Validating Research Results when Cronbach'S Alpha is Below .70: A Methodological Procedure
- 1 September 1994
- journal article
- research article
- Published by SAGE Publications in Educational and Psychological Measurement
- Vol. 54 (3) , 766-775
- https://doi.org/10.1177/0013164494054003023
Abstract
The study involved 494 auditors from five Big Six accounting firms. Although two of the cognitive tests used in the research had Cronbach's alphas in the .70 to .90 range, the Defining Issues Test had an alpha of .35. This research investigated the relationship between cognitive test elements and the sample to determine the causes of the low alpha. The procedure described in the research is a method that can be used to validate research data when Cronbach's alpha is below .70. The research findings indicate that unbundling both the test and the sample provide statistically sound justification for continuing to use all research data for a highly homogeneous sample.Keywords
This publication has 5 references indexed in Scilit:
- Ethical reasoning and selection-socialization in accountingAccounting, Organizations and Society, 1992
- Auditor independence judgments: A cognitive‐developmental model and experimental evidence*Contemporary Accounting Research, 1990
- The Reliability and Validity of Objective Indices of Moral DevelopmentApplied Psychological Measurement, 1978
- Generalized expectancies for internal versus external control of reinforcement.Psychological Monographs: General and Applied, 1966
- Coefficient alpha and the internal structure of testsPsychometrika, 1951