Friendship Quality and Peer Attachment as Predictors of Adolescents'''' Subsequent Academic Achievement

AbstractThis study investigated the hypothesis that peer influences in adolescence, specifically friendship quality and peer attachment, predict improved academic achievement over time..

Running Head: ACADEMIC ACHIEVEMENT

Friendship Quality and Peer Attachment as Predictors of Adolescents' Subsequent Academic

AchievementElizabeth A SpavinsDistinguished Majors ThesisUniversity of VirginiaMay, 2007

Advisor: Joseph P Allen

Second Reader: Nicholas D Reppucci

AbstractThis study investigated the hypothesis that peer influences in adolescence, specifically friendship quality and peer attachment, predict improved academic achievement over time

Participants included a diverse sample of 145 adolescents (M age = 13.3 years) In support of the

hypothesis, participants who reported greater levels of peer attachment at age 13 displayed improved academic achievement at age 17, controlling for academic achievement at age 13 Greater levels of certain aspects of friendship quality, such as validation and caring, as reported

by participants at age 13, also predicted greater academic achievement for participants at age 17

Friendship Quality and Peer Attachment as Predictors of Adolescents' Subsequent Academic

Achievement School and “school-related activities” occupy more than one-third of the “typical” American student’s day (Larson & Verna, 1999, as cited in Steinberg, 2005, p 200).Beyond a commitment of time, school success is pivotal to outcomes later in life Academic achievement predicts decreased juvenile delinquency (Chavez, Oetting, & Swaim, 1994), while educational attainment is correlated with increased future income (Murphy & Welch, 1995) and decreased substance use (Swaim, Beauvais, Chavez, Oetting, 1997) Thus, it is important to investigate what factors contribute to and detract from academic success for adolescents

Peer influences and interactions are likely candidates for factors that are important in adolescents’ academic achievement Previous research has found evidence for the importance of peer relations in a variety of areas of adolescent functioning Allen, Porter, McFarland, Marsh, and McElhaney (2005) report that adolescents who were well-liked by many peers displayed higher levels of ego development and secure attachment, as well as better interactions with their mothers and best friends (p 747) Popularity was also linked to “minor levels of…delinquency” (p 747) but less hostile behavior toward peers (Allen et al., 2005, p.757) Furthermore,

adolescents’ susceptibility to peer pressure from their close friends predicted future responses to

negative peer pressure, decreases in popularity, and increased depressive symptoms Susceptible teens also rated themselves as less competent in their close friendships (Allen, Porter, &

McFarland, 2006).Peer pressure susceptibility was also cross-sectionally correlated with deviant behavior and substance use, especially if a close friend had experimented (Allen et al., 2006) According to Santor, Messervey, and Kusumakar (2000), peer pressure and peer conformity are better than popularity for predicting antisocial behavior such as substance abuse, delinquency,

and worse school achievement Berndt (1979) finds that peer conformity increases from third grade until the ninth grade, at which point it begins to decline.

To be influenced by one’s peers does not uniformly predict negative outcomes, however Research has shown that the tendency to seek advice from peers over parents had no long-term consequences for early adolescents (Fuligni, Eccles, Barber, & Clements, 2001) In addition, attachment style influences the link between friendship and negative outcomes Non-dismissing attachment style has been found to moderate the link between general friendship quality and a teen’s delinquency (McElhaney, Immele, Smith, & Allen, 2006) It has also been demonstrated that adolescents’ friends’ behaviors have positive correlates Prinstein, Boergers, and Spirito (2001) report that teens were less likely to demonstrate antisocial behaviors, specifically violenceand substance use, when they had high proportions of friends who demonstrated prosocial behavior Thus, depending on the adolescent, the characteristics of the relationship, and the extremeness of the susceptibility, teens may be differentially affected by their peers

Due to the numerous findings of consequences of peer relationships, the characteristics ofadolescent friendships have been studied from various angles Variability in friendship and groupstatus is normal (p 1343), and friendships become more stable with age (Cairns, Leung,

Buchanan, & Cairns, 1995).The finding of friendship variability implicates that at earlier ages, adolescents may have multiple salient friendships As a result, studying friendships in general in addition to studying one specific friendship may reveal more information about peer influences For both specific friendships and friends in general, further research needs to be done to

determine whether these early friendships, although shifting, have long-lasting impacts Such research would complement existing research on adolescent friendships, which does not examinethe effects of friendship over significant spans of time

Dishion found that one-on-one friendships between antisocial adolescent boys were characterized by low quality and satisfaction and the presence of coercive behavior These findings did not suggest that such friendships were characterized by a lack of positive behavior (Dishion, Andrews, & Crosby, 1995) It is plausible that the presence of these negative

friendship factors has a negative impact on overall adolescent functioning Further research is needed to examine the link between friendship qualities and areas of functioning that may suffer

as a result of antisocial behavior, such as academic functioning Previous research has found homophily, the inclination to associate with similar people, characterizes adolescent

relationships, specifically in elements of antisocial behavior such as delinquency (Chavez et al., 1994) and drug use (Kandel, 1978) Homophily in adolescent friendships may apply to other domains of behavior as well, such as academic success Academic achievement is a likely candidate for similarity to one’s friends given that most adolescents attend school with their peers

Beyond their relationships with these others areas of adolescent functioning, peer

influences have been shown to predict academic functioning in adolescence One avenue in which research in the field of peer relations and academic achievement has been conducted has been the effect of peer orientation on academic achievement Evidence shows that peer

orientation may have an effect on academic achievement at the farthest end of the spectrum Extreme orientation toward peers, involving willingness to ignore parents’ rules, schoolwork, and one’s own skills for the sake of popularity, has been linked to greater problem behavior in seventh, tenth, and twelfth grade and lower academic achievement in seventh and tenth grade (Fuligni et al., 2001) The same study found evidence of homophily in antisocial behavior, in thatthe proportion of an adolescent’s friends who drank alcohol, used drugs, and skipped class was a

strong predictor of problem behavior and weaker predictor of academic achievement However, the proportion of a teen’s friends with academic orientation was not predictive on its own of academic achievement (Fuligni et al.) This could be evidence for a lack of peer influence on academic achievement, or for a need to investigate other aspects of friendships beyond similarity

in academic success Fuligni et al.’s research primarily focused on the effect of negative aspects

of an early adolescent’s friendships on academic achievement Friendships in adolescence are not only characterized by negative aspects, however Further research is needed on the positive aspects of friendship and their effects on academic achievement

The relationship between group norms and academic success has also been investigated Chen, Chang, and He (2003) examined the association between peer factors and academic achievement of Chinese children Peer groups of Chinese children are known to be homogenous with respect to academic achievement, thus the goal of the investigation was to determine the role of these group-wide characteristics, if any, in the academic and social success of individuals within the group The results showed that teens’ academic achievement was positively associatedwith popularity, measured by sociometric nominations Group norms for academics played the role of mediator between social success and academic achievement (Chen et al., p.722) Wentzel and Caldwell (1997) also found that group membership was a predictor of grades over time in a sample of early adolescents These results are encouraging because they provide evidence of peerinfluence bolstering positive teen outcomes, namely school success These results do not go so far as to determine the mechanism by which group norms incite this effect

The influence of peer relations on academic achievement appears equally strong across both genders Chen et al (2003) found no gender effects on the relationship between academic achievement and social functioning Similarly, Fuligni et al (2001) found no gender variation in

the long-term effects of extreme peer orientation Peer influences on academic achievement thus may not differ significantly by gender

Chen et al (2003) found that academic achievement-oriented groups were more socially adaptive, meaning prosocial behavior was encouraged along with academic performance in thesegroups The opposite was true in groups that were lower in academic achievement In these groups, negative behavior and low achievement were endorsed (Chen et al., p 722) This link between prosocial behavior, academic achievement, and peer success has been examined further.Wentzel (1991) found that for early adolescents, socially responsible behavior predicted both academic achievement and social abilities In a later study, Wentzel and Caldwell (1997) found that those aspects of friendships that were related to academic achievement were connected by the common basis of prosocial behavior These findings point to one specific antecedent of academic and peer success

Research concerning the impact of peer orientation and group norms on adolescents is one lens with which to examine peer relationships From a different angle, previous research has investigated the impact of peer rejection on academic achievement Research shows that peer rejection at all points of time (both recent and in the past) has negative effects on school

attendance, as well as externalizing and internalizing behavior problems (DeRosier, Kupersmidt,

& Patterson, 1994) This finding supports the idea that one’s peers are salient in academic

success because they are ever-present during instructional and testing time Peer rejection could cause a fear of participating in class or general dread of school, both of which would impede academic success Further, peer rejection need not be recent to be harmful (DeRosier et al.) Complementing these findings that peer influence can have negative academic effects, the

current research aims to uncover the relationship between positive peer effects and academic success

Interestingly, adolescents neglected by their peers have a different outcome than

adolescents rejected by their peers Wentzel and Asher (1995) find that socially neglected early adolescents may fare better than average in terms of academic achievement This finding

highlights the importance of further breaking down the elements of adolescent friendships to determine which are salient to academic success If peer rejection can prevent a student from wanting to go to school, and peer neglect does not cause academic failure, then it should be investigated whether positive peer relationships can compel students toward academic

adolescents’ views of their friendships is important because these views target more closely what

is perceived by the adolescent as the quality of their friendships overall Previous research has demonstrated the importance of friendship groups in antisocial behavior conformity (Berndt, 1979) and cigarette use (Urberg, Degirmencioglu, & Pilgrim, 1997) This research will

investigate the role of several qualities of friendships in general on academic achievement

Close friendships may matter even more than friendships in general when predicting adolescent academic achievement Adolescents have different relationships and interactions withtheir closest friends than with their group of friends in general Thus, factors in the closest

friendship may be more salient to adolescent functioning overall, or may impact the adolescent

in different ways than friends in general The differential influence of the closest friend versus the friendship group is supported in the literature Urberg et al (1997) found closest friends influence adolescents in different ways than teens’ friendship groups in terms of cigarette

smoking and alcohol use Urberg et al also posited that academic achievement is a more

“sensitive” respondent to friendships than substance use Based on these findings and logic, it is important to investigate the possibly differential effects of the friendship group and closest peer

on academic achievement

Evidence in these closest friendships of prosocial behavior characteristics linking

individuals to positive academic achievement would also support and extend previous research

of Wentzel and Caldwell (1997) Beyond the hypothesis of prosocial behavior linking peer success with academic success, the examination of close friendships may elucidate more factors that bring about academic success One’s close friendships may be more important than overall peer status in academic functioning, and may link academic achievement to peer success in different ways than general popularity does It is important to investigate both types of

friendships so that these effects are not inappropriately lumped together and can be most clearly understood

The current research seeks to answer several questions regarding peer effects and

academic achievement One aim of this study is to assess peer influences on adolescent academicachievement during more of this critical developmental period Although early adolescence is an important time period for examining peer effects, it is not sufficient The current research

assesses participants’ academic achievement over the course of five years, because it is also important to examine impact over time Little research has been conducted to determine whether

specific peer factors in early adolescence affect academic outcomes beyond two years If a causalrelationship is to be inferred, results beyond a short length of time must be found Without such evidence, it is difficult to determine whether the effect of friendships on academic achievement

is a temporary phenomenon that only holds in early adolescence or a more persistent

relationship Longitudinal evidence of the effects of early friendship elements on later academic achievement would strongly support the overall argument that peer factors are important for academic success

Other studies measured peer acceptance using sociometric nominations to assess

objective peer acceptance However, many aspects of friendship quality are personal and thus perhaps better measured subjectively The use of teens’ self-report of friendship quality is thus

an important departure from the previous literature The current research seeks to determine whatself-report friendship qualities are those elements of adolescent friendship that determine

academic success

Based on these previous findings, teens with better friendships are expected to have betteracademic outcomes over time First, it is expected that higher levels of peer attachment will predict higher levels of academic achievement over time Second, it is expected that higher levels

of friendship quality with the close friend will also predict higher levels of academic

achievement over time These specific links have not been previously assessed, and are

important to consider if the complex relationship between peer influence and academic

achievement is to be understood In conducting this research, interactions between gender and peer effects on academic success will also be investigated It is hypothesized that there will be noeffects of gender on the relationship between peer factors and academic achievement

Participants

Participants in this study were part of a larger longitudinal investigation of adolescent development in familial and peer contexts Participants included 145 seventh, eighth, and ninth

graders who had baseline data available for academic achievement at age 13 (M age = 13.3, SD =

0.62) The sample was racially and socioeconomically diverse Of the participants included, 88 identified themselves as Caucasian (61%), 41 as African American (29%), and sixteen as being from other ethnic groups (10%) Adolescents’ parents reported a median family income in the

$40,000 – $59,999range

Participants nominated their closest same-gender friend to be included in the study each year Close friends were defined as, “people you know well, spend time with and who you talk toabout things that happen in your life.” In all cases, adolescents were able to name at least one close friend using these criteria Participants completed measures about their relationship with

this close friend The mean age of close friends was 13.4 years (SD = 0.85) Of the close peers

who reported gender, 69 were male (48%) and 74 were female (52%) The self-identified racial background of the close friends was 63% Caucasian (89 participants), 29% African-American (41 participants), and 8% from other ethnic groups (12 participants) Close friends reported that they had known the adolescents for an average of 4.1years(SD = 2.93) at age 13

Formal attrition analyses revealed differences between adolescents who did versus did not have data for academic achievement at age 17 Adolescents who did have academic

achievement data at age 17 differed from those who did not have these data in age, family income, level of academic achievement at age 13, and minority status Participants who did not have academic achievement data at age 17 were older, had lower median family incomes, lower

levels of academic achievement at age 13, and were more likely to be minorities than participantswho did have these data at age 17 Adolescents who did versus did not have academic

achievement data at age 17 did not differ in gender

Adolescents were recruited from the seventh and eighth grades at a public middle school drawing from both suburban and urban neighborhoods in the Southeastern United States One cohort of eighth graders was included, and two different cohorts of seventh graders were

included in successive years The school was part of a system in which students had been

together as an intact group since fifth grade Students were recruited through an initial mailing toall parents of students in the school along with follow-up contact efforts at school lunches Adolescents who indicated they were interested in the study were contacted by telephone Of all students eligible for participation, 63% agreed to participate either as target participants or as peers providing collateral information Adolescents provided informed assent and their parents provided informed consent before each interview session Interviews took place in private officeswithin a university academic building

Procedure

Study participants were assured that all information would be kept confidential and that their parents would not be informed of their answers Data were protected by a confidentiality certificate issued by the United States Department of Health and Human Services, which

protected information from subpoena by federal, state, and local courts Transportation and child care were provided if necessary

Measures

Peer Attachment (Age 13) The Inventory of Parent and Peer Attachment (IPPA;

Armsden & Greenberg, 1987) was used to measure adolescents’ perceptions of their friendships

at age 13 Participants rated 25 items relating to the level of trust, communication, and alienation with their friends on a 5-point Likert scale Sample items included: “My friend accepts me as I am” (trust); “My friend helps me talk about difficulties” (communication); and “My friend understands me” (alienation) The responses were summed (reversing the alienation items) to create the adolescent’s total attachment to friends score Cronbach’s alphas measuring internal consistency for the three subscales were 91, 88, and 86, respectively, and 92 for the composite,

or total attachment, score This questionnaire has been shown to have good test-retest reliability and has been related to other measures of family environment and teenagers’ psychological functioning (Armsden & Greenberg, 1987)

Friendship Quality (Age 13) The Friendship Quality Questionnaire (FQQ; Parker &

Asher, 1993) was used to measure adolescents’ perceptions of the quality of their friendship withtheir closest friend at age 13.Participants rated 40 items measuring six different domains of their

relationship with their closest friend on a 5-point Likert scale The validation and caring

subscale, hereafter referred to as validation, measured caring qualities of a friendship, and had 10

items The conflict resolution subscale measured aspects of the friendship related to solving problems, and had 3 items The conflict and betrayal subscale measured how much disagreement took place in the friendship, and had 7 items The help and guidance subscale assessed how much the dyad helped one another, and had 9 items The companionship and recreation subscale assessed the activities the friends did together, and had 5 items The intimate exchange subscale

measured disclosure between the participant and close peer, and had 6 items Sample items from each domain included: “We make each other feel important and special” (validation); “We talk about how to get over being mad at each other” (conflict resolution); “We fight a lot” (conflict and betrayal); “We help each other with school work a lot” (help and guidance); “We go to each

others’ houses” (companionship and recreation); and “We tell each other private things”

(intimate exchange) Cronbach’s alphas measuring internal consistency for the six subscales were 90, 73, 84, 0.90, 0.75, and 0.86, respectively

Academic Achievement (Ages 13 through 17) An annual, weighted grade point average

(GPA) was used to measure participants’ academic achievement This average was calculated at each grade level using scholastic records from the participants’ high schools The initial rating scale was zero to four points per class, with a grade of A worth four points, a B worth three points, a C worth two points, a D worth one point, and an F worth zero points Adjustments weremade in the calculations depending on course level Advanced or honors courses were worth one extra point; applied or special courses were worth half a point less; failed courses were factored

in as worth zero points Grades were also adjusted so the number of credits awarded by the school for each course was taken into consideration Thus, one semester-long course was worth half as much as one full-year course All classes taken, electives and core subjects such as English, math, and social studies, were included in the GPA calculation

Results

Preliminary Analyses

Means and standard deviations for all demographic factors and variables examined in thisstudy are presented in Tables 1 and 2 Initial analyses examined the role of gender and family income on the primary measures Family income had a slight correlation with primary measures

in the study in several instances As a result, family income was entered into analyses whenever

it was significantly related to any other variable in the analysis Because adolescent-peer

groupings were automatically segregated by gender (participants could only bring in gender friends), gender was entered routinely as a covariate in all analyses Possible moderating

same-effects of all of these demographic factors on each of the relationships described in the primary analyses below were also examined No such moderating effects were found beyond what would

13.4130.85Gender

12 (8.4%)

Conflict & Betrayal -25.17 3.74

Validation & Caring 40.50 6.69

Help & Guidance 32.51 7.08

Correlational Analyses

For descriptive purposes, Tables 3, 4, and 5 present simple correlations among all

primary constructs These analyses indicate numerous simple correlations between elements of adolescents’ friendships and academic achievement, which are explored further in the next section These analyses also indicate that the elements of adolescents’ friendships being

considered are moderately highly correlated, and thus provide measures of different aspects of the same broad construct of friendship quality

GPAAge 14

GPAAge 15

GPAAge 16

GPAAge 15

Friendship Quality as Measured by the Friendship Quality Questionnaire Correlated with Scholastic Achievement

Friendship

Quality

GPAAge 13

GPAAge 14

GPAAge 15

GPAAge 16

GPAAge 17