National Norms and Psychometrics for the Pubertal Development Scale
Liu, Y., Bonny, A. E., Perez-Algorta, G., Youngstrom, E. A.
pediatrics
Quick verdict
First national PDS norms with rigorous psychometrics; convincing overall, though shared-method bias and floor imprecision temper the clinical claims.
Summary
This preprint reports psychometric validation and continuous national norms for the Pubertal Development Scale (PDS) using a stratified YouGov panel sample aligned to 2023 US ACS estimates: parent reports for 2,251 children aged 6-18, youth self-reports for 1,006 aged 12-18, and ~19-day retest subsamples (N=171 parents, N=90 youth). One-factor CFA models fit well once the prepubertal 6-11 floor-effect band was excluded; internal consistency was alpha .77-.89; IRT-based conditional reliability exceeded .80 across most of the 5-20 score range; test-retest reliability was r=.90-.96; item-level DIF by race and parental education had negligible test-level impact (DTF <= .17 points). Cross-informant agreement was unusually high (ICC=.87-.88). GAMLSS (BCCG) models yielded age- and sex-specific percentile curves, tables, and a Shiny app. In regressions, earlier norm-referenced pubertal timing predicted parent-reported depression (beta=.26), mania (beta=.46), and sleep disturbance (beta=.40) beyond age and sex, with negative age-x-timing interactions; youth-report models were null despite similar point estimates. Key limitations: no Tanner-staging criterion validation, mono-informant symptom regressions vulnerable to shared-method variance, sample composition deviating from ACS targets (75.3% White vs 60.5%; 47.5% parental bachelor's vs 36.2%), and selecting BCCG over the better-fitting Beta-Binomial for youth norms.
Review complete
Assessment at a glance
5
risk flags identified
● high 0
● medium 2
● low 3
5 strengths · 5 weaknesses · 5 author questions
Levels reflect verification complexity — not a quality score. Evolve never rates your paper.
Key claims under review
- The PDS measures a single latent pubertal-development factor in both parent and youth report once prepubertal floor effects are accounted for by age-band restriction.
- PDS scores are reliable: internal consistency alpha .77-.89, ~19-day test-retest ICC(2,1) .89-.96, and conditional reliability above .80 across roughly 13 of the 15 possible score points.
- Isolated items show significant DIF by race and parental education, but Differential Test Functioning is negligible (<=.17 PDS points), so total scores are interpretable across these subgroups.
- Parent-youth agreement on PDS totals is excellent (ICC .87-.88), supporting flexible informant selection at ages 12-18.
- Earlier norm-referenced pubertal timing independently predicts parent-reported depression, mania, and sleep disturbance beyond age and sex, most strongly in younger children.
Strengths
First representative continuous norms for the PDS
Prior US norms dated to Petersen et al. (1988) with 355 adolescents aged 11-14; this study provides GAMLSS percentile curves (5th-95th) and reference tables for parent report ages 6-18 (N=2,251) and self-report ages 12-18 (N=1,006), plus a Shiny app for score-to-percentile conversion.
Modern, multi-method measurement battery beyond alpha-only literature
CFA with polychoric/WLSMV, IRT-based conditional reliability (GRM + 2PL for menarche, EAP mapping onto observed scores), IRT-LR DIF with FDR plus Differential Test Functioning, ICC(2,1) and Bland-Altman plots for both retest and cross-informant agreement.
Transparent handling of prepubertal floor effects
Demonstrated incremental validity of the normed score
Reproducibility posture
Weaknesses
No criterion validation against Tanner staging
Shared-method variance in the symptom regressions
Precision is weakest exactly where screening is promoted
Distribution choice overrides model fit for youth norms
Representativeness gaps and unweighted analyses
Full Review Report
$1.99 · one-timeUnlock the complete preliminary analysis Evolve produced for this paper.
- All 5 strengths & 5 weaknesses, analyzed in detail
- 5 risk flags with severity, section & reviewer questions
- Methods & statistical evidence assessment
- 5 questions a peer reviewer would ask
- Full six-dimension evaluation report & recommendations
One-time purchase · this paper only
One-time purchase, this paper only. AI-generated preliminary analysis — not peer review.
Frequently asked questions
What are the national norms for the Pubertal Development Scale (PDS)?
This 2026 medRxiv preprint provides the first US nationally representative continuous norms: age- and sex-specific GAMLSS percentile curves (5th-95th) for parent-reported PDS ages 6-18 (N=2,251) and youth self-reported PDS ages 12-18 (N=1,006), with integer reference tables and a free Shiny app that converts a PDS total score into a percentile given informant, age, and sex.
Is the PDS a reliable and valid measure of puberty?
In this nationally representative sample the PDS showed a single-factor structure (once prepubertal floor effects were excluded), internal consistency alpha .77-.89, ~19-day test-retest reliability ICC .89-.96, negligible score-level impact of race/education DIF, and very high parent-child agreement (ICC .87-.88). It remains a parent/self-report proxy and was not validated here against clinician-rated Tanner staging, so it is positioned as a research measure or pre-screener rather than a diagnostic tool.
Does early puberty cause mental health problems according to this study?
No causal claim is possible from this cross-sectional survey. The study found that children whose pubertal development was early relative to same-age, same-sex peers had higher parent-reported depression, mania, and sleep disturbance scores (standardized betas .26-.46), but these associations explained only about 1-2% of variance beyond age and sex, were strongest in younger children, and did not reach significance in youth self-reports.
How can I convert a child's PDS score to a percentile?
Use the Shiny app linked in the paper (hosted on shinyapps.io) or the percentile reference tables in Figures 4C and 5C, selecting the correct informant type (parent report vs youth self-report), sex, and age. Replication code is available on the first author's GitHub repository.
About this review
This is an AI-generated preliminary review by Evolve.Pub, produced from the openly available text of the paper. It is not peer review and does not represent the authors' views. The canonical record of the paper is at the publisher link above.
- Figures (percentile curves, Bland-Altman plots, conditional reliability curves, worm plots, DIF item curves) were embedded as images and are not analyzable from the extracted text; judgments about curve shapes rest on the reported numbers only.
- Supplemental materials (Table S1, Table S2, Figures S1-S3) were not included in the extraction, so GAMLSS fit diagnostics and retest-sample demographics could not be verified.
- The youth-report GAMLSS distribution choice and administration logistics for parent-child dyads are described only briefly in the extracted text; details may exist elsewhere in the supplement.
- medRxiv license watermarks interleave with body text, but the narrative appears complete from abstract through conclusion and author note.
Go deeper on this paper
Wrote a paper? Stress-test it before reviewers do.
Upload your manuscript for the same structured scrutiny — reasoning, statistics, evidence, figures, and reproducibility.