A Meta-Analysis of the relationship between the Rorschach Ego Impairment Index (EII) and psychiatric severity

The present study examined the relationship between the Rorschach Ego Impairment Index (EII) and psychiatric severity. Search procedures yielded 13 independent samples (total N = 1402, average n = 108, standard deviation = 90) for inclusion in the meta-analysis. Inter-rater reliability analyses demonstrated that coding of effect sizes and moderator variables was completed with good to excellent reliability. Results indicated that higher EII scores were associated with greater psychiatric severity, with an overall weighted effect size of r = 0.29, p = 0.000002 (95% confidence interval = 0.17-0.40), supporting the EII's validity as a measure of psychological impairment. Publication bias analyses did not indicate any significant cause for concern regarding the results. The data were demonstrably heterogeneous (Q = 56.82, p = 0.0000001), and results of post-hoc tests indicated that effect sizes with dependent variables obtained via researcher ratings were significantly larger than any of the following: effect sizes with dependent variables obtained via clinician ratings, informant ratings, information about level of treatment or placement status or self-report ratings (p's = 0.0005, 0.003, <0.001, <0.001, respectively). In addition, there was a trend for effect sizes based on performance-based measures to be larger than those based on information about level of treatment or placement status (p = 0.098) as well as those based on self-report measures (p = 0.076). Other moderator analyses were non-significant (p 's > 0.10). Copyright © 2010 John Wiley & Sons, Ltd. Key Practitioner Message: • The Rorschach Ego Impairment Index (EII) demonstrated validity in measuring psychiatric severity across a range of normative, outpatient, residential, and inpatient samples. • The degree of the EII's validity in assessing psychiatric severity compared favorably to the overall validity of the Rorschach and the MMPI. • The EII appears to be most valid in capturing psychiatric severity as measured by researcher ratings of social competency or estimated ego impairment.