Factor Analysis of the Socio-Cultural Itemsa
|7. People look up to those who run own firms||0.77|
|2. Running own firm gives high social status||0.75|
|14. Starting business generates respect||0.68|
|17. Starting business does not bring prestige||0.66|
|13. Owner whose company fails brings shame||0.8|
|4. When company fails, this brings shame||0.75|
|10. All right to start a business, not acceptable to fail||0.69|
|19. People may avoid starting business from fear of shame of failure||0.51|
|12. People focus on succeeding at work||0.79|
|5. Work is very important to most people||0.76|
|3 High value on innovativeness||0.59|
|9. People who invent are highly respected||0.57|
a. All factor loadings above .50 are included in the table.
In Table 2, the first factor contained the four social status items, the second factor contained the four shame of failure items, and the third factor combined two value of innovation items with two value of work items. Since we believed that these latter two scales measured different constructs, we used them separately in subsequent analysis. We constructed a two-item scales for both innovation and the value of work. Cronbach alpha reliabilities for the social status, innovation, shame of failure, and value of work scales were .73, .74, .69, and .71, respectively. A complete listing of the socio-cultural items is included in the Appendix.
The dependent variables of feasibility, desirability, and intent were also measured through multi-item scales developed for this research project. The desirability of starting a business and intent to start one both were measured with four-items scales. The intent scale was placed in the first part of the questionnaire while the desirability scale was placed in the last part. Sample intent items included, "I intend to start my own firm within the next ten years," and "I have no intention of starting my own firm in the future," (the latter was reverse scored). The coefficient alpha reliability for the intent scale was .91. Sample desirability items included, "A major dream in my life is to start a business of my own," and "To initiate a new business venture would give me much satisfaction." The coefficient alpha for the desirability scale was .91.
In measuring the feasibility of starting a business, the methodology used was similar to that used by researchers measuring self-efficacy (e.g., Lee & Bobko, 1994). In fact, feasibility was viewed as synonymous with self-efficacy. Self-efficacy involves beliefs about one's ability to perform specific tasks to completion and with effectiveness (Bandura, 1986). In essence, the measure of feasibility is the respondents' self-assessments of their ability to successfully start a business. As such, respondents were first asked to indicate with a simple yes or no response whether they believed that they could perform each of seven tasks related to starting a business. Sample tasks included: "develop a good concept on which to start a business," "raise enough funds to start a business," and "find enough skilled employees to start and run a business." Then, for each of the tasks they had answered "yes," respondents were asked to indicate "how confident you are that you can perform the activity" on a scale of 0 to 10. This confidence indicator was used as the measure of feasibility. The reliability of this seven-item scale was .85.
The fifteen items used to construct the three dependent variables were included in exploratory factor analysis using varimax rotation for each of the countries in the sample. Since the results were similar across countries, the data were combined and an overall factor analysis conducted. Examination of the scree plot indicated that the two-factor solution provided the best fit, with all the desirability and intent items loading on the first factor and the seven feasibility items loading on the second factor. Nevertheless, since Shapero and Sokol's (1982) theory presents desirability and intent as separate steps in the process of decision making to become an entrepreneur, the two variables will be used separately in the analysis to follow.
As Table 1 indicated, demographic items were included in the questionnaire
to measure age, sex, marital status, years worked, and whether an individual
ever owned a business. Since age and years of work experience correlated
.88, only age will be used in subsequent analysis.