13.07.2015 Views

Understanding and Measuring Creative Thinking in Leaders

Understanding and Measuring Creative Thinking in Leaders

Understanding and Measuring Creative Thinking in Leaders

SHOW MORE
SHOW LESS

Create successful ePaper yourself

Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.

194 THE 71F ADVANTAGEskewed (-.41 versus .11), but also considerably less leptokurtic (-.15versus .42), mean<strong>in</strong>g that the preponderance of MM openness scoresare far more positively lean<strong>in</strong>g than those for the IPIP–NEO, as well ashav<strong>in</strong>g a flatter or wider distribution. These results raise the questionwhether the two scales are measur<strong>in</strong>g the same phenomena with respectto our population, at least <strong>in</strong> terms of how they differentially distributeour subjects.It is <strong>in</strong>structive to compare our sample’s openness means on thetwo <strong>in</strong>struments to those from Sauciers’s (1994) MM sample of 1,125community residents of Eugene-Spr<strong>in</strong>gfield, Oregon, <strong>and</strong> Johnson’s(2006) IPIP–NEO sample of 20,993 Web users. For the m<strong>in</strong>i-markers,Saucier reports an openness scale mean of 6.6, or nearly the same asours (6.4). Johnson, however, f<strong>in</strong>ds an openness mean that is decidedlyhigher than ours (3.6 versus 3.1). This comparison alone, however, doesnot make clear whether the differences are due to the <strong>in</strong>struments themselves,or rather the variable nature of the sample populations.The descriptive statistics for the <strong>in</strong>dividual openness MM items<strong>and</strong> IPIP–NEO facets are shown <strong>in</strong> table 8–10. These statistics help clarifythe sources of variability on the respective measures. While thereare statistically significant mean differences among the MM items—forexample, between “<strong>in</strong>tellectual” <strong>and</strong> “philosophical” (t = 7.8, p < .05)—differences are more pronounced on the IPIP–NEO. To cite two examples,means on both O1–Imag<strong>in</strong>ation <strong>and</strong> especially O6–Liberalism arebelow the scale midpo<strong>in</strong>t (neutral) <strong>and</strong> are significantly lower than theO5–Intellect mean (t = 11.1, p < .05 <strong>and</strong> t = 21.0, p < .05, respectively).(Statistics for the <strong>in</strong>dividual IPIP–NEO items are found <strong>in</strong> appendix B.)Multivariate AnalysisA multivariate regression analysis was conducted to help answerthe critical question of how well the MM openness scale represents therelevant facets on the IPIP–NEO openness scale. St<strong>and</strong>ardized betas arepresented to show the relative predictive power of each IPIP–NEO facetvis-à-vis MM scale scores. The results are presented <strong>in</strong> table 8–11.The conclusion from the descriptive analyses that the IPIP–NEOopenness scale might well be measur<strong>in</strong>g at least somewhat differentphenomena than the MM openness scale is considerably re<strong>in</strong>forcedby these results. The MM openness scale scores are most clearly predictedby O5–Intellect with an extremely strong st<strong>and</strong>ardized beta orbeta weight of .62. There is only one other def<strong>in</strong>itely significant effect foran associated IPIP–NEO facet, O4–Adventurousness, though its relative

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!