Micallef J, McGlangeaud-Freudenthal N, Aurran Y, Julian-Reynier C
INSERM Unité 379, Institut Paoli-Calmettes, Marseille.
Rev Epidemiol Sante Publique. 1998 Nov;46(5):383-9.
Clinical epidemiology and psychosocial studies often take into account the measurement of anxiety. Our objective was to develop and validate a short form scale of anxiety state from the Spielberger's State Trait Anxiety Inventory (STAI-Y), one of the most widely used scales.
This study was carried out on French samples of women with different levels of anxiety. The first step was to select from the original 20-items scale, the items susceptible to compose two short forms scales: an 8-items (18-STAI) and a 5-items (15-STAI) one. It was carried out on patients attending cancer genetic clinics (n1 = 160, n2 = 41). The second step was carried out on 3 other samples of women (n3 = 150, n4 = 167, n5 = 105) and measured the psychometric characteristics of the short-form scales obtained previously.
The correlation coefficients between the short-forms and the complete 20-items form were very high (> 0.90) in particular for the 18-STAI (r > 0.95). Both short forms are sensitive to change. The internal consistency measured by Cronbach-alpha was comparable to the original scale. The results obtained with the 18 scale and another 10 items Spielberger's short-form scale are comparable and better than those with the 15 scale.
The 18 French short-form scale, balanced with positive and negative items, can be recommended for the measurement of state anxiety when the complete form cannot be used.
