Impact of the Number of Response Categories and Anchor Labels on Coefficient Alpha and Test-Retest Reliability

By Li-Jen Weng. 2004

430 citations recorded by the source corpus.

Subfields: Management Science and Operations Research, Statistics and Probability

View Paper: https://www.semanticscholar.org/paper/551bd53af553a9e1ce5b7043b6df909c3ca04df4

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Management Science and Operations Research

1. Betty Becker-Kurz
2. Thomas Goetz
3. L. E. Garrido
4. Amery Wu
5. Vincent Ng
6. T. Thomas
7. F. J. Abad
8. A. Frenzel

In Statistics and Probability

1. D. McCoach
2. J. Adelson
3. Thomas Goetz
4. L. E. Garrido
5. Amery Wu
6. Vincent Ng
7. T. Thomas
8. F. J. Abad

Citations in the corpus

• Measuring the Mathematical Attitudes of Elementary Students: The Effects of a 4-Point or 5-Point Likert-Type Scale (2010, background): “Hypothesized explanations for the variability in the effects of the number of response categories on internal consistency reliability include the scale’s item homogeneity (Bourke & Frampton, 1992; Komorita & Graham, 1965; Weng, 2004) and the variability of opinions (Masters, 1974; Weng, 2004).”
• Measuring Teachers’ enjoyment, anger, and anxiety: The Teacher Emotions Scales (TES) (2016, background): “The 4-point scale also made it possible to meaningfully verbally label each of the answers, which has been shown to increase the psychometric quality of scales (Weng, 2004).”
• The Impact of Outliers on Cronbach’s Coefficient Alpha Estimate of Reliability: Ordinal/Rating Scale Item Responses (2010, background): “However, it is not yet clear how it affects the associated psychometric properties (e.g., Krieg, 1999; Weng, 2004).”
• National cultures and safety orientation: A study of seafarers working for Norwegian shipping companies (2007, background): “An even number of response alternatives forces commitment from the respondents, and research indicates that scales with more response categories tend to result in higher reliability (Weng, 2004).”
• A Review of Key Likert Scale Development Advances: 1995–2019 (2021, methodology): “The advice in this table was taken from Mackenzie (2003), Locke (2012), and Podsakoff et al.”
• Effect of the Number of Response Categories on the Reliability and Validity of Rating Scales (2008, background): “Some authors (Aiken, 1983; Boote, 1981; Brown, Widing, & Coulter, 1991; McCallum, Keith, & Wiebe, 1988, Weng, 2004) maintain that although reliability tends to increase as the number of response options rises, when the number of alternatives exceeds five or six, reliability hardly increases further.”
• Response Styles in Survey Research: A Literature Review of Antecedents, Consequences, and Remedies (2013, methodology): “Response scales have also been examined with many other criteria— for example, reliability, information recovery, distribution of scale means, and ease of use (see Preston & Colman, 2000; Weng, 2004)—resulting in similar recommendations with respect to the optimal number of response categories.”
• Validity: Meaning and Relevancy in Assessment for Engineering Education Research (2015, background): “Educational measurement experts have extensively discussed the appropriate use of the coefficient alpha as a gauge for internal consistency and ways to increase consistency (Cortina, 1993; Schmitt, 1996; Weng, 2004).”
• Impact of the Number of Scale Points on Data Characteristics and Respondents’ Evaluations: An Experimental Design Approach Using 5-Point and 7-Point Likert-type Scales (2016, neutral): “; L.J.Weng, (2004).”
• Relation between active mediation, exposure to Daniel Tiger’s Neighborhood, and US preschoolers’ social and emotional development (2016, background): “…options consistent with response options for other measures in the survey, and to coincide with best practices for the appropriate number of response options (between 3 and 7) to achieve greatest validity and reliability (Krosnick & Presser, 2010; Lozano, García-Cueto, & Muñiz, 2008; Weng, 2004).”
• Performance of Velicer’s Minimum Average Partial Factor Retention Method With Categorical Variables (2011, methodology): “Furthermore, we opted for a maximum of 7 response categories because the reliability and validity of ordinal measures appears to level off after reaching this number of scale points (Preston & Colman, 2000; Weng, 2004).”