Abstract
It is unknown whether perceived pubertal timing changes as puberty progresses or whether it is an important component of adolescent
identity formation that is fixed early in pubertal development. The purpose of this study is to examine the stability of perceived
pubertal timing among a school-based sample of rural adolescents aged 11–17 (N = 6,425; 50% female; 53% White). Two measures of pubertal timing were used, stage–normative, based on the Pubertal Development Scale, a self-report scale of secondary sexual characteristics, and peer–normative, a one-item measure of perceived pubertal timing. Two longitudinal methods were used: one-way random effects ANOVA models
and latent class analysis. When calculating intraclass correlation coefficients using the one-way random effects ANOVA models,
which is based on the average reliability from one time point to the next, both measures had similar, but poor, stability.
In contrast, latent class analysis, which looks at the longitudinal response pattern of each individual and treats deviation
from that pattern as measurement error, showed three stable and distinct response patterns for both measures: always early,
always on-time, and always late. Study results suggest instability in perceived pubertal timing from one age to the next,
but this instability is likely due to measurement error. Thus, it may be necessary to take into account the longitudinal pattern
of perceived pubertal timing across adolescence rather than measuring perceived pubertal timing at one point in time.
identity formation that is fixed early in pubertal development. The purpose of this study is to examine the stability of perceived
pubertal timing among a school-based sample of rural adolescents aged 11–17 (N = 6,425; 50% female; 53% White). Two measures of pubertal timing were used, stage–normative, based on the Pubertal Development Scale, a self-report scale of secondary sexual characteristics, and peer–normative, a one-item measure of perceived pubertal timing. Two longitudinal methods were used: one-way random effects ANOVA models
and latent class analysis. When calculating intraclass correlation coefficients using the one-way random effects ANOVA models,
which is based on the average reliability from one time point to the next, both measures had similar, but poor, stability.
In contrast, latent class analysis, which looks at the longitudinal response pattern of each individual and treats deviation
from that pattern as measurement error, showed three stable and distinct response patterns for both measures: always early,
always on-time, and always late. Study results suggest instability in perceived pubertal timing from one age to the next,
but this instability is likely due to measurement error. Thus, it may be necessary to take into account the longitudinal pattern
of perceived pubertal timing across adolescence rather than measuring perceived pubertal timing at one point in time.
- Content Type Journal Article
- Category Empirical Research
- Pages 1-12
- DOI 10.1007/s10964-011-9720-0
- Authors
- Jessica Duncan Cance, Department of Kinesiology and Health Education, The University of Texas at Austin, 1 University Station D3700, Bellmont Hall 222, Austin, TX 78712, USA
- Susan T. Ennett, Department of Health Behavior and Health Education, Gillings School of Global Public Health, University of North Carolina at Chapel Hill, Chapel Hill, NC, USA
- Antonio A. Morgan-Lopez, Department of Psychology, L.L. Thurstone Psychometric Laboratory, University of North Carolina at Chapel Hill, Chapel Hill, NC, USA
- Vangie A. Foshee, Department of Health Behavior and Health Education, Gillings School of Global Public Health, University of North Carolina at Chapel Hill, Chapel Hill, NC, USA
- Journal Journal of Youth and Adolescence
- Online ISSN 1573-6601
- Print ISSN 0047-2891