Utility of inferential norming with smaller sample sizes

Jianjun Zhu, Hsin-Yi Chen

Research output: Contribution to journalArticle

14 Citations (Scopus)

Abstract

We examined the utility of inferential norming using small samples drawn from the larger Wechsler Intelligence Scales for Children-Fourth Edition (WISC-IV) standardization data set. The quality of the norms was estimated with multiple indexes such as polynomial curve fit, percentage of cases receiving the same score, average absolute score differences, score distributions, score correlations, and sensitivity in identifying children who are gifted or intellectually disabled. Norms developed using inferential norming method and sample sizes of 50 and 75 per age group had qualities comparable to the norms derived using larger sample size. So, when large sample sizes are infeasible due to practical constraints, N = 50 per group can be considered a lower bound to derive decent norms using inferential norming method. Cautions for implementing inferential norming with smaller sample sizes and suggestions for future research are discussed.

Original languageEnglish
Pages (from-to)570-580
Number of pages11
JournalJournal of Psychoeducational Assessment
Volume29
Issue number6
DOIs
Publication statusPublished - 2011 Dec 1

Fingerprint

Sample Size
Gifted Child
Wechsler Scales
Intelligence
Age Groups
age group
intelligence
Group

Keywords

  • WISC-IV
  • inferential norming
  • sample size

ASJC Scopus subject areas

  • Education
  • Clinical Psychology
  • Psychology(all)

Cite this

Utility of inferential norming with smaller sample sizes. / Zhu, Jianjun; Chen, Hsin-Yi.

In: Journal of Psychoeducational Assessment, Vol. 29, No. 6, 01.12.2011, p. 570-580.

Research output: Contribution to journalArticle

@article{b7a81e631de24c4385ae911381b1b3a8,
title = "Utility of inferential norming with smaller sample sizes",
abstract = "We examined the utility of inferential norming using small samples drawn from the larger Wechsler Intelligence Scales for Children-Fourth Edition (WISC-IV) standardization data set. The quality of the norms was estimated with multiple indexes such as polynomial curve fit, percentage of cases receiving the same score, average absolute score differences, score distributions, score correlations, and sensitivity in identifying children who are gifted or intellectually disabled. Norms developed using inferential norming method and sample sizes of 50 and 75 per age group had qualities comparable to the norms derived using larger sample size. So, when large sample sizes are infeasible due to practical constraints, N = 50 per group can be considered a lower bound to derive decent norms using inferential norming method. Cautions for implementing inferential norming with smaller sample sizes and suggestions for future research are discussed.",
keywords = "WISC-IV, inferential norming, sample size",
author = "Jianjun Zhu and Hsin-Yi Chen",
year = "2011",
month = "12",
day = "1",
doi = "10.1177/0734282910396323",
language = "English",
volume = "29",
pages = "570--580",
journal = "Journal of Psychoeducational Assessment",
issn = "0734-2829",
publisher = "SAGE Publications Inc.",
number = "6",

}

TY - JOUR

T1 - Utility of inferential norming with smaller sample sizes

AU - Zhu, Jianjun

AU - Chen, Hsin-Yi

PY - 2011/12/1

Y1 - 2011/12/1

N2 - We examined the utility of inferential norming using small samples drawn from the larger Wechsler Intelligence Scales for Children-Fourth Edition (WISC-IV) standardization data set. The quality of the norms was estimated with multiple indexes such as polynomial curve fit, percentage of cases receiving the same score, average absolute score differences, score distributions, score correlations, and sensitivity in identifying children who are gifted or intellectually disabled. Norms developed using inferential norming method and sample sizes of 50 and 75 per age group had qualities comparable to the norms derived using larger sample size. So, when large sample sizes are infeasible due to practical constraints, N = 50 per group can be considered a lower bound to derive decent norms using inferential norming method. Cautions for implementing inferential norming with smaller sample sizes and suggestions for future research are discussed.

AB - We examined the utility of inferential norming using small samples drawn from the larger Wechsler Intelligence Scales for Children-Fourth Edition (WISC-IV) standardization data set. The quality of the norms was estimated with multiple indexes such as polynomial curve fit, percentage of cases receiving the same score, average absolute score differences, score distributions, score correlations, and sensitivity in identifying children who are gifted or intellectually disabled. Norms developed using inferential norming method and sample sizes of 50 and 75 per age group had qualities comparable to the norms derived using larger sample size. So, when large sample sizes are infeasible due to practical constraints, N = 50 per group can be considered a lower bound to derive decent norms using inferential norming method. Cautions for implementing inferential norming with smaller sample sizes and suggestions for future research are discussed.

KW - WISC-IV

KW - inferential norming

KW - sample size

UR - http://www.scopus.com/inward/record.url?scp=80055107156&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=80055107156&partnerID=8YFLogxK

U2 - 10.1177/0734282910396323

DO - 10.1177/0734282910396323

M3 - Article

AN - SCOPUS:80055107156

VL - 29

SP - 570

EP - 580

JO - Journal of Psychoeducational Assessment

JF - Journal of Psychoeducational Assessment

SN - 0734-2829

IS - 6

ER -