A Longitudinal Investigation of Adolescents' Changing ...

1 downloads 0 Views 957KB Size Report
College of Health and Human Development. Pennsylvania State University. Julia A. Graber. Pennsylvania State University. Anne C. Petersen. College of Health ...
Copvright 1991 by the American Psychological Association, Inc. 0012-1649/91/S3.00

Developmental Psychology 1991, Vol. 27, No. 4, 580-586

A Longitudinal Investigation of Adolescents' Changing Perceptions of Pubertal Timing Julia A. Graber

Judith Semon Dubas

Pennsylvania State University

College of Health and Human Development Pennsylvania State University

Anne C. Petersen College of Health and Human Development Pennsylvania State University This study examined, longitudinally, adolescents' self-ratings of timing of puberty and how these perceptions are related to an objective measure of pubertal timing, pubertal status, and feelings about their pubertal timing during 7th, 8th, and 12th grades. There is moderate consistency in self-reported pubertal timing across adolescence. Perceptions of pubertal timing at 12th grade were predicted by both actual pubertal timing and 8th-grade perceptions. Furthermore, the direct comparison of perceived timing with an objective timing measure indicated that perceptions became more accurate by 12th grade. Feelings about pubertal timing were related to perceived timing but not to the objective measure of pubertal timing. These results suggest that actual and perceived timing are overlapping but distinct timing measures that reflect different biological and psychosocial processes.

pubertal status is the five-stage Tanner criteria based on observed changes in primary and secondary sexual characteristics (Marshall & Tanner, 1969,1970). Because this technique, designed to be used by health professionals, cannot always be used in research, several self-report scales of pubertal status have also been developed (e.g., Duke, Litt, & Gross, 1980; Morris & Udry, 1980; Petersen, Crockett, Richards, & Boxer, 1988) and validated (Brooks-Gunn, Warren, Rosso, & Gargiulo, 1987; Dorn, Susman, Nottelmann, Inoff-Germain, & Chrousos, 1990; Petersen et al., 1988).

Pubertal development, characterized by rapid physical growth, large increases in levels of hormones, and the appearance of secondary sexual characteristics (Petersen & Taylor, 1980), is considered one of the most important transitions that occur in adolescence. Two aspects of pubertal development have received considerable attention: pubertal status and pubertal timing. Pubertal status refers to the current level of physical development experienced by the adolescent relative to the overall process of pubertal change, whereas pubertal liming refers to whether an adolescent's pubertal process is occurring on time, early, or late relative to his or her peers. Thus, pubertal status is an absolute measure, whereas pubertal timing is a normative measure. Changes in pubertal status have been associated with both intra- and interpersonal changes in such areas as self-concept, cognitive abilities, and family relationships (Crockett & Petersen, 1987; Simmons, Blyth, Van Cleave, & Bush, 1979; Steinberg & Hill, 1978; Tanner, 1962). The timing of pubertal development has also been associated with changes in these domains (Jones & Mussen, 1958; Mussen & Jones, 1957; Newcombe & Dubas, 1987; Petersen & Crockett, 1985; Steinberg, 1987). The most widely accepted standard for the assessment of

In most cases, pubertal timing has been assessed by comparing an individual's pubertal status (or retrospective reports of age of development) on some indicator (eg., menarche) with either national or local (i.e., sample) norms, using a trichotomization procedure or standard (Jeviation cutoffs for early, ontime, and late classifications. Self-report measures of pubertal timing have also been used (Rierdan & Koff, 1985; Sanders & Soares, 1986; Silbereisen, Petersen, Albrecht, & Kracke, 1989). Whereas the validity of self-report measures of pubertal status has been established, little is known about the validity and stability of self-report measures of pubertal timing either during pubertal development or after pubertal development. At times researchers have used retrospective recollections of puberty to examine whether the influence of pubertal timing is maintained into adulthood (Sanders & Soares, 1986). Thus, it is important to establish whether the same effects of timing of puberty are found when the ratings are obtained during or after development. One problem with self-assessment of timing during the pubertal process is that the comparison person and the reference group are changing. Thus, one might perceive oneself as early one year and on time the next, depending on the pubertal status of the peer group. Hence, retrospections of pubertal timing

This research was supported by Grant MH 30252/38142 to Anne C. Petersen. Portions of these data were reported to the Society for Research in Child Development, Kansas City, Missouri, April 1989. We gratefully acknowledge the work of the entire staff of the Early Adolescent Study, as well as the cooperation and contributions of the young adolescents who participated in the research. Correspondence concerning this article should be addressed to Judith Semon Dubas, Penn State Adolescence Study, 104 Henderson Building, Pennsylvania State University, University Park, Pennsylvania 16802. 580

581

PUBERTAL TIMING may be tied to pubertal status measures rather than, or in addition to, how one perceived his or her pubertal timing at a particular grade. In addition, retrospections of pubertal timing may be targeted to specific pubertal measures at different grades in school according to gender, because boys on the average mature 1.5 to 2 years later than girls do, depending on the indicator. Thus, investigators studying adolescents might have to take into account both the grade and gender of the adolescents. The purpose of this study was to longitudinally examine adolescents' self-report measures of pubertal timing and how these perceptions relate to objective measures of pubertal timing and self-reported pubertal status as well as feelings about pubertal timing. The longitudinal design of this study allowed us to examine the relationship between high school seniors' retrospections of their pubertal timing and their perceptions of puberty reported during seventh and eighth grades. Because boys and girls differ in their pubertal timing, we also examined whether our findings varied by sex of the adolescent. Specifically, the questions this research addressed are: 1. Is there consistency across adolescence in how individuals perceive their timing of puberty? 2. Are perceptions of timing of puberty related to an objective measure of timing of puberty? 3. Do recollected perceptions of pubertal timing in 12th grade reflect earlier perceptions of pubertal timing or status more than objectively measured pubertal timing? 4. Is there a relationship between how an individual rates his or her pubertal timing and how that person feels about their timing?

Method Subjects Subjects were 62 boys and 69 girls whose data were drawn from a larger cohort-sequential longitudinal investigation of biological and psychosocial development in adolescence (Petersen, 1984) and whose puberty data were available in the spring of 7th and 8th grades and during the 12th-grade follow-up study. The larger study from which these subjects were drawn interviewed the participants twice annually, in the fall and spring of each school year, over a 3-year period from the 6th through 8th grades, with an additional follow-up interview conducted at 12th grade. The sample was almost entirely White and from middle- to upper-middle-class families living in the suburbs of a large midwestern city. The adolescents were above average in intelligence (mean IQ = 115 at 8th grade), and their parents were well educated (83% of the fathers and 67% of the mothers had graduated from college). In this study, the subsample reported on was no different from the larger adolescent sample in any of the demographic or pubertal measures.

Pubertal Measures

Objective pubertal timing. Age at peak height velocity (APHV) was estimated from height data with the technique developed by Bock et al. (1973; See Petersen & Crockett, 1985, for reliability information and a more detailed description of this technique). Individuals were classified as early, on time, or late on the basis of two methods: trichotomized scores and standard deviation cutoffs within sex (i.e., individuals whose APHVs were more than 1 SD below the mean were classified as early, those whose APHVs were within 1 SD of the mean were classified as on time, and those whose APHVs were more than 1 SD above the mean were classified as late developers). These classifications were based on the total early adolescent sample