Journal.

The Best Writers3091https://doi.org/10.1177/0149206318779127Journal of ManagementVol. 45 No. 8, November 2019 3091 3113DOI: 10.1177/0149206318779127© The Author(s) 2018Article reuse guidelines:sagepub.com/journals-permissionsLearning off the Job: Examining Part-time Entrepreneurs as Innovative EmployeesDavid R. MarshallUniversity of DaytonWalter D. DavisClay DibrellAnthony P. AmmeterUniversity of MississippiIn this paper, we explain and examine how engaging in part-time entrepreneurship (creating and managing side businesses while remaining employed for wages in existing organizations) uniquely positions individuals to exhibit innovative behavior in employee roles. To study this phenomenon, we integrate the literatures on entrepreneurial learning, knowledge and learning transfer, and employee innovation. We hypothesize that part-time entrepreneurship provides an opportunity for individuals to acquire knowledge and skills conducive to enacting innovative behaviors as employees. Multilevel regression analysis of a sample of 1,221 employee responses across 137 organizational units provides evidence to support our positive transferal hypothesis. Further, we find that individual differences in goal orientations and work-unit climates for innovation strengthen these relationships.Keywords: employee innovation; part-time entrepreneurship; learningTo thrive in competitive markets, firms innovate by exploring new ideas and exploiting existing knowledge and resources (Levinthal & March, 1993; March, 1991). Recent perspec-tives assert that firm innovation is dependent, at least in part, on innovative behaviors exhib-ited by employees (Gibson & Birkinshaw, 2004; Mom, Van den Bosch, & Volberda, 2007; Raisch & Birkinshaw, 2008). However, the contexts and mechanisms through which employ-ees learn knowledge, skills, and abilities leading to innovative behavior at work are less clear (Gibson & Birkinshaw; O’Reilly & Tushman, 2013). Beyond developing new skills withinCorresponding author: David R. Marshall, University of Dayton, 300 College Park Dr., Dayton, OH 45469, USA.E-mail: [email protected]779127 JOMXXX10.1177/0149206318779127Journal of ManagementMarshall et al. / Learning off the Jobresearch-article20183092 Journal of Management / November 2019one’s employee role, studies suggest abilities can also be developed outside primary work roles (e.g., Eriksson & Ortega, 2006). In this study, we advance the intriguing possibility that employees engaged in entrepreneurial venturing activities outside of (in addition to) wage-employment roles are uniquely positioned to develop, refine, and transfer innovative capa-bilities from entrepreneurial to employee roles.We introduce part-time entrepreneurship as a context and entrepreneurial learning as a mechanism through which employee innovation is enhanced. Entrepreneurship, as a process of creating ventures to pursue value-added opportunities (Bygrave & Hofer, 1991), offers participants a unique learning context. Indeed, entrepreneurial learning theory describes the development of knowledge and unique skills through entrepreneurial experiences (Wang & Chugh, 2014). In particular, individuals learn to exploit existing knowledge to incrementally improve products and services and explore new knowledge to generate new ideas (Politis, 2005). Thus, part-time entrepreneurs are primed for developing innovative capabilities.Part-time (hybrid) entrepreneurship refers to simultaneous participation in wage-employment and entrepreneurship wherein wage-employees engage in entrepreneurial activities outside of primary work roles (Folta, Delmar, & Wennberg, 2010). While traditional management practice frowns on employees holding multiple jobs (Jamal & Crawford, 1981), creating and managing a new business while holding a primary employment position may be a particu-larly special case of multiple job holding given that an individual takes on a new entrepre-neurial role rather than simply an additional employee role (Stevenson & Sahlman, 1987). In this case, part-time entrepreneurs are founders and owners in unique learning contexts (Dyer, 1994) that may facilitate the development and transfer of innovative capabilities.Theories of interrole learning transfer explain that positive interrole exchanges occur because individuals view knowledge and skills generated in one role as particularly valuable to performance in another (Baldwin & Ford, 1988; Greenhaus & Powell, 2006; Noe & Schmitt, 1986). Therefore, we hypothesize that the extent to which an individual is engaged in part-time entrepreneurship is associated with greater exhibition of innovative behavior in primary employee roles (see Figure 1). However, individual motivation to transfer learning is influenced by learning and achievement preferences such as one’s goal orientation (Chadwick & Raver, 2015; Chiaburu & Marinova, 2005) and contextual factors such as work-group climate (Mathieu, Tannenbaum, & Salas, 1992; Wallace, Butts, Johnson, Stevens, & Smith, 2016). Accordingly, we examine moderating effects of individual goal orientations and work-unit climate for innovation on relationships between part-time entrepreneurship and innovative behavior in employee roles.By introducing part-time entrepreneurship as a context and entrepreneurial learning as a mechanism through which employee innovation can be developed and transferred to employee roles, we contribute to a growing body of literature focused on understanding the ways in which employees can develop and improve innovative skill sets (Liu, Gong, Zhou, & Huang, 2017; Sung & Choi, 2014). Our integration of employee innovation research, entrepreneurial learning theory, and learning transfer research extends our understanding of drivers of employee innovation (Anderson, Potočnik, & Zhou, 2014; Wallace et al., 2016). While there may be costs associated with simultaneous employee-entrepreneurial role par-ticipation (Hobfoll, 1989), our model suggests that managers who recognize the value of part-time entrepreneurship might seek ways to facilitate the transfer of learned knowledge and skills. Therefore, we shed light on the part-time entrepreneur as an important member of organizational life who is distinctly situated to contribute to organizational innovation.Marshall et al. / Learning off the Job 3093Theoretical Background and HypothesesMoving beyond research treating employee and entrepreneurial roles as mutually exclu-sive (e.g., career-choice studies; Douglas & Shepherd, 2002), both scholars and practitioners acknowledge the existence of wage-employment and entrepreneurship career overlap. For example, Entrepreneur magazine regularly publishes articles highlighting the benefits of par-ticipating in part-time entrepreneurship (Goodman, 2005; Zwilling, 2014). Likewise, research in labor economics reports that as many as 44% of entrepreneurs mix their time between organizational and self-employment contexts (Folta et al., 2010; Parker, 1997). Accordingly, there is an emerging call for research that focuses on “everyday” forms of entrepreneurship (Welter, Baker, Audretsch, & Gartner, 2017) and the part-time entrepre-neurs who perform these roles.Much of the scant research on part-time entrepreneurship assumes that simultaneous entrepreneurial and employee role engagement is a transitional stage toward eventual full-time entrepreneurial entry. Yet other studies indicate that full-time entrepreneurial entry may take place only after achieving high levels of self-employment to wage-employment income (Folta et al., 2010) or when an individual is willing and able to commit sufficient time to new venturing (Burmeister-Lamp, Lévesque, & Schade, 2012). Recently, scholars proposed that in many cases, part-time entrepreneurs never fully transition, opting instead to remain in a state of career hybridity (Thorgren, Sirén, Nordström, & Wincent, 2016). Regardless of whether part-time entrepreneurship is transitional or permanent, it offers participants the “potential to learn” that might not otherwise exist (Folta et al., 2010: 265). Part-time entre-preneurship as a “learning-by-doing” context (Petrova, 2012: 489) can result in increased entrepreneurial skills and abilities (Petrova, 2010; Raffiee & Feng, 2014). Thus, individuals engaged in entrepreneurship, even in part-time capacities, participate in important entrepre-neurial learning.Figure 1Innovative Behavioral Transfer From Entrepreneurial to Employee Roles3094 Journal of Management / November 2019Entrepreneurial LearningEntrepreneurial learning theory emerged at the intersection of organizational learning and entrepreneurship research (Harrison & Leitch, 2005) and from an individual perspective, explains the process by which entrepreneurial experiences transform into entrepreneurial knowledge (Wang & Chugh, 2014). Entrepreneurial experiences encompass the participation in or observation of the creation or maintenance of a new venture (Cope, 2005; Cope & Watts, 2000). To explain how individuals learn through entrepreneurial experiences, scholars draw from research and theory related to exploratory and exploitative approaches to learning (March, 1991). Exploitative learning involves the use of existing knowledge structures to guide the acquisition of additional information that can be used to improve entrepreneurial practices. This type of learning often involves limiting variations in practice (e.g., correcting mistakes) to improve the efficiency of operations. Exploratory learning involves experimen-tation with new variations in practices that may result in desirable results. Oftentimes, the efficacy of such exploratory learning is unknown until after the results of these variations are observed. Individuals engaged in entrepreneurship develop new knowledge as they acquire information that complements existing knowledge structures (exploitative) and experiment with new practices to develop fundamentally different knowledge structures (exploratory) (Bingham & Davis, 2012).Given that individuals participating in entrepreneurship are engaged in exploratory and exploitative learning, scholars suggest the key outcome of entrepreneurial learning is entre-preneurial knowledge and the ability to recognize and act on opportunities for innovation (Corbett, 2005; Politis, 2005). Alvarez and Busenitz (2001: 762) conceptualize entrepreneur-ial knowledge as “the ability to take conceptual, abstract information of where and how to obtain undervalued resources, explicit and tacit, and how to deploy and exploit these resources.” Thus, entrepreneurial knowledge increases an individual’s capacity to exhibit innovative behaviors.Transfer of Innovative Learning From Entrepreneurial to Employee RolesTransfer of learning occurs when knowledge obtained in one setting results in exhibition of related behaviors in another setting (Baldwin & Ford, 1988; Noe, 1986). Successful trans-fer of learning between roles typically transpires through an instrumental path allocation process in which valuable resources generated in one role become instrumental to enhancing performance in another (Greenhaus & Powell, 2006). For example, learning transfer from training sessions to real working situations is most effective when employees learn knowl-edge and skills that are relevant to behaviors needed to accomplish role tasks (Axtell, Maitlis, & Yearta, 1997). In our model, part-time entrepreneurs gain learning experiences through entrepreneurial participation that provides increased capabilities to innovate in employee roles.As previously reviewed, part-time entrepreneurship offers a useful and low-risk context through which an individual can increase entrepreneurial knowledge and innovative capa-bilities (Raffiee & Feng, 2014). Engaging in entrepreneurship requires the development of experience and knowledge necessary for the exploration and exploitation of opportunities (Shane & Venkataraman, 2000). Therefore, learning to be more entrepreneurially minded may have important developmental benefits in terms of increasing innovative behavioralMarshall et al. / Learning off the Job 3095capabilities. Indeed, learning experiences in entrepreneurship are primarily transformed into useful knowledge via innovative decision making modes (Politis, 2005). Consequently, based on our review of entrepreneurial learning theory, we contend that entrepreneurial learning is a useful mechanism through which part-time entrepreneurs generate and refine abilities to exhibit innovative behaviors.Innovation is defined as the “intentional introduction and application within a role, group, or organization of ideas, processes, products or procedures, new to the relevant unit of adop-tion, designed to significantly benefit the individual, the group, organization or wider society” (West & Farr, 1990: 9). Innovative behavior is increasingly seen as important to the enactment of employee roles (Scott & Bruce, 1994). As individuals acquire entrepreneurial knowledge through their part-time entrepreneurship experiences, they are likely to exhibit innovative behavior in their roles as employees as they see opportunities to improve performance.Individual innovative behavior can be both exploratory and exploitative in nature. Exploratory behaviors include the introduction of new ideas, solutions, services, or products associated with one’s work role, while exploitative behaviors include reliance on existing knowledge to make incremental improvements to work procedures (Lee & Meyer-Doyle, 2017). While at the firm level it is difficult to engage in both exploratory and exploitative activities (March, 1991), research suggests individuals may be more adept in enacting both (Rogan & Mors, 2014; Taylor & Greve, 2006; Zacher, Robinson, & Rosing, 2016), which are valuable to performance of employee role tasks (Lee & Meyer-Doyle). We argue that indi-viduals simultaneously engaged in entrepreneurial activities while maintaining employee roles may be more adept than other employees at demonstrating exploratory innovative behaviors in the workplace by looking for new ways to improve organizational processes, expand or go beyond management demands, and looking for new technologies to improve how work is accomplished. Likewise, these individuals may be more capable of undertaking exploitative innovative behaviors by making existing processes more efficient and finding synergies among resources to accomplish work more effectively.Hypothesis 1: The extent to which an individual engages in part-time entrepreneurship is positively associated with his or her innovative behavior in employee roles.Individual Differences and Contextual ModeratorsEmployee behavior is often viewed as a function of ability, motivation, and opportunity (Gardner, Wright, & Moynihan, 2011). Therefore, the extent to which capabilities developed in entrepreneurial roles are transferred and exhibited in primary employee roles depends on both individual and contextual factors affecting motivation and perceived opportunities to act. Learning through experience is not an automatic process, and not all individuals are motivated to learn (Colquitt & Simmering, 1998). Individuals must have the desire to trans-form and integrate experiences into new and existing knowledge structures and identify opportunities to do so.We contend that the relationship between part-time entrepreneurship and innovative behavior is moderated by goal orientation (motivation) and work-unit climate for innovation (opportunity). Goal orientation is often studied as one of the most fundamental individual traits guiding motivation in both learning and achievement settings (Button, Mathieu, &3096 Journal of Management / November 2019Zajac, 1996; Kanfer, Frese, & Johnson, 2017). Scholars are increasingly focusing on achieve-ment goal theory (Dweck & Leggett, 1988) for an understanding of the individual’s role in organizational learning processes, and individual differences in goal orientation are often seen as a primary determinant of learning and learning transfer (Chadwick & Raver, 2015). Other researchers argue that organizational climates arising from supportive policies, rules, and leaders are critical in providing the opportunity for engaging in innovative behavior (Scott & Bruce, 1994; West & Richter, 2008).Goal orientation. Goal orientation refers to relatively stable dispositions toward developing personal capabilities and demonstrating task competence. Goal orientation arises out of one’s view of whether ability is fixed (perhaps innate) or can be improved through learning, task practice, and experimentation with new approaches to task accom-plishment (Button et al., 1996). Three dimensions of goal orientation include: learning orientation, performance-proving orientation, and performance-avoiding orientation. Individuals higher in learning orientation view ability and task mastery as something that can be improved through learning and experimentation and therefore tend to seek out challenging experiences that provide opportunities to learn (Ames & Archer, 1988) and acquire new knowledge useful for developing new capabilities (VandeWalle, Brown, Cron, & Slocum, 1999). Furthermore, increased learning orientation can lead to the elab-oration of task strategies, which is the purposeful cognitive experimentation with task schemas, leading to the development of procedural knowledge (Steele-Johnson, Beaure-gard, Hoover, & Schmidt, 2000).Given their desires for learning and the subsequent development of knowledge, we con-tend that part-time entrepreneurs with higher learning goal orientations will experience greater learning through entrepreneurial experiences than lower learning-oriented individu-als. Hence, part-timers with higher learning orientations have greater capacities for exhibit-ing innovative behaviors in employee roles. Furthermore, we expect higher learning orientations to provide increased motivation for part-timers to transfer their knowledge and skills to the exhibition of innovative behaviors at work as higher learners tend to be less deterred by the risky and/or uncertain performance situations associated with innovative activities (Hirst, Van Knippenberg, & Zhou, 2009). Indeed, people with higher learning ori-entations often engage in more risky innovative learning (Chadwick & Raver, 2015) and are more inclined to transfer knowledge into actual behavioral settings despite high risks of failure (Bell & Kozlowski, 2002).Hypothesis 2: Learning goal orientation moderates the association between engagement in part-time entrepreneurship and innovative behavior in employee roles such that the association will be more positive for persons with higher learning goal orientations.Individuals with higher performance-proving orientations tend to seek out opportunities to improve performance by rehearsing existing task strategies rather than experimenting with new approaches (Steele-Johnson et al., 2000) and are inclined to engage in learning experi-ences that enable them to refine existing capabilities (Chadwick & Raver, 2015). Therefore, part-timers with higher proving orientations likely develop greater innovative capabilities than those with lower proving orientations and are uniquely equipped to draw on their newly developed entrepreneurial knowledge and skills in enacting innovative behaviors in employee roles. Further, persons with higher performance-proving orientations are more motivated toMarshall et al. / Learning off the Job 3097demonstrate their competence and gain favorable judgements from others than lower prov-ing-oriented people (VandeWalle, Cron, & Slocum, 2001). Thus, we expect that part-timers with higher proving orientations experience increased motivation to transfer entrepreneurial learning to the enhancement of their performance in employee roles through greater innova-tive behaviors, thereby garnering favor with organizational leaders and coworkers.Hypothesis 3: Performance-proving goal orientation moderates the association between engage-ment in part-time entrepreneurship and innovative behavior in employee roles such that the association will be more positive for persons with higher performance-proving goal orientations.An individual with higher performance-avoiding goal orientation views ability as rela-tively fixed, such that experimentation with new approaches to task accomplishment is unnecessary and perhaps even risky (Hirst et al., 2009). As such, higher avoidance goal-oriented individuals tend to evade opportunities to engage in learning (Chadwick & Raver, 2015). Accordingly, we posit that higher avoiding goal orientations make it harder for part-time entrepreneurs to develop innovative behavioral skill sets such as those acquired by high learners and provers. We further contend that part-time entrepreneurs with higher avoiding goal orientations than others lack the potential to make unique innovative contributions in employee roles. Additionally, performance-avoiding goal orientation is characterized by a strong desire to avoid negative performance perceptions (VandeWalle, 1997). Because the outcomes of innovative behaviors are often uncertain and can result in failure, part-time entrepreneurs who are also higher in avoidance orientation may be less motivated to transfer entrepreneurial learning to the exhibition of innovative behavior in employment roles. These individuals may fail to recognize the important information embedded in learning experi-ences and are less likely to utilize this information in new situations.Hypothesis 4: Performance-avoiding goal orientation moderates the association between engage-ment in part-time entrepreneurship and innovative behavior in employee roles such that the association will be more positive for persons with lower performance-avoiding goal orientations.Work-unit climate for innovation. Climate refers to the shared perceptions of the types of behaviors that are encouraged in a work-unit through meanings attributed to policies, practices, and norms (Schneider, 1990). Climate is typically studied as an attribute of a group or an organization rather than at the individual level (Schneider, Ehrhart, & Macey, 2013). When group policies and norms reward and encourage innovative behavior, group members are more likely to share perceptions of opportunities to engage in innovative behavior (Scott & Bruce, 1994; West & Richter, 2008). Specifically, a climate for innovation consists of a shared vision of innovative outcomes, perceived safety to participate in the innovative process, high collective standards of task performance, and shared perceptions of support for innovation in the form of needed resources (Anderson & West, 1998; Somech & Drach-Zahavy, 2013).When employees in a work-unit share similar perceptions that innovative behavior is highly encouraged, rewarded, and supported, it is more likely that part-time entrepreneurs will be motivated to transfer their entrepreneurial knowledge and skills toward innovation in3098 Journal of Management / November 2019employee roles. Perhaps most importantly, when a work-unit provides the psychological safety to promote and champion innovative ideas, part-timers will feel free of the risk and uncertainty that often accompany innovation (Choi, 2007). Additionally, innovative group climates can help part-time entrepreneurs recognize how their entrepreneurial skills might be applicable in their work tasks by conveying clear messages about what innovative behaviors look like within the unit (Sung & Choi, 2014). However, a group climate lacking in support of innovation may suggest to part-timers that learned entrepreneurial knowledge does not pertain to employee roles or is not valued, and therefore, part-time entrepreneurs may be less inclined to transfer entrepreneurial learning to innovative behavior at work.Hypothesis 5: Work-unit climate for innovation moderates the association between engagement in part-time entrepreneurship and innovative behavior in employee roles such that the association will be more positive under conditions of higher innovative climate.MethodologyDesign and SampleWe tested our hypotheses on a sample of employees of a large logistics and security com-pany working in eight locations throughout the United States. This firm is useful for testing hypotheses regarding part-time entrepreneurship and employee innovative behavior for sev-eral reasons. First, this company emphasizes team-based approaches to organizing and dis-tributing workload, thereby facilitating the testing of group-climate effects. Specifically, the sample consists of 137 unique teams distinguished by office codes assigned by the personnel department. Each office, or work-unit, is composed of at least two employees of varying functional areas (43 occupational codes; e.g., finance, HR, maintenance, R&D, supply chain, business development, etc.). Second, to avoid biased results, the sample firm provides employee participants of varying degrees of innovative capabilities (as opposed to employ-ees of high-tech innovation firms of Silicon Valley) to accurately capture the effects of part-time entrepreneurial engagement on innovative behavior within primary employing roles. As part of its vision, the firm aims to provide innovative and cost-effective capabilities to cus-tomers. Thus, while the company emphasizes the importance of innovation, its reputation is not one of high-tech or necessarily high innovation and provides a valid context in which to test our hypotheses.The survey instrument was distributed via the organization’s internal survey system to roughly 5,500 workers. Useable responses totaled 1,221, a response rate of 22.2%. The sam-ple has the following characteristics: 71% male, average age of 37 years, average tenure of 11 years, 53% holding a college or advanced degree, and 69% considered white-collar or non maintenance personnel.Construct Measurement and ValidityEngagement in part-time entrepreneurship. Entrepreneurship is a broad concept with many varied conceptualizations and definitions, making its measurement difficult (Gart-ner, 1990). Recently, scholars have warned against narrowly defining entrepreneurial activity in scholarly research to avoid restricting empirical evaluation (Welter et al., 2017).Marshall et al. / Learning off the Job 3099Consistent with this logic but also to distinguish between part-time entrepreneurship, other forms of secondary wage-employment work, and other activities that fail to generate eco-nomic value (e.g., clubs, societies, and/or other hobby-based groups), we adopt a view of entrepreneurship shared by many scholars that value-creating opportunities are realized through the creation, implementation, and management of new business entities (Alvarez & Busenitz, 2001). Thus, part-time entrepreneurship reflects engagement in the creation and/or management of an entrepreneurial venture while maintaining a role as an employee in an existing firm.To avoid restricting the range of possible entrepreneurial ventures undertaken in our sample, we allowed respondents to determine the extent to which they were actually involved in entre-preneurial venture creation and management. That is to say, we did not provide a formal defini-tion of entrepreneurship to respondents. Rather, through three items, individuals identified the extent to which they were currently engaged in entrepreneurial activities (related to starting and/or managing their own entrepreneurial ventures) outside of primary employment roles. A 7-point Likert-type scale was used for scoring each item. These activities included “founding an entrepreneurial business,” “running a part-time business,” and “creating a for-profit or non-profit entrepreneurial venture.” We scattered distractor items related to one’s involvement in other various activities outside of work but unrelated to entrepreneurial pursuits throughout the scale to aide in the distinction of part-time entrepreneurship and other types of activities such as “involvement in civic organizations or clubs.” To better distinguish entrepreneurial activities from other profit-generating activities not typically considered as entrepreneurial, such as con-sulting or driving for Uber, we included an “involvement in self-employment” activities item (full scales reported in the Appendix). Distractor/distinguisher items were not used in the final analytical model. Rather, these items helped to discriminate between entrepreneurship and non-entrepreneurship activities. Alpha reliability for the measure is 0.97.Innovative behavior (in employee roles). We relied primarily on Jansen, Van Den Bosch, and Volberda’s (2006) questionnaire to generate a list of potential items to adapt to the individ-ual level given that few studies have dealt explicitly with exploratory and exploitative behavior at the individual level. The Jansen et al. survey was designed to assess unit-level innovative behaviors based on managerial responses. Where appropriate for our research sample and set-ting, we maintained similar verbiage of the items while replacing each unit-level referent such as we and our with I and replaced statements of unit-level services with more appropriate indi-vidual-level tasks or other established items of innovative behavior (Scott & Bruce, 1994). We reviewed and revised items to ensure the measures accounted for multiple stages of innovative behavior (i.e., idea generation, coalition building, and implementation; Scott & Bruce, 1994). To enhance construct validity, we then pretested items through interviews with six senior-level managers and four members of the personnel department (responsible for maintaining posi-tion descriptions of each employee) of our sampled firm. Each interviewee was asked to indi-cate item phrasing that was relevant to the types of work processes performed by the typical employee of the firm and suggest changes to improve overall understandability of each item for their employees.We followed the approach of other scholars in relying on self-reported assessments of innovative behavior (Dorenbosch, Van Engen, & Verhagen, 2005; Jansen et al., 2006) because peer judgements may not adequately capture the full range of innovative behaviors3100 Journal of Management / November 2019of an employee given that much of the innovation process takes place in one’s mind (Janssen, 2000). Likewise, peers and/or supervisors may not recognize incremental innovations employees make in individual work tasks and processes as these changes may be “subtle” in nature (Janssen 2000: 292). Meta-analyses empirically demonstrate that “others-rated” posi-tive workplace behaviors typically offer negligible incremental explanation over self-reported measures (Carpenter, Berry, & Houston, 2014). The final scale for innovative behavior con-tained 12 items. We anchored each item to the employee role by asking respondents to rate on a 7-point Likert-type scale the degree to which they disagreed/agreed with statements such as “In my primary, full-time job: I often expand the services I offer customers/constitu-ents through new offerings” (full scale included in the Appendix).To assess the validity and reliability of adapted scale items, 127 contracted employees working at a satellite office of the sample firm completed a survey that included the 12 inno-vative behavior items, a 9-item self-reported bricolage behavior measure (Hmieleski & Corbett, 2006), and a shortened, 7-item self-reported organizational citizenship behavior (OCB) measure (adapted from Moorman & Blakely, 1995). Exploratory factor analyses of all items demonstrate convergence among innovative behavior measures while discriminating them from similar behavioral (bricolage) and other discretionary performance (OCB) con-structs. A rotated component matrix revealed three factors with innovative behavior items ranging from 0.73 to 0.89. The highest cross-loading with innovative behavior is 0.26, a bricolage item. Alpha reliability for innovative behavior is 0.95.Goal orientation. The 13-item measure established by Brett and VandeWalle (1999) assessed the three dimensions of individual goal orientation. Alpha reliabilities for each ori-entation variable are 0.98, 0.94, and 0.90 for learning (5 items), proving (4 items), and avoid-ing (4 items), respectively.Work-unit innovative climate. A short-form, adapted instrument assessed climate for unit-level innovation based on previously established measures (Anderson & West, 1998; Scott & Bruce, 1994). Unidimensional, short-form scales consistently provide robust indica-tions of innovative climates in other studies (e.g., Somech & Drach-Zahavy, 2013; Sung & Choi, 2014). Five items asked respondents the degree to which they agreed/disagreed with statements referring to unit-level innovativeness such as “Employees in my work-unit regu-larly encourage others to explore new ideas and try new ways of doing things” (full scale provided in the Appendix).Several assessments of agreement justify unit-level aggregation. The mean rwg(j) (James, Demaree, & Wolf, 1984) for work-unit innovative climate is an acceptable 0.86 (Bliese, 2000). Intraclass correlations (ICC) for climate as suggested by Bliese (2000) are 0.19 for ICC(1) and 0.69 for ICC(2), both consistent with acceptable ranges for multilevel research (Bliese, 2000). Additionally, an ANOVA test indicated significant between-group variance in climate (F = 3.03).Controls. Our analyses account for other variables that might create disturbances in the hypothesized relationships. Chief among these variables are proactive and creative personali-ties that potentially lead to both entrepreneurial engagement and innovative behaviors. In many studies, individuals possessing a proactive personality demonstrate greater innovative behavior than others (Chen, Farh, Campbell-Bush, Wu, & Wu, 2013; Kim, Hon, & Crant, 2009; Seibert,Marshall et al. / Learning off the Job 3101Crant, & Kraimer, 1999). Likewise, proactive personality is a strong predictor of entrepreneurial intent, behavior, and success (Becherer & Maurer, 1999; Brandstätter, 2011; Crant, 1996; Rauch & Frese, 2007). We control for proactive personality with an established five-item scale (Kickul & Gundry, 2002). In our study, this measure revealed good reliability with an alpha of 0.91. Because of the close ties between creativity, innovative behavior, and entrepreneurial behavior (Bharadwaj & Menon, 2000; Shane & Nicolaou, 2015), we control for creative personality using Gough’s (1979) 30-adjective checklist scale. Alpha reliability for this measure is 0.72.To further isolate learning effects, we control for overall, individual job performance as rated by each respondent’s immediate supervisor. The data for this measure originated with the organization and are part of the most recent employee performance appraisal. We used company-assigned employee identification numbers to match respondents with their respec-tive supervisor-rated performance. Each respondent’s immediate supervisor rated employees (from 1 = very poor to 9 = outstanding) in nine areas of performance such as work effort, adaptability, and work relationships. Alpha reliability for this measure is 0.93. We also con-trol for the number of years an individual has been engaged in entrepreneurship (P/T ENT tenure) as this might influence learning effects such that longer tenured entrepreneurs accu-mulate more innovative skills. Given the potential for outside entrepreneurial endeavors interfering with successfully carrying out employee role tasks, we control for an indicator of commitment to one’s venture pursuits by asking respondents to indicate intentions for entre-preneurship as a full-time occupation. Intentions for full-time entrepreneurship (F/T ENT intentions) are assessed by three items (scored on a 1-7 Likert-type scale from strongly dis-agree to strongly agree; “intend to run/own a business as your primary job; intend to run a business full-time; will eventually be a full-time entrepreneur”). Alpha reliability is 0.95.We conducted confirmatory factor analysis (CFA) to assess the validity of the multilevel measurement model. Comparing several theoretically viable models showed the hypothesized model with six focal multi-item constructs as the best fit for our data (χ2 = 5,249.34, df = 481, p .05) levels.Figure 2Part-Time Entrepreneurship × Learning Goal Orientation InteractionFigure 3Part-Time Entrepreneurship × Avoiding Goal Orientation InteractionMarshall et al. / Learning off the Job 3105Post Hoc AnalysesThe single time period nature of the data sample may raise questions regarding the inter-nal validity of our inferences. Therefore, we employ several analytical techniques to ease concerns of common methods bias and support the logic of learning effects. We analyzed a measurement model that included the six hypothesized latent variables and an additional orthogonal latent common factor that failed to converge, suggesting poor fit. We also tested a single-factor model resulting in significantly poorer fit than the hypothesized model (χ2 = 33,064.56, df = 496, p