Toward using confidence intervals to compare correlations.

Abstract

Confidence intervals are widely accepted as a preferred way to present study results. They encompass significance tests and provide an estimate of the magnitude of the effect. However, comparisons of correlations still rely heavily on significance testing. The persistence of this practice is caused primarily by the lack of simple yet accurate procedures that can maintain coverage at the nominal level in a nonlopsided manner. The purpose of this article is to present a general approach to constructing approximate confidence intervals for differences between (a) 2 independent correlations, (b) 2 overlapping correlations, (c) 2 nonoverlapping correlations, and (d) 2 independent R 2 s. The distinctive feature of this approach is its acknowledgment of the asymmetry of sampling distributions for single correlations. This approach requires only the availability of confidence limits for the separate correlations and, for correlated correlations, a method for taking into account the dependency between correlations. These closed-form procedures are shown by simulation studies to provide very satisfactory results in small to moderate sample sizes. The proposed approach is illustrated with worked examples.

Expectations Are More Predictive of Behavior than Behavioral Intentions: Evidence from Two Prospective Studies

“…Evidence for discriminant validity between behavioral intention and expectation is provided by the intercorrelation at baseline, r = .68, which is significantly weaker than unity (i.e., by more than twice the standard error, SE = .05). Both behavioral intention and expectation were significantly correlated with subsequent alcohol consumption, but the correlation between subsequent alcohol consumption and expectation was stronger, r = .41, p < .01, than that between subsequent alcohol consumption and behavioral intention, r = .22, p < .01; a difference that was statistically significant (95%CI = 0.07, 0.31, p < .05, [13]). …”

Conscientiousness and procrastination predict academic coursework marks rather than examination performance

“…We used Zou's (2007) approach to calculating confidence intervals for differences between overlapping correlations. In the case of conscientiousness, the difference was .27, one-tailed 95% CI [.17, 1.00]; for procrastination, the difference was -.15, [-1.00, -.03].…”

Programas Informáticos Para Comparaciones Entre Correlaciones: Muestras Independientes

“…Recientemente, Zou (2007) desarrolló un método basado en intervalos de confianza para examinar las diferencias entre dos correlaciones independientes. La premisa es que el intervalo de confianza informa la magnitud y la precisión de la relación, mientras que la prueba de hipótesis estándar mezcla estas dos características.…”

“…La premisa es que el intervalo de confianza informa la magnitud y la precisión de la relación, mientras que la prueba de hipótesis estándar mezcla estas dos características. (Zou, 2007). Un programa interactivo y amigable, llamado COMPCOR (Silver, 2009) Pero supongamos que uno quiere incluir otros dos grupos, por ejemplo, profesores de educación técnica y de Universidad; entonces uno podría probar la hipótesis nula que las cuatro correlaciones son equivalentes.…”