Entrepreneurial Orientation Bidding Strategies
Hermann Frank/Alexander Kessler/Matthias Fink*
Entrepreneurial Orientation and Business Performance – A Replication Study**
A bstract We examine the effect of entrepreneurial orientation (EO) and hypothesize that EO has a positive impact on business performance. We create a contingency model and a configuration model and compare their results to those derived from a main-effects model. The study replicates the work of Wiklund and Shepherd (2005) and tests the validity of their results in a different national context. Our analysis indicates a positive connection between EO and business performance only in cases in which a dynamic environment is combined with high access to financial capital and when a stable environment is combined with low access to financial capital. Our analysis also indicates that EO may have a negative effect on performance in certain configurations. JEL-Classification: M13. Keywords: Business Performance; Configurational Approach; Entrepreneurial Orientation; SMEs.
1 I ntroduction Although the topic of entrepreneurial orientation (EO) has attracted increasing interest, the vast majority of publications in this field has come from U.S. authors. So far, there have been almost no empirical results that focus on Europe. The works of Wiklund (1998; 1999), Kemelgor (2002), Kreiser, Marino, and Weaver (2002a; 2002b; 2002c), Marino et al. (2002), Wiklund and Shepherd (2003), Harms and Ehrmann (2003), and Wiklund and * Hermann Frank, WU Vienna University of Economics and Business, Augasse 2-6, A-1090 Vienna, Austria, e-mail:
[email protected]. Alexander Kessler, FH Wien University of Applied Sciences of WKW, Waehringer Guertel 97, A-1180 Vienna, Austria and WU Vienna University of Economics and Business, Augasse 2-6, A-1090 Vienna, Austria, e-mail:
[email protected]. Matthias Fink, WU Vienna University of Economics and Business, Augasse 2-6, A-1090 Vienna, Austria, e-mail:
[email protected]. ** The authors would like to thank two anonymous referees for their valuable comments and suggestions. We would also like to express our gratitude to Johan Wiklund, who readily answered our queries in the course of this study. Any errors are, of course, the responsibility of the authors.
SBR 62 April 2010 175-198
175
H. Frank/A. Kessler/M. Fink
Shepherd (2005) and some published doctoral dissertations, such as those by Harms (2004) and Haid (2004), are exceptions, but not all of the works mentioned above are empirical studies. In any case, the sparse results available do not permit generalizations on the practical importance of EO and its contribution to business performance in a European context. Kemelgor’s study (2002) compares firms in the Netherlands and the U.S. and concludes that EO is characterized by cultural differences. He finds significant differences in the intensity of EO between firms in the Netherlands and the U.S., as well as a significant correlation between EO and business performance. Kreiser, Marino, and Weaver (2002a; 2002b; 2002c) and Marino et al. (2002) work with an international sample that contains data from different European countries, but they do not compare the contribution of EO to business performance in individual countries. Wiklund and Shepherd (2005) also identify this lack of comparisons and the difficulty of generalizing the relation between EO and business performance in a European context. One implication of their study is the question of whether their findings are specific to Sweden (or all the Scandinavian countries), or whether their results are more universal. Hence, these authors propose comparative research in other countries to test the universality of the findings in other business cultures. In this paper we follow the suggestion of Wiklund and Shepherd (2005), replicating their study with firms in Austria. The value of this suggestion is highlighted by a meta-analysis conducted by Rauch et al. (2004), who conclude that “national culture is a powerful moderator in EO-performance relationships”. The GLOBE study (House et al. (2004)) provides further support for the existence of differences between Swedish and Austrian business cultures. The benefit of our replication lies in the fact that it tests the generalizability of previous empirical findings. Thus, we review whether the findings generated in one context (e.g., a certain business culture) concur or conflict with the results obtained in another context. The replicative nature of our study results from borrowing the measurement instrument, hypotheses, and evaluation method from the original study. As does the original study, in this paper we pose the question of whether EO is always an appropriate strategic orientation, or whether its relation to business performance is more complex. Therefore, we test whether a configuration model (three-way interaction), which involves the simultaneous and joint consideration of strategy, organizational characteristics, and environment, can produce a greater understanding of the impact of EO on business performance than can a maineffects-only model (universal model) and a contingency model (two-way interactions). The paper is organized as follows: In Section 2 we define and delimit the term EO and discuss the benefits arising from replication studies. In Section 3 we describe our hypotheses and the original work of Wiklund and Shepherd (2005) and we discuss the similar
176
The studies conducted by Wiklund (1998; 1999) and Wiklund and Shepherd (2003; 2005) are based on the same Swedish sample. The studies carried out by Kreiser, Marino, and Weaver (2002a; 2002b; 2002c) and Marino et al. (2002) use a joint international sample which also includes European countries.
SBR 62 April 2010 175-198
Entrepreneurial Orientation
ities and differences between our and the original study. In Sections 4 and 5 we present the methods we apply, and the results of our study. Section 6 provides a discussion of our results against the backdrop of the original study. Section 7 concludes. 2 E ntrepreneurial O rientation 2.1 D elimination
and
D efinition
Referring to Guth and Ginsberg (1990), Zahra and Covin (1995) describe EO as “a potential means for revitalizing established companies. This is accomplished through risk taking, innovation, and proactive competitive behaviors”. Lumpkin and Dess (1996) rely on a definition in which the “essential act of entrepreneurship is new entry” (…) and “EO refers to the processes, practices, and decision-making activities that lead to new entry (…)”. The key dimensions that characterize EO include a propensity to act autonomously, a willingness to innovate and take risks, and a tendency to be aggressive toward competitors, and proactive with regard to marketplace opportunities. Due to the replicative character of our study, we do not take into account the two additional dimensions, competitive aggressiveness und autonomy, that Lumpkin and Dess (1996) cite as components of the definition. For our replication, we define EO uniformly as a firm’s strategic orientation, one which captures the specific entrepreneurial aspects of decision-making styles, methods, and practices. EO is a combination of three dimensions: innovativeness, proactiveness, and risk taking. These three dimensions are based on Miller (1983). Miller uses the following characterization: “an entrepreneurial firm is one that engages in product-market innovation, undertakes somewhat risky ventures, and is first to come up with ‘proactive’ innovations, beating competitors to the punch. A nonentrepreneurial firm is one that innovates very little, is highly risk averse, and imitates the moves of competitors instead of leading the way”. In addition to this opposing delimitation, it is necessary to consider the fact that EO represents a continuum ranging from more or less conservative to entrepreneurial firms, that is, it expresses a fundamental strategic posture. 2.2 M ain Findings
of
P revious R esearch
Replicating the work of Wiklund and Shepherd (2005) raises the question of the extent to which its results, as well as those of our replication study, might possess universal validity. Therefore, we review other available study results. For this purpose, we rely on the metaanalysis by Rauch et al. (2004), which includes nearly 40 studies. We also consider the study by Dess, Lumpkin, and Covin (1997), since their paper is based on contingency and configuration models and uses a similar methodical approach to the one applied by Wiklund and Shepherd (2005). Wiklund and Shepherd’s study (2005) compares and evaluates the explanatory power of all three approaches (main effects, contingency, and configuration) for the relation between EO and business performance.
Wiklund and Shepherd (2005) with reference to Lumpkin and Dess (1996). See Barringer and Bluedorn (1999).
SBR 62 April 2010 175-198
177
H. Frank/A. Kessler/M. Fink
Although most of these studies were conducted using very specific samples and approaches, they still paint a consistent picture: EO generally has a positive effect on business performance. However, several studies show that this main effect is context-sensitive. For example, Zahra and Covin (1995) identify the degree of hostility in the business environment as an intervening factor. In assessing international entrepreneurship, Zahra and Garvis (2000) argue that EO’s contribution to business success depends on the degree to which executives perceive their firm’s international business environment. These few examples demonstrate the need to consider the background in which the relevant firms are embedded. Most studies rely on a main-effects or a contingency approach; only few use a configuration approach (Dess, Lumpkin, and Covin (1997); Lee, Lee, and Pennings (2001); Wiklund and Shepherd (2005)). The meta-analysis conducted by Rauch et al. (2004) reveals that the dimensions of EO vary independently of performance. As part of this meta-analysis, the study carried out by Lumpkin and Dess (2001) analyzes two independent factors comprising EO (proactiveness and competitive aggressiveness) and argues that their relative impacts on business performance vary over the different stages of the industry’s life cycle. According to Yoo’s study (2001), which is also part of this meta-analysis, the three factors comprising EO (innovativeness, proactiveness, and risk-taking) are closely interrelated. Similarly, Rauch et al. (2004) identify differences in the relation between EO and objective compared to perceptual performance measures, thus indicating that EO is only related to selected indicators of business success. In their meta-analysis, Rauch et al. (2004) also find that the relation between EO and performance varies substantially according to national culture. However, only a few studies are based on or include a European sample; several are based on a Swedish sample, while others use an international database that covers several European countries. Only one of the studies included in the meta-analysis contains data from Germany. Apart from that study, we found only one other study (Harms and Ehrmann (2003)) that is based on a sample from a German-speaking country. The studies give the impression that EO exhibits a context-sensitive positive relation to business performance. Thus, EO represents a promising strategic posture that also merits greater attention in the European research context, especially in light of the few results available to date. 2.3 R eplication R esearch
and S cientific
P rogress
Replication studies play an important role in the advancement of empirical research. Replications can be carried out with and without extensions (Hubbard and Amstrong (1994)).
178
For example, Kemelgor (2002) reported that while EO was positively related to all three performance measures used in the U.S. sample (i.e., innovations, number of patents, and return on sales), the results from the Dutch sample indicate that EO was only positively related to the number of patents and return on sales. Easley, Madden, and Dunn (2000) present four types of replication studies that show the alterations between the original study and its replication.
SBR 62 April 2010 175-198
Entrepreneurial Orientation
Replication studies without extensions are important for establishing the reliability and validity of empirical findings (Amir and Sharon (1990); Lindsay and Ehrenberg (1993)). However, those that introduce extensions (i.e., studies that use the same conceptual relations as the original study, but alter some aspects of the initial design, such as geographical areas, industries or populations) aim to evaluate the generalizabillity of earlier research findings (Hubbard and Amstrong (1994); Hubbard, Vetter, and Little (1998)). Although exact replications are important for the purpose of assessing original findings, replication studies with extensions make a far more substantial contribution (Rosenthal (1990)). Given the importance of replication studies, their small share in published empirical work is surprising (Neuliep and Crandall (1990); Hubbard, Vetter, and Little (1998); Baumgarth and Evanschitzky (2005)). This fact can be due to any of several reasons: First, researchers assume that the information needed to perform a replication study is difficult to obtain (Hubbard and Amstrong (1994)), as intensive communication with the author(s) of the original study is required. Second, replication studies are less likely to be published (Eden (2002)), due to a negative bias among editors of high-tier journals (Hubbard, Vetter, and Little (1998)). Third, in line with Kane (1984), Hendrick (1991) reports that researchers often see replication studies as less creative. However, a strong case can be made for replication studies. Unreplicated empirical results might only lead to “one-shot” theories of unknown scope and limitations (Hubbard, Vetter, and Little (1998)), and this problem might result in a fragmented mosaic of results. From a critical, logical point of view, we must test the robustness of empirical results if we are to achieve scientific progress (Tsang and Kwan (1999)). Confidence in the insights derived from studies in the social sciences can only grow after a tradition of replication has been established. Replicating an original study advances our knowledge: (1) It may contradict the original outcomes and raise doubts, thereby providing a sound reason to continue testing the original hypothesis. Replication studies that fail to confirm the outcomes of original studies contribute to the scientific process by earmarking previous empirical findings that might be based on Type 1 errors (erroneous rejection of the null hypothesis) or false correlations (Amir and Sharon (1990); Lindsay (1994); Hubbard, Vetter, and Little (1998)). Therefore, replication research can serve as a form of quality control. Second, replication might also generate the same findings, which serves to enhance confidence in the generalizability of the original findings (Eden (2002)). Therefore, replication research is an indispensable element of scientific progress (Furchtgott (1984); Lindsay and Ehrenberg (1993)). This study is positioned as a replication with extensions. For specific information on the characteristics of the replication performed here, please refer to Section 3.2.
See Diekmann (2004).
SBR 62 April 2010 175-198
179
H. Frank/A. Kessler/M. Fink
3 C onceptual B ackground 3.1 The H ypotheses
of the O riginal Study
Wiklund and Shepherd (2005) base their study on the assumption that there is a positive relation between EO and business performance: the higher the EO, the higher the firm’s performance will be. These authors describe EO by using three dimensions: innovativeness, proactiveness, and risk-taking. The innovativeness dimension represents the aspect of a firm’s strategic posture that refers to the firm’s willingness and ability to question – and abandon – existing or given circumstances, and to create room for creativity, new ideas, and experiments. The objective is to think innovatively, which can manifest itself in the launch of new products, in the exploration of new markets, and in process innovations. The proactiveness dimension reflects the aspect of a firm’s strategic posture that refers to the firm’s willingness and ability to anticipate new developments as early as possible and to act as a “first mover” vis-à-vis competitors, rather than to wait for new developments and trends and then react to them. The risk-taking dimension represents the aspect of a firm’s strategic posture that refers to the firm’s willingness and ability to devote increased resources to projects whose outcome is difficult to predict. These three dimensions shape the EO construct; we assume that the items in these three dimensions covary. As do Wiklund and Shepherd (2005), we treat EO as a single joint construct and therefore regard it as a reflective indicator. The four hypotheses below and their derivations are borrowed from Wiklund and Shepherd (2005, 74-79). Due to increasingly short product and market life cycles, if they are to succeed, then businesses generally need to orient their strategic posture toward increased EO. The following hypothesis postulates a relation between EO and performance regardless of environmental influences: H1: EO has a universal positive effect on business performance.
This hypothesis disregards any internal and external conditions that may be specific to the firm, which can moderate the impact of EO on business performance. In stable business environments, proactive and risk-raking innovative behavior could consume more resources than the additional benefits it actually creates. But in a dynamic competitive environment, EO is an important prerequisite that the firm itself must exert if it is to
180
Cf. the discussion in Wiklund and Shepherd (2005) and the diverging opinion expressed in Lumpkin and Dess (1996).
SBR 62 April 2010 175-198
Entrepreneurial Orientation
have an active influence on market developments. Accordingly, the following hypothesis can be put forward: H2: The relation between EO and business performance is moderated by environmental dyna-
mism. Business performance increases with EO, but at a faster rate for those in dynamic environments.
To “live” EO as a strategic posture, firms need resources to develop and test new ideas, and then to position them on the market. In this context, financial capital is considered a universally deployable resource through which other resources, for example know-how, can be acquired. Thus, the following hypothesis can be put forward: H3: The relation between EO and business performance is moderated by access to financial
capital. Business performance increases with EO, but at a faster rate for those that have greater access to financial capital.
Hypotheses 2 and 3 refer to a two-way interaction model that describes the interactions between EO and the environment/resources and follow the contingency approach, which broadens this perspective. This approach assumes that in terms of strategic characteristics, organizations can only be described by using combinations of characteristics, and that those characteristics must work together harmoniously if they are to make an actual contribution to business performance. In this context, it is important that this synergy refers not only to the firm’s internal characteristics, but also to its interaction with the external environment. This interaction demonstrates how configuration analysis expands on the contingency approach (Mugler (1998)). It means, for example, that a firm that operates in a highly aggressive competitive environment, and has a certain degree of EO but is not willing or able to provide the resources necessary for innovation projects, will be less successful than a firm that does make such resources available. Therefore, we propose the following hypothesis: H4: (a) Business performance is explained by configurations of EO, access to capital, and envi-
ronmental dynamism. (b) Business performance is highest among firms with a high degree of EO, greater access to financial capital, and in dynamic environments than it is for other configurations. (c) Business performance is lowest among firms with a high EO, little access to financial capital, and in a stable environment than it is for other configurations.
3.2 The E xtensions
to the O riginal Study
To ascertain whether the results of the hypotheses are transferable to a different context, we apply the measurement instrument and statistical analysis methods used in the original study (Wiklund and Shepherd (2005)) to a sample from the Austrian Electrical and Electronics Industry (EEI). However, to establish the generalizability of results for other
See Wiklund and Shepherd (2005).
SBR 62 April 2010 175-198
181
H. Frank/A. Kessler/M. Fink
countries, we must consider that countries and industries can develop their own business cultures (House et al. (2004)), and that the development of measurement instruments may already be subject to “cultural bias”. If we do not ensure cross-cultural reliability and validity, then there is a risk that we could draw inaccurate conclusions and make the wrong recommendations. In international studies, the dimensions of the scales are also important. If a scale is to be used internationally, its factor structure and factor loadings should be identical, or at least similar, across all cultures. One study, in which the cross-cultural validity and reliability of an EO scale similar to the one used in this paper were tested on French- and English-speaking firms in Canada, identified a high level of intercultural reliability and validity in the form of internal consistency and convergent and discriminant validity10. Another study (Kreiser, Marino, and Weaver (2002b)) using the EO scale applied by Wiklund and Shepherd (2005) includes six countries (Australia, Finland, Mexico, Netherlands, Norway, and Sweden). It tested and confirmed the crosscultural validity of the EO scale. Therefore, we rely on this proven measurement tool in our study. However, it has been argued that national culture influences the EO-performance relation (e.g., Shane (1994); Busenitz and Lau (1996); Rauch et al. (2004); Todorovic and Ma (2008)). Wiklund and Shepherd (2005) suggest that a comparison should be performed in a different country, one which does not belong to the Nordic Europe Cluster. The GLOBE study (House et al. (2004)) shows that there are numerous differences, such as performance orientation, future orientation, and assertiveness, and several similarities, such as visionary leadership and uncertainty avoidance, between the geographical setting of the original study (Sweden) and that of the replication study (Austria) that are assigned to different regional clusters (Germanic Europe Cluster, such as Austria, and Nordic Europe Cluster, such as Sweden). Apart from using a geographical setting that is different from that of the original study, we introduce the following variations as extensions in the replication study. One difference lies in the fact that the original study’s sample contains a mix of various industries, such as knowledge-intensive manufacturing, labor-intensive manufacturing, professional services, and retail trade, but our sample includes only members of the trade association for the electrical and electronics industry. Our focus on a single industry has the advantage of reducing unobserved heterogeneity, which is a major problem in entrepreneurship research. However, we note that the Austrian EEI represents an assortment of firms with varying business purposes. We select this industry on the basis of its environmental conditions and the thorough documentation of the overall population. In their sample, Wiklund and Shepherd (2005) included only independent companies with 10 to 49 employees, that is, small businesses according to the EU definition. Our sample includes microbusinesses (fewer than ten employees), small businesses (10-49 employees), and medium-sized businesses (50-249 employees). Hence, we examine the relation between EO and business performance not just for a single size class of businesses, as is the case in the Swedish sample, which uses only businesses classified as small according to the EU definition, but for the entire range of small and medium-sized enterprises. Another difference is in the performance measure we use. One final difference between our study and 10
182
See Knight (1997). See Knight (1997).
SBR 62 April 2010 175-198
Entrepreneurial Orientation
that of Wiklund and Shepherd (2005) lies in the fact that no longitudinal data is available in our case. 4 M ethod 4.1 S ample The Austrian Electrical and Electronics Industry comprises 12 fields, the most significant of which are components; distribution and control apparatus; motors, generators, and transformers; and communication technology. With some 57,000 employees (2004), Austria’s EEI is among the largest – and thus one of the most important – manufacturing industries in Austria. In terms of the number of employees, this industry is Austria’s second-largest industrial employer. In 2004, exports provided decisive stimuli for the Austrian EEI sector. Compared to the previous year, exports rose by 5.2% and accounted for 11.7% of Austria’s total exports that year. Within the industry, the share of exports makes up 70% of total revenues. The most important export market is the EU, primarily Germany11. A comparison of the years 2000 and 2004 shows that the number of employees in this industry declined by about 7,000, mainly in the years 2001 to 2003. The fact that employees were laid off in nearly half (44.6%) of the businesses included in the analysis sample points to a difficult business environment12. The substantial decrease in the number of manual workers (–20.7%) is conspicuous compared to the number of salaried employees, which remained fairly stable in a five-year comparison. This decline can be explained by the transfer of production facilities to low-wage countries. At the same time, revenues dropped substantially year to year (2001: –7.3%; 2002: –5.8%; 2003: –3.4%) and did not begin to increase until 2004 (+6.1%). These figures show that the Austrian EEI sector has been exposed to dynamic environmental conditions in recent years, and at the same time they underline this sample’s high suitability for reviewing the effects of EO on business performance. The basis for this study is a survey carried out in the summer of 2005 among the 266 firms in the Austrian EEI. We obtained contact information for the survey from the membership directory of Austria’s EEI trade association. In Austria membership in the trade associations is obligatory. Therefore, the survey can be regarded as representing a complete inventory. We addressed our survey to the owner-managers of the firms, since strategy formulation is generally handled by the owner-managers of firms in this size range (Mugler (1998)).
11 12
See Association of the Austrian Electrical and Electronics Industries (2006). In the sample used by Wiklund (1998), which forms the basis for Wiklund and Shepherd (2005), the proportion of shrinking businesses was 17.9%.
SBR 62 April 2010 175-198
183
H. Frank/A. Kessler/M. Fink
The survey yielded n = 125 (47%) in returns and analyzable questionnaires. Out of the 118 firms that indicated their size, six (5%) are microbusinesses with up to nine employees, 26 (22%) are small businesses with 10 to 49 employees, 53 (45%) are medium-sized businesses with 50 to 249 employees, and 33 (28%) are large businesses with more than 249 employees. Thus, the final sample comprised a total of 85 (72%) SMEs, including the six microbusinesses. For the purposes of our analysis, we rely on the subsample of 85 small and medium-sized businesses (including microbusinesses). As was done in the original study, we excluded large businesses from the analysis. The distribution of micro-, small-, and medium-sized businesses roughly matches the structure of the overall population. Approximately 80% of the respondent firms were over ten years old, and 45.3% were over 40 years old at the time of the study. 4.2 Variables
and M easures
Business performance We combine two measures to capture business performance, sales growth and cash-flow growth relative to competitors (Cronbach’s α: 0.84). We measure both factors by using seven-point scales ranging from “much less than our competitors” to “much more than our competitors”. This measure constitutes a reduced version of the measure used by Wiklund and Shepherd (2005) which also included the elements of gross margin and profitability (relative to competitors), and sales and employee growth (as differences in the information provided at two different times by the firms surveyed and in relation to competitors). We use fewer components in the performance measure because longitudinal data was not available for this study, meaning that we could not identify any differences between two survey times. Another reason for this reduction lies in the special characteristics of Austrian business culture. Due to the general lack of publication requirements for small businesses and the respondents’ resulting lack of access to information on competing firms, it did not appear sensible to ask about gross margin and profitability relative to competitors. Because small and medium-sized businesses are especially reluctant to vary their number of employees in case there should be fluctuations in orders13, we use sales growth as a performance measure that is more closely linked to competition. However, since sales growth is not a disposable value, we use cash-flow growth as a complement. Entrepreneurial orientation To measure EO, we use the eight-item scale developed by Miller (1983; 1987a), and used by Wiklund and Shepherd (2005). The scale consists of three items that measure innovativeness, two items that measure risk-taking, and three items that measure proactiveness, with each item articulated as a pair of opposite statements on a seven-point scale. The psychometric properties of the EO scale (e.g., Kreiser, Marino, and Weaver (2002c); Lumpkin and Dess (2001)) have also been a topic of discussion, mainly regarding the question of whether EO is a reflective indicator (unidimensional measure) or a formative 13
184
See Mugler (1998).
SBR 62 April 2010 175-198
Entrepreneurial Orientation
indicator (multidimensional measure). Since Wiklund and Shepherd (2005) treat EO as a reflective indicator, our replication uses the same approach. The reflective nature of the scale is underscored by a Cronbach’s α of 0.86. Access to financial capital As do Wiklund and Shepherd (2005) we use a subjective measure of the small business owner’s/manager’s level of satisfaction with his/her access to financial capital. We measure this factor on a seven-point scale with the opposite statements “insufficient and a great impediment to our development” and “fully satisfactory for the firm’s development”. Environmental dynamism For this dimension, we also rely on the items used by Wiklund and Shepherd (2005). We operationalize environmental dynamism by using four items borrowed from Miller (1987a; 1987b), which pairs opposite statements such as decrease/increase in growth opportunities; changing/remaining technology; decrease/increase in rate of innovation; and decrease/increase in R&D activity. In our study, the scale exhibits acceptable reliability (Cronbach’s α: 0.73). Control variables Like Wiklund and Shepherd (2005), we use firm size and firm age as control variables, because these characteristics can also impact success. Since the data set in the original study consisted of firms from knowledge-intensive manufacturing, labor-intensive manufacturing, professional services, and retail, Wiklund and Shepherd (2005) also used the industry as a control variable. We derive our sample from an industry conglomerate that formally operates as an association. However, this association includes firms with varying purposes. At the same time, the number of cases is too small to classify the firms by purpose, so we did not use the “industry” control variable in our data analysis. 4.3 A nalysis In empirical research on strategic management and entrepreneurship, there are three basic perspectives on how independent variables relate to each other and to business performance: the main-effects approach, the contingency approach, and the configuration approach14. The simplest approach – the main-effects approach – describes the relation between the independent variables and business performance as a function in which the independent variables do not interact with one another, e.g., business performance = f (firm size, firm age, environmental dynamism, access to financial capital, EO). In addition, the main-effects approach assumes that if variables are not included in the analysis, it is because they have no impact. This approach implies that the relation between the independent variables and performance are valid under all circumstances. The contin14
See Wiklund and Shepherd (2005).
SBR 62 April 2010 175-198
185
H. Frank/A. Kessler/M. Fink
gency approach goes one step further and accounts for selected interactions between two variables in the form of two-way interactions. Thus, the contingency approach assumes that the extent and direction of an independent variable’s impact on a dependent variable will vary under certain circumstances. We express this relation as follows: business performance = f (firm size, firm age, environmental dynamism, access to financial capital, EO, EO * environmental dynamism, EO * access to financial capital, environmental dynamism * access to financial capital). The configuration approach considers “relationships among elements or items representing multiple domains”15. Therefore, a formal relation can be expressed as business performance = f (firm size, firm age, environmental dynamism, access to financial capital, EO, EO * environmental dynamism, EO * access to financial capital, environmental dynamism * access to financial capital, EO * environmental dynamism * access to financial capital). Thus, configuration analysis accounts for both two-way interactions, and a three-way interaction when there are three independent variables. In using the interaction terms, we mean-center all of the independent variables (Cohen et al. (2003)). Because we ask if EO is always an appropriate strategic orientation, or if its relationship with performance is more complex, we test whether a configurational model (three-way interaction), which involves the simultaneous consideration of strategy, organizational characteristics, and environment, can produce a greater understanding of the impact of EO on business performance than can a main-effects-only model and a contingency model. Like Wiklund and Shepherd (2005), we use hierarchical linear regression analysis to test whether the main effects, contingency, or configurational model best fits the data. In each step of the hierarchical analysis we add the next higher order of interaction and evaluate the incremental R2 and F-tests of statistical significance. An interaction effect exists if the interaction term yields a significant contribution over and above the direct effects of the independent variables16. Due to missing data, the number of cases we include in the regression analysis reduces to n = 75. Although the sample we use for the regression analysis is small, it meets the statistical requirements for multiple regression studies (Milton (2001), Urban and Mayerl (2006)). Further, we analyze a homogeneous population or sample. Doing so limits the risk of unobserved heterogeneity and also reduces sample size requirements. Our results indicate that the size of the sample is sufficient. An explained variance of 39% (p < 0.01) combined with a sample size of n = 75 yields a statistical power of 0.99, which is well above the conventional threshold (0.8) (Cohen (1988)). 5 R esults Table 1 displays the correlations of the variables. First of all, we note that with the exception of EO and environmental dynamism (0.50**), the correlations between the independent variables are relatively low. However, this value does meet our expectations, because 15 16
186
Dess, Newport, and Rasheed (1993). See Cohen et al. (2003).
SBR 62 April 2010 175-198
Entrepreneurial Orientation
the description of the sample and industry indicates high environmental dynamism and at the same time the nature of the industry itself permits us to assume a certain degree of innovativeness. Therefore, this result does not detract from our further analysis. We find a slightly negative correlation between firm age and EO, and between environmental dynamism and firm age. These findings indicate that newer firms consider their environment to be more dynamic than do older firms. There is a relatively high and significant correlation between business performance, which we use as a dependent variable, and access to financial capital (0.50**). We also find a significant correlation (0.23*) between business performance and firm age. The low correlation between business performance and EO (0.05) is surprising in light of other studies, such as those by Rauch et al. (2004), and Kemelgor (2002). Table 1: Means, standard deviations, and correlations for quantitative variables
Mean
S.D.
(1)
(2)
(1) Performance
4.85
1.94
1
(2) EO
34.15
8.83
0.05
1
(3) Environmental dynamism
21.00
4.31
0.16
0.50**
(4) Access to financial capital
4.81
1.84
(5) Firm size
85.27 68.35
(6) Firm age
32.64 28.43 0.23* –0.14 –0.22* 0.11
0.50** 0.20*
0.16
0.05
(3)
(4)
(5)
(6)
(7)
(8)
(10)
1
0.20*
1
0.07
0.21*
1 0.22*
1
(7) EO * env. dynamism
–0.01 –0.06 0.21*
0.05
–0.10
0.03
1
(8) EO * capital
0.12
–0.10
0.05
–0.11
0.06
0.19
0.03
(9) Dynamism * capital
0.07
0.06
–0.04 –0.22* 0.04
0.00
–0.17 0.41**
0.29**
0.12
–0.07 0.40** –0.06 –0.10
(10) EO * dynamism * capital
(9)
1 1
0.13 –0.34** –0.14
1
** p < 0.01; * p < 0.05
To check for multicollinearity in the regression analyses, we calculate the variance inflation factor (VIF) for the individual predictors (e.g., firm age). The values are just over one, meaning that they are far below critical values. In Table 2 we perform the hypotheses tests in a manner analogous to the original work of Wiklund and Shepherd (2005). We first add the control variables (results reported in Column 2), then the independent variables (main-effects model in Column 3), then the
SBR 62 April 2010 175-198
187
H. Frank/A. Kessler/M. Fink
two-way interaction terms (contingency model in Column 4), and finally the three-way interaction term (configuration model in Column 5). Table 2: Performance: main effects, contingency, and configuration model (n = 75) Control variables
Main-effects model
Contingency model
Configuration model
β
S.E.
β
S.E.
β
S.E.
β
S.E.
Firm size
0.111
0.003
0.009
0.003
–0.016
0.003
0.005
0.003
Firm age
0.201*
0.138
0.193*
0.126
0.182
0.129
0.215*
0.126
EO
–0.095
0.026
–0.121
0.026
–0.154
0.026
Environmental dynamism
0.154
0.054
0.172
0.057
0.242*
0.057
Access to financial capital
0.469***
0.112
0.521***
0.114
0.404***
0.124
EO * dynamism
–0.063
0.005
–0.113
0.005
EO * capital
0.065
0.013
0.141
0.013
Dynamism * capital
0.157
0.030
0.133
0.029
0.268**
0.003
EO * dynamism * capital R2 (significance of F-test)
0.062(*)
0.301(***)
0.342(***)
0.390(***)
Adjusted R2 (sign. of F-test) 0.036(*)
0.250(***)
0.262(***)
0.305(***)
∆R2 (change in sign. of F)
0.239(***)
0.041
0.062(*)
0.048(**)
Standardized regression coefficients are displayed in the table. *** p < 0.01; ** p < 0.05; *p < 0.10
The control variables (firm size and age) explain 6.2% of the variation in performance and the model barely attains statistical significance in this step (p = 0.099). Firm age turns out to be a slightly significant predictor (p = 0.091). Our next step is to review the universal influence of EO, environmental dynamism, and access to capital on business performance. These three variables account for an additional 23.9% of the variation in performance (p < 0.01) and for the adequate statistical significance of the model (p < 0.01). Thus, the explained variance attains a value of 30.1% (adjusted R2 = 25.0%). Access to financial capital (for business development) has a
188
SBR 62 April 2010 175-198
Entrepreneurial Orientation
significant positive relation (p < 0.01) to business performance. As was the case with our control variables, firm age turns out to be a slightly significant predictor (p < 0.1). In contrast to Wiklund and Shepherd’s (2005) study, EO does not prove to be a significant predictor. Its beta coefficient is even negative. This result means that H1 must be rejected: we cannot identify a universal relation between EO and business performance in this sample. But as in the original study, access to capital/resources proves to be a significant predictor. With an explained variance of 34.2%, the contingency model achieves a statistical significance overall (p < 0.01), but the explained variance gained (∆R2) of 4.1% does not meet the requirements for statistical significance (p = 0.26). Also, none of the three two-way interactions added (EO * environmental dynamism, EO * access to financial capital, environmental dynamism * access to financial capital) are statistically significant. This result means that H2 and H3 are also rejected. These results concur with the findings of Wiklund and Shepherd (2005). However, since the model itself is highly significant, we assume that the individual predictors do have a combined effect, but that for individual factors this effect is too nonspecific to reach the requisite level of statistical significance. The positive influence of access to financial capital on business performance is maintained (p < 0.01). Again, we find a certain degree of influence in firm age (p = 0.10). Including the three-way interaction term significantly increases explained variance to 39% and the adjusted R2 to 30.5% (∆R2 = 0.05; p < 0.05). This result supports H4 (a). Compared to the contingency model, the configuration consisting of EO, environmental dynamism, and access to financial capital exhibits greater explanatory power. As in the Wiklund and Shepherd (2005) study, the interaction term EO multiplied by environmental dynamism multiplied by access to financial capital also proves significant in our configuration analysis (p = 0.03). In addition to the three-way interaction term, firm age (p < 0.1) and access to financial capital (p < 0.01) once again turn out to be significant predictors. Furthermore, we identify environmental dynamism as a significant variable (p < 0.1). We interpret the consistently significant influence of firm age and access to financial capital as a sign that the results are robust. Access to financial capital in particular constitutes a critical resource for the realization of development and innovation activities. If we consider these results as a whole, then to interpret them we must study both the main effects and interactions. For higher-order interactions, all lower-order interactions and main effects must be considered (Aiken and West (1991)). Based on the coefficients in our regression analysis, we plot the effect of EO on business performance for given values of environmental dynamism and access to financial capital. We use values of environmental dynamism and access to capital one standard deviation above and below the mean, and we use a range of values for EO17. Figure 1 shows the results.
17
See Wiklund and Shepherd (2005).
SBR 62 April 2010 175-198
189
H. Frank/A. Kessler/M. Fink
Figure 1: Configuration analysis
High
Performance
Type 4: Dynamic Environment & High Access to Capital
Type 1: Stable Environment & Low Access to Capital Low
EO
High
Type 2: Stable Environment & High Access to Capital
Type 3: Dynamic Environment & Low Access to Capital
Low
The depiction of the interactions reveals two configurations, Types 1 and 4, in which EO has a positive effect on business performance. Hypothesis 4 (b) is confirmed by EO’s highly positive effect on performance in Type 4. Studies such as those by Miller (1988) and Zahra (1993) also confirm this assumption empirically. In this context, it is striking that EO has a similarly positive effect in Types 1 and 4, which represent two opposing combinations in terms of environmental conditions and access to financial capital. At the same time, the illustration shows two configurations, Types 2 and 3, in which EO has a negative effect on business performance. This finding not only runs counter to the hypotheses we develop in this paper, but also contradicts the widely held belief that there is a consistently positive connection between EO and business performance (e.g., Naman and Slevin (1993); Zahra and Covin (1995); Sapienza and Grimm (1997); Lee, Lee, and Pennings (2001); Runyan, Huddleston, and Swinney (2006)). In contrast, our replication shows that EO may even have a negative effect on business performance in certain configurations. This negative effect is most clearly evident in the case of a dynamic environment with low access to financial capital (Type 3). This finding underscores the importance of access to financial capital for the purpose of exploiting opportunities.
190
SBR 62 April 2010 175-198
Entrepreneurial Orientation
6 D iscussion
and
C onclusion
6.1 R esults of the R eplication O riginal Study
against the B ackdrop of the R esults of the
Wiklund and Shepherd’s study (2005) identifies a positive relation between EO and business performance. However, this is not the case in our replication study, which shows a negative, albeit not statistically significant, relation between EO and business performance in certain configurations. These results raise doubts as to whether the findings of Wiklund and Shepherd (2005) can be generalized. A different response to the question of generalizability should be based on a comparison of the configuration results, because, as clearly shown in Wiklund and Shepherd (2005), configuration analysis constitutes a superior instrument for analysis. This advantage manifests itself in the significantly higher explanatory power and in the exposure of EO’s effect on business performance in different contexts. A closer examination of the configuration types reveals both similarities and differences. There are parallels between the findings of the Wiklund and Shepherd study (2005) and the replication for Types 1 and 4 (stable environment/low access to capital; dynamic environment/high access to capital). For both types, we can confirm the positive effect of EO on business performance, even in the context of a different business culture. For Type 4, EO’s effect on business performance can be explained by the fact that dynamic environments open up opportunities that involve large capital requirements, and which call for proactive innovation behavior. EO’s effect on business performance in the case of Type 1 can be explained on the basis of a resource-based view, because EO provides an effective differentiation mechanism under resource constraints and stable market conditions (Wiklund and Shepherd (2005)), with the scarcity of financial resources encouraging the efficient deployment of funds. This result contradicts Hypothesis 4 (c), in which we postulate that business performance is lowest among firms with high EO, low access to financial capital, and a stable environment. Thus, we must reject Hypothesis 4 (c), as was the case in Wiklund and Shepherd (2005). However, there are also differences in the other two configuration types. Our study shows that EO can even have a negative impact on business performance. This impact arises under stable environmental conditions if there is high access to financial capital (Type 2) and is especially clear under dynamic environmental conditions with low access to capital (Type 3). These findings contradict those of Wiklund and Shepherd (2005), who find that EO has a positive effect on business performance in those two configurations. The negative effect of EO on business performance in the Type 2 configuration can be explained by the fact that in a stable environment, too few profitable entrepreneurial opportunities can arise (or be created) that can generate an appropriate return on the capital invested. Therefore, innovative behavior might consume resources in excess of the additional benefits it actually creates for the business. However, the negative effect on business performance in this configuration might be explained by an excessively generous use
SBR 62 April 2010 175-198
191
H. Frank/A. Kessler/M. Fink
of the available financial resources, because stable, predictable environmental conditions tend to lower the pressure for efficiency and differentiation from competitors. The especially clear negative effect of EO on business performance in Type 3 can be explained by the fact that entrepreneurial orientation must be combined with high access to capital in a dynamic environment (see Type 4). This result means that efforts to innovate that are undertaken with excessively low capital investment will be rendered ineffective, thus creating a negative cost/benefit relation. This tendency is exacerbated by the fact that obtaining additional financial resources is associated with relatively high costs in the case of low access to capital. Access to financial capital appears to play a decisive role and may generate different effects on performance depending on the specific configuration (degree of environmental dynamism, degree of EO). In light of the significant positive contribution of access to capital identified in all main effects, contingency, configuration models, this explanation seems entirely plausible. In environments that are dynamic and thus potentially rich in opportunities, access to capital becomes especially significant with regard to EO’s effect on business performance, while access to capital shows a less pronounced effect on performance in stable environments. Thus, we also note that the businesses in our sample operate in an especially dynamic environment. On a more theoretical level, the impact of EO on business performance in the context of environmental dynamism and access to financial capital, which is generally regarded as positive, turns out to be more complex. Hence, we reconsider the positive impact of EO on business performance and the catalytic role of high environmental dynamism and access to financial capital postulated in theory (Wiklund and Shepherd (2005)). Our identification of a partially negative impact of EO on business performance in a shrinking industry acts as an indication that we must take industry-related business conditions into account. In addition, the phase the industry has reached in the life cycle should be integrated into theories on the EO-performance relation (Lumpkin and Dess (2001)). The more differentiated picture drawn by our results can be understood as support for the claim that too much EO can have a negative impact on business performance. Businesses with excessive EO may exaggerate their risks and devote funds to R&D that are not in line with their market environments or the circumstances in which they compete (Madsen (2007)). 6.2 D ifferences
in Study
D esign
and E ffects on R esults
Business culture: In both Kemelgor (2002) and the meta-analysis conducted by Rauch et al. (2004), a nation’s business culture is regarded as an important factor influencing the EO-business performance relation. The GLOBE study shows that Austria exhibits a higher degree of performance/future orientation and assertiveness compared to Sweden (House et al. (2004)). These differences in business culture can be interpreted to mean that a high level of EO interacts syner-
192
SBR 62 April 2010 175-198
Entrepreneurial Orientation
gistically with higher performance/future orientation and assertiveness, thus generating a more substantial effect on business performance than in the Wiklund and Shepherd study (2005). Since this result contradicts the findings of both the Wiklund and Shepherd study and our replication study, we cannot consider it as a viable explanation. The differences in the results can also be interpreted to mean that the weaker effect of EO on performance in the Austrian sample arises from the fact that EO is less differentiated in a context of higher performance/future orientation and assertiveness than in cases in which those values are low. Therefore, the argument citing the significance of cultural differences in the EO-performance relation (Rauch et al. (2004)) is not refuted, but the direction of national culture’s influence can not be clearly determined. Industry: It is difficult for firms in the Austrian EEI that we analyze to achieve high performance levels. The fact that industry revenues dropped over several years and employees were laid off in nearly half (44.6%) of the businesses in the sample points to a difficult business environment. In contrast, in the sample used by Wiklund (1998), which forms the basis for Wiklund and Shepherd (2005), the share of shrinking businesses was only 17.9%. Widespread downsizing in an industry is generally accompanied by poor business performance and difficult access to financial capital at the company level. This fact restricts the firm’s ability to be proactive and to take on the riskier innovation activities that can affect performance. Hence, there are sound arguments for a differentiated view of EO’s impact in different configurations. These arguments include not only company growth and stabilization, but also negative growth; in this case, layoffs. Our replication study is based on only one industry, and the difficult situation in that industry clearly manifests itself in our case. The mix of industries analyzed in Wiklund and Shepherd (2005) makes it possible to compensate for this problem. This consideration suggests that “industry effects” might be another reason for the differing results. Cross-sectional data: In addition to the decline in the industry, another reason for the divergent results may be that no longitudinal data is available for this study. For example, Zahra and Covin (1995) show that the strength of the EO-performance relation increases over time. However, we should not overestimate this effect in our study, which is based on businesses in the electrical and electronic industry, because, according to its own indications, this industry makes a substantial contribution to R&D expenditures in Austria and we can assume continuity in R&D activities. Only when R&D expenditures and innovations are discontinuous does a longitudinal perspective become especially relevant. Thus, we can assume that cross-sectional data is not a decisive reason for the differences in the results. Performance measure: As Rauch et al. (2004) note, the composition of performance measures is very important in the EO-performance relation. In contrast to Wiklund and Shepherd (2005), whose performance measures are sales and cash flow growth, gross margin and profitability relative to competitors, and sales growth and employee growth measured at two different times, our replication relies only on cash flow and sales growth relative to competitors. Based on the large share of shrinking companies (in terms of staff and revenues) in the industry and in our replication sample, the measurement of performance relative to the competition, and the need to compare performance with the best in
SBR 62 April 2010 175-198
193
H. Frank/A. Kessler/M. Fink
the industry, we assume that performance assessments generally tend to be critical. Against this backdrop, the weakened effect of EO on business performance appears plausible. In contrast to Wiklund and Shepherd (2005), our study uses a performance measure based on only two dimensions, which are both sensitive to business downsizing processes. This effect is presumably quite strong and can be regarded as a significant reason for the differences in the results. Firm size: In contrast to Wiklund and Shepherd (2005), whose sample consisted of small businesses with ten to 49 employees, the sample in our replication study is based on businesses with one to 249 employees, i.e., micro-, small-, and medium-sized businesses. Especially in medium-sized businesses, there is a greater need for clear structures, routines, task assignments, and standards. For such businesses the effect of EO may be counteracted by the need for formalization measures (Mugler (1999)). This effect may also contribute to the weaker impact of EO in our sample. The inclusion of medium-sized businesses in the replication study can thus also be regarded as a reason for the differing results. These considerations suggest that the differences in our results cannot be explained by differences in national business cultures alone, but that the industry context, performance measurement, and firm size also have a major influence on the results. The misgivings expressed on the generalizability of results can be partly eliminated by citing differences in study design. Another surprising outcome of this study compared to the original study, and to empirical EO-performance research in general, is the fact that EO can have a negative effect on business performance in certain configurations. 6.3 M anagerial I mplications These results imply that EO does not constitute a “secret weapon” under all environmental conditions and developments in a firm (Lackner (2002)) but that EO might preferably be pursued in rapidly changing environments that offer new opportunities, and in which the firm has sufficient financial resources at its disposal to take advantage of those opportunities using a portfolio of innovation activities. Using EO is not advisable in cases in which a dynamic environment is combined with low access to financial capital. Should a firm’s financial resources be depleted due to adverse business conditions, it is first advisable to restore those resources, e.g., by streamlining, before taking EO measures. 6.4 R esearch I mplications The usefulness of replication studies has been confirmed by the fact that similar findings indicate generalizability, while differing results make it necessary to refine the original theory, or in this case, to address the effects of EO on business performance in a distinctly different manner. According to our theory-based claim, empirical studies must also incorporate the business life-cycle construct and industry-related business conditions.
194
SBR 62 April 2010 175-198
Entrepreneurial Orientation
In contrast to most prior results, our findings show that EO can also have a negative effect on business performance. This finding calls for further investigation of the conditions under which the EO-performance relation is negative. Therefore, special attention should be paid to differentiating between specific industry contexts and developments. Thus, it becomes apparent that to better understand the EO-performance relation, configuration analyses should be used more extensively in future research. Comparative international studies are also necessary if we are to improve our ability to assess whether country-specific research findings can be generalized. To enable a methodically grounded analysis of the impact of national business culture, such studies should also account for the national business and corporate cultures as variable(s). This claim is especially important because there are no studies that explicitly examine how national culture moderates the EO-performance relation (Rauch et al. (2004)). In addition, a stronger emphasis on replication studies is desirable, since such studies can enhance the validity, reliability and, in the case of extensions, the generalizability of empirical findings. Finally, it would also be advisable for research to discuss suitable performance measures, because the EO-performance relation is sensitive to the composition of such measures in main-effects and configuration analyses. R eferences Aiken, Leona S. and Stephen G. West (1991), Multiple Regression: Testing and Interpreting Interactions, Newbury Park: Sage Publications. Amir, Yehuda and Irit Sharon (1990), Replication Research: A “Must” for the Scientific Advancement of Psychology, in James W. Neuliep (ed.), Handbook of Replication Research in Social Science [Special Issue], Journal of Social Behavior and Personality 5, 51-69. Association of the Austrian electrical and electronics industries (2006), Die Infrastruktur-Branche, Jahresbericht der österreichischen Elektro- und Elektronikindustrie, Wien: Wirtschaftskammer Österreich, Bundessparte Industrie. Barringer, Bruce R. and Allen C. Bluedorn (1999), The Relationship between Corporate Entrepreneurship and Strategic Management, Strategic Management Journal 20, 421-444. Baumgarth, Carsten and Heiner Evanschitzky (2005), Die Rolle von Replikationen in der Marketingwissenschaft, Marketing – ZFP 27, 253-262. Becherer, Richard C. and John G. Maurer (1997), The Moderating Effect of Environmental Variables on the Entrepreneurial and Marketing Orientation of Entrepreneur-led Firms, Entrepreneurship Theory and Practice 22, 47-58. Busenitz, Lowell W. and Chung-Ming Lau (1996), A cross-cultural cognitive model of new venture creation, Entrepreneurship Theory and Practice 20, 25-39. Cohen, Jacob (1988), Statistical Power Analysis for the Behavioral Sciences, 2nd ed., Hillsdale, New Jersey: Lawrence Erlbaum Associates. Cohen, Jacob, Patricia Cohen, Stephen West, and Leona S. Aiken (2003), Applied multiple regression/correlation analysis for the behavioral sciences, Mahwah, London: Lawrence Erlbaum Associates. Dess, Gregory G., Stephanie Newport, and Abdul M.A. Rasheed (1993), Configuration Research in Strategic Management: Key Issues and Suggestions, Journal of Management 19, 775-795.
SBR 62 April 2010 175-198
195
H. Frank/A. Kessler/M. Fink
Dess, Gregory G., G. T. Lumpkin, and Jeffrey G. Covin (1997), Entrepreneurial Strategy Making and Firm Performance: Tests of Contingency and Configurational Models, Strategic Management Journal 18, 677-695. Diekmann, Andreas (2004), Empirische Sozialforschung: Grundlagen, Methoden, Anwendungen, 12th ed., Reinbek bei Hamburg: Rowohlt. Easley, Richard W., Charles S. Madden, and Mark G. Dunn (2000), Conducting Marketing Science: The Role of Replications in the Research Process, Journal of Business Research 48, 83-92. Eden, Dov (2002), From the Editors, Replication, Meta-Analysis, Scientific Progress, and AMJ’s Publication Policy, Academy of Management Journal 45, 841-846. Furchtgott, Ernest (1984), Replicate, again and again, American Psychologist 39, 1315-1316. Guth, William D. and Ari Ginsberg (1990), Guest Editors’ Introduction: Corporate Entrepreneurship, Strategic Management Journal 11 [Special Issue], 5-15. Haid, Dirk (2004), Corporate Entrepreneurship im strategischen Management. Ansatz zur Implementierung des Unternehmertums im Unternehmen, Wiesbaden: Deutscher Universitätsverlag. Harms, Rainer (2004), Entrepreneurship in Wachstumsunternehmen, Wiesbaden: Deutscher Universitätsverlag. Harms, Rainer and Thomas Ehrmann (2003), The Performance Implications of Entrepreneurial Management: Linking Stevenson’s and Miller’s Conceptualization to Growth, Paper presented at the Babson Kauffman Research Conference on Entrepreneurship, MA: Babson College. Hendrick, Clyde (1991), Replications, Strict Replications, and Conceptual Replications: Are They Important?, in James W. Neuliep (ed.), Handbook of Replication Research in Social Science [Special Issue], Journal of Social Behavior and Personality 5, 41-50. House, Robert J., Paul J. Hanges, Mansour Javidan, Peter W. Dorfman, and Vipin Gupta (2004), Culture, Leadership, and Organizations – The GLOBE study of 62 societies, Thousand Oaks: Sage Publications. Hubbard, Raymond and Scott J. Armstrong (1994), Replications and Extensions in Marketing: Rarely Published but Quite Contrary, International Journal of Research in Marketing 11, 233-248. Hubbard, Raymond, Daniel E. Vetter, and Eldon L. Little (1998), Replication in Strategic Management: Scientific Testing for Validity, Generalizability, and Usefulness, Strategic Management Journal 19, 243-254. Kane, Edward J. (1984), Why Journal Editors Should Encourage the Replication of Applied Econometric Research, Quarterly Journal of Business and Economics 23, 3-8. Kemelgor, Bruce H. (2002), A Comparative Analysis of Corporate Entrepreneurial Orientation between Selected Firms in the Netherlands and the USA, Entrepreneurship and Regional Development 14, 67-87. Knight, Gary A. (1997), Cross-Cultural Reliability and Validity of a Scale to Measure Firm Entrepreneurial Orientation, Journal of Business Venturing 12, 213-225. Kreiser, Patrick M., Louis D. Marino, and K. Mark Weaver (2002a), Reassessing the Environment-EO Link: The Impact of Environmental Hostility on the Dimensions of Entrepreneurial Orientation, Academy of Management Proceedings 2002, G1-G6. Kreiser, Patrick M., Louis D. Marino, and K. Mark Weaver (2002b), Assessing the Relationship between Entrepreneurial Orientation, the External Environment, and Firm Performance, Frontiers of Entrepreneurship Research, 268-282. Kreiser, Patrick M., Louis D. Marino, and K. Mark Weaver (2002c), Assessing the Psychometric Properties of the Entrepreneurial Orientation Scale: A Multi-Country Analysis, Entrepreneurship Theory and Practice 26, 71-94. Lackner, Stefan (2002), Voraussetzungen und Erfolgsfaktoren unternehmerischen Denkens und Handelns: Eine empirische Analyse mittelständischer Unternehmen, Hamburg: Dr. Kovac. Lee, Choonwoo, Kyungmook Lee, and Johannes M. Pennings (2001), Internal Capabilities, External Networks, and Performance: A Study on Technology-Based Ventures, Strategic Management Journal 22, 615-640. Lindsay, Murray R. (1994), Publication System Biases Associated With the Statistical Testing Paradigm, Contemporary Accounting Research 11, 33-57.
196
SBR 62 April 2010 175-198
Entrepreneurial Orientation
Lindsay, Murray R. and A. S. C. Ehrenberg (1993), The Design of Replicated Studies. American Statistician 47, 217-228. Lumpkin, G. T. and Gregory G. Dess (1996), Clarifying the Entrepreneurial Orientation Construct and Linking it to Performance, Academy of Management Review 21, 135-172. Lumpkin, G. T. and Gregory G. Dess (2001), Linking two Dimensions of Entrepreneurial Orientation to Firm Performance: The Moderating Role of Environment and Industry Life Cycle, Journal of Business Venturing 16, 429-451. Marino, Louis, Karen Strandholm, H. Kevin Steensma, and K. Mark Weaver (2002), The Moderating Effect of National Culture on the Relationship Between Entrepreneurial Orientation and Strategic Alliance Portfolio Extensiveness, Entrepreneurship Theory and Practice 26, 145-160. Madsen, Einar (2007), The significance of sustained entrepreneurial orientation on performance of firms – a longitudinal analysis, Entrepreneurship & Regional Development 19, 185-204. Miller, Danny (1983), The Correlates of Entrepreneurship in Three Types of Firms, Management Science 29, 770-791. Miller, Danny (1987a), Strategy Making and Structure: Analysis and Implications for Performance, Academy of Management Journal 30, 7-32. Miller, Danny (1987b), The Structural and Environmental Correlates of Business Strategy, Strategic Management Journal 8, 55-76. Miller, Danny (1988), Relating Porter’s Business Strategies to Environment and Structure: Analysis and Performance Implications, Academy of Management Journal 31, 280-308. Milton, Sande (2001), A sample size formula for multiple regression studies, Public Opinion Quarterly 50, 112-118. Mugler, Josef (1998), Betriebswirtschaftslehre der Klein- und Mittelbetriebe – Band 1, 3rd ed., Wien: Springer. Mugler, Josef (1999), Betriebswirtschaftslehre der Klein- und Mittelbetriebe – Band 2, 3rd ed., Wien: Springer. Naman, John L. and Dennis P. Slevin (1993), Entrepreneurship and the Concept of Fit: A Model and Empirical Tests, Strategic Management Journal 14, 137-153. Neuliep, James W. and Rick Crandall (1990), Editorial Bias Against Replication Research, in James W. Neuliep (ed.), Handbook of Replication Research in Social Science [Special Issue], Journal of Social Behavior and Personality 5, 85-90. Rauch, Andreas, Johan Wiklund, Michael Frese, and G. T. Lumpkin (2004), Entrepreneurial Orientation and Business Performance: Cumulative Empirical Evidence, Frontiers of Entrepreneurship Research, 164-177. Rosenthal, R. (1990), Replication in Behavioral Research, Journal of Social Behavior and Personality 5, 1-30. Runyan, Rodney C., Patricia Huddleston, and Jane Swinney (2006), Entrepreneurial Orientation and Social Capital as Small Firm Strategies: A Study of Gender Differences from a Resource-Based View, International Entrepreneurship and Management Journal 2, 455-477. Sapienza, Harry J. and Curtis M. Grimm (1997), Founder Characteristics, Start-Up Process, and Strategy/Structure Variables as Predictors of Shortline Railroad Performance, Entrepreneurship Theory and Practice 22, 5-24. Schmidt, F. L. (1992), What Do Data Really Mean? Research Findings, Meta-analysis, And Cumulative Knowledge in Psychology, American Psychologist 47, 1173-1181. Shane, Scott (1994), The effect of national culture on the choice between licencing and direct foreign investment, Strategic Management Journal 15, 627-642. Todorovic, Zelimir W. and Jun Ma (2008), Entrepreneurial and market orientation relationship to performance. The multicultural perspective, Journal of Enterprising Communities 2, 21-36. Tsang, Eric W. K. and Kai-Man Kwan (1999), Replication and Theory Development in Organization Science: A Critical Realist Perspective, Academy of Management Review 24, 759-780. Urban, Dieter and Jochen Mayerl (2006), Regressionsanalyse: Theorie, Technik und Anwendung, 2nd ed, Wiesbaden: VS Verlag für Sozialwissenschaften.
SBR 62 April 2010 175-198
197
H. Frank/A. Kessler/M. Fink
Wiklund, Johan (1998), Small firm growth and performance: entrepreneurship and beyond, Jönköping: Jönköping International Business School. Wiklund, Johan (1999), The Sustainability of the Entrepreneurial Orientation-Performance Relationship, Entrepreneurship Theory and Practice 24, 37-48. Wiklund, Johan and Dean Shepherd (2003), Knowledge-Based Resources, Entrepreneurial Orientation, and the Performance of Small and Medium-Sized Businesses, Strategic Management Journal 24, 1307-1314. Wiklund, Johan and Dean Shepherd (2005), Entrepreneurial Orientation and Small Business Performance: a Configurational Approach, Journal of Business Venturing 20, 71-91. Yoo, So-Jin (2001), Entrepreneurial Orientation, Environmental Scanning Intensity, and Firm Performance in Technology-Based SMEs, Frontiers of Entrepreneurship Research, 356-367. Zahra, Shaker A. (1993), Environment, Corporate Entrepreneurship, and Financial Performance: A Taxonomic Approach, Journal of Business Venturing 8, 319-340. Zahra, Shaker A. and Dennis M. Garvis (2000), International Corporate Entrepreneurship and Firm Performance: The Moderating Effect of International Environmental Hostility, Journal of Business Venturing 15, 469-492. Zahra, Shaker A. and Jeffrey G. Covin (1995), Contextual Influences on the Corporate Entrepreneurship-Performance Relationship: A Longitudinal Analysis, Journal of Business Venturing 10, 43-58.
198
SBR 62 April 2010 175-198
