Jesus Sanz, Department of Clinical Psychology, Universidad Complutense de Madrid
14 May 2005
Cronbach’s alpha for the Greek version of the Zung Depression Rating Scale (ZDRS) was surprisingly low (0.09) and is not consistent with alpha coefficients reported in literature. Previous studies have found high alpha coefficients around 0.70s- 0.80s for a variety of samples: psychiatric patients, medical patients, university students, adults from the general population, and so on (see, for example, Conde López, V., et al., 1975, Revista de Psicologia General y Aplicada, 30: 903-913; Dugan, W. et al., 1998, Psycho-Oncology, 7:483-493; Olukayode Jegede, R., 1976, The Journal of Psychology, 93:27-30; Tanaka-Matsumi, J., et al., 1986, Journal of Consulting and Clinical Psychology, 54:328-333; Kivelä, S.-L., et al., 1987; Journal of Clinical Psychology, 43:318-327; Dunn, V. K., et al., 1989, Psychology & Aging, 4:125-126; Knight, R. G., et al., 1983, British Journal of Clinical Psychology, 22: 245-249). Furthermore, test-retest correlation for the Greek ZDRS was so high as 0.92, whereas alpha was so low as 0.09. Although test-retest reliability and internal consistency reliability are not just the same, both indices are reliability estimates and, therefore, such a high discrepancy does not make sense. Given that the ZDRS is comprised of 10 direct items (e.g., “I feel down-hearted and blue”) and 10 inverse items (e.g., “I eat as much as I used to”), a failure to reverse the inverse items prior to calculate the alpha coefficient may explain the very low alpha reported in this study. I would suggest authors to review the calculation of Cronbach’s alpha and check if the 10 inverse items of the ZDRS were properly reversed prior to calculate alpha.
A possible error in calculating alpha coefficient
14 May 2005
Cronbach’s alpha for the Greek version of the Zung Depression Rating Scale (ZDRS) was surprisingly low (0.09) and is not consistent with alpha coefficients reported in literature. Previous studies have found high alpha coefficients around 0.70s- 0.80s for a variety of samples: psychiatric patients, medical patients, university students, adults from the general population, and so on (see, for example, Conde López, V., et al., 1975, Revista de Psicologia General y Aplicada, 30: 903-913; Dugan, W. et al., 1998, Psycho-Oncology, 7:483-493; Olukayode Jegede, R., 1976, The Journal of Psychology, 93:27-30; Tanaka-Matsumi, J., et al., 1986, Journal of Consulting and Clinical Psychology, 54:328-333; Kivelä, S.-L., et al., 1987; Journal of Clinical Psychology, 43:318-327; Dunn, V. K., et al., 1989, Psychology & Aging, 4:125-126; Knight, R. G., et al., 1983, British Journal of Clinical Psychology, 22: 245-249). Furthermore, test-retest correlation for the Greek ZDRS was so high as 0.92, whereas alpha was so low as 0.09. Although test-retest reliability and internal consistency reliability are not just the same, both indices are reliability estimates and, therefore, such a high discrepancy does not make sense. Given that the ZDRS is comprised of 10 direct items (e.g., “I feel down-hearted and blue”) and 10 inverse items (e.g., “I eat as much as I used to”), a failure to reverse the inverse items prior to calculate the alpha coefficient may explain the very low alpha reported in this study. I would suggest authors to review the calculation of Cronbach’s alpha and check if the 10 inverse items of the ZDRS were properly reversed prior to calculate alpha.
Competing interests
No competing interests