Entrepreneurial self-efficacy: refining the measure.

A growing number of studies on entrepreneurial motivation, intentions, and behavior include entrepreneurial self-efficacy (ESE) as an explanatory variable. While there is broad consensus among researchers on the importance of including ESE in an intentionality model, there remain inconsistencies in the definition, dimensionality, and measurement of ESE. This study

takes an important step toward refinement and standardization of ESE measurement. Within a new venture creation process framework, a multi-dimensional ESE instrument is developed and tested on a diverse sample that includes nascent entrepreneurs. Implications for entrepreneurship theory and entrepreneurship education are discussed.

Introduction

Entrepreneurship-oriented intentions are considered precursors of entrepreneurial action (Bird, 1988; Kolvereid, 1996; Krueger & Brazeal, 1994; Krueger, Reilly, & Carsrud, 2000). In order to further develop entrepreneurship theory, researchers need an understanding of the factors that might influence the intentions of those considering entrepreneurship for the first-time nascent entrepreneurs (Carter, Gartner, & Reynolds, 1996; Reynolds, Carter, Gartner, & Greene, 2004; Rotefoss & Kolvereid, 2005). Factors that would influence one to become an entrepreneur are many, and consist of various combinations of personal attributes, traits, background, experience, and disposition (Arenius & Minniti, 2005; Baron, 2004; Krueger et al.; Shane, Locke, & Collins, 2003).

One of these personal attributes, entrepreneurial self-efficacy (ESE), appears to be a particularly important antecedent to new venture intentions (Barbosa, Gerhardt, & Kickul, 2007; Boyd & Vozikis, 1994; Zhao, Seibert, & Hills, 2005). Simply stated, ESE is a construct that measures a person's belief in their ability to successfully launch an entrepreneurial venture. ESE is particularly useful since it incorporates personality as well as environmental factors, and is thought to be a strong predictor of entrepreneurial intentions and ultimately action (Bird, 1988; Boyd & Vozikis). Moreover, recent research suggests that an individual's ESE may be elevated through training and education; thus, potentially improving the rate of entrepreneurial activities (Florin, Karri, & Rossiter, 2007; Mueller & Goic, 2003; Zhao et al.).

While the ESE construct is quite promising, it remains empirically underdeveloped and many scholars have called for further refinement of the construct (e.g., Forbes, 2005; Kolvereid & Isaksen, 2006). Three issues, in particular, appear to warrant further investigation and serve as the motivation for this study. First, there remains some debate on whether an ESE construct is even necessary. Several scholars (see Chen, Gully, & Eden, 2004) advocate the use of a general measure of self-efficacy instead of a domain-specific ESE construct. Second, the dimensionality of the construct has yet to be fully established. While most scholars acknowledge the multi-dimensional nature of the ESE construct (e.g., Wilson, Kickul, & Marlino, 2007; Zhao et al., 2005), very few researchers have explicitly examined the underlying dimensions that make up the actual construct by using some type of theoretical model of entrepreneurial activity and tasks. Moreover, several scholars have simply relied on single survey questions to capture an individual's level of ESE. Finally, very few studies have included a sampling of nascent entrepreneurs (Forbes). Rather, most of the initial studies of ESE relied on samples of university students or samples of small business owners (Chen, Greene, & Crick, 1998; De Noble, Jung, & Ehrlich, 1999; Drnovsek & Glas, 2002; Mueller & Goic, 2003).

The current study attempts to advance the understanding of ESE and its effect on venture intentions by developing a multi-dimensional measure of ESE within a four-phase venture creation process framework. The instrument's reliability and validity is then tested, using a diverse sample that includes nascent entrepreneurs--individuals who are engaged in activities that are intended to result in a new business--and non-nascent entrepreneurs.

Measuring ESE

Self-efficacy refers to an individual's belief in their personal capability to accomplish a job or a specific set of tasks (Bandura, 1977). Self-efficacy is a useful concept for explaining human behavior as research reveals that it plays an influential role in determining an individual's choice, level of effort, and perseverance (Chen et al., 2004). Simply stated, individuals with high self-efficacy for a certain task are more likely to pursue and then persist in that task than those individuals who possess low self-efficacy (Bandura, 1997).

Self-efficacy, when viewed as a key antecedent to new venture intentions, is referred to as ESE (Boyd & Vozikis, 1994; Chen et al., 1998; Krueger & Brazeal, 1994). Although the literature on ESE is quite robust, there remain at least three obstacles that impede further development and effective application of the construct. First, disagreement exists as to whether the ESE construct is more appropriate than general self-efficacy (GSE). Second, there is inconsistency in the manner in which researchers attempt to capture the dimensionality of the ESE construct. Third, ESE researchers appear to be overly reliant on data collected from university students and practicing entrepreneurs. Each of these potential obstacles is discussed in the following paragraphs. A summary of relevant empirical studies is also provided in Table 1.

GSE Versus ESE

There remains fundamental disagreement regarding the very need for an ESE construct. Some theorists argue that a GSE construct is sufficient, as it is a relatively stable, trait-like, generalized competence belief (Chen et al., 2004). GSE captures an individual's perception of their ability to successfully perform a variety of tasks across a variety of situations. In other words, GSE refers to an individual's confidence in meeting task demands, regardless of those demands.

Researchers advocate the use of a measure of GSE because entrepreneurs require a diverse set of roles and skill sets; therefore, they believe it would simply be too difficult to identify a comprehensive, yet parsimonious, list of specific tasks explicitly associated with entrepreneurial activities (Markman, Balkin, & Baron, 2002). From a purely pragmatic perspective, it is much easier to measure GSE than to explicitly capture the nuances of ESE. In any event, several empirical studies have measured self-efficacy by eliciting responses about an individual's confidence in various areas not specific to entrepreneurial activities (e.g., Baum & Locke, 2004; Baum et al., 2001; Utsch & Rauch, 2000).

Bandura (1977, 1997), however, argued that self-efficacy should be focused on a specific context and activity domain. The more task specific one can make the measurement of self-efficacy, the better the predictive role efficacy is likely to play in research on the task-specific outcomes of interest (Bandura, 1997). Gist (1987) suggested that researchers aggregate a number of related but domain specific measures rather than relying on an omnibus test. While a composite measure of self-efficacy would be arguably more convenient, a number of scholars have sacrificed convenience in favor of greater predictive power (e.g., Begley & Tan, 2001; Chen et al., 1998; De Noble et al., 1999; Forbes, 2005; Kolvereid & Isaksen, 2006).

The majority of existing ESE measurement scales have been developed in a similar manner. Salient literature is referenced to identify tasks associated with core entrepreneurial activities or skills, such as opportunity recognition, risk and uncertainty management, and innovation. This list of tasks is then reviewed by academic experts and/or entrepreneurs to ensure appropriateness. Factor analytic techniques are subsequently used to identity final measurement items. Chen et al. (1998), for example, developed an ESE scale by referencing 36 entrepreneurial roles and tasks which, in turn, were reduced to a 26-item measurement instrument. Factor analysis identified 22 items that loaded on five distinct dimensions: (1) marketing, (2) innovation, (3) management, (4) risk taking, and (5) financial control. Such techniques produced viable task-specific ESE measurement instruments that allowed researchers to distinguish entrepreneurs from nonentrepreneurs (Chen et al.), better understand entrepreneurial decision-making processes (Forbes, 2005), and effectively predict entrepreneurial intentions (De Noble et al., 1999).

Unidimensional Versus Multi-dimensional Measures of ESE

While most theorists argue that ESE is best conceptualized as a multi-dimensional construct, much of the empirical research has relied on limited-dimensional or even unidimensional measures of ESE (Arenius & Minniti, 2005; Baum & Locke, 2004; Baum et al., 2001; Kristiansen & Indarti, 2004). At the extreme, several scholars claim to have measured ESE by simply asking subjects to respond to one or two questions regarding their confidence in starting a new venture. As an illustration, in a recent study by Tominc and Rebernik (2007), respondents were asked to provide a yes or no response to the question, "Do you have the knowledge, skills, and experience required to start a new business?"

Even those studies attempting a broader approach to measuring ESE dilute the multi-dimensionality of the construct by relying on a "total ESE" score rather than focusing on the underlying dimensions (Chen et al., 1998; De Noble et al., 1999; Forbes, 2005; Zhao et al., 2005). In their study of whether managers and entrepreneurs exhibited differing levels of ESE, Chen et al. identified five underlying factors or dimensions of the ESE construct but relied on a total ESE score (i.e., an average of 22 items). Although this technique allowed them to effectively distinguish entrepreneurs and managers, their results offered little insight on the importance of the construct's specific underlying dimensions (e.g., marketing, innovation, etc.). In other words, a total or composite measure of ESE fails to provide insight into what specific areas of self-efficacy are most influential. For example, it is impossible to determine whether a high level of self-efficacy in risk-taking or marketing is more influential in creating entrepreneurial intentions than a high level of self-efficacy in finance.

Print Get the Mag Weekly Updates