Our initial analysis took the form of a regression model in which Time Utilization was regressed on Primary Goal, Management Level and Organization Size. We found all coefficients to be significant. However, to improve our analyses we employed path analysis using OLS regression (Cohen and Cohen, 1983). We adopted this approach primarily for two reasons. First, the relationship among variables in the study suggested both direct paths and indirect explanatory paths. Second, the regression results suggested the presence of multicollinearity, rendering the estimated coefficients of the regression analysis unstable. By employing path analysis, we were able to address these issues. Within a given structural model, path analysis, unlike multiple regressions, can determine what part of a correlation is due to the direct effect and what part is due to indirect effects (Kerlinger, 1973).

Figure 1 shows the expected direct and indirect relationships between management level (Mgmtlev), primary goals of the entrepreneur (Primgoal), and time utilization by the entrepreneur (Timeutl). The control variable of initial organizational size (Qemp1) is also shown. Figure 1 also shows the expected direct and indirect relationships between these variables and organizational performance (Takeot). As is the standard practice in path-analytic methodology, we have reported standardized coefficients for each of the paths (see Figure 2). Standardized coefficients are reported in path-analytic applications because of their ease of interpretability (Pedhazur and Schmelkin, 1991). Since we have variables with considerably different units of measurement, standardized coefficients make interpretation much easier.


Definitions of Variables

Dependent Variables


Time allocated by entrepreneurs as a percentage of their total work hours to craft activities (selling/customer contact, production, maintenance, and dealing with suppliers) and to administrative activities (dealing with employees, arranging finance, record keeping, and planning).

Operationalized as the percentage of time spent on craft activities; values range from 0 to 100.

Independent Variables


Defined as a function of the entrepreneurs’ goals when going into business. Goals include items "to do the kind of work I want," "avoid working for others," comfortable living is enough," "build something for family," and "earn lots of money."

The first two items and the second three items were added and standardized separately. These standardized scores were then added and were standardized again to obtain a measure with mean of zero and unit variance.


  1. Worker
  2. Supervisor
  3. Manager
  4. Owned business


Number of employees in the initial year


Amount of cash taken out by entrepreneurs in year three and is measured on 0-7 scale, with 0 indicating less than $10,000 and 7 indicating greater than $75,000


Descriptive Statistics and Correlation Coefficients

  Takeot Timeutl Primgoal Mgmtlev Qemp1
Mean 4.09 62.45 0.00 2.53 4.64
Std. Dev. 1.70 21.80 1.00 1.07 4.59
Takeot 1.00 -0.17 -0.17 0.17 0.28
Timeutl   1.00 0.26 -0.17 -0.36
Primgoal     1.00 -0.16 -0.25
Mgmtlev       1.00 0.17
Qemp1         1.00

N = 432

Path Analyses - Significant Paths

(Not available)


As hypothesized in Hypothesis 1, we expect entrepreneurs emphasizing craftsman goals (Primgoal) to allocate relatively more time to craft type activities (Timeutl) and less time to administrative activities. We found support for this hypothesis, suggesting that these entrepreneurs emphasize those activities which appeal to them.

In Hypothesis 2, we hypothesized that entrepreneurs’ primary goals are likely to be influenced by their previous management experience (Mgmtlev). The analysis provided support for this hypothesis, suggesting that entrepreneurs with little management experience are more likely to pursue the goals of craftsmen entrepreneurs, while those with higher management experience are more likely to pursue the goals of administrative entrepreneurs.

We posit a negative relationship between management experience and time applied to craftsman activities in Hypothesis 3. We found support for this hypothesis, suggesting that entrepreneurs with experience at higher levels are more likely to devote time to administrative activities than those with experience at lower levels. Craftsmen entrepreneurs are likely to spend most of their time on craft activities such as selling and producing, while ignoring administrative tasks. The control variable of organizational size (Qemp1) showed the expected relationships. Firms which were larger at startup were less likely to have owner-managers with craftsman goals. These larger startups were also less likely later to have entrepreneurs who devoted time to selling, producing, and other nonadministrative activities.

In regard to new venture performance, it was expected that firms whose entrepreneurs devoted relatively more time to administrative activities would show better performance. The findings were only suggestive at the .10 level, with those firms whose owner-managers devoted more time to craftsmen activities (Timeutl) showing somewhat lower performance (Takeot) (see Figure 2).

To examine the goodness of fit of the path-analytic model, we employed the Q statistic (Pedhazur, 1982). The Q statistic is based on the sum of all the squared residual paths (1-R2) for each of the regression equations used in the path model. Values of Q can range from 0 to 1, with higher values indicating better fit. In this analysis, the value of Q was 0.99, indicating a good fit.



The entrepreneur is at the center of any new venture. Initially, the strengths and weaknesses of the firm reflect the capabilities and interests of the entrepreneur. Thus, this paper considers differences in how entrepreneurs allocate their time, whether these differences are related systematically to antecedent variables and whether this time allocation makes a difference in organizational performance.

Drawing upon the extant literature on entrepreneurial typologies, we utilize the constructs of craftsmen-entrepreneurs and administrative entrepreneurs. The former emphasize non-economic goals when starting, such a doing work they want to do and avoiding working for others. They are less likely to have had previous high level managerial experience. Subsequently, as they manage their ventures, we find that they devote relatively less time to administrative activities and more to the "doing" activities of sales, production, etc. In contrast, administrative entrepreneurs emphasize economic goals when starting and are less likely to see the business as a vehicle for a particular lifestyle. They are more likely to have had previous high-level management experience. As they develop their ventures, they delegate more of the "doing" activities and allocate more of their time to administrative activities. Prior levels of managerial experience have both direct and indirect effects on time allocation. In regard to performance, there is some indication (p=.10) that their firms are more successful, in the sense that they are able to draw more cash out of their businesses.

This is, we think, the first study to examine explicitly relationships between entrepreneurial typologies, antecedent experience, time allocation, and venture performance. There is a strong suggestion that entrepreneurs tend to do that which they are most familiar with. Entrepreneurs and their advisors should recognize that this may or may not be in the best interests of their ventures.

It should be recognized that there are a number of limitations to this research. Time allocation data are self-reported and may not be accurate. Furthermore, the way in which the time is allocated is measured only one way. The research on managerial work (Mintzberg, 1975) is a reminder that there are many ways to measure and classify managerial work. There are also other antecedent and contextual variables not considered here which may influence the way entrepreneurs choose to spend their time. These include the specific functional experience and management style of the entrepreneur, the strategy of the firm, and the extent to which partners or outside professionals are utilized in developing the business. Thus, there are many opportunities for further research on entrepreneurial behavior and the ways that entrepreneurs allocate their time.

Previous Page | Main Menu | Next Page



1997 Babson College All Rights Reserved
Last Updated 1/15/97 by Geoff Goldman & Dennis Valencia

To sign-up for the Center for Entrepreneurial Studies' publication lists,
please register with the
Entrepreneurship WebTeam.