Document Type



Doctor of Philosophy


Adult & Higher Education

Date of Defense


Graduate Advisor

Steven Spaner, Ph.D.


John Henschke, Ed.D.


Galovski, Tara

Pi Chi Han, Ed.D.


ABSTRACT This study utilized confirmatory factor analysis (CFA), canonical correlation analysis (CCA), regression analysis (RA), and correlation analysis (CA) for first-round validation of the researcher¿s Dimensions of Intuition (DOI) instrument. The DOI examined 25 personal characteristics and situations purportedly predictive of intuition. Data was collected from 302 respondents, ages 20-79, from differing occupations and educational backgrounds nationwide. Hypothesis 1: CFA disconfirmed the theorized 3- and 21-factor intuition models, finding 15 factors, accounting for 65.6% of the variance, to be the most efficient capture of intuition. Hypothesis 2: CCA tested the relationship between the 15 factors and the brain quadrants, as measured by the Herrmann Brain Dominance Instrument® (HBDI®). Seven factors loaded on quadrant A; nine each on quadrants B, C and D, confirming this hypothesis. Hypothesis 3: RA was used to test the relationship between the 15 factors and the HBDI® brain hemispheres. An R-squared value of .667 was found for the right/left hemispheres; .575 for the cerebral/ limbic hemispheres, confirming this hypothesis. Hypotheses 2 and 3 findings provided some evidence of intuition as a whole-brained functionality, with right/left scores providing the most discriminative value. Hypothesis 4: CA was utilized to examine the relationship between the DOI total and variable T scores and the six subscales of the Personal Style Inventory (PSI). Expected directions were found for 47 of 54 significant correlations between the variable scores and subscales (87% hit rate). Significant correlations in expected directions were also found between the DOI total score and the Control, Vision and Insight subscales. The overall conclusion supports the DOI¿s validity and reliability; though additional validation studies with other populations and other statistical methods, including structural equation modeling and multi-dimensional scaling, are recommended.

Included in

Education Commons