Robust estimation of Cronbach's alpha

Abstract

Cronbach's alpha is a popular method to measure reliability, e.g. in quantifying the reliability of a score to summarize the information of several items in questionnaires. The alpha coefficient is known to be non-robust. We study the behavior of this coefficient in different settings to identify situations where Cronbach's alpha is extremely sensitive to violations of the classical model assumptions. Furthermore, we construct a robust version of Cronbach's alpha which is insensitive to a small proportion of data that belong to a different source. The idea is that the robust Cronbach's alpha reflects the reliability of the bulk of the data. For example, it should not be possible that some small amount of outliers makes a score look reliable if it is not.

Ethical Decision Making in Organizations: The Role of Leadership Stress

“…Since the sample was small, more robust measures were taken into account as suggested by Christmann and Van Aelst (2006). Thus the Cronbach's α based on Tukey's S-estimator (S bw ) was 0.63.…”

“…Without the time consuming The mean value with the sample on this part in Study 1 was 1.24 (SD = 0.43), and the Cronbach's α based on the classical approach was 0.57. Since the sample was small, more robust measures were taken into account as suggested by Christmann and Van Aelst (2006). Thus, the Cronbach's α based on Tukey's S-estimator (S bw ) was 0.60.…”

Pharmacoethics: Development and Validation of Instrument to Measure Knowledge, Perception and Awareness of Graduating Pharmacy Students on Pharmacy Code of Conduct in Malaysia

“…Cronbach's alpha is one of the statistical techniques (Christmann & Aelst, 2006), that was used to determine the reliability in this study.…”

Aykırı değer varlığında sağlam konum ve ölçek kestiricilerinin faktör analizinde kullanımı

“…Veri setinde aykırı değer olması durumunda bu iki kestirici etkileneceğinden klasik faktör analizinin performansı da buna bağlı olarak kötüleşecektir [4]. Tek bir aykırı değer varlığı korelasyon kestirimlerinin yanı sıra Cronbach alfa ölçüsünü, kestirimi yapılan faktör analizi model parametrelerini ve modelin uyum iyiliği ölçülerini bile bozabilir [5][6][7]. Sonuç olarak aykırı değerlerin yanlış model belirlenmesine yol açabildiği ve modele ekstra bir faktörün eklenmesine bile sebep olabileceği ileri sürülmüştür [8].…”