The Effect of Family Structure and Family Functioning on Adolescents' Perceptions of Intimate Time Spent with Parents, Siblings, and Peers

25

0047-2891/97/0200-0025$12.50/0 O 1997 Plenum Publishing Corporation

The Effect of Family Structure and Family Functioningon Adolescents' Perceptions of Intimate Time Spentwith Parents, Siblings, and Peers

Barry J. Falion1 and Terry V. Bowles2

Received February 14, 1995; accepted February 23, 1996.

In previous research, family structure variables have been operationalized asfamily size, birth order, sibling spacing, and sibling density. These structurevariables have been linked to parental strictness, reasonableness, andsupportiveness [J. S. Kidwell (1981), "Number of Siblings, Sibling Spacing,and Birth Order: Their Effects on Perceived Parent-Adolescent Relationships,"Journal of Marriage and the Family, Vol. 43, pp. 315-333]. Other researchhas drawn attention to the link between the amount of time adolescents spendwith parents and peers and the influence of family relations variables —intensity, duration, and frequency of conflicts. [R. Montemayor (1982), "TheRelationship Between Parent-Adolescent Conflict and the Amount of TimeAdolescents Spend Alone and with Parents and Peers," Child Development,Vol. 53, pp. 1512-1519]. More recent research has related family structure(family size, sibling spacing, and gender) effects to the amount of timeadolescents spend with parents, a measure of relationship quality, and a setof measures of discipline [R. A. Richardson et al. (1986), "Parent-ChildRelationships in Early Adolescence: Effects of Family Structure," Journal ofMarriage and the Family, Vol. 48, pp. 805-811]. To date, the questions asked byresearchers have not simultaneously explored whether it is the family relations or thefamily structure variables that contribute most powerfully to the amount of time thatfemale and male adolescents spend with family members and peers. In this researchadolescents were asked to report the amount of time they spend with their mother,

1Senior Lecturer, Department of Psychology, University of Melbourne, Melbourne, Australia.Received his Ph.D. in social psycholody from State University of New York at Buffalo.Research interests include interpersonal relations and the work-family interface.

2Doctoral candidate, Department of Psychology, University of Melbourne, Melbourne,Australia. Received Post Graduate Diploma in adolescent and child psychology from theUniversity of Melbourne. Research interests include self-motivated change, time orientation,work-family relations, and adolescent academic achievement.

Journal of Youth and Adolescence, Vol. 26, No. 1, 1997

father, siblings, and peers, and to respond to a questionnaire that has threefactors that define family functioning: Parenting Style (democratic decisionmaking), Intimacy, and Conflict [P. Noller et al. (1992), "Parent and AdolescentPerceptions of Family Functioning: A Comparison of Clinic and Nonclinic Family,Journal of Adolescence, Vol. 15, pp. 101-115]. The analyses revealed that complexand interpretable family structure and family functioning factors differentiallyinfluence whether males and females spend time with family members and peers.

INTRODUCTION

The purpose of the present study was to investigate whether family struc-ture variables and/or family functioning variables best predict the amount ofintimate time adolescents spend with family members and peers. Researchusing family structure variables is based on an assumption that such variableshave an effect on the interactions between family members. A further assump-tion that is often made is that more interaction between parents and theirchildren is better throughout childhood and adolescence. The fewer the num-ber of children, or the more spaced the children are in age, the greater thepotential, and therefore probability of experiencing actual contact between parentsand children. Research involving such family structure variables began withinvestigations into the influence of birth order and moved to the more complexconstruct of sibling density (Kidwell, 1981). Richardson et al. (1986) has linkedfamily structure variables with the amount of time adolescents spend with fam-ily members and the quality of that time. Montemayor (1982) found complexrelations between the amount of time adolescents spent with family membersand peers and the influence of the family functioning variable of conflict. Inneither study were the family structure and functioning variables used concur-rently to ascertain which had the greatest influence on time with others.

Family Structure

Research findings using family structure variables are compelling be-cause of their simplicity and face validity. Birth order is reported to berelated to a variety of other variables. For example, first borns have moreexpected of them, are confronted by parents more often, are controlledmore coercively than later borns, and are more aggressive toward youngersiblings (Ernst and Angst, 1983). Pfoutts (1980) reported the effects of birthorder and sibling spacing (defined as the average age difference betweenthe target adolescent and his or her siblings) on a number of family rela-tions variables and IQ. Pfoutts (1980) found that in families with two sons,

Fallon and Bowles26

the closer they were in age the more positive their relationship with theirbrother and father. Koch (1955) found that children in closely spaced fami-lies were influenced more, both positively and negatively, by siblings thanby their parents. Koch also reported that as spacing increased parental in-fluence increased and sibling influence decreased.

With 1700 males from a national sample, Kidwell (1981) investigatedthe family structure variables of sibling spacing, birth order, sibling density(defined by Kidwell, 1981, as the reciprocal of the number of years betweenthe target child and each sibling), and sex on perceptions of parental puni-tiveness, reasonableness, and supportiveness. Kidwell maintained that siblingdensity was important because it was a combination of sibling spacing andbirth order that had previously been shown to influence parent-adolescentrelations. She used the reciprocal of the spacings to give decreased weightto siblings who were spaced further apart on the time line. Using a seriesof regression equations to predict adolescent's perception of parental puni-tiveness, reasonableness, and supportiveness, Kidwell reported that havinglarge numbers of siblings increased the perceptions of parental punitivenessand decreased the perceptions of parental reasonableness. Males spacedclose together (less than 12 months) and distant (more than 4 years apart)reported more positive relations, with the optimal spacing being 4 to 5 years.Middle boms reported being more punitively treated and less reasonablyand supportively treated than older and younger borns.

Two elements of Kidwell's (1981) research bear further attention. Sib-ling density was not included in all analyses as it did not account for a sig-nificant amount of variance and that sibling spacing and number of siblingswere adequate measures. This explanation is counterintuitive in that, whileit may be qualitatively different, if spacing and number of siblings explainvariance so ought density, as it is the mathematical combination of both.Second, Kidwell's findings were based only on males. Taken together thefindings and arguments regarding density would suggest that it is at leastimportant to ascertain whether density is an influence for both males andfemales and whether Kidwell's (1981) original finding can be replicated.

Time

Some research has examined the relationship between family function-ing and the time adolescents spend with parents and peers (Montemayor1982, Richardson et at, 1986). Prior to considering their findings, it is im-portant to substantiate the use of time as an appropriate dependent variablein adolescent research.

Family Structure and Family Functioning 27

In previous research, time has been used in a way that correspondsclosely to the Newtonian temporal framework of time as objective, con-tinuous, universal, and reductive (Slife, 1993). Piaget's (1969) researchshows that by the age of 11 children are capable of reversibility, and ofmaking and recalling an account of a rational succession of events, basedon operational modes of thought, that are no longer intuitive or as ego-centric as are the processes used in earlier developmental stages. Harner's(1982) summary of psychological research into perceptions of past and fu-ture perspectives of time in adolescents and adults reports similar findingsto those of Piaget (1969). She maintains that by the end of the middleyears of childhood, approximately 6-11 years of age, children have mas-tered the basic system of time relations and the linguistic structures thatare necessary for meaningful articulation of events from the past, or in thefuture. Taken together, the conclusions of Piaget (1969) and Harner (1982)suggest that it is appropriate to ask adolescents after the age of approxi-mately 11 to recall their perception of the quality and duration of time inthe company of others. Further, these findings would suggest that suchhindsight recollections would be expected to be as accurate as other selfreport measures.

Montemayor (1982) used verbal self-reports during telephone inter-views on three randomly selected evenings approximately a week apart toinvestigate the link between the time adolescents spend with parents andpeers and parent-adolescent conflict. Montemayor found that adolescentshad three times more conflicts with their mother than their father. Femaleadolescents had more conflicts than males. The majority of the conflicts thatfemales had were with their mothers, were longer and more intense thanthey had with fathers, as well as longer and more intense than those reportedby males (Montemayor, 1982). The respondents were also asked about timeduration, operationalized as "Doing something with someone else, and feel-ings as if you and the other person were together and not just at the sameplace at the same time." Montemayor (1982) drew three conclusions fromthe research. First, adolescents spent approximately equivalent amounts oftime with parents and peers, but used their time differently with parentsand peers. Second, for females there appeared to be competition betweenparents and peers, while for males there appeared to be competition be-tween being with parents or being alone. Finally, for both males and females,disagreeable relations with mothers corresponded with greater time withmother and not with peers.

Richardson et al. (1986) gathered data from young adolescents fromintact families with at least two children to explore the link between familystructure, family functioning, and time spent with parents. Adolescents whowere further in age from their siblings spent more time with their fathers

28 Fallon and Bowles

than those who were closely spaced. Consistent with previous research(Kidwell, 1981), adolescents with a mean sibling spacing of less than fiveyears indicated that fathers were more strict than those with wider spacing.Further, when sibling spacing was wider, adolescents perceived their par-ents' disciplinary behavior as more fair. Sibling spacing did not influenceperceptions of mother strictness, but males rated their mothers as morestrict than did females. The respondents were also asked to indicate howmuch time they spent with parents together and individually. Females spentmore time with both parents present and mothers alone than did males,but there was no significant sex difference for time with fathers. This effectwas mediated by family size, in that females spent more time with theirmothers as the family size increased with little change for males as familysize increased.

Age Differences

Montemayor (1982) pointed out that parents and peers have differentroles or functions for adolescent males and females, but did not investigatewhether the function changes as adolescents age. Research has shown thatthe function of the peer group changes with age and gender. Brown et at.(1986) found that females considered peer affiliation more important thanmales. The positive reasons for seeking affiliation were enhancement ofsocial support, friendship, and social activities. The negative reasons asso-ciated with peer affiliation related to the emphasis placed on conformityand unnecessary commitment to friendships.

Given these findings, it is inappropriate to view relationships betweenadolescents and their parents and adolescents and their peers as an"either-or" choice. In line with the findings of Montemayor (1982), theexplanation of Noller and Callan (1991) provides a more accurate descriptionof the parent-adolescent-peer triad. Rather than the parents and peers beingin competition it may well be that they complement each other. Such adescription of interaction stresses the interrelatedness of the relationshipbetween the adolescent, family, and peers in a mutually rewarding andaffirming fashion. This conception of family and peer interaction is supportedby Sebald and White (1980) and Wilks (1986), who found that while parentswere consulted for issues such as work orientations and future orientations,peers were consulted for more immediate considerations. In short, one setof interactions did not deny the other. It fulfilled a different purpose — apoint that has been made previously by Biddle et al. (1980) and Montemayor(1982).

29Family Structure and Family Functioning

Family Functioning

Accepting the notion of functional differences in the interaction betweenadolescents, parents, and peers does not deny that adolescents may experienceinconsistent levels of contact with family members and peers that may be re-lated to the combined influence of family structure and family functioning.This draws into question the Kidwell (1981) proposition that family structurevariables directly influence family relations (parenting practices) and in turnadolescent behavior. The amount of time adolescents spend with family mem-bers and peers (Montemayor, 1982; Richardson et al,, 1986) would be onesuch behavior. The model suggests that greater amounts of time correspondto more positive family relations, after the influence of family structure factors,such as the number of siblings or density of siblings is taken into account. Itis possible that both family relations (family functioning variables) and familystructure variables influence relationship quality simultaneously. In the presentresearch the three-factor (intimacy, parenting style, and conflict) Family Func-tioning Scale (Noller et al, 1992) will be used to measure family relations.

Quality Time

Quantity of time with others may not meet the needs of the adolescents.Quality time with others is not dependent only on quantity. It is dependenton having opportunities to interact with someone with whom there is atrusting or intimate type of relationship. Simply spending time in the companyof others may not be satisfying the need for assertive care, or provide theknowledge that care is available, if desired. In his research, Montemayor(1982) used amount of time, qualified as not just being in the same placebut "feeling as if you and the other person were together," with parents, peers,and alone as the variable of interest. While intimacy has been considered animportant dimension of relationships between adults (Walker and Thompson,1983), it is not a term that is used very often in referring to adolescent-parentrelationships. Of interest in the present study was not simply the amount oftime adolescents spend with parents, siblings, and peers, but the amount oftime that would be characterized by intimacy and trust — the type of timeduring which issues of importance to either party to the interaction can beexplored and discussed. Intimacy has been variously defined as comprisingelements of affection (Berger and Calabrese, 1975), altruism, solidarity(Levinger and Snoek, 1972), a heightened sense of the importance of therelationship (Huston and Burgess, 1979), openness (Altmann and Taylor,1973), and commitment (Huston and Burgess, 1979).

30 Fallon and Bowles

Of interest here is the amount of time adolescents spend with theirfamily members and their peers when they feel free to talk about thingsthat are especially important to them, and to ask questions about and dis-cuss things they would not want any other person to know, or to knowthey were speaking about. In this type of interaction there is a directlyintimate element. Hall et al. (1995) in their research had children providefour self-report items concerning time with their parents. One of these issimilar to that used in the present study — time for having private talks.While our proposed operationalization of time spent with parents, siblings,and peers is similar to Montemayor's (1982) operationalization, with regardto the intimacy dimension, it was found in some pilot work with youngadolescents that the proposed definition was more easily understood. Timein the current study was defined in this way because, with the exceptionof Montemayor (1982), no previous research was found that examined thisimportant dimension of the quality of adolescent's relationships with sig-nificant others.

Research Questions

We were primarily interested in whether the amount of "intimate"time adolescents spend with their mother, father, siblings, and peers canbe best explained as a function of the family's structure, functioning, orwhether it is a combination of some elements of family structure and familyfunctioning. Second, as suggested by previous research, we were interestedin whether there are differences in the ways males and females are influ-enced by family structure and family functioning variables. Specifically, itwas hypothesized that, in line with Richardson et al. (1986), family size willbe positively related to the amount of time adolescents spend with familymembers, particularly mothers.

It was also hypothesized that sibling density will influence the amountof time adolescents spend with family members and peers. That is, low den-sity will be related to more time with family members and less with peersand the reverse for high sibling density. Extending Montemayor's (1982) re-search by defining time more specifically, we expect some differences be-tween the age groups of adolescents in the amount of "intimate" tune spentwith family members and peers.

Family Structure and Family Functioning 31

METHOD

Respondents

Two hundred and ninety-nine adolescents, 144 females (48.16%) and155 males (51.84%) who were attending secondary schools in the Mel-bourne, Australia, metropolitan area participated in the study. The samplewas limited to adolescents with one or more siblings and two parents/caregivers. The schools were in middle class suburbs. In terms of race, all ofthe respondents would be described as white. The respondents were fromone coeducational and two single-sex (one male, one female) metropolitanhigh schools in Melbourne. The mean age of the 155 males that partici-pated was 14.26 years (SD = 1.72) and the mean age of the 144 femalesthat took part was 14.11 years (SD = 1.87). There were three age groups,a young group comprising 58 females and 51 males with a mean age of12.24 years (SD = 0.57), a middle age group of 35 females and 54 maleswith a mean age of 14.30 years (SD = 0.43), and an older group comprising44 females and 49 males with a mean age of 16.35 years (SD = 0.56).There were eight respondents whose age was not known.

Measures

Time

In addition to demographic variables of age and sex, the questionnairemeasured the following constructs. The respondents were asked to reportin hours/minutes the amount of time spent alone with their mother, theirfather, their siblings, and their peers. They were informed that "amount oftime" was defined as the time spent, in the last week, with that person,"when if you wanted to, you would feel free to talk about things that wereespecially important to you or them; ask questions about and discuss thingsyou would not want any other person to know, or to know you were speak-ing about."

Family Structure

The Family Structure variables of number of siblings, birth order, sib-ling spacing, and sibling density were constructed from information aboutthe ages of the respondent's sibling(s). Following Kidwell (1981) the num-ber of siblings was calculated as the total number of brothers and sisters,

32 Fallon and Bowles

including those no longer living at home with those reporting 4 or moresiblings forming a single group. The mean number of siblings in the samplewas 2.23 (SD = 1.38) with a range of from 1 to 9. Birth order was codedinto three categories, first, middle, and last born (Kidwell, 1981). Siblingspacing was operationalized in line with Kidwell (1981) as the average agedifference between the target adolescent and Ms or her sibling(s)(mean = 3.47, SD = 2.31) ranging from 0 to 14. Sibling density was op-erationalized as the sum of the reciprocals of spacings between therespondents and each sibling (Kidwell, 1981). The mean sibling density was0.900 (SD = 0.31) ranging from 0.07 to 2.58.

Family Functioning

The Family Functioning Scale (Noller et al, 1992) has 30 items — forexample, "People in our family help and support each other" -- requiring Lik-ert responses (1: Totally Disagree, 6: Totally Agree). It comprises three scales:Intimacy (12 items), the extent of sharing and closeness., as well as expressive-ness and openness in communication; Parenting Style (10 items), the extentto which family members have a say in rules and decisions, and are encouragedto operate independently; and Conflict (8 items), the extent of misunderstand-ing and interference, and difficulty in solving problems and making plans. Thevalidity and reliability of the instrument has been reported (Noller et al., 1992)with alpha coefficients of .92 for intimacy, .68 parenting style, and .82 for con-flict, with test-retest reliabilities reported at .77 for intimacy, .79 conflict, and.81 for parenting style.

Procedure

Two hundred and forty-six respondents were administered the ques-tionnaire in class groups in their normal class settings. The other 53respondents who were not able to complete the questionnaire in classgroups were administered the questionnaire under the supervision of psy-chology students in one-to-one interview settings in the adolescent's home.

RESULTS

Preliminary analyses of the dependent variables indicated that therewere no significant differences between respondents from the single sexand coeducational schools. Nor were there any significant differences on

Family Structure and Family Functioning 33

the dependent variables between respondents from the two data gatheringmethods. The reliabilities for the Family Functioning scales were .92 forIntimacy, .87 for Parenting Style, and .74 for Conflict, which are consistentwith the high reliabilities previously reported. Correlational analysis re-vealed that multicollinearity was not a problem.

Age and Sex Differences

The first set of analyses tested for age and sex differences in theamount of time spent with individual family members and with peers. Table Ishows the mean amount time adolescents reported. There were no multi-variate or univariate main effects for sex. A multivariate analysis of variancewith the four tune variables as dependent variables and age and sex as theindependent variables showed that there were no multivariate sex by agegroup interactions. At the univariate level there were interactions for theamount of time adolescents reported spending with their fathersF2,285 = 4.02, p < 0.05). Post hoc comparisons showed that younger agedfemales and the mid-aged males spent significantly more time with theirfathers than did the mid-aged group females (t285 = 2.18, p < 0.05;t285 = 2.18, p < 0.05, respectively). There was also a univariate interactionfor the time adolescents spend with their mothers (F2,285 = 3.05, p < 0.05).Younger females indicated spending significantly more time with their moth-ers than did younger males t285 = 2.08, p < 0.05), mid-aged femalest285 = 2.53, p < 0.05), older aged females t285 = 2.14, p < 0.05) and olderaged males (t285 = 3.02, p < 0.05).

Pillais multivariate test showed an effect for age group for amountof time spent with peers (F8,140 = 2.08, p < 0.05). Univariate analyses in-dicated that the amount of time adolescents spent with peers increasedwith age (F2,.285 = 3.78, p < 0.05) with significant differences between theyounger and older age groups (t285 = 2.50, p < 0.05).

Analysis of the intercorrelations revealed that significant differenceswere present between males and females for the correlations of the amountof time spent with father and siblings (z = 2.29; p < .05) and time withmother and siblings (z = 3.38; p < .01; see Table II). These show thatthere is a greater relationship between tune with parents and siblings formales than for females. That is, more time with parents for males corre-sponds to more time with siblings.

34 Fallon and Bowles

Family Structure and Family Functioning 35

Table I. Means and Standard Deviations of the Number of Minutes Spent byAdolescents with Family Members and Peers in the Previous Week

Family members and peers

Sex and age ofadolescent

Younger age

FemaleMeanSD

MaleMeanSD

CombinedMeanSD

Middle ageFemale

MeanSD

MaleMeanSD

CombinedMeanSD

Older age

FemaleMeanSD

MaleMeanSD

CombinedMeanSD

For entire sampleMeanSD

Father

442.65645.64

254.76492.14

354.74583.96

167.80211.75

435.26854.43

330.08688.68

229.25332.34

314.63448.68

274.24398.11

321.47567.76

Mother

742.961185.37

405.181109.37

584.911157.57

281.74279.23

548.98917.53

443.88744.57

378.18405.19

360.21462.22

368.68433.92

472.67857.58

Brothersand sisters

659.861043.13

402.04947.86

539.231003.46

338.86372.77

657.591024.97

532.25843.17

396.14483.31

477.49740.88

439.00630.27

505.06847.83

Peers

727.26856.10

669.26903.56

700.18874.98

665.20668.67

844.17827.98

773.78770.31

1174.931051.69

887.121003.00

1023.291030.88

825.95906.41

Predicting Amount of Time

The next set of analyses used multiple regression to predict theamount of time spent with family members and with peers. Initial analysesindicated that birth order did not contribute to the variance, which couldbe accounted for in any of the analyses and so it was dropped from furtheranalyses. The curvilinear relationship between sibling order (Kidwell, 1981)and other variables were not replicated; instead the relationship was linear,which indicated that it was appropriate for inclusion in the regression equa-tions. Therefore the family structure variables that were included in theprediction of times were sibling spacing, sibling density, and number ofsiblings. The family functioning variables of Intimacy, Conflict, and Parent-ing Style represented the family functioning variables. Age in years wasincluded as a possible predictor in the regressions. Separate analyses wereconducted for males and females.

Table III presents the variables found to be significantly relatedto the criterion variables of time for males. The regression analyses re-vealed that 18% of the variance of the amount of time adolescents spendwith their fathers is accounted firstly by level of democratic parenting,then density of the siblings, then intimacy of the family, and then thenegative influence of conflict. Males' time with mother and siblings wassimilarly predicted by parenting style and sibling density. Whereas theamount of time with peers was predicted weakly and negatively by familyconflict.

By contrast, Table IV indicates that family structure variables werethe more significant predictors of time spent with family members andpeers for females. That is, the best predictors of intimate time with fathersfor females was lower sibling density, being younger and having fewer

36 Fallon and Bowles

Table II. Intercorrelations of the Amount of Time Spent with Fattier,Mother, Sib!ing(s), and Peers for Males and Females"

Males

FatherMotherSibling(s)Peers

Father

.76*

.72*

.35*

Females

Mother

.77*

.72b

.35b

Sibiing(s)

.56*

.46*

.46*

Peers

.33*

.34*

.35*

a Correlation coefficients above the diagonal are for females, those beloware for males.

* Significance less than .01 (2 tailed).

siblings, in that order. The equation accounted for 19% of the variance.Accounting for a similar amount of the variance for the amount of timeadolescent females spend with mothers is lower sibling density, beingyounger and less democratic parenting. Sixteen percent of the variance oftime with siblings was accounted for by lower sibling density and being ayounger adolescent. Lower sibling density and lower levels of familyintimacy was related to spending more time with peers.

In summary, significant age group and sex effects were found for theamount of time adolescents spend with individual family members and withpeers. For males, family functioning variables were more important predic-tors of time than were the family structure variables while for females, thereverse was found.

DISCUSSION

The purpose of this research was to investigate the influence of familystructure and family functioning variables to ascertain which were influentialon male and female adolescents' perception of intimate time spent with in-dividual family members and peers. In line with the hypotheses, the findingsshowed that the amount of time that adolescents spend with their mother,father, siblings, and peers varies, at least in part, as a function of the ado-lescent's age and sex. The results also show that males were influenced more

37

Table III. Regression Analyses Predicting the Amount of Time AdolescentsSpend with Parents, Siblings, and Peers for Males

Parenting style

Sibling density

Intimacy

Conflict

R2 (adjusted)F Ratio

Father

.24a

2.85b

.354.36

.273.53

-.21-2.18

.189.73

Mother

.293.55

.212.50

.077.20

Sibling(s)

.323.90

.212.55

.088.43

Peers

-.19-2.34

.035.49

a Standardized regression coefficients (beta weights).bt values.

Family Structure and Family Functioning

by family functioning factors and females more by family structure factors.Across all four relationships there were significant positive correlations formale and female adolescents. The strongest correlations were between timewith mothers, fathers, and siblings for males and mothers and fathers forfemales. Predictions of whether family structure or family functioning vari-ables are related to the amount of intimate time with significant othersshowed that for males family functioning and sibling density variables werethe most influential, whereas for females family structure variables and agewere the most influential. Overall substantial support for the hypotheses wasfound. A range of important issues emerge from these findings.

The means and the standard deviations of this intimate time with sig-nificant others is consistent with previous time research reported byMontemayor (1982) and indicates that this intimate element of time withothers may be correctly interpreted as a subset of total time adolescentsspend with family members and peers.

Analysis of the mean amount of time spent with individual familymembers and peers revealed some consistent age group and sex patterns.For females, in all situations, it was the middle-aged group of adolescentswho spent the least amount of time with family or friends. For males, themiddle-aged group spent the least amount of tune with their mothers whereasfor fathers, siblings and peers, it was the youngest age group of males who

Fallon and Bowles38

Table IV. Regression Analyses Predicting the Amount of Time AdolescentsSpend with Parents, Siblings, and Peers for Females

Sibling density

Age

Number of siblings

Parenting style

Intimacy

R2 (adjusted)F Ratio

Father

-.39a

-5.05b

-.24-3.11

-.17-2.19

.1912.06

Mother

-.465.58

-.22-2.70

-.20-2.45

.1912.02

Sibling(s)

-.405.04

-.18-2.32

.1613.33

Peers

-.29-3.63

-.18-2.19

.098.00

a Standardized regression coefficients (beta weights).bt values.

spent the least amount of time. These periods where little time is spent withothers in this intimate fashion may indicate a developmental phase shiftand/or be a period of vulnerability when the adolescent; will engage inbehaviors to enhance the possibility of deeper associations with parents orsiblings or peers to attract attention from such individuals.

The correlations showed that more time spent with parents for malesis associated with more time with siblings, whereas these findings were sig-nificantly different from the correlation of the time females spend withparents and siblings. The consistency of the correlations indicate that forboth males and females time with parents is most highly correlated, thentime with parents and siblings, and the least but still moderate correlationoccurs between family members and peers. These findings differ from thoseof Montemayor (1982), who reported that the correlations of time betweenmothers and peers for females was consistently negative but less consis-tently so for males. The correlations in the present study show that timewith peers is not negatively related to time with either parent. Thus thereis little evidence of a parent-peer conflict model operating. This furtherreinforces the notion that those who spend the least amount of this specialtime with family members and peers are likely to be vulnerable as all thecorrelations were positive. Thus, less time with family members is associ-ated with less time with peers.

The consistency of these findings and the association between timespent with family members and peers suggests that rather than having a groupof adolescents defined by their association with peers or parents, what actuallymay be occurring is a different polarization during adolescence. This moredisturbing situation would have those who spend "quality time" with familymembers and peers at one end and those who are socially isolated or lonelyat the other. Research into loneliness has shown that 22% of adolescents inAustralia responded positively to the statement "I am so very lonely"(Atkinson, 1988). This group is likely to be represented in this research asthose with very little time with all others. The complexity and importance ofthe attributions made about the quality of such contact and other influentialfactors such as personality variables has been noted in previous research(Boldero and Moore, 1990; Inderbitzen-Pisaruk et al, 1992). Future researchlinking the perceptions of intimate time spent with others and such variableswould assist in the understanding of loneliness.

The predictors of intimate time with parents and siblings for maleswere democratic parenting style and high sibling density, while for females,low sibling density and being of younger age were important. Weak pre-dictors of time with peers for females were lower sibling density and lessintimacy in the family, whereas for males the only predictor was low conflictin the family. Of interest also is the direction of the predictors of time

39Family Structure and Family Functioning

with peers. For males ail of the variables that distinguish time with familyand peers are positive (with the exception of conflict where the negativebeta weight implies a positive outcome, i.e., less conflict) while every factorfor females was negative. That is, males are positively influenced mostlyby the family's functioning as well as sibling density, whereas family struc-ture factors are negatively influential for females. These findings concurwith the assertion that larger families function more like a group and areless democratic, particularly for females (Kidwell, 1981). Importantly, andcontrary to previous research (Kidwell., 1981), the other family structurevariables did not consistently and significantly contribute to prediction oftime with family members and peers.

Central to the descriptions of adolescent interaction with family andpeers is the accepted notion that adolescents communicate less with parentsand more with peers as they age (Fasick, 1984; Vernberg et al., 1993).These findings have been partially supported by this research. The meantime with fathers, mothers, and siblings for the entire sample show thatadolescents spend the most intimate time with peers, then siblings, mothers,and fathers. One explanation offered for this transition from dependenceon parents to peers is that families simply spend less time together and,by comparison, adolescents spend more time with peers (Miller and Lane,1991). Another explanation is the onset of the peak period of puberty anda corresponding distancing in family relations. Studies have shown that thisperiod is characterized by more interrupting language in interaction be-tween males and parents (Steinberg and Hill, 1978; Steinberg, 1981, 1988)and greater conflict between adolescents and their mothers (Hill et \aL,1985).

Another explanation is that parents, siblings, and peers differ in theirfunctional roles (Lempers and Clark-Lempers, 1992). Mothers and fathersare sources of affection, instrumental aid, and alliance, whereas peers pro-vide intimacy and companionship as do siblings. Lempers and Clark-Lem-pers' (1992) research also associated conflict with parent-adolescentsrelations, and Montemayor (1982) concluded that disagreement between fe-males and their mothers was related to spending greater amounts of timetogether. In this present study conflict was associated with less time withfathers and peers for males but not for females. Of interest also was thefinding that less democratic parenting style was associated with spendinggreater amounts of time with mothers for females. Investigation of thecauses and outcomes of such family functioning is necessary.

The changes in the amount of intimate time adolescents spend withfamily members varied across adolescence, which is in line with findingsfrom research into emotional disclosure by adolescents to parents andpeers. The findings show the need for families to adapt to the needs of

40 Fallon and Bowles

adolescents at various adolescent stages ( Brown et  al, 1986; Papini et al.,1990). Adapting to such needs does not necessarily mean parents spendingmore time with adolescents. It is possible that more time with parents isnot necessary or welcome and may not relate to better adolescent adjust-ment or well-being. In line with the findings of Miller and Lane (1991), itmay be that the function of interaction with parents at various stages ofadolescence has not only a different effect but also a different quality andtherefore a differential salience for the adolescent. Therefore increasingthe absolute amount of time with a significant other, such as a parent with-out some purpose, may have a detrimental influence on the relationship.It is more likely that a flexible approach, responsive to the adolescents'needs in the context of other competing social and developmental goals,is appropriate.

Some words of caution are needed regarding the present study eventhough the selection of respondents and the data gathering techniquesused were not unusual in this type of research. While there is no evidenceto suggest a biased sample from the one coeducational and two single-sexschools used, respondents were not selected randomly and therefore mayor may not be representative of the schools populations or of thepopulation of Australian schools. Sampling of individual adolescents' thatwould have permitted interviews rather than pencil-paper responses werenot permitted by school policies. No specific information concerning thesocioeconomic status of the families was available; however, the schoolswere located in middle class suburbs. The usual caveats that apply toself-report data are also required; for example, while the respondents wereinstructed that the time which was being asked about was one-on-one time,there was no way to ensure that the adolescents complied with thisinstruction.

A number of possible research questions emerge from the study. Fu-ture research into the nature of interactions of those adolescents who arelow on quality time with both family members and peers is necessary. Ex-ploration of the predictors of time with different family members may helpexplain the sex differences. Future research into the values and importanceof the qualitative differences in the amount of time spent with family mem-bers and peers would be appropriate as it is too simple to assume thatmore (for example more time with peers with age) implies a better rela-tionship or a more meaningful relationship. Such research would helpdefine why family intimacy influences time with peers negatively. Ascer-taining whether the periods of less time with family members and peers,in early adolescence for males and midadolescence for females, are actuallytimes of loneliness and vulnerability is also important.

Family Structure and Family Functioning 41

REFERENCES

Altmann, L., and Taylor, D. (1973). Social Penetration: The Development of InterpersonalRelationships. Holt, Rinehart, Winston, New York.

Atkinson, R. (1988). Respectful, dutiful teenagers. Psychol. Today 26: 22-23.Berger, C., and Calabrese, R. (1975). Some explorations in initial attraction and beyond:

towards a developmental theory of interpersonal communication. Human Commun. Res.1: 99-112.

Biddle, B., Banks, B., and Marlin, M. (1980). Parental and peer influences on adolescence,Social Forces 58: 1057-1079.

Boldero, J., and Moore, S. (1990). An evaluation of the de Jong-Gievald's loneliness modelwith Australian adolescents. J. Youth Adolesc. 19: 133-147.

Brown, B., Eicher, S. A., and Petrie, S. (1986). The importance of peer group ("crowd")affiliation in adolescence. J. Adolesc. 9: 73-96.

Ernst, C., and Angst, A. (1983). Birth Order: Its Influence on Personality. Springer-Verlag, NewYork.

Fasick, F. (1984). Parents, peers, youth culture and autonomy in adolescence. Adolescence 19:143-157.

Hall, L. D., Walker, S. J., and Acock, A. C. (1995). Gender and family work in one-parenthouseholds. /. Marriage Family 57: 685-692.

Harner, L. (1982). Talking about the past and the future. In Friedman, W. J. (ed.), TheDevelopment of the Psychology of Time. Academic Press, New York.

Hill, J., Holmbeck, G., Marlow, L., Green, T., and Lynch, M. (1985). Pubertal status andparent-child relations in families of seventh grade boys. / Early Adolsec. 5: 31-44.

Huston, T., and Burgess, R. (1979). Social exchange in developing relationship: an overview.In Burgess, R. I., and Huston, T. L. (eds.), Social Exchange in Developing Relationships.Academic Press, New York.

Inderbitzen-Pisaruk, H., Clark, M., and Solano, C. (1992). Correlates of loneliness inmidadolescence. X Youth Adolescence 21: 153-167.

Kidwel!, J. S. (1981). Number of siblings, sibling spacing, sex, and birth order: Their effectson perceived parent-adolescent relationships. J. Marriage Family 43: 315-333.

Koch, H. (1955). The relation of certain family constellation characteristics and the attitudesof children towards adults. Child Develop. 26: 13-40.

Lempers, J., and Clark-Lempers, D. (1992). Young, middle and late adolescents comparisonsof the functional importance of five significant relationships. /. Youth Adolesc. 21: 53-96.

Levinger, G., and Snoek, J. (1972). Attraction in Relationship: A New Look at InterpersonalAttraction. Genera! Learning, Morristown, NJ.

Miller, J., and Lane, M. (1991). Relations between young adults and their parents. J. Adolesc.14: 179-194.

Montemayor, R. (1982). The relationship between parent-adolescent conflict and the amountof time adolescents spend alone and with parents and peers. Child Develop. §3: 1512-1519.

Noller, P., and Cailan, V. (1991). The Adolescent in the Family. Routledge, London.Noller, P., Seth-Smith, M., Bouma, R., and Schweitzer, R. (1992). Parent and adolescent

perceptions of family functioning: A comparison of clinic and nonclinic families. J.Adolesc. 15: 101-115.

Papini, D., Farmer, F., Clark, M., Micka, J., and Barnett, J. (1990). Adolescence 25: 960-975.Pfoutts, J. H. (1980). Birth order, age-spacing, IQ differences, and family relations. /. Marriage

Family 42: 517-530.Piaget, J. (1969). The Child's Conception of Time. (A. J. Pomerans, trans.). London: Routledge

& Kegan Paul. (Original work published 1927.)Richardson, R. A., Abramowitz, R. H., Asp, C. E., and Petersen, A. C. (1986). Parent-child

relationships in early adolescence: Effects of family structure. J. Marriage Family 48:805-811.

Sebald, H., and White, B. (1980). "Teenagers' divided reference groups: Uneven alignmentwith parents and peers. Adolescence 15: 979-984.

42 Fallon and Bowles

Slife, B. D. (1993). Time and Psychological Explanation. State University of New York Press,New York.

Steinberg, L. (1981). Transformations in family relations at puberty. Develop. Psychol. 17:833-840.

Steinberg, L. (1988). Reciprocal relations between parent-child distance and pubertalmaturation. Develop. Psychol. 24: 122-128.

Steinberg, L., and Hill, J. (1978). Patterns of family interaction as a function of age, the onsetof puberty and formal thinking. Develop. Psychol. 14: 683-684.

Vernberg, E., Beery, S., Ewell, K., and Abwender, A. (1993). Parents' use of friendshipfacilitation strategies and the formulation of friendships in early adolescence: Aprospective study. J Family PsychoL 7: 356-369.

Walker, A., and Thompson, L. (1983). Intimacy and intergenerational aid and contact amongmothers and daughters. /. Marriage Family 45: 841-848.

Wilks, J. (1986). The relative importance of parents and friends in adolescent decision making.J. Youth Adolesc. 15: 323-335.

Family Structure and Family Functioning 43