed a decline. the third group showed very little change in their friendships. and the last group showed a pattern of growth (Parker & Scal. 1996 ). All of these were found to be bt:haviorally different than children \\"ho wac chronically (Parker & Seal. Jl)96 ). These patterns of friendship interactions caused by behaviors and reputations are extremely helpful because they can be used to isolate behaviors and their ti"iendship consequences. This not to say all lack of formations or overabundant need to .:orrected or wert: caused by had behavior that can he predided. Some natural friendship .:ycles can occur, hut it is to say that many can he predicted and changed if unhealthy. 22


The Complexity of Social Context flo II' Popularity is D(ffere/ll tlwn Hurillg Friend\ (Put in Social Context 1 This seLl ion delves into the complexity of social roles darifying the dilTcrcnccs bet wcl.!n friendships contact and quality. having friends at schooL social popularity and trouble v.ith peers. Popularity and reputation arc difkrcnt but related. Peer acceptance. social preference. and reciprocal friendships arc all measured and produced thwugh ditkrcnt social skills and one's placement and reputation in a social network. These factors irnpacr the chii,rs or adolescent's social life as the individuals treated differently and have different pressures depending on levels of acceptance. status and rccipro.:al they have. As already stated, McGuire and Weisz 1 I YH2) assert that thl' general implications of their show that investigators must not equate friendship and friendliness with p(1pularity. This conclusion was formed as the data sho\\-"ed that children with 'chums had higher levels altruism displayed than those without friends while popular and unpopular children showed no dilTcrl.!nce in their displays of altruism. Thenfon: altruistic behavior prediction hct\vcen popularity and friendship was not congruent. The distinctions betwcl.!n these variables arc contingent ro understanding dynamics. Ausubcl ( 195.1) found that rhcre was not always reciprocity between till' individual and the group when it came to aL:ccpL;:mcc. The group L:ould accept someone. but no one in the group were 'friends'' with thc.:m. or someone within the group could be an individuals' friend. but the group didn't accept them. This dynamic illustrates the difference between acceptance and friendship. The strength or this almost universal desire to he accepted and admired may reproduce the inequalities in social universes. The reality of reputation may reinforce the norms of the hierarchy as 23


similar statused childr\!n group together and do not want!(' include less popuh1r .:hildren in their play ,group for fear of unpopular contamination. Oftt:n ostracized and rejected children form with children like themselves. unskilled

literature rl'View to provide a background knowh:dgc about some of the causes and problems inherent in the variable trouble with peers. It also provides evi(kncc sh()INing the importanCl' of rejection and peer trouble a" a social \ ariablc. Rejection is prevalent and there is not "imply one type of rejected child. Many variahb. come int(l play. induding thi.!ir skills. their reputation, their behaviors and their look:. or abilities. Even though therl.! is nnt only one type of rejc.:tcd child. there arc 1\.ey behaviors th

Peer Influences on a Child's Social Development The a child socializes with arc a meaningful factor in a child's development. There arc three ways that a child's peer group shapes their heha\ior and t the!'.e children !Synder & Uranowitz, 1978Lln classroom observations it was found that when problem children exhibited pro-social behavior it was ignored by their peers. but when they exhibited deviant behavior their peers focused 26


uron actions (Solomon & Wahkr. 1 '17 ). Rcjcded chi ldrcn ar..: also cxclud..:J h..:caus..: of their peers fear that they will he with the lack of porularity and rejection. Children arc usually aware of the rcrutalion-. and social positions of their peers and arc aware of the potential .,ocial risks that they would he taking if they hefricnu or stanu up for a rejected child . Erwin ( 1993) addressed thrl!e interrelated aspects of social cognitive development. Thl! first named was empathy and role taking which is the ability to understand ohjl!cls and events through other people's eyes. This is needed to relate to Nhcrs feelings and Empathy is needl!d for many differing functions like prediction, sympathy and comliJrt. Secondly. the child's understanding of friendships' inherent rul..:s, obligations and benefits is ne..:declto inform their ability to understand the motivations and actions of friends versus in the context of friendship structures. It can guide their actions in reciprocation and loyalty and can impact their friendship quality. Erwin ( 1993) talks about the significance of whether children attribute their SIICial success to luck or behavior. If they attribute social succco.;s to behavior it will help them evaluate 27


nther;.' actions wdl as their own in a social context and allnws or disallows to be if an ouH;ome is unwanted tErwin, I 941) Many interrelated but distinct sncial skills and aspects pf social competence arc needed f(lr ;.ocial success. It is important to remember that social arc not the S(llc to popularity hut an.: contributors. Putallaz aml Gottman. (I 981 l revcalcd that popular and unpopular childrcn behave dillerently when trying tl) interact or enter into intcra.::tions with peers. They found unpopular children where more disagreeable and less lil..cly to provide good reasons for of their peers. They also found that during entry bids unpopular children were more likely to try to draw attention to themselves. ask informational questions or try to insert thcmsches by talking about their opinions. whereas popular children watL:hed and learned the game first :1nd tried to integrate themselves into the game without these atLention calling strategies. Entry bids art' concre.te behavioral differences between popular and unpopular children. Another behavior difference was found by French and Waas ( 1985) who looked at rejected. neglected, popular and average-status children. measured and grouped them through socil)mcric and had their teachers and parents complete the school behavior checklist and child hchavior checklist respectively. On both of the it rejcctcd children that exhibited more behavior than the other children. Neglected children did not have a higher occurrencc than average status children. This finding points out that rejected and neglected children an; not all the same and their prohlerns with bullying, popularity and friendships arc also varied. The Ro(lls of Social Skills Deficits The root of the child s social problems is another factor needed to be analyzed to add tu a greater understanding of their pt:er rejection. It may be the child does not know when certain behaviors 28


are appropriate. they may lack the skills themschcs. they may lack the ahility to intcrprct others social cues. (1\:lize & Ladd. 1990) or the child might have anxiety or 1llhcr emotional factors that inhibit their usc of appropriate social skills iCox & Gunn. Whatever the cause or causes of the problems, the problems need to he addressed sooner than later hccausc peer rejection itself cause<, more peer adjustment prohlcrns. It can start Vvhen a child is first st:nl away from play and thlir hids for entry are ignort:d by pt:ers. can happen due to overly aggressive bt:haviors or behaviors that arc withdrawn or anxious. Then the child is len playing alone. with children of the same low skill set. or younger children that have not yet acquired the more skills that the older peer needs to learn (Ladd. This deprives the child of the social skills learning they need (Coie, 1990) to be included in their peer group. When these already isolated and deficient peers arc targeted and victimized this creates more resentment and feelings of loneliness. alienation. depression and anxiety. which will further inhibit skilled interactions and damage mcntal health. Finally, thc end result is th:.1t the pea problems become a chronil issue the damaging child's or adolescent's functioning (Coie. 1990). The NeJ;utii'C Spiral and .'ielf.,ystems systems. derived from past experience. help a chilLI understand and manage their social world (Crick & Dodge. 1994 ). Research that these systems arc not changed without because they are rooted in the child's past and social experiences and are how they understand these experiences (Bierman. 2004). point out there are important functions that these systems serve. 29


A well functioning system should (I) give a foundation from which to control and predict the ouhiJe world, !2) he CllJHinuous and keep the sense of self stahle. (3,1 help self-protecti\ln. problem solving. and keeping a favorahle sense of self (Cairns. 1991; Epstein. 1991; Harter. J99R). Selfsystems sene fum:tions that cannPt be ignored hy that have cataloged a lot of poor slleial and negati\e social experiences can bias informatilln processing and foster prophecies as children. adokscems and adults who fear rejection. interpret it before it comes and act in \vays that increase its likelihood (Cairns. 1991: Epstein. 1991: Harter. 19LJ!l). Victimization Children who becomes victims often arc bullied. harassed and intimidated. (Graham & Juvoncn. 199R) as well as gossiped ahout and or socially excluded t Crick & Gr\ltp<:tcr. 1995 ). For the children a "'negative spiral' is created keeping them trapptd in isola1inn and victimization. This pain is damaging emotionally and It must he addressed and for its harmful fnr all involved. Its place in the social emotional context of the adolescent as part of the variable trouble with peers will be analyzed in tht: later stati'>tical analysis. The discussion of peer trouble is long in this literature review because of the !mgt: impact that it has on the lives of auokscents and children. It is hopt:d that by understanding peer troubles in the context of thcother social variables interventions can be designed to help solve the problems and dissolve the pain it causc:s. Mihashi ( 19!l7 J explored bullying and discrimination in Japan, specitically the crudty found in bullying and the isolation it causes. He proposes. '"it could be said that the bully sees in the victim the inferiority that he tears. It is widely known that some take part in bullying out of fear of becoming 30


the next victim"' (Mihashi. I icr, llJ95) plunging them 31


further into isolatiLln whirh can haunt them. and bewme a reoccurring theme in their Being lonely i:-harmful. hut hl'in_g alone is not. Being alonL' cause and pain. hut in some instances it may also be nnrmal. It dear that aloneness hurts in the short term and long term. that social health and interaction is correlated to hettcr functioning and happiness and that good social development and experiences should be pursucd and fostercd. One of the rcasons fL1r this study is to find ways to help mitigate the negativc emotions of socially rejecwd adolescents and to look at social causes of cleprcssion. The section on pecr troubles provides support for the that intaventions arc needed. Blaming Social Victims Amhert 1 IYlJ4) viewed peer abuse as phenomenon thai is sociocullural. and analy 1ed it through the recollections of college It was noted that childhood r..:jection and victimization is usually linked, in the literature reviewed for Amberl"s study. to children's characteristics. possibly coming from their parents background or behavior in most of the typical research they reviewed. This is criticized hy Amhert because shc pointed out that: Children can be victimized lor vcry trivial and random reasons that have absolutely nothing to do with their own personalitics (but may occasionally have something to do with their physical appearance \)f clothing for instance.) It follo\vs that any normal and well-adjusted child can be peer abused. Under such circumstances, studying which childrearing activities by parents contribute to their c.:hild"s victimization might not be as fruitful an approach as it was. The important question may well haw become: what cultural and structural conditions give rise to peer ahuse'! (Ambert. 1994, p. 128). No one deserves to be vk:timized. It is true that often rejection can lead to a negative spiral. It is also true that there are negative traits often exhibited by unpopular and rejected individuals. An unpopular child can have avoidant behaviors. inallentive and immature behaviors, or aggressive and 32


disruptive behaviors to name a few n from unpopular to average or popular, from unaccepted to accepted, from friendless and rejected to included .into a group of friends is a positive change that the rejected child and adolescent yt:arns for. In Kinney's ( 1993) article he looked at when this transformation took place without outside intervention. Kinney interviewed recovning 'nerds'' and documented how they were 33


ahle to overcome the stigma of the lahel.'' !Kinney. 141)3). During the transition from middle to high s..:hool these students were able to gain sdf-contidence and find friendship support. New activities and new enabled them to he inten:sted in new things with new people. Once seeing themselves in a new light they reinvented themselves and left out the negati\c 'trendy .. judgments that hound them to their stigma or simply lkcided not care ahout the label (Kinney. llJ93). lt was this change of self image and the linJing of interests and friends through which nerds gained confiden.:e and This possibility ft)r transformation. in part, drives the hypothesis. Can friendship overcome stigma and labels thrt)ugh giving the individual confidence and enjoyment of socialiLing? Can other social help make a .. nerd'' happy'! lt is also important to address where the problem behaviors of the child are coming from and address these roots from more than one angle if possible. This is especially vital v.:hen a social problem is combined with another pwhlem like ADHD or a conduct disorder. \Vhcn this is the case multifaceted interventions are needed. In the behavioral approach to intervention there is more than one technique to use in reteaching the child or adoll!sccnt. Therl! is shaping. which i:-. modifying the child's behavior through various actions. then: is ITitldeling. which is vicarious or observational learning occurring through watching an adult. <.:hild or film that sho\vs pro-social behavior and its rewards. anJ then: is com:hing, which is the direct teaching and practice of social skills (Erwin. 1993). Through using of these techniques differing successes have been noted. When assessing the needs of a child or adolescent for intervention it is for the intervention to calculate how their friendships as well as general peer status and bullying dynamics impact the <.:hild and his/her needs CBeinnan. 2004 ). In fa<.:t, there are four dimensions of peer relations that a child could he having trouble with and all four should he taken seriously. A child's assessment 34


should include all four of these: (I) how IntH.:h a ..:hild is liked or disliked hy their peers. ( 2) how many Jiiendshi["JS they haq: and their ljUality. (l) social nctVitlrks plal:ement and if thl:y :JSSlll.:iate with troublesome peers, and (.f) if they arc hy peers (Bierman, 2004). During dimension needs to be kept in mind and dealt with diflercntly. Sol:ial factors arc already differentiated when planning interventions. The variety nf factors and stKial skills involved in the social worlds of children and adolescents arc addrcs:-.ed in most but their interactions upon the cmotiom of adolesl:ent as a group or factors arc not yet dcscribl:d. This study will describe the emotional impal:tS and interactions for the academil: and prJl:tical understanding of social variables '"'ithllUt hi as or assumptions about if being popular or having friends are more imptlrtant. or if bullying is the most important social facwr. In movies and books there are many perspectives on popularity and friendship. It would be good to take a statistical look at tlu:se factors. There have heen detailed interventions outlined in hooks to help children ovcr..:ome their social inadequacies. Ther.: arc also studies detailing the placement nf children in cliques and dusters aml the interactions of these groupings. difkrentiation of' having friends and being liked brought clarity to the reputation saddled upon children. It has been shown Ill be annther hurdle unpopular children face. Sncial of l:hildrcn and arc complex and fascinating. They also can hold key factors toward hdping development for rejected peers. Social inclusion for youth is for the social skill development they need for future relationships. It is also part of their self-system The literature review has covered many aspects of the socialization of and adolescents on three main factors: popularity, friendships and bullying. The context of these factors should be taken and applied to the later statistical analysis. 35


Gender a control variahk in the analysis. Gender di!Terences in so..:iali;ing cannot he ign111Td and end up hcing a meaningful variahlc. There much rcsear..:h the diiTcrences between the genders in sncial Grady ( 1979) argues that given the same social situations, expectations and reinforcements the would act n!ry much in the same way. This may or may not he true. but it must he acknowledged that males and fcmaks arc treated differently and often act dillercntl). More often than not males and females arc in differing situations and arc held to differing: f<>r arrropriate or efrectivc behaviors (Erwin, llJ9:\ J. This sets the scene for differing social worlds for males and females. Falling within the sex-role stereotype is aggressive-assertive hehavior. It was found that ten year old boys expected less part:ntal dio.;approval and guilt when being aggressive than girls (Perry, Perry & Weiss. 19R9). Erwin ( 1991) g11es on to add that same-sex peers usually have a greater compatibility of behavioral styks. Another common dillcrencc in the literature between hoys and girls is that girls have more intimalc and exclusive friendships than hoys. Along with this linding is the fact that boys often play in larger groups (13rendt, Since males and 1\:rnales have dissimilar styles of interaction. they also have fricndshir costs and benctits that diller. The support that girls find in their friendships relates to the lower numbers of (in elementary schooi'J and also their reciprocity (Frankel, 19!)0). The uneven number of friendships. where boys have mme friends and girls less. is no longer found in adolescence. hut the kmale friendships are still more intimate than the males friendships tBrendt, 1982). These li"iendship characleristics in the initial crisis of divorce caused differing outcomes for the boys and girls The boys in the study were hettcr able to use their friends as distractions 36


during this lime of turmoil (Wallerstein & Kdly. 1980). These arc gender dillerenct., found in fricnlbhip and popularity stuJi..:s. The gender differences found in co-rumination also well here h..:cau ... c of its possible implications of dillcrenccs in functioning for the genders. lls finding-; fit with 1hc above conclusions about the imrospcclivc styles of l'cmak fri..:ndships that \:an lead to friendships that arc less distracting from problems. Gender is a control in this study. which illlcrplays with emotions and social factors. hut is not the main fi.lcus. 37


CHAPTER 3 METHODS Statistil:al was the mdhod in this study to Jearn ahout social fw.:tors upon adolescent emotions. In the statistical variables were created to measure social components. the independent variables, then their emotional impaetf, and interactions were measured analyzing the dependent variable. adolescents emotions. in an ordinary least squares regression. The National Longitudinal Study of Adolescent Health (Add Health) was used as the data set because of its large sample and broad reaching questions about social contexts of and their emotions. Data The Add Health study is a nationally representative longitudinal study researching adolescents and their health. The study looked into the social conte.xts of the adolescents' lives as well as other facets 111" their hl.!alth aml behaviors. They looked at the people around them in their schools. neighborhoods. communities and their families and friends. The) also measured the behaviors. altitudes. knowledge. emotions and health. In 1994 and 1996 two waves or information gathering were conducted and in 200 I and 2002 there was a third wave to study the effects nr adolescence on early adulthood. The study used an in-school questionnaire for grades 7 through 12. and questionnaires for peers. family members and school staff. In addition, a more in depth in-home questionnaire was administered to approximately 12.000 students. Data from both the in-school and in home questionnaires were used here. The current study used data from wave I only. The 38


yuestionnaires were broad and detailed and enahblthis siUdy' !'.author to create multiple variables to use in the analysis. The sample was pulled from 80 high schools and 52 middle schools. This US sample used systematic methods and implicit stratification to make sure that the sample represented the different regions or the country. the di rrercnt and types nr schools. different and levels of urbanicity (Harris. 2003 ). There were approximately 6,500 in the Add Health data set public usc version and that was what of this study. Variables Dependent Variables The first variables that were created were the dependent variables. These variables measured the emotions of the adlllescent. These variables were created from section I 0 of the survey. titled Feelings Scale. Thi:-. section was given to all the r...:spondcllls and asks ab(\Ut emotions (Udry. 2003). The rcspond...:nts wen: supposc.:d to answer these.: questions while eonsidering how they felt in general within the last we...:k. This reflects the ''c.:motional well-being"' that this study was seeking as a dependent v ariablc. Section I 0 is comprised of nineteen statements. The respondent was given four responses that they could pick for each statement. For example. statement one said. "You were bothL:red by things that usually dun 't bother you'' (Udry. 200_1 ). The respondent could .:heck off the response-;, 'never or rarely. sometimes. a lot of the time. most of the time or all of the time'' (Udry, In the study never or rarely was coded as a 0. sometimes as a I. a lot of the time as a] and most of the time or all of the time as a 3 (Udry. 20m). This made fm a scale that could be added togdher to give a range that 39


the fel'lings of the respondent. The e.xact wording. the scales used in and how statements were coded to creatl' the dependent all feeling-; can he found in the tahk 4. I in Methods of this Facror Loading Since the covered a large of feelings it seemed possible that different statements might be capturing different of the adole<.cents feelings. A factor loading showed differentiation hctween emotions. This distinction between emotions adds depth to the analysis because wh(,)e emotion scale could be used. as well as sub feelings from the scale t'nabling the study to look at differences between types of emotions and social impacts uron them. An initial factor loading different categories. hut overlapping Then a rot:lled compunt:nt matrix was run excluding all variables with factor loadings below .5. In the matrix some of the statements no longer appeared and were dropped from the and no statements appeared in more than one factor. This factor loading put the statements into four distinct fadors. The first factor included the statements about being bothered, feeling the blut:s, feeling depre-.scd, feeling life was a failure. feelinf h)nely. feeling sad, and feeling liiC was not \\Wth living. These were labeled as the exclusively depressed and unhappy variable scale. The factcw included: ICcling just as good as others. feeling hopdul. feeling happy, and enjoying life. These were labeled as the happy and wonhy scale. Factor three included the statements about people being unfriendly. and being disliked. Factor four inducted the statements ahout.being too tired. trouble focusing. and it being hard to do things. The last two factors were not used in t.he analysis. After the factor analysis there were three dependent variables, one that focused on depression. one on happy feelings and onl' including all the feelings. The 40


statements exaet wording. the scale:-. used in and hrm the statements were coded [(I create the dependent happy and depressed fcdings can he found in the 4.2 an 4.3 in the Methm.b section of thi!-> thesis. Independent Variahles There an: many indepemlcnt variables that were created for this analysis. The main independent variables in this analysis arc the variables named male friend contact and fcmak friend contact. Friend contact, is a component of friendship and is being used here as a variable that represents having a friend and having contact with that friend. Section 20 measures friends and their level of contact ( L'dry. 2003 ). In this section all of the respondents were asked to name their first or hest friend. one of each sex. Then they were askecl five 4uesti(1ns about contact. Partial selections of the rcsponclents were asked 111 list their first five friends in each gender group and then outline their contact. The group that answered ahout all live friends was not because using the group that listed all their friends would dimini:-.h the sample size. Creating variables representing the having one or two best or main friends and the contact with them, rather than a variable representing their overall friendship network and social activity simplifies the analysis and keeps the sample size large. At first. friend contact with friends or both genders combined and used for analysis, but it was found later that ditlerent produced different results; so they were used separately in the 41


The Cremion of Friend1hip Conracr Variahle.1 To create the friendship contact variables the live yes or no qucstiom about cnntact time \verc totaled. The higher the total meant more contact. Ira n.:spondent a t.en1 it meant either they had no contok at their frequencies you can see the genders are similar in distribution and when combined they have a relatively standard normal distribution. This can be seen in ligures and 42


Male Friendship Contact 1.50 1. 0.00 2.00 4.00 600 Figure 3. I The Distribution of the Variable Male Friendship Contact 43 Mean =2.51 Std Dev. =1 .777 N =6.498


Female Friendship Contact 0.00 2.00 4.00 6.00 Mean =2.58 Std. Dev. =1.81 N =6.496 Figure 1.2 The Distribution of the Variabk Female Friendship Contact 44


Male and Female Friendship Contact 0.00 3.00 6.00 9.00 12.00 Mean =5 10 S1d. Dev. =2.795 N =6.493 Figurt: Distribution or the Variables Male and Female ContaL"t Combined 45


The Creation o{Gcndcr and Other Sociodl'l1/uf!.mJ)hic Variahlcs The next variahk created controlled for gender. wac codcu as zcw and lcmales wen: coded l. Next. race was made into a control variable. A respondl'nt was c:lassilicd as Hispanic, Asian. African American, Native White or Other. Family structure, categorized as living with hnth biological parenh. living with your biological mother and a father figure. living with your biological mother only and other family structures. as well as parental educational allainment were created control variables. The parcmal educational auainment variable was created using the highest education obtained hy the respondents' There were three categories, than a high school diploma or GED. a high school diploma or GED and more than a GED or high school diploma. The Cremion ofSuhstance Use, Delinquency, Grades a/UI Parental Closeness Variahles The usc or substances was eon trolled for, using three variables: one that measured the amounts of eigarl!ltcs smoked. the second measuring the amount of alcohol consumed and the third measuring the amount of drugs tried. controlled for by totaling the amount of .. delinquent" aetivities participated in and how often. from seetion 29 called the Delinquency scale ( Udry. 2003\. An average of the most recent grades became another comrol variable. Lastly. parental closeness wa:-. controlk'd for. because relationships with parents obviously impact adolescents' emotional states. It is informative to know the impact of parent closeness a5. compared to diff'crelll social variables. Parental doseness was created from section 16. Relations with Parents (Udry, 2003). During the creation of this variable the fact that that many respondents come from dissimilar family structures had to he taken into account when creating an parental closeness variable. So, whichever questions the respondent answered (answering for the father figure or mother figure) 46


became default scon: and if both parental ligures had scores the highest useu. In this way, the tonk intn account the impact of family structure once. anu not again through a different unconnectcu variable, parental The Creation of the School Trouhle \iariahle Through using section five. labeled Academics and Education tUdry 2003). an overall picture or the school cornpt,nent or the adolescents' world was created. The questions all were prefaced with ''how often did you have trouble ... : The referred to "getting along with your teachers'!" The second asked about, 'paying attention in school'!'' The last. was about ''getting your homework done'!" There was a scale of zero to four with zero designating the frequency of never and with four designating the frequency of "everyday'' (Udry 20fn ). TIKsc questions created the variable measuring overall schooltrouhlc. The Creotion qf the Other Social Variahles The next variables were created. like the friendship contact, to he analyzed more intensively to look for their interactions as social variables and impact with emotion!>. Social context is important in adolescents' lives. There an: many components that comprise a social world. For this study four social'' variables were created. The amount that the adolescents think their friends carl! about them is a social variable. This variable comes from section Protective Factors. The question asks. "How much do your friends care about you'!'' !Udry. 2003). This variahlc is closely related to, if nut defined as friendship quality. 47


The discussion of friendshir quality in the literatun: review demonstrates that there arc many conceptions tlf friendshir quality and how to measure ittGI!orge & Hartman. 19<.)6). Believing that your lril!nds care about you wi II represent friendship quality and friendship affection in this analysis. It should he remembered that this variable is created from a question that asks the respondents to answer using their pen.:cption. It is not a variable created by outside observation. It also docs not usc more than one component of the concept friendship quality in its measurement. A variable was created to measure the impact of having your lirst named friends at school with you. This was using the section on friends, where r0spondcms arc askeJ if their lirst listed fri0nd went to their school. Does having a close ally at school make an impact on well heing'! This variable was included because of the literature (George & Hartman. 19l)6) that found socially unpopular adokscents and chi ldrcn often friends with children outside of school. or that were in a ditlacnt age group. The next variable i.., this study's 'pupular .. variable whilh is represented as healthy socialization and indusion. This variable measures the rcspnndenb' happiness with and their social inclusion at school. The three statements comprising this variahlc arc. I. You reel close to people at your school. 2. You fed like you arc a part of your school. You arc happy to he at your school (Udry. 200J). This variahlc represents popularity through inclusion and exclusion. Those that arc in12luded arc in till: in-group, the popular group. and those not included arc in the out group. Th0 last so..:ial variabl0 is concerned about trouble getting. along with other It was created from one question. The question asks how often they have trouble with other students (Udry. 200J). This variable represents. in this analysis. the occurrence uf hullying and victimization in the adolescents' so<.:ial woriJs or at kast "trouble .. or a disruption with their peers. These variahles were analyzed using an ordinary least squares regression. The details on the creation of the independent 48


variable:-rlpre:-.enting friendship qualit). if your lirst friends are at popularity and troubh: with peers can be found in tahles 4.6. 4.7. 'US and 4.lJ in the l'vlethods Sl!Cti1111 nf this thesis. Brief Description of the Analysis 1\-Icthods Ordinary least squares regressions arc U'-l!d to evaluate the association of friendship contact with adolescents' emotional adjustment. controlling for the other faLtors described above. Control variables arc added in conceptually distind hloch in ordl!r to determine whether accounting for a given concept attenuates the apparent association hetween friendship contact and adjustment. To look at the interplay and changes or the impacts. seven blocks. or groupings of variables v.ere crl'atcd. Block onl' included male friend contact. female friend cor:nact and the gender of the respondtnt. Block two included race, parental educational attainment and structure. Block three included the amount of cigardte;. consumed. the amount of alcohol consumed and the amount of drugs tried. Block four includes the total of delinquent behavior;.. respondents' total schooltrouhle, how close they arc to a parent and their grades. Block five included friendship quality (friends caring) and if your liiends go to your school. Block six was composed of popularity. The last block. block scvl!n. includes trouhh: getting along with other studl!nts. Three th!pendent variables were used in this analysis and with each dependent variable all seven blocks described above were added in the same order to look for whether changes occurred when analyzing the cfl'ects of social upon dirlcrent types or emotions. All together there are twenty-one output models to annlyze. 49


CHAPTER 4 HYPOTHESIS AND RESULTS Hypothesis The hypothesis of this study is that the emotional benefits of friendship is as strong as the other benclits or wsts of other social like being popular or having troubles with rcers. Therefore friendship is a beneficial emotillllal intlucn.:e in an adolescent's life that can mitigate or lessen the other possible negative social variables. This is tested using a ordinary least squ:ucs regression with emotions as the dependent variables and social components as the independent variahles to sec their effects. the size of their effects and hov.' their ctTccts intera.:ted with each other. The hyrothcsis relics on the litnature that investigated the importan.:e of friendship to adole:-.ccnts and children and the impacts of bullying and popularity. Specilically. the functions of friendship like support. allegiance. help, allcction and skill development. make fril.!ndships a beneficial .:omponent in adolescent social life (Georg.: & H:lrlrnan. 1996 ). It was also found that friendship and relations were linked to children s understanding of their adjustment during.carly auolesccnc.: anu (Gaute et al.. 1996) establishing a close fri.:ndship was found to be mon.: important during early adoksccncc than younger ages (Buhrmestcr. 1990). It is these components that inform the hypothesis which asserts friendships in a child's life were as beneficial as other social in impacting emotions. 50


The Dependent Yariahks Coding and Interpretations In the analysis then.: are a total of three dependent variables. Each dependent variable represents a di llerent emotion and with L'a-=h erntion the same seven blocks of variables were added in the same order to look for effects. First contact, then demographic attributes. illicit behaviors, school adjustment, friendship quality. status at school and1inally, peer relational adjustment arc added and tested. In the results, the dependent variables were coded so that the higher or the more positive a number an independent variable received meant the more it was contributing to that feeling. For example. the first seven models analyzed with the dependent variable named all feelings.'' All feelings was created from all of the in the section on feelings. More of tht: :-tatements were worded in a deprcssing or unhappy way, so all feelings is a coded so that a higher represents lower .::motional adjustment. The questions th:tt were positivdy worded Wi!re ri!coded to reflect a negative'' inllucnce on emotions. It is important when viewing the results to not he confused hy language. Thl!se negative" emotions will he visually represented as numerically positive on the all feelings scak and will be analyted as 'detrimental." lf the results show a low or a negative numerical score it is good or 'beneficial" impacts. For the second dependent variable. happy feelings, the original code was used signaling a higher or more positi' e numher repn:sents more happy fet:lings. lnstead of calling these impacts 'positive they will he called 'beneficial' and the lack of these feelings will again be called '"detrimental.'' The last dependent variable focusing on depression or "negative' feelings is like both of the others. A higher mort: positive score signals more contribution to the variable feelings. in this 51


depression. Therefore when using feelings as a tkpendent variable a higher number contributing to more depression. The language usr.:J in these dcscriptiDns docs not fully describe the measurements used in analysis. It is important to look at the qucstitlns used to create thr.: variahks and thr.: follov.'ing discussion about their full m...:aning. This is true not only for the dependent variables. hut also for th...: independent The labels of "bencfit'ial' and 'detrimental'' arc beuer understood in of the created variables. but the variables themselves should he studied to bclter understand whal 'bcnelicial ''and 'detrimental" stands for in each of the models. Visual Representations of the Three Dependent and Six Social (Independent) Variables Below are visual representations of thr.: three d...:pcndt:nt \'ariahles and the six social variahles in tables 4.1. 4.2. 4.3. 4.4. 4.5 and 4.6. (Not all of the control variables' creation tahks wert: included for brevity. Sine...: they are not thr.: focus. thr.:y are not Jiscuss...:d in kngth.J At the end of these tables the descriptiv...: statistics of all or variables of the study are induded. The wording in tables 4.14.6 all come from thr.: Add Hr.:alth survey (Udry, 1943 ). 52


Tahk -1-.1 A table shnwing the ..:reatinn and coding of the c.h:pcndent variable all feelings Variable name A.ll.f('c/inR.I (Dependent Variable# I J A list of the questions/statements composing the variable prefaced with. 'These questions will ask about ho\V you feel emotionally and about how you feel in generaL How often wa:-. each of the following things Lrlll: c.Juri ng the past week?"" I. You were bothered by rhinR.\ thm u.1uallr don't hot her _I"U. 2. }'ou didn't j(;ellike eatinR. your appetite um poor. 3. You felt that you could not shake ojfthe blues. even H"ith hdp jinm yourjiunilr and friends. 4. You felt that you 1rere just as good os other people. 5. You fwd tmuhle keeping your mind on what you were doing. 6. You jdt depressed. 7. You felt that you were roo tirl:'d to do things. 8. You felt hopejitl about the future. 9. You tlwuRht your/if fwd he en a failure. I 0. You felt fewjitl. II. You happy. 12. You tulkl!d les.1 than u.mal. 13. You felt lonely. 14. People were unji"imdly to you. 15. You enjoyed lij(. 16. You jdt sud. 17. You felt tina people disliked you. 18. It m1.1 hard to get swrteJ doiiiR tlti11g.1. 19. You jdt life 1ras not worth living. 53 Rec:nded ( rerersed) or original coding Original scale (0 ne1er or rarely, I wmetime.1. 2 u lot of the time. 3 most or all of the timl') Original Original Recod'J reversed (0 most or all of the time. 1 u lot of the tinw. 2 sometime.\, 3 11ever or rarely! Original Original Origi11al Remded Origi11ol Origi11ol Recoded Origi11al Original Origi11al Recodl'd Origi11ol Origi/1(1! Origi11al Orightal


Tahle 4.2 A table shlnving the ereatinn and coding or the dcpemknt variahk happy feelings Variahlc name Happy feelings I Dependent variable #2) AI of composing the variahlc with. 'These questions will ask about how you ll:cl emotionally and ahout ho\v you fcd in general. How often \vas each of the following things true Juring the past week?'" I. You felt that you were just as good as other p:orlc. 2. You felt hopeful about the future. 3. You were happy. 4. You enjoyed life. Recoded t n:vcrsed J or original coding Original scale 10 never or rarely. I sometimes, 2 a lot of the time, 3 most or all of the time) Original Original Original Table 4.3 A tahle showing the creation and wding of the dependent variahle depressed kclings Variable name A list of questions/statements cornpusing the variahle prefaced with. questions will ask how you feel emotionally and ahout how you feel in general. How often was each or the following things true Juring the past week"!"' Depn.:ssed feelings I. You \vere bothered by things that usually don't bother (Dependent you. variahle #3) 2. You felt that you could not shake otT the blues. even with help from your family and friends. 3. You felt depressed. 4. You thought your life had been a failure. 5. You felt lonely. 6. You felt j;ad. 7. You fdtlifc was not worth living. 54 Rce()(kd (reversed) or original coding Original scale (0 never or rardy. I sometimes. 2 a lot of the time .1 most or all of the time) Original Original Original Original Original Original


Tahk 4.4 A table showing the cn::atinn and coding of the independent variable mak friendship contact Variable nam.: Male friendship Contact (Independent) ''' 7 was put into the 0 category becaus.: it meant that the respondent did not hang out with that friend for a kgit rea ... on(like th.:y didn't name one and that wa-.; an important possibility to measurethe or the friendship) A list of composing the variable prefaced with. 'The next 4uestions arc about your I will ask you for their names in ord.:r to k.:.:p track or th.:rn but their names wi II be erased from the computer as soon as we finish this section or the interview. These statement<> were for the listed 'FIRST or ONLY l\1ALE FRIEND .. I. Did you go to (NAMES} s house during the past seven days'! 2. Did you meet (NAME} after school to hang out or go somewhae during the past scvcn days'! .1.. Did you spend time with I NAME l during the past wcckend'! 4. Did you talk to I NAME) about a problem Juring the past (]Jys? 5. Did you talk to [NAME} nn the telephone nnthe past seven days'! 55 Recoded ( rc versed) or original coding Original code (0 no, I yes. 6 refuscu. 7 kgitima.te skip)* Note in variahle column Original Original Original Original


Tahk 4.) A table showinf the creation and coding of the indepcmknt variable female friendship contact Variable name Female friendship Contact (independent) A list of ..:ornposing the variable prefaced with. 'The next questions arc about your lricnds. I will ask you for their names in order to keep track or them but their names will he erased from the computer as soon as we finish this se<.:tion or the interview. These statements were for the listed FIRST or ONLY FEI'v1ALE FRIEND .. I. Did you go tu [NAMES} 's house during the past seven Rewded (rt.:versedl or original ..:oding Original ..:ode (0 no, I yes. h refused. 7 legitimate skip)* Same as in table 4.4 7=0 2. Diu you meet {NAME l afler sdwol to hang nut or Original go somewhere during the past seven Jays'.' l. Diu you spend time with {NAME} during the past Original weekend? 4. Did you talk to {NAME} about a prohkrn during the Original past scvt:n day:-? 5. Did you talk to {NAME} on the telephone on the Original past seven days? 56


Table 4.6 A table the .:reatilln and coding or thl! inlkpendent variable friends caring Variable name Friends caring ahoutyou (! ndependent) A list or composing the variable prefaced with. This section ask[ s l questions about the support you re..:cive from the people around you.'' I. How much do you feel that your liiends care about you? Rl'COl.kd (rewrsed) or original coding Original code (I not at all. 2 very little. 3 sornev.hat. 4 quite a bit. 5 very much) Table 4.7 A table showing the creation and coding of the independent variable first listed friends at school Variable name A list of composing the \ari:J.hle prefaced with. 'The next questions arc about your I will ask you for their names in order to keep track of them hut their \viii be erased from the computer as soon as we finish section of the intervJe\v." These statements were for the list.:d "r:IRST or ONLy MALE/ f-EMALE FRIEND First Friends (Male 1. Does NAME} go to your sLhonl? and Female) at School (Independent) 57 Recoded (reversed lor original coding Original code (0 no, I yes. 7 legitimate skip)


Tabk: 4.R A table sh1l\ving the l:f(!atinn and c:oding of the indept:ndcnt variabk popularity Variable mum; Social Status (Popularity) ((ndepencknt) A list of questions/statt:nh..:nts composing the variable with. 'The questions that follow ask about your educational How much do you agree or disagree with the following statcm(!nts: 1. You feel close to people at your s.:hool. fl f summer] year, you tdt dose to people at your 2. You ft:cllike you arc a part of your [If summer] Last yt:ar, yllu felt like you are a part of your school. :;. You are happy to be at your school. [If mer] Last year, you v.ere happy to _he at your school. Recoded (reversed) or original coding Recodedreversed (4 :-.trongly agree. 3 agree. 2 neither agree or disagn.!e, I disagree) Recodedreversed Rccodcdreversed Table 4.9 A table :-.bowing the creation and coding of the independent variable trouble with peers Variable name A li:-.t of questions/statements cornptlSing the variahk prefaced with, '"Sin.:e sdmol started this year. how often have you had trouble: fl r summer] During the 1994-1995 school year. how often did you have trouble:'' Trouble with Peers I. Getting along with other students'! (Bullvin!.! or isolation) ( 58 Recodcd (reversed) or original ending Original scale lO never, I just a few times. 2 about once a week, 3 almost everyday. 4 everyday)


Descriptive Tahle 4.10 A table (\f the descriptive stati..,tics for all the variables used in analysis N l'vlinimum Maximum Mean Depmdent Variables 5.)46 .on 50.00 I0.7n7 All feeling!'. Happy 5346 .00 12.00 8.20 I g DepressL'U feelings S::l4ti .00 21.00 2.6g39 lndepenJent \/ariahies f'.-1ale Friendship Contact 5346 .00 5.00 2.5417 Female Contact 5346 .00 5.00 2.64g.s Gender (!'v1alc =0. Female =I ) .00 1.00 .5215 Race 1ras com rolled jin hy purring tlte responde/Its into six racial categories. ( \Vhite was the excluded l'ariable. i Hispanic 5346 .00 1.00 .1135 Asian 5346 .00 1.00 African American 5346 .00 1.00 .181g Native American 5346 .00 1.00 .0204 59


Tahle 4.10 (Con"t.J Otha Family structure Has controlledfor hr m!ling the rt'SIIOI!dents inro four lwnilr .\tmcturc cmcgorics. ( Li1ing With Biological Mother and Biological Farha \HIS the excludt'fl variable.) Living With Other Family Structure Living \Vith Biological i\:1(lther and Father Figure Living With Biological !\')other Only Parental Educmion Alfainmcnr IW/.1 contrr,lled j()/' by pulling the re.1pondcnts inro three parental education altainmellt categories. (Parental Education at the High School Level Ha.\ the excluded l'ariahle. J Parental Education at the Less Than High School Level Parental Educational the More Than High School Levl'l N Minimum Maximum !\-lean 5346 .00 1.00 .0077 5346 .00 1.00 .1057 5346 .00 1.00 .1324 5346 .00 I .00 .2000 5346 .00 1.00 .1122 5346 .00 1.00 .5041 60


Tahk 4.10 !Con"L) N rvtinirnurn !vtaximurn Mean Controls/(,,. Suhstancc Abuse 5346 .00 97.00 6.6160 Total Usc Tnlal Alcohol Usc 5346 .00 Y5.00 4.1453 Total or Drug Tries 5346 .00 5.00 .4300 Other Controls Tol

IJCfJendent Varioh/e Crcmion Di.\cu.\.\ion Tables .:1.1 to -1.9 show the exact vvording and coding that created the variables. The dependent variables measure differing emotions that arc all impPrta111 to the \veil being of an adolesccnh emotional life. It is an nvcrsimplilication to say that happy feelings only measures happiness because when looking at the qucstinm that created it there is more being measured. It measures happiness along with self esteem. hope and enjoyment of life. All of these can be categorized as needed for good emotional adjustment, hut r.:annot only be seen as happiness even though that is the given variable name. This is also true for the dependent variable named depression. It measures depression along vvith irritation. feelings of failure at life, sadness. and feelings or life not being worth living. All of the separate questions/statements measuring the fet:lings contribute w the respondents" score on the scales equally. As explained earlier lhc words hencricial and to be added and the main variables to he u'ed in the analysis are male and fernak friendship contact. How they were created was discussed in the methods section and visually represented in the tables .:1.4 to 4.S. The variables were created as friendship contact scores representing having a friend and spending with that friend. 62


This variabk having and time with a friend without measuring the friendship's quality. This important hLc:ausc the hypothesis asserted that having a friend irsclf is a gPod thing and did not make the quality of that friendship a requirement. That being many of the components making up friendship quality arc the n:asons why friendships make feel good. To measure and account for this impm:t friends c:aring ahout you was UM.!U as a measure of friendship quality. Possibly these two variables could have been combined. hut it added more depth to the analysis to look at them scparatdy because with the separation the study analyzing friendship existence and time spent together. and li"icndship quality as separate variables. The rest of the sol:ial \ariables and tlll'ir creation is self evident from the Methods section. table and labels. It should he noted that popularity and trouble with peers (bullying and were created from the questions available and were made as imperfect representations that are general. hut together can account for the concepts covered in the literature review. Interpretation of the Results ;Hodds Usinx Dependent Variable One (All Falings) Tables 4.11 to .f../7 In table 4.1 I. the loeflkicnts indicate that an adolescent \.vho had t:ontact \.vith their lirst listed male friend during the last week has a predicted emotional adjustment score .207 higher than an adolescent that did not have that contact. This means that male friendship contact is associated with poorer emotional adjustment using the all feelings scale. This effect is significant al the .00 I level which means that with a 9t).t)<;;. confidence level it can he stated that this result did nnt occur at random 63


and reflel'l-; a true emotional hetween friemhhip and poor emotional A similar ellecl was found for female friendship contact with a slightly lower significam.:e l-.!velof p<.O I. It \\as surprising tn see th

Table 4.15 saw thl.! addition of thl.! first two social variabll.!s. \-vhich rcprl.!sent fril.!ndship quality or aflix:tinn and having at sch110I."" These arc predictably cmotinnally benclicial variables. When adtkd they both had significantly bcnl.!ficial effects. Friends caring was hend"icial to an adnlcscent" s emotions at thl.! p<.OO 1 lcvd and having first friends at school was beneficial at thl! p<.05 level. With thl.!sc additions thl! dTects of friendship contact werl! again significantly positive (detrimental) at the p<.05 level for male friendship contact and at the ["J<.O I level for !Cmalc contact meaning that friendship quality and having. friendships at schnnl arc benl.!ficial componl.!nts 1\.)und in friendship n'ntact. Adding. pupularity in table 4.16 brings about a divergence in the effect of friendship contact by the friend s gender. In table 4.16 the addition of popularity. causes male friendship ..:on tact's detrimental effects to decrease by .003. but female friendship contact"s detrimental effects to increase by .028 and change frum a p<.O 1 to a p<.OO I significance kvel indicating that social popularity is a large bl.!ncficial Ctlmponent in females friendship and wntact time, hut not for male friendship contact time. The last variable to be addl.!d. troubll.! with pcl.!rs (table 4.17). also had difTI.!r..:nt cffl.!cts hctwl.!cn thl.! genders. With its aduition male friendship contact had its detrimentall.!ffccts incn;asl.! from .119 to .142, as female friendship contact slightly decn.:ased its detrimental impact from .194 to .191. This indicates that like popularity for female fricmlship contact. trouble with peers is a large beneficial component of male friendship contact. In the end, in tabh! sl'vcn. both friendship contacts produce detrimental cf!Ccts to a signi ficam:l.! or p<.05 upon the emotions or the adoles..:ent meaning that with a 95t;;(cnntidcnee it can he predicted using thl' all feelings dependent variable that more friendship ..:ontact will cause poorer emotional adjustment even when controlling for sociodemographic and other social factors. The possihle interpretations of these results will be found in the Discussion section. 65


The Data Tahh:s for the First Dependent Yariahlc Tahle 4.11 Ordinary results predicting adjustment as a function nf and gender with unstandardized and standardized using all feelings as the variahh: Variables Unstandardizcd Coefficients Std. Error Standardized Coefficients Male Friendship Contact 0.062 0.0497 Female Friendship Contact 0.19R*':' 0.06:' 0.04R Gender 1.628*'!'* 0.216 0.1 I 0 (Male =0. Female =I) *r<.05. **p<.O I. ***p<.OO I Tahlc 4.12 Ordinary least squares regression results predicting adolcsl:ent emotional adjustment as a function of friendship contact, gender. ral:e. family structure. and parental education attainment with unstandardized and standardized coefficients using all feelings as the dependent variable Variables Male Friendship Contact Female Friendship Contact Gender 1 Male =0. Female = 1) Race tms com rolled for br putting the respondents info six racial categories. ( H-'hirc was thl:' excluded l'(lriab!t. J Hispanic lJ nstandari zed Std. Coefficients Coefticients Error 0.062 0.062 0.061 0.063 1.561 *** 0.206 0.106 1.52*** 0.393 0.065 66


Tahk 4.12 1Con'1.J Variahks Asian Al'ri..:an American Native American Other Family structure control/cdfor hr putting the responde111s intu.f(wrfamilr structure categories. ( Lil'illg With Biologicaii'v!other rwd Biological Father H'as the excluded rariable.) Living With Other family Structure Living \Vith Biolugical Mother and Father Figure Living With Biological 1\lother Only Parental Education Attainment was controlled .fiJr hy f111Tting the rcspondellts into three parental educarinn arrainment categories. (Parental Education at the high school 1nu the excluded rariahle.) Pan:nt::tl Edul:ation at Lhl! Less Than High Sr.;hl)nl Level Parental Education at the More Than High School Level *r<.05, **p<.O I, '"** p<.OO I lJnstandarizcJ Std. Standari7cd Coefficients Cnt.:llicit:nts Error .L521':":'* 0.64 0.095 1.2R5*** 0.315 0.067 U9 0.879 0.027 -1.459 1.004 -0.017 0.42lJ O.R72** 0.3-+lJ 0.04 0.086 I .689'''** !U56 0.072 -0.101 67


Tahk 4.13 Ordinary least squarts regression pn:dit:Ling emotional adju:-.Lment a;. a funt:tion of friemhhip wntat:l. gender. ran:. family strut:Lurc. parental education attainmclll and suhstan(;e vvith unstandardized and standardized (;Oeffit:icnts all as the dependent variable Variables Male Friendship Contat:t Female Friendship Contact Gender (Mail: =0. Female= I) Race 1vus controlled jin by puttinR the respondents into six racial cmegories. (White H'os the excluded variable.) Asian African Amnican Native Amerit:an Other Fwnil\' stmcture 1ras controlled for hy purring the respondent.\ illto .Jilltrfamily structure categories. ( Liting With Biological lvlother wtd BioloRical Fathr:r 11us the excluded variable.) Living With Other family Strut:lllre Living With Biological Mother and rather Figure Living With Biological Mother Only Unstandarized Coc1Ticicnh 0.102 0.104 1.807"** 1.603**''' 3.752'"** I.IJ2.'F** I.:B8 -1.012 0.417 1.101"** 68 Std. Stwulardi::ed Coejjlcients Errors 0.058 0.025 ().()58 0.025 0.206 0.122 0.388 0.06') 0.611 0.101 0.311 0.100 0.811 0.03 0.978 -0.0 II 0.405 0.057 fU67 0.019 0.276 0.06


Tahlt.: !Con't) Variables Parentall:'ducation AttainmPnt IH/.1 cnntrnlfedfor hy putting the respo11dents into three twrental education attainment cotegories. ( Purmtul Education at the hiph school /nelwas the exduded l'llriab/e.) Parental Education at the Than High School Level Pan.:ntal Education at More Than High School Level Total Cigaretle Use Total Alcohol Use Total of Drug Tries *''-p<.O I. '''**p<.OO I Unstandarizcd Cudlicients 1.653*'''* o.

Tahk 4.14 tCon"l.) Variahlt.:s U nstandarizl!d Stu. Standardized Coefficients Cnd"ficil!nts Errors Hispanic (1.336 0.059 Asian O.:'i72 0.09 Afiican American 1.54-V'<'' 0.10H O.OX I Native Aml'rican 0.355 0.7()9 0.007 Other -0.956 1.065 -0.012 Family structure was cnntrolledfor by putting the respondents into jimrj(lmi!y .\trurture categories. ( Liring With Hio!ogical Mother and Hinlogical Father was the excluded I'C1riable.) Living /ith Othl!r Family Structurl' 0.7.<7 0.37:'i 0.031 Li\ing With Biological l'vtother and Father 0.232 OJ> II Figurl! Living With Biological Mother Only o.o06':' 0.26 o.o:n Parental Education Attainment was com rolled j{1r by putting the respondents imo three parental edtl("ation attainment categories. (Parental Education at the high school/ere/ was the excluded rariahle .. J Parental Education at the Less Than High 1.41 *''":' 0.319 0.06 School Level Parental Education at the Morl! Than High -1.202*** 0.196 School Level Total Cigarcllc Usc (l.02* 0.009 ().037 Total Alcohol Usc 0.002 O.OIH ().002 Total of Drug Tries fU56* O.IH 0.04 70


Tahle 4.1-+ !Con.l.) Variabks Total 0f Delinquenl Aclivitics Total School Trouble An Average of the Rcl:cnt Grades The Closeness the Respondent Feels to a Parent *'''p<.O I, ''"''*p<.OOI LJ nstandarizl!d 0.134'''*''' 0.6g I''":":' -0.7H7''"''' -1.282*':"' Std. Standardized Errors 0.03 0.09 0.047 0.219 0.138 -0.08 O.IIS -0.169 Table 4.15 Ordinary least squares regression results predicting adolesccnl emotional adjustment as a function of friendship contact. gender, race. family structure. parental education atlai nment. suh&tance use. dt!linquency. school trouble. grades, parental closeness. friends caririg about you and first friends at school with unstandardized and standardized coefticients u&i ng all feelings as the dependent variable Male Friendship Contact Female Friendship Contact Gender (Male =0. Female= I) Race 1\"U.I controlled jin by putting the re.1pondents imo six racial categories. (White 1m.1 the excluded variable.) Hispanic Asian African American Native American Other Unstandarized CoetTicil!nls 0.122':' 1.248''"'":' 3.141 '''** 1.144'''** 0.244 -1.039 71 Coefficients 0.058 0.029 0.059 0.041 0.219 0.177 0.328 0.054 0.549 0.085 0.298 0.06 0.783 0.005 1.1 -0.0 I 2


Tahle .:.1-.15 ICon't.) Variahks Unstandaritcd Std. Standardized Coefficients Coc Erwrs Family .1tmcturc was contro!!edfor bY p11tring the respondents into ji!llrfamily structure catl!gorics. ( Li1ing With Biological Mother and Bioloximl Father was the excluded vuriab/1!.) Living With Other f-amily Structure 0.672 fU72 0.028 Living With Biologil:al Mother and f-ather 0.118 0.341 0.005 Figure Living With Hiologil:al Mother Only 0.608* 0.255 o.crn Parental Education Attainment 11as comrolled./(1r by putting the re11wndents i11to three parental education c1ttainment catexories. (Parental Edutarion at the hiKh school/eve! was the excluded mriable.) Parental Education at the Less Than High 1.221*'''* 0.126 0.052 School Level Parental EJucation at the More Than High -1.117'''** 0.188 -0.076 School Level Total Cigarette Use 0.021 0.009 0.039 Tlllal Alcohol Usc 0.002 0.017 0.002 Total or Drug Tries 0.178 0.037 Total of Delinquent Activities 0. I 2fl'''** 0.03I 0.084 Total School Trouble 0.657*** o.o4g 0.2 I 1 An A vcrage or the Most Recent Grades -0.7IY''* O.I4 -0.073 The Closeness tht' RcsponJem Feels to a -1.14!P** 0.122 -0.151 Parent 72


Tahk4.15 (Con't.) variabks Std. S tandarJ i zed ITicit:nts Error CnctTicit:nts Friends Caring Ahout You -1.177':'** 0.161 -0.124 First Friends (Male and Female) at Schnol -0.426** 0.149 -fl.041 *p<.05. *'''p<.O I, ***p<.OO I Table 4.16 Ordinary least squares regression results predicting adolescent emotional adjustment as a function of friendship contact. gender, race. family structure, parental education allainmcnt. substance use. delinquency, school trouble. grades, parental closeness. friend1caring about you, first friends at school and popularity with unstandardized and standardized coefticients using all feelings as the dependent variable Variables Male Friendship Contact Femak Friendship Contact Gender (JI.-1ale =0. =I) Race 11as colltrolledfi>r by purring the respondent.\ i11to six racial cuTefr hy purTillg rhe respondents imo jintrfwnily .\tructure categories. ( Li1inx With Bioloxical Mother and Biulogicul Father was The excl11ded mria.hle. J 73 Unstandardized CnctTicicnts 0.119* 0.194*** 2.421 *** 1.337*** 3.201 *** l.lt13*** 0.404 -1.05 Std. Standardized Error Cocfficil!nts 0.05X 0.029 0.057 0.048 0.216 O.IM 0.327 0.057 0.52 O.OH6 0.291 0.061 0.761 O.OOX 1.112 -0.012


Tahle 4.16 !Con't.) Variahles Living With Other family Structure Living With Biological Mllther and Father Figure Li' ing With BiologiL:al Motha Only Parental Education Anainnum was controlled for hy puttin:;; the respondents inro three parental education allainmelll categories. (Parental Educatirm at the high .1clwol l{'le/11as the excluded variable. Parental Education at the Than High School Llvd Parental Education at the More Than High School Level Total Cigarette Use Total Alcohol Use Total of Drug Tries Total of Delinquent Activities Total School Trouble An Average or the Most Recent The the Respondent Feels to a Parent Friends Caring About You First Friends (Male and Female) at School Popularity *p<.05. **p<.O I, ***p<.OO I 74 linstandardized Coel'f'icienb 0.615 0.074 0.561 I.IR5*** --1.124**'1 0.016 -0.002 0.23 0.124**''' 0.551*** -0.61**: -1.026*''"'' -0.971**''' -0.231 -0.42g*''"'' Std. Standardi1.ed Error Coefficients {).026 tu.n 0.0(13 0.24H 0.010 0.321 0.051 0.192 -0.076 O.OOlJ O.

Tank 4. I 7 Ordinary least results predicting ernPtinnal adjustment as a function of friembhip contact. gender. race, family structure. parental education attainment. substance usc, dclinquerh:y, school trouble, grades. parental fril'nds caring about you. first friends at school. popularity and trouble with peers with and stanuardizcu coefficients using all feelings as the ucpendcnt variahk Variabk:s Mall' Friendship Contact Female Friendship Contact Gcnucr (Malt.: =0. Female= I) Rua Hus controllcdfor by puttinM the respondents into six racial categories. (White H'US the excluded l'(lriablc.) Hispanic Asian African American Native American Other Family structure was comrolfedjiJr by putting the respondent\ into fourfamily structure categoric.\. ( Liring With Biological Mother and Biological Father lVG.\ the excluded l'ariahle.) Living With Other family Structure Living With Biological Mother and father figure Living With Biologi..:al Mother Only Parental Education Atwinm('llt \\'as controlled for hy putting the respondents into three parental education attainment categories. ( Parenllll Edumtion at tile high school levelli'U.\ the excluded variable.) U ted cod"licients 0.142':' 0.19 I ': : :oo:o 1.537*':'* 3.272''*':' 1.156''*''' 0.304 -1.103 0.496 -0.01 0.494* 75 Std. Standardized coefficients Errors 0.(156 0.057 0.047 0.217 0.157 0.32 0.066 0.511 (1.088 0.281 0.060 0.773 11.006 1.146 -0.013 0.359 0.021 0.3.H -0.001 0.243 0.027


Tahk 4.17 !Con"!.) Variahles Parental Education at tlu: Less Than High School Level Parental Education at the !\ore Than Hi,gh School Level Total Cigarette Usc Tntal Akohol Usc Total of Drug Tries Total of Ddinquent Activities Total School Trouhlc An Average of the Most Recent The Closem;ss the Respomk:nt Feels to a Pan:nt Friends Caring About You First Friends (Male anJ Female) at School Popularily Trnuhlc With Peers (Bullying or Isolation) *p<.05. *'''p<.O I. ***p<.OO I Unslandardizcd Std. Coefficients Error lU25 0.192 0.017 0.009 0.002 0.01X CU19 0.175 0.105':"'"'' 0.031 0.4 16';'** 0.051 -0.651:\':<:-0.153 -0.26 0.147 0.047 1.033''''"'' 0.123 76 Standardized fl.04X -0.07 0.012 0.002 ().()35 0.071 0.137 -0.131,1 -0.f)l)2 -0.026 -0.132 0.132


Bela Lists For the Fint Dependent Variahle Before moving on to the next dependent variable. the magnitude with which the variables impacted cmlltions positively or negatively with all feelings as a dependent variable will be listed in order. The above section used the unstandardized cocllicients for its analysis. The next section analyzes the output the betas. The unstandarized coefficients c.xplain the variance or impact of a given independent variable upon the dependent variable within each model, but cannot be compared in a magnitude relation with the other unst:mdardized coefficients in the model because they are composed of different units. A beta score is standardized and can be used to compare the magnitudt:: of its t::l'fcct with other beta scores. Higher beta mean more variance in the dependent variable is explained hy that independent variable. Beta comparison is not preferred by some because it is a comparison made with variables of differing units. but both unstandarized and standardized codTicients are being analyzed in this study for a fuller description and each type of comparis\lns has its The beta slorcs arc labeled as the standardit.ed cocfticicnts in tables 4.11 to 4.17. These ranking:s arc especially important to lnok at dosdy as they arc answering part of the hypothesis. The sratcs that fiicndship not only beneficial emotionally. but it will be as significant. as important, in magnitude as other social !actors. These rankings will show if the second part of the hypothesis is correct and what differing impacts the different social and control variables have upon emotions. 77


Remember that for all feelings a higher s.:ore represents poorer emotional adjustment. (So a negative sign bcucr em111ional adjustmtnt.) Bendi.:ial effects (All l I. pan:ntal dosencss ( -.139o:l) 2. popularity (-.13105) 3. friends care (-.09167) Detrimental effects (All variables) I. gender ( .15963R l 2. school trouble (.133615) 3. trouble with peers ( .13207) Top Six effects (All variables,) I. gender (.15963R) 2. parent closeness ( -.13903 l 3. school trouble (. 4. trouble with peers (.13207) 5. popularity(-.13105) 6. tiiends care (-.09167) Top Six Social efli:.:ts

arc henl!ficial would still he hencficial and those that Wl!rc dl!trimt:ntal would still he detrimental. hut this is not thl! case. The em1ltinnal impact that friendship contact had upon the adoll'sccnts emotions changed from detrimental to henel"icial. None nr the other variables did this. Originally. the emotions wen: broken down to sec if there were different impacts hctwecn social variable;, because it was thought dillen:nces might shnv. in tht: beta comparison'> hut. this sh1lWS an even more important ditTcrencc for fiiendship contact whl'n looking at difkrcnt aspects of emotions. Tablc 4.1 S whid1 includes gender. male friendship contact and kmalc friendship contact shows both male fiiendship contact and female fiiendship contact having bendicial effects contributing to good behavioral adjustment. Female friendship contact was significant at the p<.05 Incl. but the male ti'icndship contact was not signilicant. ln table 4.19 race, parental educational attainmcnt, and family structure were added. ln this model these control variables lessened the beneficial effects, or increased tht: detrimental effects. of friendship contact, as did with all thn:e depcndcll! variables. In table 4.20. illicit behaviors wl!re added and like in all the models. thl!se controls l!nhanced the hcncl'icial of friendship contads showing that illicit behaviors impact f'riembhip l:Ontact in a detrimental way, hurting the adnleseent's \!motional wcll-heing. Controlling for substances maue the male contact significant at the p<.O I level and female contact at the p<.OO I level making them both signi tic ant predictor!; of be Iter emotional adjustment. As delinquency, grades. and pare mal closeness was controlled for in table 4.21. a ui!Terent result ocr.:urrcd for male friend contact and female friend contact which only occurs with this dependent variabk. Mal!! fril!nd contact increas!!d in its beneficial impact from a p<.05 impact to a p<.OI. while female friend contact lessened slightly without changing its significance. This may indicate that for male friendship contact when happy emotions are involved. delinquency. school 79


trnublc, grades and pan:nl (or a lack their o[') largely inhibit emolional v.-ell being when looking at happy kl'lings. The addition of the first two social variables, friends caring about you and tirst friends (male and female) at your school in table 4.22 cause both friend contacts to fall in their beneficial ellects and become nonsigni licant. although still beneficial. The addition of pnpularity in table 4.23 impacted the friendship contacts difk:rcntly as malc contact"s beneficial ellects wne bringing it to a significance at the p<.05 Icwl. ami female friendship contacts beneficial effects decreased from a .042 in table 4.22 to a .033 in tab1c 4.23. This difference in the genders of the friendship contacts was found with the last dependent variable. depressed feelings as well. In table 4.24 with the addition of trouble with peers (bullying and isolation) male friendship's beneficial effects decreal'.ed by .003 changing. it from having significance at p<.05 to not and female beneficial contact went up by .00 I. still leaving it insignificant. In this last hlock neither friend contact. male or female. was significant. but both remained beneficial. The Data Tables for the Second Dependent Variahlc Table 4.18 Ordinary least squares regression results predicting emotional adjustment as a function of fril:ndship contact and gender with unstandardized and standardized cocf"lkients using happy feelings as the dependent variabk Variables Unstandarized Coefficients Std. Errors Ma1c Friendship Contact 0.04 0.027 Femalt.! Friendship Contact ().058* 0.025 Gender (Tvlak =0. Female= I) -0.327*'''* 0.085 *p<.05. **p<.O I. '''**p<.OO I 80 Standardized Cocflicicnts 0.027 0.04 -0.062


Tahlc 4.19 Ordinary kao.;t s4uarcs results predicting adolescent ernutional adjustment as a function of" friendship contact. gender. race, family structure, and parental education attainment with unstanJardi1ed and standardit.ed col'lficients using happy feelings as the dependent variahk Variahks Male f-riendship Contact Female friendship Contact Gender il'v1ak =0. Female =I ) Ruce was controllni{{Jr hy t>urting the respondents into six racial (White was the e.rcludetllariab/e.) Hispanic :\sian African Native American Other Family structure was controlled j{>r by putting the respondel!ts into four family structure categories. I Li1ing With Bia/agical Mother wui BioloRicul Father was the excluded variahlc.) Living With Other Family Structure Living With Biological Mother and Father Figure Living With Biological Mother Only Parel!tal Education Attainmellt It'll.\" contrnlledfor hy putting the respol!dents into three parental education attainment categories. (Parental Education at the hiRh schoollnel was the excluded l'r.Jriab/e.) UnstandaridizcJ Coefficients 0.025 0.04 -0.299**''' -0.493''-*''' -0.111 0.1 tl9 0.141 -0.5::!4*** -0.214 -0.347*** 81 Std. Standardized C\lclficil.!nts Ern1rs (1.()26 0.017 0.024 0.027 -0.0511 0.151 -(!.059 0.203 -(1.065 0.12Q -0.016 0.304

Tabk 4.10 (Con't.) Variables Parental Education at the Less Than High School Level Parental Education at the More Than High School Level '''p<.05, I. ''"''*p<.OO Unstandaridizcd Std. Standardized Coefficients Coeflicienls Errors -0.54'"*''' 0.1 )() -0.064 0.6X6**''' 0.084 0.129 Table 4.20 Ordinary least squares regression results predicting adolescent emotional adjustment as a function of friendship contact, gender. race. family structure. parental education attainment and substance usc with unstandardized and standardized coefficients using happy feelings as the dependent variable Variables fl.ale Friendship Contact Female Friendship Contact Gender (Male =0. Female =I) Race 11'0.1 controlled for by putting the respmulents into six racial cu.rep,nries. ( \Vhite was the excluded variahle. J Hispanic Asian African American Nati\'C American Other Fumily structure I HIS controlled ji11 by putting the respondent.\ illfo jimrfamily structure categories. (Living With Biological Mother and Father was the excluded vuriable.) U nstandard ized Coefficients 0.0115* 0.079**''' -0.3 59**,,. -0.514**'1 -0.922**''' -0.275''' 0.202 0.02 82 Std. Standardized Coefficients Errors 0.025 0.044 0.023 0.054 O.OSI -0.067 0.153 -0.062 0.197 -0.069 0.13 -0.04 0.291 0.011 0.438 0.00 I


Tahk 4.20 (Con"t.) Living \Vith Otht.:r Family Structure Living With Biological fvtnther and Father Figure Living With Biological Mother Only Parental Edrwation Allainment wu.1 controlled./(!/ by putting the respondent.\ into three parental ed11cation attainment carcgories. I Parental Educatinn at the high schonl/cFelwas the excluded variahle. J Parental Education at the Less Than High School Level Parental Education at the t\lorc Than High School Level Total Cigarelle Use Total Alcohol Use Total of Drug *p<.05. *'''p<.O l. ***p<.OO I Unstandaridi1cd Cocf1icients -(U32''' -O.OlJ -0.215* -0.0 [ 2*''' -0.007 Std. Codficienrs Errors 0.155 -0.0.1R 0.1 :I-t -(1.() [ 2 0. [ 04 0.135 -0.063 0.081 .0.124 0.004 -0.062 0.007 -O.OIR 0.059 -0.123 Table 4.2 l Ordinary least squarl.'s results predicting adokscent emotional adjustment as a function of friendship contact. gender. race. family structun.:, parental eJucation allainmcnt. substance usc, delinquency. school trouble. grades and parental closeness with unstam.lardizccl and standardized weflicients using happy feelings as the dependent variable Variables Male Friendship Contact UnstandariLcd Coefficients 83 Std. Errors 0.025 Standardized 0.049


Tabh: 4.11 iCon't.) Variahlcs female Friendship Contact Ciender 1 l'vtak =0, Female =I ) Race 1ra.1 comrol/cd_f('r In puTting the resprmdents into six mciol caregorics. 1 Whirc was the excluded \'uriah/c. J Asian American Native American Other Familr stmcture controlled j{1r by putting the respondent.\ info j(,urfamily strucTure categories. ( Lit'illg With Biologica!JHother a11d Biological Father tws the excluded l'ariable.) Living. With Other Family Structure Living. With Biological Mother and Father Figure Li,ing With Biological Mother Only Parental Education Attainment 1ras contro/ledfor by putling the respondents into three pare11tal education attainmenT categories. (Parental Education at the High school fele/ was the excluded 1'{/riah/e.) U nstandaridized Std. Standardized Coefficients Cocrticiems Errors 0.075*** 0.022 0.051 -0.479*'1"'' O.OX4 -0.09 -0.-126** 0.146 -0.051 -O.R 1''-** 0.181 -0Jl61 -0.115 0. I -0.017 0.44lJ 0.2S2 0.024 -0.021 0.469 -0.00 I -0.1 -0.012 -0.018 0.132 -0.002 -0.057 0.102 -0.009 84


Tahlt: .:1-.21 1Con"t.) Variables Parental Education at the Less Than High School Level Parental EJucation at the More Than High School Level Total Cigarette Total Alcohol Usc Total of Drug Tries Total of Delinquent Activities Total School Trouble An Average of the !\:lost Recent Grades The Closeness the ResponJent Feels to a Parent *p<.05. **p<.O I. ***p<.OO I Unstandaridizcd Std. StandarJized Coefficients Coer1i.:ients Errors -0.435'"*''' 0.131 -0.0.52 0.55.:1-*** O.OR6 fl.I0-1 -0.005 0.004 -0.026 ().()04 0.006 0.010 -0.19*'1 ().()64 -0.059 -0.022':' 0.0 II -0.041 -(!.097**':' 0.016 -O.OR7 0.477**: 0.053 0.137 0.435**'' 0.043 0.159 Table 4.22 Ordinary least squares regression results predi.:ting adolescent emotional as a function of friendship contact. gender. race. family structure. parental edunllinn allainment, substance use. delinquency. school trouble, grades. parental closeness, friends caring about you and first tiiends at school with unstandardized and coeffi.:ients using happy feelings as the dependent variable Variables Male Friendship Contact Female Friendship Contact Gemkr (Male =0, Female =I) Race was controlled for by Jllllfing the respondents into six racial categories. (White \l'as the excluded variable.) Unstandardized Coefficients 0.052 ().()42 -0.574*** 85 Std. Standardized Error Coefficients 0.026 ().035 0.023 0.029 -0.108

PAGE 100

Tahk 4.22 (C:on"t.) Variables Unstandardized Std. Coefficients Error Coefficients -0.384*''' 0.142 -fl.046 Asian -0.737*'1";' 0.18 -0.055 African American 0.033 0.127 0.005 Native American 0.493 0.267 0.026 Other 0.013 0.001 Family strttcture 1ms controlled j(Jr /Jy pul!inp the n:spondents into j(wrfamily strucrure categories. ( Liting With Biological Mother and Biological Farha the excludNl l'ariable.) Living \Vith Other Family -0.08 0.1-l.) -0.009 Structure Living With Biological Mother 0.017 n. 1 31 0.002 and Father Figure Living With Biological Mother -0.062 -().()()9 Only Parnrtal Education Attain me/IT was controlled fnr by putting the respondent.\ into three parellfal education attainment categories. r Parental Eduration at the high schoollevd tht' excluded mriab/e.) Parental Education at the Less -0.37 0.135 -0.044 Than High School Level Parental Edu-:ation at the More 0.5 I 9**'!' 0.085 0.098 Than High School Level Total Cigarette Usc -0.005 0.004 -(1.026 86

PAGE 101

Table 4.22 tC:on't.) Variables Total All:ohol Usc Total of Drug Tries Total of Delinquent Activities Total School Trouble An A of the Most Recent Grades Thl Closeness the Respondent feds to a Parent Friends Caring Ahout You First Friends (Male and Female) at School *p<.05. *':'p<.O I. *'''*p<.OO I Cocllicients 0.004 -0.185** -O.OIQ -0.087*':,-:, 0.454*** 0.179**';' 0.506**';' 0.064 Std. Standan.Iized Error Coefficients 0.006 0.010 0.064 -0.057 0.011 -o.m5 0.016 -0.078 0.053 0.11 0.042 0.119 0.06 0.148 0.061 ().()18 Tahle 4.23 Ordinary least squan.'s regn.'ssion predicting adolescent emotional adjustment as a function of friendship comact. gender. race. bmily structure. parental education attainment, substance use, delinquency. school trouble, grades. parental doseness, friends caring ahout you. first friends at school and popularity with unstandardized and stam.Iardized coefficients using happy feelings as the Jcpendcnt variable Variables Male Friendship Contact Female Friendship Contact Gender (Male =0, Female =I) Race was control In/for by putting the respondents into six racial categories. (\Vhite was the excluded l'ariaiJ/e.) U nst::mdari zed C:ocl'ficients 0.053* 0.031 87 Std. Error Standarditcd Coefficients 0.026 0.035 0.022 0.022 0.09 -0.096

PAGE 102

Table 4.23 (Con't.'J Unstanuardited Std. Error S tandaru i zed Codlicients Coefficients Hispanic -OA 12'"'' 0.145 -0.049 Asian -0.7 56':":":' 0.177 -0.057 African American 0.027 0.124 0.004 Native American 0.:257 ().0:24 Other 0.016 (J.OO I FamilY .\fmcture was controlled for hy putting the re.1pondents into jiJUrfwnily structure categories. ( LiFing With Biological Motht'r and Biolo!!,ical Father 1\'a.\ the ext luded variable.) Living With Otha f-amily Structure -0.062 0.147 -0.007 Living With Biologicall\other and O.Oll 0.129 0.()()4 f-athcr f-igure Living With Biological Mother Only -0.04!-1 0.10 I -0.007 Parental Education Attainment 1ra.1 controlled for by putting the respondent.\ into three parenwl education attainmellt cate!!,ories. (Parental Education at the high sclwol leFe/wos the excluded variable.) Parental Edm:atinn at the Lcss Than -
PAGE 103

Tahl.: 4.2.' tCon't.) Variahh:s Total of Ddinquent Activities Total Schot)l Trouble An A vcrage of the Most Recent Grades The Closeness the Respondent Feels to a Parent Friends Caring About You First Friends (Male and Female) at School Popularity *p<.05. *''p<.O I. *'''*p<.OO I Unstandan.lizcd -0.018 -0.054*** 0.4 2 i) 0.34''":'* 0.441 *':"'' 0.002 0.136''"'"'' Std. Error 0.012 0.016 0.055 0.041 0.057 0.06() 0.017 Standarditcd Codficicnts -O.O:B -0.048 0.123 0.124 0.129 0.001 0.133 Table 4.24 Ordinary least squares regression n:sults predicting adolescent emotional adjustment as a function of friendship contact. gender. race, family structure. parental education attainment, substance use. delinquency. school trouble, grades, parental caring about you. first friends at school. popularity and troubk with peers with unstandardized and standardized coefficients using happy fedings as the dependent variable Variables Male Friendship Contact Female Friendship Contact Gender (Male =0, Female= I) Race was controlled for hy putting thl' resprmdellt.\ into six racial categories. ( White was the excluded variah/e.) Unstandardized Coefficients 0.05 0.034 -0.505*''"i' 89 Standardized Coe llil:ients 0.026 0.()3_, 0.(122 0.023 0.09 -0J)95

PAGE 104

Tahlc 4.24 1Cont"t.) Variabks Std. Error Standardized Coe rlicicn -0.44 ''* 0.146 -0.053 Asian -0.766*''"'' 0.174 -0.057 African Aml'rican o.o2g 0.123 0.004 Native American 0.456 0.261 0.024 Otha 0.023 0.456 0.001 Family structure Has romrolledfiH by putting the rCSJ>Oil{/ent.l into jimrfamily structure catep.ories. ( Liring With JJiologiw!JHother and Biological Fmher ll"liS the excluded variable.) Living. With Other Family Structure -0Jl45 0.144 -(!.005 Living With Billlogical Mother and 0.042 0.12(} 0.005 Fathcr Figun; Living With Biological M0thcr Only -O.Olg 0.10 I -0.006 Parenwl Education Attainment H'as controlled for by pulling the respondent.\ into three parental education attainment catcg(Jrit'.\. (Parental Education at the high school/eve! the e.rclwled 1ariablc. J Parental Education at the Less Than -CU49** 0.134 -0.042 High School Level Parental Education at the More Than 0.508**''' o.ms6 0.095 High S<.:hool Level Total Cigarette Usc -0.004 0.004 -0.021 Total Alcohol Usc 0.005 0.006 0.013 90

PAGE 105

Tahlc 4.24 tCon't.) Total of Drug Tries Total of Ddinquent lu:tivitics Total School Trouble An Average of the !\:lost Recent GraJes The Closeness the Respondent Feels to a Parent Friends Caring Ahout You First Friends (Male and Female J at School Popularity Trouble With Peers (Bullying or Isolation) *p<.05. *'''p<.O I, '''** p<.OO I U nstandan1 i zed -O.IM* -0.0 15 -!Un61 0.432'''** 0.344';'''''1 0.427'''':'* 0.006 0.12X*'''* -0.141 *'!' Std. Error 0.065 0.0 I I 0.017 0.055 0.041 0.057 0.065 0.017 0.052 Bew Lists for the Second Dependent \-'ariable Standardited Coet'licients -0.051 -0.028 -0.032 0.124 0.126 0.125 ().()()2 0.125 -(l.05 Bcl'ore moving on to the next dependent variable, the order of the variables impacts will be ranked :tgtin. Rcmemhcr that for happy fedings a higher score represents better emotional adjustmcnl. effect!> (All variahles) I. parental closeness (.125779) 2. friends care (.124943) 3. popularity (.124887) 91

PAGE 106

Detrimental effects 1 All !.gender (-.IN47XI 2. Asian( -.057:\i 3. Hispanic (-.0531 Top dlcus (All I. parent closencs.., (.125771)1 2. friend.., care (.1249-B) 3. popularity (.124gS7) 4. gender (-Jl9478J 5. more than high school (.0954) 6. Asian (-.05n) Top Six Social effects (Only social variables) I. friends care 1.124443) 2. popularity 3. peer U"\uhlc ( -.05006) 4. mak friend contact 1 .o:rn 19 J 5. female friend contact (.0230lJ8) 6. friends at <.001674) Modr:/.1 Using Dependent Variable Three (Depressed Tahles 4.26 ro 4.31 A higher swn.: in these models. like in the tahks 4.1 I to -L 17, represent a poorer emotional adjustrncnt when looking spccilically the 'depressed" dependent variable. With this new tlepentlent variable table 4.26. male friendship contact's unstanclardized coefficient score was a .151 at a p<.OO I significance level and female friendship contact s unstanclardized coefficient score was a .147 with a significance level at p<.OOI. That is recent friendship contact predicts. at a 9lJ.l)l7( level. an in.:rease in adolescents" deprt:ssing emotions. These results had the largest effect found in the tirst block of all the dependent variables. The second block. table 4.27. adding race, parental educational attainment and family structure increased the depressing effects of hoth male and female friend contact. Then in block three 92

PAGE 107

hoth friendship contacts depressing (.kcrea:-ed from .168 fnr male and tt:male friendship l.'(llltact in block two. table 4.2. 7 to a .105 for males and a .I 03 fnr females, although hoth were still signi fie am at the p<.OO I level, when substances. cigarcllei>. alcohol. and drugs were controlled fnr in block 3. table 4.28. This lessening of the depressing effects was found again with the addition of ddinqu.:ncy, school trouble. grades and parent closeness, although the significance was not changed and these changes and cflccb wac to he expected and seen in the other models. The addition of the social factors of friends caring about you and first friends at school in table 4.29 changed the friend contacts by increasing the depressing effects for males and fernaks by .016 and .025 respectively. Popularity increased the depressing effects for female friendship contact by .0 I 0 and decreased the depressing effects for males by.OO I. Still both of the friendship contacts were at the p<.OO I level. Finally, the last block, table I. added trouble with peers. This lowered the depressing effects t(:>r females hy .00 I. but increased the depressing hy .008 for males. In the end, both male and female contacts are depressing impacts signiticant at the p<.OO I level. Indicating that when looking just at depressing emotions friendship contact is a strong predictor of poorer emotional adjustment. This dependent variable. depressed feelings. showed a stronger predictive impact of detrimental (depressing) feelings from the independent variables, male friendship contact and female friendship contact. than either happy feeling showed for beneficial (happy) feelings or all feelings showed for detrimental feelings. This difference of impact should be noted as a significant result of the last models using depressed feelings as a dependent variable. 93

PAGE 108

The Data Tables fur the Third Dependent Variable Table 4.25 Ordinary least squares results predicting adolescent emotional adjustment as a function of friendship contact and gender with unstandarJized and standardized coefficients using deptnsed feelings as the dependent variable Variables Male Friendship Contact Female Friendship Contact Gcndl!r (Male =0, Female =I) *p<.05, **p<.O I, I Un<.tandardizcd Coefficients 0.151*** Std. Errors Standarditcd CoeflicicnLs 0.025 0.085 0.028 0.{)84 0.099 0.151 Table 4.26 Ordinary least regression results predicting adolescent emotional adjustment as a function of friendship contact, gender, race. family structure and parental educational attainment with unstandardized and standardited coenicients using depressed fl!clings as the dependent variahlc Variables Male Friendship Contact femall! friendship Contact Gender (Male =0. Female =I) Race was to/ltrolleJ j(Jr hy putting the respondent.\ into six ruciul categories. White was the excluded variable.) Hispanic Asian African American Native American Other Unstandardi1.l!d Coerlicil!nts 0.168*** 0.494*'" 1.318''":'* 0.587*** 0.782' .. -0.528 94 Std. Errors Standardized 0.026 0.094 0.027 0.096 0.097 0.149 0.1 !n 0.049 0.259 0.083 0.129 0.071 (!.353 ().()35 0.406 -0.0 15

PAGE 109

4.26 tCon't.) Family structure was collfmlledj(w b1 putrillf! the respondents intofourfiunilr stmcture (Lil'iiiR BioloRical Mother and Biologicul Fatlter uus the excluded mriable. J Living With Other Family Structure Living With Biological Mother and Father Figure Living With Biological Mother Only Purentul Education Attainment was controlled fin by puttillf! the respondents into three purental educution artuinment categories. ( Parnlful Education at the hiR!t school level was thmriable.) Parental Edu..:ation at the Less Than High School Level Parental Education at the More Than High School Level *p<.05. *':'p<.O I. ***p<.OO 1 Std. Error-, -(1.444''''""' 0.171) 0.144 0.174 0.09 Standardized Codli..:ients 0.01)2 0.04 0.079 0.05K -0.07 Tahlc 4.27 Ordinary squares regression predicting adokscent emotional adjustment as a function of friendship ..:ontact. gender, ra..:e. family structure, parental cdu..:ational attainment and substance usc with unstandardized and standardized coeffi..:ients using depressed fedings as the dependent variable Variable Unstandarizcd S tel. Errors Standardized Coefficient Coefficient Male Friendship Contact 0.1 0.5*** 0.025 0.059 Female Friem.lship Contaet 0.103*** 0.025 0.051) 95

PAGE 110

Tabk 4.27
PAGE 111

Table 4.27 !Con'!.) Variables Parental Education at the Less Than High School Level Parental Edth.:ation at the More Than High School Level Total Cigarclle Usc Total Alcohol Total of Drug Tries '''p<.05. '"*p<.O I. '''**p<.OO I Unstandardized Std. C\ldTicients 0. I 6 -0.389'''*':' 0.085 0.022*** 0.005 0.009 0.085 Standan.lized Coefficients 0.056 -0.061 O.O(}fl 0.055 0.12 Table 4.28 Ordinary least s4uarcos rt:greso.;inn results predicting adolescent emotional adjustment function of fritndship contact. gender. race, family structure. parental educational attainment. substance use, delinquency. school trouble. grades and parental closeness with unstandardi?cd and standardized coefficients using depressed feelings as the dL'PL'ndl'nt variable Variables U nstandardized Std. Errors Standardized C oc llicients Cocflicicnh Male Friendship Contact 0.025 0.052 Female Friem.lship Contact 0.092**''' 0.025 0.053**** Gender (Male =0, Female =I) 1.222''""* 0.099 0.193 Rua wus hy pullinx tht respondent.\ illfo six racial cafcRories. 1 White nas tht excluded vuriuble.) Hispanic 0.476** 0.172 0.048 Asian 1.249*** 0.238 0.07H 97

PAGE 112

Tanlc 4.2X tCon'l.) Varianks Unstandardi7ed Std. Errors Standard i 1ed Codlicients Coefli..:ienh Arri..:an American 0.749*''"' 0.131 0.091 Native American 0.449
PAGE 113

Tahle 4.2R tCon"l.) Total or Drug Tric), Total of Delinquent Activities Total School Trouhle An A veragt:! or the Most Rcct:!nt Gradl!s The Closeness the Respondent Feels to a Parent *p<.05. **p<.O I. '''*'''p<.OO I Std. Em1rs Standardized Cleflicients Coeflicients o.12s -0.174':":' 0.06.1 -0.1142 -0.4tl5*** 0.055 -0.149 This difference occurred when rounding from six places Tahle 4.29 Ordinary least s4uares regression results predicting adolescent emotional adjustmenl as a runction or friendship contacl. gender. race, family structure, parental educational anainment, substance usc, delinquency. schooltroublc, grades. parental closeness, friends caring ahoul you and first friends being at school '.'.ith unstandanJizcd and standardized coefficients using depressed feelings as the dependent variable Variables lJ nstandard i Std. Errors Standardized Coefficients Coefficients Male Friendship Contact 0. 0.025 tl.U60 female Friendship Contacl O.IIY':'* 0.026 0.067 Gender (Male =0, Female =I l 1.253'!":'* 0.198 Race H"l.lS controlled j(Jr hy putting the respo11de11t.\ into six racial cateMOric.l. (White was the excluded variable.) Hispanic o.4.n: 0.17 0.044 Asian 1.193*** 0.232 0.075 African American O.h4'''** O.D2 0.07R 99

PAGE 114

Table 4.2lJ (Con't.) Unstandardized Std. Errors Standanlitcu Cocllicicnts Coel'fic icnts American 0.422 IU45 O.Q19 Other -0.3 71 0.4ll) -lUl I Family struct11re was controih'd ji1r by the rnpurulents illfo jlmrfamily Jtructure ( \t'ith Biological Mother and Biological Father was the excluded l'ariable.) Living With Other Family 0.429** 0.042 Structure Living With Biological Mother 0.099 0.141 0.011 and Father Figure Living With BiologiL:al Mother 0.259''' 0.127 0.0.'3 Only Parental Education ATtainment was controlled jr1r by putting the rc.\f)(JIIdent.\ into three paremal education allainment categories. 1 Parental Education ill the High schoo//ercl was the excluded mriable. J Parental Education at the Less 0.445 ** 0.16.' 0.044 Than High School Level Parental Education at the More -0.32S'"':'* 0.081} -0.052 Than High School Lcvd Total Cigarette Usc 0.0 14'1'* 0.004 0.061 Total Alcohol Use 0.001} 0.009 0.019 100

PAGE 115

Tahlc 4.29 !Con't.) Variahlcs Unstanuardizcd Sid. Standardized Total of Drug Total of Ddinqucnt A.:tivitics Total School Trnuhle An Average of the Most Recent Grade:. The Closeness the Respondent Feels to a Parent hiends Caring Ahout You First Friends (l\'lale and Female) at School ''"''p<.O I, I Coefficients 0.1 17 0.049''"'* 0.21 '''*'!' -0.148'i' -0.456*** -0.206** Cocflicienh I 0.014 ().()76 0.023 0.157 0.063 -0.036 0.056 -0.14 0.065 -0.062 0.067 -0.048 Table 4.30 Ordinary least squares regression results predicting adolcs.:enl emotional adjustment as a function of friendship contact. gender. ra.:e, family structure. parental educational attainment, substanel' usc, Jclinquency. school trouble. grades, parental doscness. friends caring ahout you. first friends Ixing at school and popularity with unstandardized and stamJardizcd coetliLients using depn::ssed feelings as the dependent variable Variables Male Friendship Contacl Female Friendship Contact Gender (Male =0, Female =I) Race was controlled j(n by putting the respondents into six racial categorie5. (White was the excluded l'ariable.) Unstandarizcd Coefficients 0.1 07*'''* 0.127*** 1.186*** 101 Std. Errors Standardized Coefficients 0.025 0.06 ().()25 0.072 ().()1.)7 0.187

PAGE 116

Tahle 4.30 tCon"L) Varianles Std. Errors Standardized Coefficients Hispanic 0.47.>*'' 0.165 0.047 Asian 1.214**''0.224 0.077 African American 0.647*'''* O.IT?. o.on Nati\ c American 0.479 lL'42 0.021 Other -0.375 0.426 -0.010 Family structure was controlled j'i!r by putting thl:' info }i111r family structure categories. I Living With Biological IV! other and Biological Father was the excluded variable) Living With Other Family Structure 0.40R* 0.15R 0.04 Living With Billlogical Mother and ll.OX4 0.142 0.001 Father Figure Living \Vith Biological Mother Only 0.24.> 0.124 0.031 Parental Education Attainmellf nas controlledfor by puttinp, the respondent.\ infO threl' parental education artainmellt categories. (Parental Educatioll ar the high .1chool level was the c.rcludC'd l'(lriahle. I Parental Education at the Less Than 0.432':'* 0.162 0.043 High School Level Parental Education at the More Than -O.J3**';' 0.09 -0.052 High School Level Total Cigan:l!c Use 0.012'1 1 !J.004 0.052 Total Alcohol Use 0.009 0.017 102

PAGE 117

Table tCon.l.) Variablt:s Total of Drug Tries TOlal of Delinquent Activities Total Sehoul Trouble An Average of the Most Recent Grades The Closeness the Respondent Feels to a Parent Friends Caring About You first Friends (Male and Female) at School Popularity UnstanJardi1.ed C 1 dlic icnts O.flR (). ()..J.f\ o;
PAGE 118

Tahk 4.31 !Con.l.) LJnstandardized Std. Errors Standanlized Cll:flicicnls Ruce 1rus controlledfor hr tlu respondents into six racial catew!l"ics. (\!v'hite wus the excluded \'{/riah/e.) 0.549**': 0.162 0.055 Asian *** 0.224 0.07X Ali"ican American 0.644*'" :o 0.127 0.078 Native American 0.441 0.341 0.02 Other -(L\95 0.452 -0.011 Family strucrure was control/uf j(Jr /Jy puuing the respondents into four fnmily struct11re categories. r Li1inx With Biological Morher and Hiolngical Father was rile excluded rariahle.) Living With Other Family Structure 0.36:::1* 0.155 0.035 Living With Biologicall\olher and 0.14 0.006 Father Figure Living With Biological l\other Only 0.217 0.121 0.027 Parl!lllal Education Attainml!nt was controlildfor hr puffing tht respondents into three parellfal ed11cation attainment categories. ( Parenwl Education at the high scltoo{ lcvellras the excluded m ria bl e. J Parenlal Education at lhc Than 0.404* 0.16:::1 0.()4 High Schon! Level Parcnlal Education at the More Than -0.294**'1 0.089 -0.046 High School Level 104

PAGE 119

Tabk 4.31 (Con't.) Variables Total Cigarelle Total Akohol Usc Total or Drug Tries Total of Delin4uent Activities Total School Trouhh: An Average of the [\lost Recer11 The the Fecb to a Parent Friends Caring About You First Friends (Male and Female) at School Popularity Trouble With Peers (Bullying or Isolation) Std. Errors O.OIF'* 0.004 0.009 0.00() 0.11.1 0.041 '''* 0.013 0.121 **''' 0.025 -0.13* -0.424**''0.057 -0.142" 0.063 -0.148* 0.067 -0.13**':' 0.024 0.39*** O.Oii I Beta Lists For the Third Depend em Variable Standardited Coc rticie nt s 0.052 0.019 0.029 O.OM (1.091 -0.03 I -0.1.1 -0.035 -0.035 -0.107 0.116 Remember that for depressed l'eelings a higher score represents poorer emotional adjustment. Benclicial effects (All variables) I. parental closeness ( -.13017 J 2. popularity (-.1065) 3. more than high school ( -.04638) 105

PAGE 120

DL'triml.!ntal efll'Cts (All 1. gender ( .1 83287) 2. trouhk with (.11627) 3. sch\lOl trouble ( .OlJ0624) Top cllc.:ts (All variables) I. gender ( .1 832X7) 2. parental closenl!ss l -.13017 J 3. trouhll.! with peers ( .11627) 4. popularity ( -.1065) 5. trouble (.090624) 6. African Ameri,.;an (.078) Top Si.x Social effects (Only social variabks) 1. trouble with peers (.11627) 2. popularity (-.1065) :;_ female friend contact (.071874) 4. male friend .:ontact (.064346) 5. friends care (-.{)3489) 6. friends at school l-.0346 7) 106

PAGE 121

CHAPTERS DISCUSSION Describing the emotional impacts or social variables has many applications, and practical. Understanding the dynamics of social variables and emotions can help to create different models for sm:ial environments in schools. hdp inform therapies and help enrich intervention strategies for the socially alienated. It also strengthens and enlivens the academic's vision of adolescent's social world anJ the connections to their emotional world. This discussion will look at the results of the study and then link these results to the literature review. possible causes and possible applications. Friendship Contact's Impact on Emotions Using All Feelings as the Dependent Variabk Much of the literature emphasized the importance of friendship for development and support. The hypothesis or this study asserts that given that friendship is important in so many ways it; must have a large emotional impact. this impact must be bendicial, its impact must be as large as the impacts or other s
PAGE 122

variables in imract. means that friendship ihclf is not a tool to help or socially alienated tCl'Ib, hut friendship quality is much more vital to interventions and emotions. The hda the imptlrtance of being popular and not hcing as priorities in the crt:ation or school environments. As stated ah1>\T when using all feelings as the dcrcmknt variable. the outcomes did not support the hypothesis that friembhip contact is emotionally bcncliciaL but did support that fricndshir quality is hcncficial emotionally. Friendship contact, male and female, was found to increase unhappy and unhealthy emotions when using all feelings as the dependent variable. This result of friendship contact increasing detrimental emotional feelings occurred in all of the blocks and models used to analyze all feelings with varying significances. Even when controlling for know rroblcms of friendships: tiiendship existence and contact was a detrimental intluence. The addition or the controls substance ahuse. delinquency and school trouhlc decreased the detrimental eftecto., of friendship. hut not cn11ugh to make it a hcnelicial factor. This fact is inten.:sting given all the supporting evidence that friends ami friendship an: good for adolescents. This signals a need to rccv
PAGE 123

2007). It possible that for both male and female friendship there is co-rumination leading to friendship contact having a depressing effect. Another possihk reason why friendship contact has l'llects is that friendship quality was not consiuereu (purposeli.tlly) in the ..:rcation of the variablc friendship contact. One measure 1lf friendship quality. friends caring about you. was measured in another variable, hut not included in friendship contact. nor where any other measures of friendship quality. lt could be the case that poor friendship quality is causing the depressive effects. As Hart up ( 1996) contends knowing the behavioral altitudes. and quality of friendship is necessary to make predictions about friendships. In fact, Windle ( 1994) found that overt and covert hostility and less reciprocal relationships in ti"iendships Wl're associated with higher levels of drinking. delinquency. depression and suicidal helwvims and (Kupersmidt et at.. 1995) found some rejected children with best friends were more likely than their rejected peers to develop antisocial problems. When looking at time spent with friends this analysis. and some of the literature. clearly demonstrates that friendship is not necessarily a good thing and that could he what is causing its detrimental effects. The of friendships should he analyzed to see if it is poor friendship quality that is making friendship contact have a depressive efrect or if it something dse about spending time with friends as an adolescent. Impucts and lwcrp/ays between Control Variohll'.l All Ftelings us the Dl'fJt"lldf!llf Variable There are some other meaningful findings in table 4.13. The addition of the control variables reprcsellling substance US(' lessened the detrimental effi:cts of friendship contact to a nnnsigniticant level meaning tbatthe impacts of smoking. drinking and doing drugs is a large detrimental factor in the lives of adolescents that leaks into their friendships and the time spent v.:ith their friends. This seems 109

PAGE 124

hut it is gnod to recngnize and think about the "obviuus" and expected impach hecausc they can he explored and better described in future studies. The addition of ddinqueney. grades (lower) and parental closeness (lack uf) all also had (detrimental) expected impacts. Their block lessened the detrimental emotional effects of friendship contact indicating that if time with friends is being used being delinquent. failing or having school trouble or adding to parental troubles it i-, emotionally detrimental their friendships well as being emotionally detrirnemal on their own. Although no new discoveries were found in this part of the analysis expected and interactions should be noted when looking at the tables. impacts and interactions between Social Variables Using Aff Feelings As the Depe11dem Variable (And All Three Dependent Variables as a \Vhole When Looking for Repcati11g Patterns) In table 4.15 tWll social were added. One was if the respomlcnt' s first friends were at their scholll. and the other measured friendship quality hy asking the respondent if they felt their friend!> cared about them. Both of these variables had the expected beneficial emotional impat:ts as variabh.:s and increased the detrimentalcllects of friendship contact when added to the model leading to the conclusion that part of the reason 'A'hy having friends make you feel good and arc emotionally beneficial is hecau-;e have support and companionship at school and you have individuals who you feel care about you. These arc obvious findings that again should be noted. ln the section above the necessity of differentiating between good friendship and had tiiendship was already discussed and should be applied here as a possible explanation of why friendship contact was detrimental, but friendship quality was beneticial. Some obvious reasons why having your first named friend at school with you is beneticial is becau:-,c: it improves your Pl'pularity provides companionship and 110

PAGE 125

provides suppnrl. Another reason why having your at school has a beneficial effect it may in indicate already existing popularity or good social adjustment. This may roc the case because a child ur adolcs..:cnt 'A'ithout friends would not haw them at school and also it was found that rejected children and adolescents often have friends that arc younger. older or found in another h1cation !George & Hartman, 1996), ThL: next social variable added was popularity. This variable caused different impacts on male vs. female friendship contact. The dctrinu.:ntal effects of male decreased slightly making friendship contact a more beneficial variable (increased hy.O!n). Sim:e the change was not large it could he simply that this variable is meaningless to the emotional impact of male friendship contact or that it in a small way make male friendship contact less beneficial. This may be because popularity is not something that adds to male fricnd!>hip and may even get in its way. It could also mean spending time with your male friends does not increase your popularity. for female friendship contact the addition of popularity as a variable caused detrimental effects to go up (increased hy and from p<.Ol to p<.001 J dra\ving the conclusion that popularity is an emotionally beneficial side t:f!Cct of female friendship This may he due to female causing the respondent to !Cel and be more and have higher popularity score. or it is simply more of a component in female friendship priorities. (This pattern may be small in some instances. hut it each time the block is added when using all three dependent variables. so it is consistent and so may be meaningful.) Another consistent hut numerically small genda difference is that each time friends caring and first friends at school were added into the model female friendship wa:o. more detrimentally impacted which may mean that female friendship contact is more beneficially impacted by having friends at school and feeling that people care about you. That may he explained by female friendships' 111

PAGE 126

need for intimacy and sharing. Genlkr diiTerences in social can he found in the literature. some (lf which could help possihly explain this panern. Brendt ( 19S2) found that girls had mPtT exclusive and intimate hPys who tended to play in larger groups. This intimacy dirfcrence could he one explanation of the stronger impact or friendship 4uality upon female friendship contact and could abo explain why the listed (preferred) friends being at school was more important to female friendship contact. Lastly. the dillercnt interaction style within male and female fiienclship could he a partial explanation. A linding from Wallerstein and Kelly ( 1980) ohservt:d that during the turmoil of divorce hoys were hella at ustng their friends as distractions than their female counterparts. It could he that since temi to he more intimatt: and exclusive friendship quality is more of a pay off, for male friendship contact maybe distraction or companionship may lwve the same effect. The different payoffs for male and female fril'ndship should he taken into account if a study were to look at friendship l(Ualitics emotional impact. \Vith the last model the variahk representing hullying. pel'r trouhlc. was added. Male friendships increased and female friendships henclicial effects incn:aseJ slightly. It looks like an inversl' of the last variahlc's dTects. Here again, the genders had different effects and here again they rnay he numerically small in some instances. hut the) arc consistent with all three dependent variahles and in table 4.24, the addition of trouhle with peers changed male fiiendship contact's happy effects from p<.05 to being nonsignificant. One could conclude that male friendship contact is bendicial in part because it helpf> mitigate peer trouble and this is more important to male friendship contact than popularity. The interaction between hoys and the possible higher rates of aggression could account for this component being more imp11rtantto hoys. The different impacts of these variables could be explained hy the different 112

PAGE 127

\\ay-. male-; and lemales into.::rad in friendships and at school. as females friendships work to garna acceptann:, intimacy and status and male l'rio.::ndships work towards bullying an1idancc. having fun and being distracted. Other Findings in Tables 4.1/to 4.17 Using All Feelings As a Dependmt \:ariab/e One of the meaningful findings in the analysis of tables ..J..ll to 4.17 is that the variabk that measures how much a respondent feels their friends care about them has the opposite emotional elleLt than having a friend and spending time with that friend when using all tt:ding as a dependent variable. This signals the necessity of recognizing the difference of friendship quality and friendship contact upon the emotions of adolescents. Even though contact and companionship may be considered a part of some measurements that make up friendship quality because there arc many measurement of friendship quality it is clearly not the same having your friends care or think they care. The friendship quality variable is much more beneficial. important anJ separate from simply having a friend than the hypothesis lir:'a contended. It was always wus hcnelil:iaf (using all three dependent variuhlcs). whereas often friendship contact was not (when using two out of three dependent variables it was detrimenlal). This finding should lead to more investigations in to why friendships in without good quality arc more harmful than nN having friends at all. Also, maybe there can be research into helping children and adolescents learn how to have quality friendships so that they can all benefit more from having friends. 113

PAGE 128

Comparing the /Jcta Varia!Jfcs ComJlllfcd P,!hilc lhing A !I Ft:clings as the Depcndmt Faria!J!c Another component of the analysis is comparing the magnitude of the emotional impacts of the variables and. if interesting. the control variables. As explained before betas let you compare the variahlc., impact., across dillcring. units. The three variables with the hcneficialernotional effects upon and adolescent arc: parental closeness. popularity and friends caring. (As can he seen on page 100.) Both parental closeness and popularity had beta impacts at -.13 <1nd friends caring had a beta impact of -.09. This shows that for their parents and their n:lationship !1> them is as important as peer upon the impact of their emotirns. This also shows quality as being one of the top variables of impact. but it is still .04 points behind popularity meaning popularity, when using all feelings as a dependent variable. ha-; a stronger effect as a variable than friendship quality. let alone friendship contact. It must be concluded that the a-;sertion that friends thcmsclvc-; arc as important a'i other social variable:-. is false in this part of the The three variables with the highest detrimental emotional effects upon adolescent emotions arc: gender (female). school trouble and trnubll' with peers. Being a female has a detrimental effect of .15 and school trouble and trouble with are tied with a The fact that being an adolescent female is so depressing is a disturbing fact, but is not the focus of this study. The fact that troubles with peers and school trouble arc almost tied for second means that trouble with peers has a huge impact upon the lives of adolescents and must be addressed when adolescent social or emotional life. Trouble with peers outweighs the strength of having friends who care and is almost tied at [.13] with popularity and parental This means thm inventions cannot just focus on helping the adolescent gain social skills to help make friends, they must also help the adolescent avoid and ] 14

PAGE 129

confront trnubks with peer and their lack of popularity land not just through the addition of' friends or contact with friends). Looking: at the social variables and their betas. page 100. are telling. Trouble \Vith peers and popularity arc nne and two at [. l 31. then comes friends can.: with a -.09 and at half that magnitude there female friendship contact at .04. male friendship contact .03 and having lirst friends at school with -.02. This beta ranking clearly shows largest social impacts on emotions bcndicially or detrimentally arc trouble with peers. popularity and friendship quality. This strongly suggests that to help improve the emotions or adolescents the social world they arc apart or at school and in their daily lives must be addressed more so their having friends. who may or may not benefit them cmotiLmally. depending on the friendships' affection and caring. New Findings in the Modds Using Happy Feelings as a Dependent Variable The second group of' models a different emotional variable l as its dependent variable. This caused an interesting and surprising finding. In tables 4.18 to 4.24 the friendship contacts became variables that had a beneficial effect on adolescents' emotions. the opposite of the last model. This is interesting because it means the emotional impact of friendship changes depending on the emotion involved. A meaning or this is that friendship amplifies the feelings or the adolescent. It could be that friends that arc reeling a certain way make their friends feel the same \\'ay hecause of the and talk they participate in. or it could be thar friends that are already feel the same way hang out together and reinforce each others emotions. This is a phenomenon that should be explored. Friendship contact is the only variable that had a change of effect. Why and how this happened should be explored. 115

PAGE 130

Like in the la:-.t groups of modeb using all feelings a:-. the derendcnt variable. whcn using happy feelings. substance use was detrimental to the beneficial effects of friendship contact as was delinquency, school trnuhlc. grades (lower) and parental (lack oi).The pattern of the gender dillercncc between p(lpularity and trouble with peers was the same well. In the end, friendship contact and the other variables had thc same patterns and outcomes with the exception of friendship contact which switched its emotional impact from detrimental to beneficial. The last model. block 7, shows friendship contact as beneficial. but not significantly sn. meaning that the possible amplification cffel:ts could be more for detrimental emotions than bt:neficial emotions. Comparing the Beta Variables Computed Vr'hile U.1ing flappv Feclings As the Dependent \' ariahle The strengthL:n of the impal:ts upon emntions of the different variabks changeu when the dependent variable became happy lcdings. meaning that dillercnt variables have a stronger effect on some emotions and not others. The three having the strnngest benelil:ial effects were parental closeness. friends caring. and popularity. All of their betas were .12 making their impacts very similar in magnitude. In this instance it is interesting to sec that popularity and friends. caring arc about equal in impacting happy emotions meaning that in the realm of happiness friendship quality and popularity an: tied. This is meaningful hecause it points to the idea that popularity problems can be mitigated by good friends. but this is only one part of the puzzle piece as a whole and need to be taken with conjunction with the other betas in the other models. On the detrimental side. gender with a-.09, and being A-;ian and being Hispanic.: with were the strongest detrimental variables. None of these 116

PAGE 131

were discussed in depth in the literature review. nor were they the focus pf the study so their impact will simply he observed. f-riends care (.I 2 ). popularity t. I 2 }, peer trouhle ( -.05 ). male friends contact 1 .03 ). temale friLnd contact (.02). first friends at schPol (.fl I J shows friendship quality competing with popularity for the greatest impact on happy kdings and then a drop oil of magnitude followed hy peer trouble. friendship contacts and friends at sctwol. This means that the beneficial impact of friendship contact was slight even when focusing on feding joy. It is also interesting that peer trouble was so far away from the top two variables as a factor when happy feeling!as a dependent variable meaning is more dependent on social factors that hring support rather than those that tear it down. New Findings in the Models Using Depressed Feelings as a Dependent Variable The third modd. exclusively looking at the more and dejected emotions mirrored the tirst model in outcomes and the gender dif!Crences already discussed. All of the variables had the same impacts and interplay<.; they did in the other models. The meaningful difll=rencc hetween this dependent variable and the others is that in all of the tables friendship contact (male and female l was a emotionally detrimental impact at the !-ignificancc level of p<.OO I making it apparent that friendship contact"s amplification effect is more detrimental when looking specifically detrimental emotions. Comparing the Beta Variables ComfJuted Using Depressed Feelings As the Dependent Variable On page 128 paremal elosencss (-.1;\J, popularity (-.10) and more than high school (-.046) are the top beneti.:ial variables. while gender 1.18 ). twublc with peers ( .116l and school trouble (.!)90) arc 117

PAGE 132

the 111p detrimental impacts. When using deprcssin!! tmotions as thl dependent \ariahle fricmJ... caring docs not makc much of an impact compared to the other in fact <.:11ntact makes a higher impact that is detrimental solidifying the fact that an: not the important social when looking at adolescent ernntinns. This leads to the conclusion that just having friends. and even friendship quality. i., not important in fighting emntions than is avoiding trouble with peers and gaining popularity and this must he integrated into the interventions and of social theories. 118

PAGE 133

CHAPTER 6 COt\CLUSIONS From qudy a few points can be concluded. The importance of popularity. bullying and quality cannot be ignored. Adolescents emotions an: profoundly impacted by the friendship quality they have with their friends, along with the popularity or bullying they feel at school. The phenomenon of emotional arnplilication is curious and should be stLu.Jied to sec if it can be reproduced. It is also imporlant to note that different emotions are impacted in different ways by the different social factors. This is important to understand when planning interventions and doing more research. A last conclusion from this study is that gender differences are profound and can impact emotions and social variabks differently. One limitation of this study was that there were not more questions on the survey that could determine friendship quality or create a variable measuring co-rumination: v.'orking with a pre am.wered took away the ability to <:reate questions to measure exact social variables as deli ned by your study. In this study thl' many measurements of the variahks came from self-reporting about how the ldt or thought ahout something and this not hacked up by direct observation. Another challenge of studying social variables is ddining what the variables popularity. trouble with peers. and friendship quality should represent and how they should he measured. In this study popularity was measured by the respondent feeling included or excluded at school. but another study might measure popularity by how many peers lahel the respondent as cool or who they're ti"iends with within the social network at school. Trying to ligure out what components of social variables to measure and anal:ze is one of the hardest aspects of analyzing social variables impact on 119

PAGE 134

emotions. This study social variables' impacts on emotions and has discovered some very interesting findings that should be rese:m.:hcd in a more in depth manner. 120

PAGE 135

BIBLOGRAPHY Adh:r. P .. & Adler P. ( 1995 ). Dynamics of inclu-..ion and exclusillll in pn::adolcs<:cnt Social Psn:hologr Quarterly. 58(3), 145-162. Adler. P .. & Adler. P. ( 200 I). Peer pmrcr: Preudolescent Culture and Power. New Brunswi<:k: Rutgers Univnsity Press. AmberL A. ( 1<)94 ). !\qualitative study of peer abuse and its effects: Theoretical and Empirical Implications. Joumal o(Murriage and the Family. 56( I). I 19-130. Asher, S. R .. & J. G. ( llJ1N). The significance of pea relation problems in childhood. In B.H. Scheider. G. Allili .1. Nadel. & R. P. Weissberg (Eds. ), Social compete/let: in dnelopmental per.1pecti1c (pp. 5-23 ). Amsterdam: Kluwcr. Ausubcl. D. ( ;. Reciprocity and assumed reciprocity of acceptance among adolescents. a sociometric study. Socilmel ry. 16( 4 l. 339-348. Berndt. T. (1 9:52). The features and efkcts and friendship in adolescence. Child De1efopnzent. 53(6), 1447-1460. Bierman. K. t 2004). Peer reicction: Dt'l'elnpmemal proce.ucs and illtenemion stmtegie.\. New York, New York: Guilford Press. Bagwell. C.. Newcomb. A .. & Bukowski. \V. ( 1()98). Prcadolt!sccnl friendship and peer rejection as predictors of adult adjustment. Child Developmellt, 69( I). 140-J5j. BPivin. M., & Begin. G. ( 1989). Peer status and self-perceptions among early elementary sdtool children: The cas.:: or the rejected children. Child Del'elopmcnt. 60(3), 592-596. Boivin, t-.L HymeL S .. & Buk.owski. \V. l\1. ( 1995). The n1ks or social withdrav., peer rejection. and victimization by peer in predicting loneliness and depressed mood in childhood. Delelopmcnt and Psychoparlwlo,;y, 7. 765-785. Bonney. 1\1. ( 1946). A sociometric study of the relationship or some factors to mutual friendships on elcrnclllary, secondary. and college level. Sociometry. 9( I). 21-47. Bonney. l\-1. ( 1943 ). Values or sociometric studies in the classroom. Sociometry. 6(3 ), 251-254. Bowerman, C., & Kinch, J. ( 1959). Changes in family and peer orientation of children between the fourth and tenth grades. Social Forces, 37! 3 ), 206-211. Bukowski. W. M., H\lZa, B .. & Boivin, M. t 1994 ). Measuring friendship quality during preand early adolescences: The development and psychomelric properties or the Friendship Qualities Scale. Journal Social and Personal Relationships, I I. 4 71-484. 121

PAGE 136

Bulhh:k. J. ( 1993 ). Children' June and their with family and peers. Family Relations. 421 I ). 46-49. Buhrmesll:r. D. ( 1990). Intimacy or friendship. interpersonal competence, and adjustment during pn.:adolcscencc and adolesL:ence. Child Development. 6 )(4 ). I I 01-11 I I. Burles11n. B .. Delia. J.. & Applegate .. J. ( 1992) ElfeL:ts of maternal communication and chiiJren 's anJ communication skills on children's acceptance by the peer group. Family Relations. -ll (1). 204-272. Cairns. R. B. ( IY91 ). Developmental epistemology and sell"-knowledge. Towards a reinterpretation of self-esteem. In G. Greenberg & E. Tobach (Eds.), Theories of the emlution o(knowing (pp. 6Y-R6J. Hillsdale. NJ: Erlbaurn. Chapin, S. ( 1950 ). Sociometric stars as isolates. The American Journal 263-267. Coie. J.D. ( 19l)0). Toward a theory of peer rejection. In S.R. Asher & .1. D. Coie (Eds.). Peer re;ection in childlioocl (pp. 365-401 ). Cambridge. England: Cambridge University Press. Condry, J ., & Siman. M. ( 1974l. Characteristics of peerand adult-oriented children. Journal of Marriage and the Family. 36t ). 5-B-554. Corsano. P .. M;tjorano. M . & Champretavy. L. I 2006). Psychological wdl-hcing in adolescence: The contribution or interpersunal rclatiuns and experience of being alone. Adolescence, 41 ( 162 l, 341-353. Corsaro. W. (2005 ). The sociology
PAGE 137

Dion, K .. & Berscheid. E. ( 1974). J>hy!->icJl altractin:ncss and peer-perception children. Sociomctn, ). 1-12. Dishion, T. 1 .. D. W., & Crosby. L. t 1995). Antisocial boys anu their friends in early aJult.:scence: quality and interactional process. Child Development. 66. 139-151. Ennett. S .. & Bauman. K. ( 199J). Peer group structure and adolescent cigarette smoking: A social net work analysis. Journal of" Health and Social Belwtior. 3-t( 3 ). 226 236. Epstein. S. t 19lJ 1 ). Cognitive-cxpericmial self theory: Implications for developmental psychology. In !Vl. R. Gunnar & L.A. Sroufe (Eds.). Minnesota Symposia on Child P.1rcholof!,y: \iol. 23, Se(f" pmcesses and derc/opment (pp. 79-123 ). Hillsdalt:. NJ: Erlbaum. Erwin. P. ( 19"1.'). Friendship and peer relations in clzilclrcn. England: John Wiley & S\1ns. K. A. ( 1990). Girts perceptions of pet:r relationship support and stress. Journal of Early Ado/escenu'. 1 0( I ), 61}-!S!S. French. D. ( 19XX). Heterogeneity of boys: Aggressive and nonagresive subtypes. Child /)el"('lopment. 59(4 ). 976-9X5. French. D. t 1990). Heterogeneity of pt.:cr-rejt:ctcd girls. Child Derelopment, 61 ( 6). 202R-2!n I. French. D .. & Waas. G. ( 19R5 ). Bt:havior prohkms as peer-nt:glectcd and pet:r-rt:jected elementary-age children: and Child DeveloJmtellt, 56( I). 24o-252. Grady. K. E. ((979). Androgyny reconsidered. ln: J. H. Williams (EJs.J Psychology Selected ReudinRS. New York: Norton. Gauzt:, C., Bukowski., \V .. Aquan-Assce. J., & Sippola. L. (1996). Interactions hctwecn family cnvironmt:nt and friendship and associations with self-perceived wdl-being during early adolcscenccs. Child Detelopmellt. 67(5 ), 2201-2216. George, T .. & Hartmann. D. ( 1996). Fnendship network!> of unpopular. average. and popular children. Child Development, 67(5 ). 2301-2316. Goslin, -D. ( 1lJ62J. Accuracy of self perception and social acceptanct:. Socioml'tl:\'. 25(3 ), 2g3-296. Graham, S .. & Juvonen, .1. t Jl)IJ!S). Self-blame and peer victimization in rniddk school: An attribute analysis. Developmental P5ychology. 34, sg7-599. Hallinan. 1'vl. ( 1976). Friendship in open and trJditional classrooms. Sociology of"EJucatioll. 49(4). 254-265. 123

PAGE 138

Hallin:.m. l\:1. ( 1979 ). Structural dkcts on children s friendships and Social P.lyclwlof!,y QuarterlY. 421 I). 4:\-54. S. (I Y81 ). Fgo d(.vdoprncnt and peer friendship net\>.'orks. Sociology oj'Educatilm. 54( I). 51-. Harri'>. K .. Florey. F .. Tabor. 1.. Ikarrnan. P .. Jones. J., & Udry, R. (2001). The National Longitudinal Study of Adolescent Ht:alth Research Design [WWW document]. URL: http :1 /www .cpc. unc .edu/projects/addhealth/dt:'>ign. Harter. S. (I Y98). The development of In W. Damon (S..::ries Eds.) & N. Eisenherg (Vol. Ed.). Ha11dhuok qf' child p.1rclwlogr: Vol. 3. Social. emotiollul. and personality de1elopment (51 h ed .. pp. 553-618). New York: Will!y. Hartup, W. ( 19%). The company they keep: Friendship5o and thl!ir developmental signiticanc..::. Child De1elopment. 67(1), 1-13. Haynie, D .. & Payne. D. (200t1). Race. friendship networks. and \iolcnt delinquency. Criminology. 44(4), 775-805. Hogan. R .. & Mankin. D. ( 1970). Determinants of interpersonal attraction: A claritication. P.1ychological Reports. 26, 235-238. Kandel, D. ( 1978). Homophily. selection. and snciali7.ation in adolescent frientbhip. The America// Journal o(Sociology. 84( 2). Kennedy. E. ( 1995 ). Correlates of peret.:ived popularity among peers: A study of race and gender among middle school The Journal of'Nef!.ro Education. 64(2), 186-195. Kinney, D. ( 1993 ). From nl!rds to normals: The recovery of identity of among adolescents from middle school to high schnol. Soc:iolof!.y of Education. 66( I). 21-40. Kupersmidt. J. B., Burchinal, M., & Pa!lerson, C. J. ( 1995) Dcvcloprnental patterns of childhood peer relations a;; predictors of externalizing prohlcms in adolescence. Child Development. 61. 1350-1362. Ladd. G. W. ( 1990). Having friends. keeping friends, making friends, and heing liked hy peers in the classroom: Predictors of children s early school adjustment'! Child DeFelopment, 61. I 081-1 ()lJO. Ladd. G. W. ( 1983). Social networks of popular. average. and rejected children in settings. /t,ferrill-Palmer Quartaly. 29. 282-307. Laird. R .. Pettie G., Mit-e. J.. Brown, G .. & Lindsey. E. 1 1994 ). Mother-child conversations ahout peers: contributions to competence. Family 43<4 ). 425-432. 124

PAGE 139

1\'larb, J. 1 1954 ). I h.:adcrship and sociometric among adlllescenl.-;. SociometiT. 17( 4 ). l'vkGuire. J. ( 1973 ). Aggn.:ssilln and sociMY 32o-:n I. Mize. J.. &Laud, G. W. ( 1990). Toward the development of sutcessful social skills training for preschool children. Developmenral Psychologr. 26(3 J. M11reno, .1. ( 1950 J. Sociometric theory of leadership and isolation in who shall survive'! Sociometry, 1314 l, Muldoon, J. 1 1955). The concentration of liked and disliked members in groups and the relatiunship of the concentrations group cohesiveness. Sociometry. 18( I), 73-81. Nash, R. 1 1973). Clique formation among primary and secondary schnol children. The British Journal of Sociology. 24! 3 ). Newcom h. A. r: .. & IJagwell. C. L. ( 1995 ). Children's friendship relation'>: A meta-analytic review. Psychology Bulleti11, 117. Northway, M. (1946 ). Sociometry and some challenging prohllms of sm:ial relationship. Sociometry, 9W.1J. 187-19R. Olweus. D. ( 1993 ). BullyinJ!, in sclwol: What we ktWH' and n-lwt lrt' can do. Oxford: Blachvcll. Z., & Nagano-Nakamura. K. ( 1992). Maternal views on aggression: Japan and Israel. Aggres.\il'l' Behavior. 18, Paley. B .. Conger. R., &Harold, G. (2000). Parents' affect. Adolescent Cognitive Representations. and A(i.Jksctnt Social Developrnenl. lo11rnal ofl'>.farriage and the Family, 62(.l. ). 761-776. Parker. J. G .. & Asher, S. R. ( 1987 ). Peer relations and later personal adjustment: Are low accepted children at risk'? P.Hchological Bulletin. I 02, 357 125

PAGE 140

Parka J. G .. & Asher S. R. ( 1!)9:\ ). Friendship and friendship LJUality in middk childhood: Links with 1x-er group acceptance and feelings of loneliness and :-ocial dissatisfaction. IJerdotmwntal Psrdwlogr, 2'-.J. 61 1-621. J. G., Ruhin. K.H., Price. 1. M., & DeRosier, M. E. 1 llJlJ5). Peer relations. child development and adjustment: A developmental psychopathological perspective. In D. Cicchetti &D. Cohen ( Eds. I Derelr!f'l1lental Psychopar!tologr: Vol. 2 Risk. disorder, anJ aJaptmion (pp.96-161 ). 1\:ew York: Wiley. Parker. 1.. & SeaL J. tl9lJ6). Forming, losing. renewing. and replacing friendships: Applying temporal parameters to the of children's friendship experiences. Chi/J De1elopment. 67(5 ), 224g-2268. Perry. D.G .. Perry. L.C .. & Kennedy. E. ( llJ92J. Contliet and the development of antisocial behavior. In C. Shantz & W Han up iEds.J. Conflict in chi!J and adolesn:lll Je1elopnzent lpp. :101-329 ). Cambridge. England: Cambridge Press. Perry. D. G .. Perry. L. C .. & R. J. ( Sex differences in the consequences children anticipate for aggression. Developmentol Psvchologl'. 25. :112-319. Potashin. Rev a. ( ll)46). A sociometric study of children s friendships. Sociometry, 9( I). 4S-70. Putallaz. M .. & Gottman .1. ( 1981). An intcrn:Hional model of children's entry into peer groups. Child /Jelclopnwnt. 52(:1), 9R6-994. Reger. R. ( 1962 l. An to integrate a group isolate .lo11rnal of' Educational Sociologr, 36(4 ). 154-15S. Rose. A., Carlson., W .. & Walll'r. E. (2007). Prospective associations of co-rumination with Friendship and emotional adjustment: Con,idering the socioemotional trade-oils of co-rumination. Developmental 43( 4 ). 1019-1031. Sehollicld. J.. & Whitley. Jr .. B. ( 19H3 ). P.:er nominations vs. rating scak measurement of children's peer preferences. Social Psrchologr Quarterly. 46(:1 ), 242-251. Schutte. 1.. & Light, J. ( 197K). The rclatiw importance of proximity and status for friendship choices to social hierarchie5>. Social Psychology. 41 (3 ), 260-264. Schwendingcr. H., & Schwendinger. 1. ( 1997 ). Charting subcultures at a frontier of knowledge. The British .lou mal of Sociology, 4H( I). 71-94. Sheldon. P. ( Jl)59). Isolation as a characteristic of highly gifted children. Journal ofEJucafional Sociology. 32(5 J. 215-221. 126

PAGE 141

Shaver. P., & C. ( 198Q). lonely, fall in love: Perspectives from allachmcnt In: M. Hojat & R. Crandall (Elis.). Lo11elincss: Theon, Rcsearclt, and ApplicaTions. Newbury Park, CA: Sage Shoohs, N. (1946). Sociumetric test in the Socinrttefly, 9(2/3). 145-146. Shrum, W .. & Check. Jr.. N. ( 1987). Social structun: during the school years: Onset of the degrouping process. /\merican Socio/op,ical Rel'it'li', 52(2). 218-223. Smucker. 0. ( 1974 L Measurement of group tension through the usc of negative sociometric data. Sociometry. I()( 4 ). 376-383. Solomon. R. W .. & Wahler. R. G. ( 1971). Peer reinforcement control or classroom problem behavior. lou mal Behavior Analrsis, 6. 49-56. Steinberg. L., & Silvaberg, S. ( 1986). The vicbsitudes of auwnomy on adolescence. Child Dnelopment. 57, 841-851. Sydner. 1\L & Uranowitz. S. W. (I Sl78). Reconstructing the past: Some cognitive consequences of personal perception. Jouma! of Personality and Social Psrcholilgy, 36. 1)41-950. Trcmhaly, R. E .. Masse, L.C.. Vi taro. F .. & Dobkin. P. L. (I Q95 ). The impact of friends' deviant behavior on early onset of dclin4uenc)< Longitudinal data from 6 to 13 years of age. Del'elopment and P.lyclwparlwlogy, 7, 649-667. UJry. J.R. (2003) The Natitmal Longitudinal Study of Adolescent Health (Add Htalth) Waves I &II. 1994-1946: \Vavl.! III. 2001-2002 [machine readable Jata tile and documentation]. Chapel Hill. NC: Carulina Population Centl.!r. University of North Carolina at Chapel Hill. Wallerstein. J. S .. & Kelly. J. B. ( 1980). Surri1'i1tg rite breakup: llrm children and tJarents cope with diForce. New York: Basic Books. Wheeler. D. ( 1961 ). Popularity among adokscents in Western Australia and in the United States of Aml.!rican. Tlze School Review, 69( I), 67-81. Windle, Michael. ( 1994 ). A study of friendship charactcrisLiLs and problem behaviors among middle adolcscl.!nts. Child Devl'lopment. 65(6). 1764-1777. Yamagu..:hi. 1\:. ( 1990). Homophily and social distance in the choice of multiple friends: An analysis based on conditionally symmt:tric log-bilinear association model. Joumal of the American Statistical Association, 85(410),356-366. Zuckerman. E .. & Jost. J. (2001 ). What makes you think you're so popular? Self-evaluation maintenance and the subjective side of the friendship paradox''. Sclwal Psychology QuarterlY, 64(:; ). 207-223. 127