estimated by Cronbach’s alpha, varies across cultures, with some items apparently being interpreted differently within and between cultures. Reliability of a large majority of the factors within cultures ranged within acceptable ranges of 0.60 – 0.90. The range for this study was also 0.6 – 0.9. In previous studies (Littrell, 2002; Schneider and Littrell, 2003; Littrell and Baguma, 2004; and Littrell and Valentin, 2005) the factor scores do discriminate between cultures for critical managerial leadership behaviours, yielding useful information for managerial leadership development and application.
Surveys for the South African sample were distributed to 550 part-time MBA students in Pretoria; 308 were returned, with 221 surveys providing sufficient demographic information to select subjects for the race and gender groups. The sample was of varied age (25 - 52 years) and from varied industry sectors. The education level of the sample was high, with 76% having a bachelor’s degree or higher, and the rest having formal education at least at the associate or professional qualification level. The breakdown by race and gender appears in Table 1.
Analyses of variance (ANOVA) were run testing the hypotheses stated below. The mean factor scores for each factor were the dependent variables, and the race x gender groups defined the levels for analysis, with eight groups (the reader should review Table 1 for sample sizes), Asian males and females, Black males and females, Coloured males and females, and White males and females. Noting the varying sample sizes, review of the assumptions of ANOVA and discussions with colleagues with many