The reliability of a two-item scale: Pearson, Cronbach, or Spearman-Brown?

Excerpt

To obtain reliable measures researchers prefer multiple-item questionnaires rather than single-item tests. Multiple-item questionnaires may be costly however and time-consuming for participants to complete. They therefore frequently administer two-item measures, the reliability of which is commonly assessed by computing a reliability coefficient. There is some disagreement, however, what the most appropriate indicator of scale reliability is when a measure is composed of two items. The most frequently reported reliability statistic for multiple-item scales is Cronbach’s coefficient alpha and many researchers report this coefficient for their two-item measure (Cuijpers et al. 2009; Löwe et al. 2005; Michal et al. 2010; Young et al. 2009). Others however claim that coefficient alpha is inappropriate and meaningless for two-item scales (Sainfort and Booske 2000; Verhoef 2003; Cramer et al. 2006; O’Brien et al. 2008). Instead, they recommend using the Pearson correlation coefficient as a measure of reliability. Still others argue that the inter-item correlation equals the split-half reliability estimate for the two-item measure and they advocate the use of the Spearman-Brown formula to estimate the reliability of the total scale (Hulin et al. 2001). As these recommendations are reported without elaborating, there is considerable confusion among end users as to the best reliability coefficient for two-item measures. This note aims to clarify the issue. …