Frontiers of Entrepreneurship Research
1996 Edition
SUMMARIES

Back to Index96
Order hard copy editions of Frontiers of Entrepreneurship Research by mail


ENTREPRENEURIAL CHARACTERISTICS INVENTORY: VALIDATION PROCESS OF AN INSTRUMENT OF ENTREPRENEURIAL PROFILES


Yvon Gasse

Centre for Entrepreneurship and Small Business
Laval University
Quebec, Canada G1K 7P4

Telephone/Fax: (418) 656-7960 / (418) 656-3337

Principal Topics

The objective of this research is to validate an instrument for measuring entrepreneurial characteristics and identifying entrepreneurial profiles. It also aims to study and compare various target groups in terms of their entrepreneurial potential.

Method

Based on previous research in the field and questionnaires developed over the years by this author, and utilized for the self-evaluation of potential entrepreneurs, a new and more elaborate instrument of 200 items has been pre-tested and analysed on a population of 567 respondents in 1994. The theoretical model from which the instrument has been derived comprised 21 dimensions, all deemed representative of entrepreneurial traits and predispositions. All the items had been proof-read by a group of specialists for their correctness in terms of language, phrasing and construct. For the purpose of this analysis, two groups have been compared; a group of entrepreneurs and a group of general public. Both groups are well matched on relevant socio-demographic dimensions. Research has shown that certain affective, cognitive and behavioral characteristics are stronger in entrepreneurs than in the rest of the population.

Major Findings

Multiple item-analysis has reduced the original instrument to 128 items, comprising 15 dimensions plus a social desirability scale. The Entrepreneurial Characteristics Inventory (ECI) has two abridged versions, 61 and 47 items, mainly used for self-assessment purposes. All three versions have been tested for their metrological performance.

Implications

Reliability

It is a measure of the extent of error instruments produce and where. The ECI92s reliability was estimated on the basis of the internal consistency of the total score and its various dimensions. The internal consistency coefficients alpha (Cronbach, 1951) were calculated and the results were satisfactory, with a total coefficient alpha 0.93 and a median coefficient for the dimensions of 0.67. Moreover, there were no negative or null item-total correlations between the individual items and their respective dimensions. Based on these results, this instrument seems to reliably order individuals.

Validity

It is an estimate of the degree to which the instruments measure what they purport to measure; however, no single validity coefficient can be computed. For instance, construct validity was first estimated by six specialists, who considered either the correspondence of the items in each of the dimensions, or the construct of the items. The convergent and content validity were then estimated using another instrument that shared some traits with the ECI. Known as the "California Psychological Inventory" (CPI), it was administered to a group of respondents immediately after they had completed the ECI. All the correlations for the main CPI and ECI dimensions showed as anticipated. The discriminant validity was assessed using a special scale developed to estimate the tendency of respondents to answer in a socially desirable way. The results for 160 subjects showed that most of the correlations were null, and therefore the ECI was not influenced by such a tendency.

Hypothesis testing

Univariate and multivariate variance analyses were performed on the entrepreneur and general public groups to see if they could be discriminated in the various dimensions in the ECI instrument. The MANOVA results confirm that the ECI is in fact able to discriminate between the two groups, since the F value probabilities are all significant at 0.05, in the direction postulated, with higher scores for the entrepreneurs. The ANOVA tests showed that the entrepreneurial dimensions all discriminated between the groups, except for the "competition" dimension, for which and F value below 0.05 was not obtained.

In sum, the validation analyses were satisfactory and supported the theoretical model. On- going research includes factorial analyses, criterion-related validity, measures of stability and additional measures of equivalence. ECI has already been used for testing entrepreneurial profiles of more than 2000 subjects. Additional comparisons between groups are being performed, such as self-employed people, students and professionals.

Back to Index96