Assessment of the maximal split-half coefficient to estimate reliability

Barry L. Thompson, Samuel B. Green, Yanyun Yang

Research output: Contribution to journalArticle

7 Citations (Scopus)

Abstract

The maximal split-half coefficient is computed by calculating all possible split-half reliability estimates for a scale and then choosing the maximal value as the reliability estimate. Osburn compared the maximal split-half coefficient with 10 other internal consistency estimates of reliability and concluded that it yielded the most consistently accurate estimate of reliability across the conditions explored. In this simulation study, the authors demonstrate that the maximal split-half coefficient can, under some conditions, badly overestimate reliability because of capitalization on chance and under other conditions appreciably underestimate reliability. Cross-validation can correct the former problem. A difficulty with the cross-validated coefficient is that it is a relatively inefficient estimator of reliability.

Original languageEnglish (US)
Pages (from-to)232-251
Number of pages20
JournalEducational and Psychological Measurement
Volume70
Issue number2
DOIs
StatePublished - Apr 2010

Fingerprint

Coefficient
Estimate
Internal Consistency
Cross-validation
Simulation Study
Estimator
simulation
Demonstrate
Values

Keywords

  • Internal consistency estimates
  • Maximal split-half reliability
  • Maximized lambda4
  • Reliability

ASJC Scopus subject areas

  • Algebra and Number Theory
  • Psychology(all)
  • Developmental and Educational Psychology
  • Psychology (miscellaneous)
  • Education

Cite this

Assessment of the maximal split-half coefficient to estimate reliability. / Thompson, Barry L.; Green, Samuel B.; Yang, Yanyun.

In: Educational and Psychological Measurement, Vol. 70, No. 2, 04.2010, p. 232-251.

Research output: Contribution to journalArticle

Thompson, Barry L. ; Green, Samuel B. ; Yang, Yanyun. / Assessment of the maximal split-half coefficient to estimate reliability. In: Educational and Psychological Measurement. 2010 ; Vol. 70, No. 2. pp. 232-251.
@article{381a047753a14de1995fc07e06cecafc,
title = "Assessment of the maximal split-half coefficient to estimate reliability",
abstract = "The maximal split-half coefficient is computed by calculating all possible split-half reliability estimates for a scale and then choosing the maximal value as the reliability estimate. Osburn compared the maximal split-half coefficient with 10 other internal consistency estimates of reliability and concluded that it yielded the most consistently accurate estimate of reliability across the conditions explored. In this simulation study, the authors demonstrate that the maximal split-half coefficient can, under some conditions, badly overestimate reliability because of capitalization on chance and under other conditions appreciably underestimate reliability. Cross-validation can correct the former problem. A difficulty with the cross-validated coefficient is that it is a relatively inefficient estimator of reliability.",
keywords = "Internal consistency estimates, Maximal split-half reliability, Maximized lambda4, Reliability",
author = "Thompson, {Barry L.} and Green, {Samuel B.} and Yanyun Yang",
year = "2010",
month = "4",
doi = "10.1177/0013164409355688",
language = "English (US)",
volume = "70",
pages = "232--251",
journal = "Educational and Psychological Measurement",
issn = "0013-1644",
publisher = "SAGE Publications Inc.",
number = "2",

}

TY - JOUR

T1 - Assessment of the maximal split-half coefficient to estimate reliability

AU - Thompson, Barry L.

AU - Green, Samuel B.

AU - Yang, Yanyun

PY - 2010/4

Y1 - 2010/4

N2 - The maximal split-half coefficient is computed by calculating all possible split-half reliability estimates for a scale and then choosing the maximal value as the reliability estimate. Osburn compared the maximal split-half coefficient with 10 other internal consistency estimates of reliability and concluded that it yielded the most consistently accurate estimate of reliability across the conditions explored. In this simulation study, the authors demonstrate that the maximal split-half coefficient can, under some conditions, badly overestimate reliability because of capitalization on chance and under other conditions appreciably underestimate reliability. Cross-validation can correct the former problem. A difficulty with the cross-validated coefficient is that it is a relatively inefficient estimator of reliability.

AB - The maximal split-half coefficient is computed by calculating all possible split-half reliability estimates for a scale and then choosing the maximal value as the reliability estimate. Osburn compared the maximal split-half coefficient with 10 other internal consistency estimates of reliability and concluded that it yielded the most consistently accurate estimate of reliability across the conditions explored. In this simulation study, the authors demonstrate that the maximal split-half coefficient can, under some conditions, badly overestimate reliability because of capitalization on chance and under other conditions appreciably underestimate reliability. Cross-validation can correct the former problem. A difficulty with the cross-validated coefficient is that it is a relatively inefficient estimator of reliability.

KW - Internal consistency estimates

KW - Maximal split-half reliability

KW - Maximized lambda4

KW - Reliability

UR - http://www.scopus.com/inward/record.url?scp=77949619498&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=77949619498&partnerID=8YFLogxK

U2 - 10.1177/0013164409355688

DO - 10.1177/0013164409355688

M3 - Article

AN - SCOPUS:77949619498

VL - 70

SP - 232

EP - 251

JO - Educational and Psychological Measurement

JF - Educational and Psychological Measurement

SN - 0013-1644

IS - 2

ER -