The questions of validity and reliability are often raised with regard to Discus and other similar systems. This section of the site reproduces a South African study, A Reliability and Validity Study on the Discus Personality Profiling System, by Karin Roodt of the prestigious Technikon Natal. Note that this study - known as The Roodt Report - relates not just to DISC in general but specifically to the Discus software.
A Reliability and Validity Study on the Discus Personality Profiling System
K. Roodt (Ms)
The purpose of the investigation was to determine whether the Discus measuring instrument could be considered a reliable and valid instrument. The test-retest method was used in the reliability study and was administered to 90 employees from a variety of companies in Kwa Zulu-Natal and Gauteng. The Pearson's product-moment correlation coefficient was used and correlation scores of 0.728 (Dominance), 0.645 (Influence), 0.730 (Steadiness) and 0.550 (Compliance) were established. The p-value in all the cases was as low as 0.0001. This indicates significance at alpha = 0.001. It can therefore be concluded with 99.9% level of confidence that the Discus instrument is reliable.
In the validity exercise criterion-related validity was used. An exploratory study was undertaken in order to determine which of the 15 Factors (Factor B excluded) of the 16-PF correlated with the four dimensions of the Discus. One hundred and twenty respondents in South Africa were involved for this purpose. The Pearson's product-moment correlation coefficient was applied. It was found that Factors Q1, X=G, L, Q1 and X=Q2, E; E, Q2 and -I show significant correlations with Dominance at the 1% and 5% level of significance. Factors A, -Q2, H, F and -Q3 show significant correlations with Influence at the 1% and 5% level of significance. Factors -E and -Q1 show a correlation with Steadiness at the 5% level of significance. Factors -E, Q2, -H, -G and O show significant correlations with Compliance at the 1% and 5% level of significance. It can therefore be concluded that the correlations were significant.