Testing the difference between two alpha coefficients with small samples of subjects and raters

Leonard S. Feldt, Seonghoon Kim

Research output: Contribution to journalArticle

29 Citations (Scopus)

Abstract

Researchers sometimes need a statistical test of the hypothesis that two values of Cronbach's alpha reliability coefficient are equal. The situation may involve scores from two different measures administered to independent random samples or from the same measure administered to random samples from two different populations. Feldt derived a test that functions well with large or moderate numbers of subjects. However, he validated this test only when the number of parts (k) of the measurement was fairly large, as it would be if the parts were individual test items. He did not consider instances in which the parts were raters, and hencek would be as small as 2 or 3. In this article, the Feldt test is investigated for such situations. It is found to function quite well in its control of Type I error.

Original languageEnglish
Pages (from-to)589-600
Number of pages12
JournalEducational and Psychological Measurement
Volume66
Issue number4
DOIs
StatePublished - 2006 Jan 1

Fingerprint

Statistical tests
Small Sample
Research Personnel
Testing
Coefficient
random sample
Population
Type I error
Statistical test
Test function
statistical test
Values

Keywords

  • Cronbach's coefficient alpha
  • Difference in reliability estimates
  • Reliability
  • Small samples

Cite this

@article{5a0671ac346e4612915ecb349f07c936,
title = "Testing the difference between two alpha coefficients with small samples of subjects and raters",
abstract = "Researchers sometimes need a statistical test of the hypothesis that two values of Cronbach's alpha reliability coefficient are equal. The situation may involve scores from two different measures administered to independent random samples or from the same measure administered to random samples from two different populations. Feldt derived a test that functions well with large or moderate numbers of subjects. However, he validated this test only when the number of parts (k) of the measurement was fairly large, as it would be if the parts were individual test items. He did not consider instances in which the parts were raters, and hencek would be as small as 2 or 3. In this article, the Feldt test is investigated for such situations. It is found to function quite well in its control of Type I error.",
keywords = "Cronbach's coefficient alpha, Difference in reliability estimates, Reliability, Small samples",
author = "Feldt, {Leonard S.} and Seonghoon Kim",
year = "2006",
month = "1",
day = "1",
doi = "10.1177/0013164405282488",
language = "English",
volume = "66",
pages = "589--600",
journal = "Educational and Psychological Measurement",
issn = "0013-1644",
number = "4",

}

Testing the difference between two alpha coefficients with small samples of subjects and raters. / Feldt, Leonard S.; Kim, Seonghoon.

In: Educational and Psychological Measurement, Vol. 66, No. 4, 01.01.2006, p. 589-600.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Testing the difference between two alpha coefficients with small samples of subjects and raters

AU - Feldt, Leonard S.

AU - Kim, Seonghoon

PY - 2006/1/1

Y1 - 2006/1/1

N2 - Researchers sometimes need a statistical test of the hypothesis that two values of Cronbach's alpha reliability coefficient are equal. The situation may involve scores from two different measures administered to independent random samples or from the same measure administered to random samples from two different populations. Feldt derived a test that functions well with large or moderate numbers of subjects. However, he validated this test only when the number of parts (k) of the measurement was fairly large, as it would be if the parts were individual test items. He did not consider instances in which the parts were raters, and hencek would be as small as 2 or 3. In this article, the Feldt test is investigated for such situations. It is found to function quite well in its control of Type I error.

AB - Researchers sometimes need a statistical test of the hypothesis that two values of Cronbach's alpha reliability coefficient are equal. The situation may involve scores from two different measures administered to independent random samples or from the same measure administered to random samples from two different populations. Feldt derived a test that functions well with large or moderate numbers of subjects. However, he validated this test only when the number of parts (k) of the measurement was fairly large, as it would be if the parts were individual test items. He did not consider instances in which the parts were raters, and hencek would be as small as 2 or 3. In this article, the Feldt test is investigated for such situations. It is found to function quite well in its control of Type I error.

KW - Cronbach's coefficient alpha

KW - Difference in reliability estimates

KW - Reliability

KW - Small samples

UR - http://www.scopus.com/inward/record.url?scp=33745965825&partnerID=8YFLogxK

U2 - 10.1177/0013164405282488

DO - 10.1177/0013164405282488

M3 - Article

AN - SCOPUS:33745965825

VL - 66

SP - 589

EP - 600

JO - Educational and Psychological Measurement

JF - Educational and Psychological Measurement

SN - 0013-1644

IS - 4

ER -