Master Thesis draft for MSc IB&M

(1)

Master Thesis draft for MSc IB&M

Entrepreneurial orientation, export performance, and the influence of

culture: a meta-analysis

University of Groningen

Faculty of Economics and Business

(2)

I

ABSTRACT

Recently, more and more studies are being published on the entrepreneurial orientation – export performance relationship. Driven by the lack of a definitive conclusion on the strength and direction of the relationship, this thesis aimed to give an accurate and definitive

estimation of the effect size of entrepreneurial orientation on export performance. To investigate the relationship between entrepreneurial orientation and export performance, a meta-analysis using 41 studies published between 2000 and 2020 including a total of 8042 firms from 21 different countries, was conducted. Additionally, this thesis investigated how several variables and measurement aspects of export performance influence the

entrepreneurial orientation – export performance relationship through a moderator analysis. The results of this thesis have shown that there is a significant and positive relationship between entrepreneurial orientation and export performance. The relationship looks to even be stronger than the already confirmed positive relationship between entrepreneurial

orientation and firm performance, indicating that entrepreneurial orientation is more effective in an export environment. The results of variables and measurement aspects influencing the relationship were found to be insignificant, except for studies using subjective measures for export performance finding a significantly stronger relationship compared to studies using objective measures.

(3)

II

LIST OF TABLES

Table 1: List of studies used ... 20

Table 2: List of variables/moderators and their measures ... 21

Table 3: Bivariate meta-analysis results ... 24

Table 4: Moderator analysis results ... 29

(6)

V

LIST OF FIGURES

Figure 1: Conceptual model ... 16

Figure 2: Literature search flow diagram ... 19

Figure 3: Culture moderator confidence intervals ... 25

Figure 4: Firm size moderator confidence intervals ... 26

Figure 5: Measure type moderator confidence intervals ... 27

Figure 6: Number of measures moderator ... 27

(7)

1

1. INTRODUCTION

Strategic Orientations are defined as broad strategic choices which are implemented to attain sustainable superior performance and are the starting point for creating employee, manager, and firm behaviour and activities which are necessary to achieve superior performance (Gatignon & Xuereb, 1997; Narver & Slater, 1990). In other words, strategic orientations describe the general mindset of the managers of a firm which translates to employee behaviour and firm activities. Strategic orientation is the collective name for three different orientations: Entrepreneurial Orientation, Market Orientation, and Learning Orientation. Each of these orientations reflects a different manager mindset which encourages behaviours and internal processes (Hakala, 2011; Krzakiewicz & Cyfert, 2019).

This thesis focuses on one of the strategic orientations, entrepreneurial orientation. The entrepreneurial mindset is best described by Miller as “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” (1983: 771). Thus, firms with an EO will put importance on innovation, are willing to take risks, and are proactive.

A large part of the research into strategic orientations is focused on linking

orientations to a myriad of firm performance measures (Baker & Sinkula, 1999; Calantone, Cavusgil, & Zhao, 2002; Ellis, 2006; Gatignon & Xuereb, 1997; Narver & Slater, 1990). Entrepreneurial orientation is no exception as it has gained a lot of attention in the fields of strategic management and entrepreneurship, where previous work has already established that there is a positive relationship between entrepreneurial orientation and firm performance (Lumpkin & Dess, 1996; Rauch, Wiklund, Lumpkin, & Frese, 2009; Soares & Perin, 2019; Wales, Gupta, & Mousa, 2013).

Recently, more studies are being published about the entrepreneurial orientation – export performance relationship (Ribau, Moreira, & Raposo, 2017; Rua, Monteiro, & Soares, 2017). While the majority of the results indicate a positive relationship, a conclusion based on meta-analysis research is unexplored territory. This raises the question of whether

(8)

2 strength and direction of the entrepreneurial orientation – export performance relationship, combined with the preliminary suggestion that this relationship might be stronger and therefore valuable to exporting firms, results in a research gap

An interesting phenomenon that has surfaced with two recent meta-analyses is

whether the effect of strategic orientations on performance is influenced by culture. Soares & Perin (2019) found no evidence that the impact of entrepreneurial orientation on performance is stronger in Western culture compared to Eastern culture. That result differs from the result of an older meta-analysis by Ellis (2006) who found that the effect of market orientation on performance was stronger in Western culture. Bıçakcıoğlu-Peynirci & İpek (2020) used the Hofstede (1980) cultural dimensions of Masculinity and Collectivism as moderators in their meta-analysis and found that both cultural dimensions influence the market orientation – performance relationship. Thus, there is mixed evidence on the role of culture in the relationship between strategic orientations and performance and much remains unknown.

For entrepreneurial orientation specifically, there is evidence that entrepreneurial orientation is more prevalent in cultures with certain Hofstede (1980) cultural dimension scores (Mueller & Thomas, 2001). This evidence proves that culture does influence entrepreneurial orientation itself, but the question remains whether culture influences the relationship between entrepreneurial orientation and different kinds of performance. While the influence of culture on the relationship between entrepreneurial orientation and

performance has been tested and found insignificant (Soares & Perin, 2019), this might not be the same for export performance. Since the presumption of this thesis is that entrepreneurial orientation plays a larger role in export performance compared to organisational performance, a potential difference in the ability to leverage entrepreneurial orientation for export

performance based on culture may come to light. Thus, while there is evidence of the influence of culture on entrepreneurial orientation, the influence on the relationship with performance remains unclear which results in a research gap as a potential for exploratory research. Through a meta-analysis this thesis will fill these research gaps as it aims to answer the following research questions:

1. What is the strength and direction of the relationship between entrepreneurial orientation and export performance?

(9)

3 The meta-analysis will be conducted by collecting previous studies containing data on the relationship between entrepreneurial orientation and export performance. These articles will be collected, coded, and saved into a database. After enough work is collected

calculations will be conducted to find the average effect of entrepreneurial orientation on export performance across all studies used. Additionally, a moderator analysis will be conducted to determine the effect of culture and other variables on the entrepreneurial orientation – export performance relationship.

This thesis contributes to the literature in several ways. The first contribution is the meta-analytic estimation of the entrepreneurial orientation – export performance relationship. While the theory agrees that entrepreneurial orientation should have a significant positive effect on export performance, the primary studies find mixed, mostly positive results. Therefore, this meta-analysis will shed light on the strength and direction of the relationship in a more definitive way. Confirming the entrepreneurial orientation – export performance relationship will allow research to progress to factors that may influence this relationship, or to investigate the effect of entrepreneurial orientation on other performance indicators. The second contribution of this thesis consists of the moderator analysis which investigates the effects of culture, firm size, and methodological factors on the entrepreneurial orientation export performance relationship. While some of these factors have been investigated in other studies, a conclusion based on multiple studies is still missing. Using multiple studies from different countries to determine the effect of these factors will increase the understanding of how both internal and external variables influence the entrepreneurial orientation – export performance relationship.

(10)

4

2. LITERATURE REVIEW AND HYPOTHESIS

DEVELOPMENT

This chapter uses theory to explain the main concepts of this thesis. Additionally, the

framework that establishes the relationship between the main concepts will also be explained. 2.1 Entrepreneurial orientation: definition and concept

Entrepreneurial orientation is one of the three major orientations which are collectively called strategic orientation (Gatignon & Xuereb, 1997). Strategic orientation is described as a “broad set of strategic choices implemented in pursuit of sustainable performance and is a predisposition for creating the proper employee, manager, and overall firm behaviour and activities for achieving superior performance” (Sorensen & Madsen, 2012: 426). Based on the description by Sorensen and Madsen (2012), two concepts can be identified within strategic orientation, business strategy and orientation. Business strategy in general outlines how firms seek to achieve a competitive advantage in their industry (Morgan & Strong, 2003), whereas the orientation is a preference toward strategies that emphasise certain key traits (Venkatraman, 1989). Essentially, the strategic orientation of a firm describes how it tends to handle opportunities and threats.

Entrepreneurial orientation, as mentioned before, is one type of strategic orientation. An entrepreneurial firm is best described by Miller (1983: 771) as “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”. With this description, Miller (1983) conceptualised three main traits that entrepreneurial firms favour, which has been adopted and developed by other influential authors in this field (Covin & Slevin, 1989; Lumpkin & Dess, 1996) namely: innovativeness, risk-taking, and proactiveness.

(11)

5 Risk-taking, in general, describes the act of accepting a certain chance of failure associated with a strategy to accomplish goals or targets (Baird & Thomas, 1985). However, a risk analysis is rarely able to identify all the factors influencing a certain strategy outcome, and therefore some degree of uncertainty will remain. Uncertainty is inherently risky which means that all strategies with uncertain elements will have some degree of risk. Investing in strategies with a high amount of uncertainty is more like gambling instead of strategic risk-taking. What separates entrepreneurial firms from other firms stems from two other types of risk-taking defined by Baird and Thomas (1985: 231–232) as “committing a relatively large portion of assets” and “borrowing heavily”. Firms with an entrepreneurial orientation are willing to commit a larger portion of their assets or make a larger investment through borrowing compared to firms with a different strategic orientation (Miller & Friesen, 1978).

Proactiveness is broadly described as having proactive behaviour when it comes to emerging market or industries, continuously looking for new market opportunities, and experimenting with possible responses to environmental trends (Venkatraman, 1989). In other words, proactive firms encourage opportunity seeking and take initiative by anticipating and acting upon new opportunities. The opposite of proactiveness is passiveness, which is indifference or the inability to act upon opportunities, whereas reactiveness is responding to changes in the environment and moves made by competitors. What separates proactive firms from their counterparts is that they instead of reacting to the environment, they try to shape the environment while at the same time reacting to competitors (Lumpkin & Dess, 1996).

After the original three traits, two additional traits have been conceptualised by Lumpkin and Dess (1996): competitive aggressiveness and autonomy. These traits are less popular and are not used as much in identifying and measuring entrepreneurial firms compared to the original three traits (Wales et al., 2013). However, these traits still explain some important elements of entrepreneurial orientation. Competitive aggressiveness

describes the tendency of the firm to directly and intensely challenge its competitors to gain entry in a market or to improve its market position, to outperform their industry rivals (Lumpkin & Dess, 1996). There is some overlap with the reactive element of proactiveness, however, the sole focus is on the competitors to gain competitive advantage and outperform rivals. What differs firms with competitive aggressiveness from their counterparts is a strong desire to remove their competitors and dominate the market, compared to a live and let live attitude from firms without competitive aggressiveness (Lumpkin & Dess, 1996).

(12)

6 1996). Constraints such as lack of resources and competitor pressures will always be present no matter how much a firm encourages autonomy. However, the idea behind autonomy is that the firm allows freedom for individuals or teams to be creative and that the firm supports promising ideas, such that the firm maintains a strong entrepreneurial spirit (Lumpkin & Dess, 1996).

To summarise, entrepreneurial orientation is the mindset of the firm. In a very general way, this mindset describes how the firm behaves to achieve its goals. Firms with an

entrepreneurial orientation will use a combination of innovativeness, risk-taking, proactiveness, competitive aggressiveness, and autonomy to achieve their goals.

2.2 Export performance

When a firm conducts export activities, the economic outcomes of these activities are referred to as export performance (Katsikeas, Leonidou, & Morgan, 2000). Export performance as a measure is supposed to indicate the success of the export activities of a firm. However, there are many ways this can be measured, for example, export sales ratio, export sales growth, export profitability, export mart share, etc. The many ways of measuring export performance may allow for versatility, but at the same time, it becomes difficult to compare results across studies as studies might use completely different ways of measuring export performance.

(13)

7 unique names makes it even more difficult to compare the results of multiple studies, even though they all use export performance as an outcome construct.

Furthermore, export performance can be measured with objective indicators, subjective indicators, or both. The review by Katsikeas et al. (2000) showed that most researchers choose objective measures but they also acknowledged that several measurement problems are posed by objective measures. The three major problems that were identified are: (1) published data by firms often do not distinguish between domestic and export business; (2) certain objective measures may be influenced by factors that complicate comparability i.e. accounting practices in different countries; (3) what constitutes as a successful or

unsuccessful firm is arbitrarily decided by researchers, often based on “the average of a sample extracted from a heterogeneous population of export firms” (Katsikeas et al., 2000: 505). The second problem is not as obvious since a lot of research on export performance is done in a one country-context, and it only becomes a problem when comparing results across countries (Zou, Taylor, & Osland, 1998). That is not to say that objective data should not be used to in favour of subjective measures, in terms of certain aspects of performance, objective financial data is more precise than a manager’s gut feeling captured on a Likert scale. Along the same line Shoham (1998) found that using both objective and subjective measures, when possible, leads to the most accurate result. When firms are not willing to share specific

financial data, subjective measures can be as an alternative, as subjective satisfaction data and subjective comparison data is found to be highly correlated to objective data (Morgan,

Kaleka, & Katsikeas, 2004; Shoham, 1998).

In another empirical review of studies using export performance, Zou et al. (1998) gathered many export performance measures and created the EXPERF scale. This scale subjectively measures export performance and can be applied to multiple countries. Zou et al. (1998) divided export performance into three dimensions: financial export performance, strategic export performance, and satisfaction with the export venture, with each dimension consisting of three statements.

(14)

8 2.3 Theoretical foundation: Resource-Based View

This section explains the background and main arguments of the Resource-Based View, which is the theory that will be used to empirically establish the relationship between the two main concepts entrepreneurial orientation and export performance.

Shortly after Porter (1980) published his book Competitive strategy, the so-called positioning school became the most popular subject in strategic management. The positioning school focused on the external environment of the firm where positioning in the market and compared to competitors create a competitive advantage for the firm (Porter, 1980). During that time Barney (1991) did the opposite and improved on the concept of creating competitive advantage with the internal environment of the firm, which was first conceptualised by

Wernerfelt (1984). In his article, Barney (1991) created the modern version of the Resource-Based View (RBV) and described the requirements for a competitive advantage and sustained competitive advantage. These advantages are created by the firm through firm resources which include “all assets, capabilities, organisational processes, firm attributes, information, knowledge etc. controlled by the firm which enables it to conceive and implement strategies that improve its efficiency and effectiveness” (Barney, 1991: 101). The firm resources can be broadly categorised into three types of resources: physical capital resources, human capital resources, and organisational capital resources. The physical capital resources are the assets, location, and raw materials, whereas the human capital resources describe the characteristics of individual managers and workers, and their relationships. The organisational capital resources include the organisational structure of the firm both formal and informal, intergroup relations, reporting system, and controlling and coordinating (Barney, 1991).

The basic argument by Barney (1991) is that firm resources are heterogeneous, meaning that there are no firms with the same resources. This argument makes sense

(15)

9 The result that firms would want to achieve with their strategies and firm resources is to gain a competitive advantage. According to Barney (1991: 102), firms gain a competitive advantage when “implementing a value strategy not simultaneously being implemented by any current or potential competitors”. When the benefits of the value strategy are unable to be copied by other firms, it constitutes a sustained competitive advantage (Barney, 1991).

However, to become a sustained competitive advantage, a firm resource needs to have four attributes, the firm resource must be: (1) valuable, (2) rare, (3) imperfectly imitable, and (4) non-substitutable. A bundle of resources can also fulfil these attributes and grant sustained competitive advantage, for simplification purposes this will also be grouped under the term firm resources.

To be of use for a firm, a firm resource needs to be valuable, which refers to the second part of the definition of firm resources, enabling a firm to create or execute strategies that will improve the efficiency and effectiveness of the firm (Barney, 1991). If a firm resource is not valuable, there is no reason to use it or to invest in it and it certainly will not result in a competitive advantage. Additionally, being valuable alone is not enough to guarantee an edge over competitors. A firm resource can be very valuable, but if other firms have access to this resource as well, it loses its purpose of granting a competitive advantage. Therefore, a firm resource must also be rare, meaning that only a few other firms have access to this resource. The fewer firms have access to the resource the higher the advantage will be. Resources which are both valuable and rare will allow the firm to have a competitive

advantage.

While valuable and rare resources allow for competitive advantage, only if the resource cannot be easily copied or imitated will the advantage last for a longer period, resulting in a sustained competitive advantage. Barney (1991) identified several factors that influence the imitability of firm resources which has been expanded over the years by other contributions (Wills-Johnson, 2008). One factor is causal ambiguity, which occurs when the firm itself does not understand or very imperfectly understands the link between the resources of the firm and its sustained competitive advantage (Barney, 1991; Dierickx & Cool, 1989). Sometimes it is difficult to identify why a certain firm performs better than its competitors, even the firm itself may not know the exact reason. This makes it hard for competitors to imitate the success of the firm and gives it a sustained competitive advantage.

(16)

10 culture, and interpersonal relations among a firm’s managers (Barney, 1991). Contrary to causal ambiguity, the firm and its competitors understand what it is that creates a competitive advantage. However, due to the social complexity, it is very difficult to imitate unless a competitor is prepared to make major changes in, for example, their supply chain, organisational structure, or organisational culture.

There are also the factor time compression diseconomies, which states that an asset developed over several years cannot be imitated exactly within a shorter period, even with the same time and resources. As an example, in the context of R&D, time compression

diseconomies implies that investing a certain amount in R&D over several years has a higher yield compared to investing twice the amount in half the time (Dierickx & Cool, 1989). In the strategic context this means that while competitors may know why the firm has a competitive advantage and can copy the resources necessary for this advantage, they will still need time to imitate the same results.

The final attribute that a firm resource must have for it to lead to a sustainable competitive advantage, is that it must be non-substitutable. In the words of Barney (1991: 111) “there must be no strategically equivalent valuable resources that are themselves either not rare or imitable”. A firm can have a resource which is valuable, rare, and difficult to imitate. However, if the same advantage that the resource grants can be achieved through a different resource which is not rare and easy to imitate, there will be no sustained competitive advantage for the original firm nor the firms using the different resource.

(17)

11

3. FORMULATION OF HYPOTHESES

This chapter formulates the hypotheses based on the literature review. The formulated hypotheses will be tested in this meta-analysis and the result can be found in chapter 5. 3.1 Entrepreneurial orientation and export performance

To theorise the relationship between entrepreneurial orientation and export performance, it is best to start at regular performance. To establish the relationship between entrepreneurial orientation and performance, the RBV will be used as the basis. As stated in the previous section, firm resources or resource bundles provide sustained competitive advantage when they are valuable, rare, imperfectly imitable, and non-substitutable. With the sustained competitive advantage, a firm will be able to increase its performance. Thus, to argue that entrepreneurial orientation increases performance, it is necessary to establish that it possesses the qualities for sustained competitive advantage.

Barney (1991) states that a resource is valuable when it enables the firm to create or employ strategies that allow for an improvement in the effectiveness and efficiency of the firm. Based on this definition, entrepreneurial orientation can be said to be valuable as it allows the firm to develop new ideas before competitors, which results in first-mover advantages in the market. Additionally, entrepreneurial orientation encourages the firm to monitor changes in the market and take advantage of newly found opportunities (Wiklund, 1999).

Furthermore, while it can be said that many firms have an entrepreneurial orientation, it is unlikely that two firms will have the same amount of innovation, risk-taking tendency, or proactiveness. Firms with entrepreneurial orientation will vary in how much risk they are willing to take, in addition to their capacity to innovate and their proactiveness in seeking out new opportunities (Miller & Friesen, 1978). With all the possible variances between firms in terms of entrepreneurial orientation, it can be argued that it is a rare resource.

(18)

risk-12 taking, and proactiveness will also influence the level of entrepreneurial orientation. For the same reasons, entrepreneurial orientation is also difficult to substitute.

To sum up, entrepreneurial orientation can be considered valuable, rare, imperfectly imitable, and non-substitutable which makes it a sustained competitive advantage for the firm. This sustained competitive advantage allows a firm to outperform its competitors and thereby increasing its performance.

As a part of the more general performance, export performance will be increased through entrepreneurial orientation with the same reasoning. Furthermore, an argument can be made that entrepreneurial orientation has an especially positive effect on export

performance compared to regular performance. This argument is based on the fact that separately, risk-taking and innovation were already found to have a positive effect on export performance (Calantone, Kim, Schmidt, & Cavusgil, 2006; Cavusgil, 1984; Dichtl,

Koeglmayr, & Mueller, 1990). The effect on export performance is further increased by the proactiveness element. Proactiveness allows firms to be the first in a market which lets them skim the market and make large profits before their competitors can catch up (Zahra & Covin, 1995). Especially in export, which relies on the opportunities that arise frequently, being proactive can result in large benefits to a firm. Thus, the following hypothesis is formulated as:

Hypothesis 1: Entrepreneurial orientation has a positive effect on export

performance. The higher a firm’s entrepreneurial orientation, the higher its export performance.

3.2 Moderators

3.2.1 Cultural differences. One benefit of collecting studies for a meta-analysis from

(19)

13 Empirically, an argument can be made as to why the effect of entrepreneurial

orientation on (export) performance would be influenced by culture. Hofstede (1980: 25) defines culture as “the collective programming of the mind which distinguishes the members of one human group from another”, he also included systems of values in his definition of culture. The values and norms that Hofstede (1980) talks about have a large influence on human behaviour. Additionally, these values are “programmed” into people from a young age. Hofstede (1980) developed four dimensions to highlight the differences in cultures: masculinity, individualism, power distance, and uncertainty avoidance. Of special interest are uncertainty avoidance, which refers to how a society deals with the unknown, and

individualism which refers to the extent that the individual is considered more important than the group. Low uncertainty avoidance societies are more open to new ideas and changes, whereas high individualism societies emphasize individual initiative and individual achievements (Efrat, 2014). Thus, in theory, cultures with both low uncertainty avoidance and high individualism should produce firms and people who are more innovative, proactive, and more willing to take risks to achieve their goals. Based on this theoretical argument it can be assumed that some cultures are ‘better’ at using entrepreneurial orientation. Being ‘better’ at using entrepreneurial orientation results in having a higher export performance at the same level of entrepreneurial orientation compared to cultures ‘worse’ at using entrepreneurial orientation.

To support this theoretical argument, the findings of Mueller & Thomas (2001) can be used. In their study of the effects of culture on entrepreneurial orientation, these authors found that entrepreneurial orientation is more prevalent in cultures with both low uncertainty avoidance and high individualism. While more firms having an entrepreneurial orientation due to culture does not mean that these firms will be better at using it, or that their

performance will be higher. However, more firms with entrepreneurial orientation will increase the competitive intensity among these firms, ensuring that only the firms who are best at leveraging entrepreneurial orientation to increase performance will survive. Based on the arguments, the following hypothesis has been formulated:

(20)

14

3.2.2 Firm size. Aside from the influence of culture on the entrepreneurial orientation

export performance relationship, firm size also needs to be considered. The reason as to why the firm size would influence the relationship is due to the impact that the organisational environment can have on the effectiveness of entrepreneurial orientation. In larger firms, the size of the firm, as well as its complexity, require a certain amount of structure in the shape of organisational layers or specialisation. This structure is simply necessary to be able to

manage the day to day affairs of a large firm (Sirén, Hakala, Wincent, & Grichnik, 2017). However, while this type of structure in firms may allow for more efficiency, at the same time it makes a firm more reluctant to change as it will complicate the existing

communication and decision-making hierarchy. The reluctance to change is also because a large firm often possesses an established resource base. This resource base gives it a competitive advantage and any change to that resource base will likely be met with fierce resistance (Hitt, Hoskisson, Johnson, & Moesel, 1996). These issues make it harder for larger firms to use entrepreneurial orientation to increase their export performance, even if they have the same level of entrepreneurial orientation compared to smaller firms.

Another downside to a heavily structured firm is that it takes more time for

information to spread and to interpret this information in knowledge due to the layers in the organisation. Additionally, having multiple departments with specialisations makes it more difficult to identify who should receive new information and knowledge (Tsai, 2002). Finally, aside from the resistance to change and the slow spread of information, the execution speed of large firms can also be described as slow. For a larger firm to act it needs more

information and pressure as several decision-makers need to be persuaded before the firm acts (Sirén et al., 2017).

On the contrary, smaller firms do not encounter many of these issues as they do not face the same pressure larger firms do to increase efficiency by structuring the firm.

Comparatively, this makes small firms more flexible and perceptive to radical change or other types of innovation. The flexibility of smaller firms thus allows for more innovation potential. Furthermore, due to the lack or lesser extend of structure in smaller firms,

spreading information is much easier. Without specialised departments and multiple decision-makers interpreting information into knowledge and acting upon this knowledge becomes a much quicker process. Thus, smaller firms can use entrepreneurial orientation more

(21)

15 In general, the more complex and structured the organisational environment, the harder it becomes to leverage entrepreneurial orientation into increased export performance. Therefore, this hypothesis will be formulated as follows:

Hypothesis 3: The entrepreneurial orientation – export performance relationship is stronger with SME sized firms compared to larger firms.

3.3 Methodological moderators

3.3.1 Subjective and objective measures. As mentioned before while discussing

export performance, there is no consensus regarding measurement and many different types are used. This raises the question of whether the measurements used to measure export performance influence the relationship between entrepreneurial orientation and export performance. First, export performance measures can be measured objectively or subjectively, with some authors using both. Several benefits and downsides have been discussed regarding objective methods and subjective methods in the literature review. To summarise, while objective measures are impartial, it is difficult to get firms to give out this data. Furthermore, due to the differences in accounting practices, comparisons between countries are more difficult. Whereas subjective measures depend on the judgemental

assessment of the respondents which makes these type of measures more reliable and flexible (Soares & Perin, 2019). Since these subjective measures use respondent data they are better suited for comparisons across firms and countries as well, since a Likert scale is easier to compare than different financial data with a potential difference in accounting practice.

Increasingly, more studies are using both measure types. This method results in the most accurate outcome according to Shoham (1998). When both measure types are used, both the impartiality of objective data and the reliability and flexibility of subjective measures are combined. Since this thesis argues that entrepreneurial orientation has a positive effect on export performance, it is expected that a more accurate measure of export performance will result in a stronger positive effect. Thus, the following hypothesis has been formulated:

Hypothesis 4: The entrepreneurial orientation – export performance relationship is stronger when both subjective and objective measures are used to measure export performance compared to either.

3.3.2 Single measure and multiple measures. Second, researchers also vary in how

(22)

16 literature review, export performance refers to the economic outcomes of firm export

activities. The implication of this measure is to determine the success of the firm in its export activities. However, there are different ways to measure success and these different measures can tell a different story. For example, if firm x and firm y compare their export profit

margin, firm x has a higher margin and would, therefore, be more successful. However, firm y purposely lowered their prices to increase their market share which turned out to be a success. As illustrated in the example, only looking at one measure gives a distorted view of the export performance of the firm and can lead to wrong conclusions.

Measuring with multiple measures gives a more complete picture and the average of multiple measures are more suitable to compare across firms and studies. This might be why multiple measures were used more frequently in the study by Katsikeas et al. (2000).

Empirically it has also been proposed that using multiple measures of export performance captures the different aspects of export performance better (Shoham, 1998). Since this thesis argues that entrepreneurial orientation has a positive effect on export performance, it is expected that a more accurate measure of export performance will result in a stronger positive effect. Thus, the following hypothesis has been formulated:

Hypothesis 5: The entrepreneurial orientation – export performance relationship is stronger when multiple measures are used to measure export performance compared to a single measure.

3.3 Conceptual model

(23)

17

4. METHODOLOGY

This chapter explains the process of how the studies included in the meta-analysis were found and selected. Additionally, this chapter also explains how the necessary data was extracted from these studies and how the variables were measured.

4.1 Literature search and search criteria

The literature search of this thesis had the goal of collecting as many independent studies related to entrepreneurial orientation and export performance as possible. The process of finding these studies can be divided into different phases: identifying, screening, and selection (Ellis, 2010). This process of the literature search in the separate phases can be found in figure 2 as a flow diagram.

The first phase, identifying related studies, consisted of scouring databases to find studies related to entrepreneurial orientation and export performance. To find relevant studies, a keyword search was conducted in mainly Google Scholar and other databases such as JSTOR and EBSChost (Ellis, 2006; Schweiger, Stettler, Baldauf, & Zamudio, 2019). Building on a previous meta-analysis on entrepreneurial orientation (Soares & Perin, 2019) and Wales et al.’s (2013) empirical research on the same topic, a variation of keywords related to entrepreneurial orientation were used: “entrepreneurial orientation”,

“entrepreneurial proclivity”, “entrepreneurial disposition”, and “entrepreneurial posture”, which refer to the same concept. The entrepreneurial orientation keywords were used in combination with one of the traits “innovativeness”, “risk-taking”, or “pro-activeness”, in addition to using “export” to limit the results to studies related to exporting. After the

database searches, the references from the entrepreneurial orientation studies were examined to find additional studies not identified in the database searches which is an often used method in meta-analyses (Schweiger et al., 2019). Finally, only studies published in the year 2000 or more recently were accepted. The year 2000 as a cut-off point was implemented by the author to ensure that only data from the previous two decades would be collected. Collecting data of the past two decades limits the impact that time may have on the results. Additionally, limiting the scope of potential studies to two decades was necessary due to time constraints.

(24)

18 relevance. After the first 200 results, the chance of finding a related study became

significantly smaller and at that point, a new keyword search was conducted. This method was based on the author’s experience in searching through databases for this meta-analysis and used to increase the efficiency which was needed due to time constraints.

The third phase, selecting eligible studies, determined which studies were going to be included in the meta-analysis. To be eligible, the studies needed to fulfil several criteria (Ellis, 2010). First, a study needed to include entrepreneurial orientation and export performance measures. Second, a study needed to report correlation coefficients of its variables. When correlation coefficients were not reported, the studies were searched for regression coefficients, path coefficients, t-values, or p-values (Ellis, 2010). Regression coefficients and path coefficients were converted to correlation coefficients with the

(25)

19 Figure 2:

Literature search flow diagram

When the literature phases were completed, the studies to be included in the meta-analysis numbered 41 with a total sample size of 8614. There is no real minimum amount of studies required in a meta-analysis, as the available number of studies varies greatly depending on search and selection criteria (Pigott, 2012). With the search criteria of this study and based on the number of studies found to be related to entrepreneurial orientation and export

(26)

20 studies used a mixed or unspecified sample. The studies included in the meta-analysis can be found in table 1.

Table 1: List of studies used

Study Year* N Country**

Abiodun & Mahmood, 2014 2014 212 Nigeria

Acosta et al., 2018 2018 161 Mexico

Ajayi, 2016 2016 225 Nigeria

Balabanis & Katsikea, 2003 2003 82 United Kingdom

Bloemer et al., 2013 2013 134 Netherlands

Celec & Globocnik, 2017 2017 99 Slovenia

Dana et al., 2016 2016 100 Italy

De Clercq & Zhou, 2014 2014 158 China

Deligianni et al., 2016 2016 216

Dimitratos et al., 2004 2004 152 Greece

Etchebarne et al., 2010 2010 88 Chile

Fernandez-Mesa & Alegre, 2015 2015 150

Gibbons & O'Connor, 2003 2003 336 Ireland

Hernández-Perlines et al., 2016 2016 174 Spain

Holterman, 2011 2011 60 Netherlands

Hosseini et al., 2018 2018 198 Iran

Imran et al., 2018 2018 364 Pakistan

Jalali, 2012 2012 154 Iran

Jantunen et al., 2005 2005 299 Finland

Javalgi & Todd, 2011 2011 150 India

Jin & Cho, 2018 2018 470 Korea

Kaya, 2008 2008 150 United States

Knight, 2001 2001 268

Lisboa et al., 2011 2011 254 Portugal

Mac & Evangelista, 2016 2016 128 China

Mavrogiannis et al., 2008 2008 103 Greece

Monteiro et al., 2017 2017 265 Portugal

Mutlu & Aksoy, 2014 2014 33 Turkey

Okpara, 2009 2009 89 Nigeria

Patel & D'Souza, 2009 2009 270 United States

Ribau et al., 2017 2017 147 Portugal

Riviezzo & Garofano, 2017 2017 186 Italy

Roxas & Chadee, 2001 2011 175 Philippines

Rua et al., 2017 2017 247 Portugal

Samson & Mahmood, 2015 2015 230 Nigeria

Sundqvist et al., 2012 2012 783 Finland

Swoboda & Olejnik, 2016 2016 604 Germany

Taghavi Shavazi et al., 2015 2015 120 Pakistan

Yeon-Sung, 2013 2013 101 Korea

(27)

21

Zhang et al., 2012 2012 117 China

* Year refers to the year that the literature has been published **Country refers to the country in which the sample was taken

4.2 Coding process

To gather the information necessary to test the hypothesis, a coding scheme which included the necessary information was created. Of every study the following information was gathered: correlation coefficient, sample size, and the Cronbach Alpha ( ) as the

measurement reliability test. The above information was vital in conducting the meta-analysis test on the entrepreneurial orientation – export performance relationship. For the

methodological moderators the following information was gathered: firm size (“SME” or

“other”), country of study (later aggregated into the regions “Africa”, “Asia”, “Europe”,

and “North America”), export performance measure type (“objective”, “subjective”, or

“both”), amount of export performance measures in a study (“single measure” or “multiple

measures”). Additionally, as a control variable, the year of data gathered was included to check whether the differences in time had a significant effect on the sample. The year of data gathered variable was operationalised by splitting it in three parts “<2000”, “2000-2009”, and “2010-2019”.

Table 2:

List of variables/moderators and their measures Variable/Moderator Measure

Entrepreneurial Orientation

Correlation data from studies

Export Performance Correlation data from studies

Culture Measured by aggregating the country of study into the regions "Africa", "Asia", "Europe", and "North America" Firm size Measured by splitting the sample into "SME" and "Other" Measurement type Measured by splitting the sample into "Objective",

"Subjective", and "Mixed"

(28)

22 4.3 Bivariate meta-analysis

After the process of searching for usable studies and coding them to obtain all the necessary information, it was time to start calculating the necessary data for the analysis. The process of calculating the necessary data is loosely based on the method by Ellis (2010). First, the correlation coefficient of each study needed to be weighted by the sample size of the study. This step ensures that studies with low sample size and extremer results do not excessively influence the results of the meta-analysis. Afterwards, the correlation coefficients weighted by the sample sizes were adjusted for the reliability Cronbach’s alphas of these studies

(Geyskens, Krishnan, Steenkamp, & Cunha, 2009). In some of the studies, the authors did not disclose the reliability of their measures. In this case, the average reliability of all studies was used as the reliability for that study (the reliability average for this meta-analysis is 0.855 for Export Performance, and 0.793 for Entrepreneurial Orientation). Once the sample weighted and reliability corrected correlation coefficient was calculated, the significance of

correlations between variables could be determined. To determine the significance, a 95% Confidence Interval was calculated as, according to Ellis (2010), when 0 is not in the confidence interval, the correlation between variables is determined to be significant.

To account for potential publication bias, a publication bias analysis was conducted. This analysis uses a trim-and-fill procedure where a funnel plot is used which estimates how many studies are required for the funnel to be symmetrical (Suurmond, van Rhee, & Hak, 2017). To measure for publication bias, the correlation coefficients are transformed to z-scores with the Fisher Z-transformation (Suurmond et al., 2017). In the case of this meta-analysis, two studies have been imputed on the left side of the funnel plot due to publication bias. The imputed studies adjusted the observed average z-correlation from 0.41 to 0.36 and the confidence interval from 0.30/0.51 to 0.24/0.48.

To determine the amount of heterogeneity within the sample, two variables were used: The Cochran’s Q (Cochran, 1954), and the I2 (Higgins & Thompson, 2002). The

Cochran’s Q is the weighted sum of squared differences between the observed effects and the weighted average effect (Suurmond et al., 2017). Note that the Q-statistic is not a direct measure for heterogeneity as it only measures the variation around the average. The Q-statistic comes with a p-value (p(Q)) which serves as an indication of the degree of

heterogeneity. Criticism of the Cochran’s Q is that it is influenced too much by the number of studies in a meta-analysis, as meta-analyses with a low number of studies will automatically result in low Cochran’s Q scores (Higgins & Thompson, 2002). A better measure to

(29)

23 observed variance that reflects real differences in effect size (Suurmond et al., 2017). The I2

is a relative measure and expresses the heterogeneity as a percentage between 0 and 100 per cent. The I2_{calculates the variance within the sample which results from heterogeneity and}

the variance resulting from chance ( = 100% × ) (Higgins & Thompson, 2002). In general, one could state that a value of 25% is considered low, 50% medium, and 75% or more is considered high in terms of heterogeneity. When the I2_{goes above 75%, a moderator}

analysis is recommended to explain the variance (Higgins & Thompson, 2002). In the case of this meta-analysis, the I2_{value is at 95% which means that there is a significant amount of}

(30)

24

5. RESULTS

This chapter shows the results of the meta-analysis and the moderator analysis. 5.1 Bivariate results

The sample weighted average reliability adjusted correlation coefficient ( ) was calculated. The determinant of the relationship and its strength is the and its statistics, as it has been corrected for reliability. The 95% confidence interval and the z-value were used to analyse the results and draw conclusions on statistical significance. To examine the heterogeneity of the sample, both the Cochran’s Q (Q test) and the I2 will be reported. However, as the I2 is a better indicator for heterogeneity, only this statistic will be used for interpretation.

The summary results of the bivariate meta-analysis can be found in Table 3.

Hypothesis 1 predicted that entrepreneurial orientation would have a positive effect on export performance. This hypothesis was tested with 41 studies ( = 41) and a combined sample size of 8,614 ( = 8614). The bivariate result shows that there is a significant positive relationship between entrepreneurial orientation and export performance ( = 0.39; < 0.01). Furthermore, the z-value is 7.89 which is far beyond the necessary critical value of 2.576 ( / = 2.576) for a significance level of 0.01. Thus, the positive relationship between

entrepreneurial orientation and export performance has been proven with a statistical confidence of 99%.

The testing for heterogeneity within the sample shows that there is quite a large difference within the sample. The high levels of heterogeneity ($ = 892.04; %$ < 0.01; = 95.52) imply that most of the variation within the sample is not because of chance. Other factors may have a significant influence on the entrepreneurial orientation – export performance relationship. The high amount of heterogeneity in the sample calls for a moderator analysis to identify the source of the variance within the sample.

Table 3:

Bivariate meta-analysis results

Relationship Number of effects (k) Total Sample size (N) Corrected Mean* (rc) Standard Error SD(ρ) 95% Confidence

Interval Q test I²

EO > EP 41 8614 0.39*** 0.32 0.29 0.47 892.04*** 95.52

* p < 0.1 ** p < 0.05 *** p < 0.01

(31)

25 5.2 Moderator analysis

This section presents the results of the moderator analysis. Table 4 contains the results of the moderators discussed in the literature review in addition to the control variable. Figures 3 to 7 visualise the confidence intervals of each moderator and the control variable. As can be seen in Table 4, none of the moderators as well as the control variable has a significant effect on the entrepreneurial orientation – export performance relationship. This can be seen from the overlap in the confidence intervals as well as the p-value of the analysis of variance total.

Hypothesis 2 predicted that there would be a difference between regions due to the differences in culture based on the combination of the cultural dimensions Uncertainty

Avoidance and Individualism. Due to the studies and sample sizes within Africa ( = 4; 756 and North America 3; 581 being limited in number in this meta-analysis, they are unusable for interpretation since the data is not representative. Most of the studies in the sample are from Asia 13; 2530 and Europe 18; 4175 . As can be seen in figure 3, the confidence intervals of both these regions are very similar &

0.22/0.58; & 0.21/0.51 . This implies that the entrepreneurial orientation – export performance relationship is not significantly influenced by regional differences, also implying that cultural differences do not have a significant influence on the strength of this relationship 0.253 . This result was also found by Soares and Perin (2019) in their meta-analysis on entrepreneurial orientation and firm performance when they compared West to East.

Figure 3:

(32)

26 The firm size moderator distinguishes between studies who examine SMEs 17;

3455 and studies having a mixed sample 24; 5159 . While the of the SME sample studies 0.31 is somewhat lower compared to the mixed sample studies

0.44 , figure 4 shows that the confidence interval of both moderators overlaps largely (SME & 0.14/0.47; Mixed & 0.33/0.54 resulting in a non-significant difference

0.461 .

The measure types moderator consists of objective measures 7; 1203 , subjective measures 25; 6069 , and a mix of both objective and subjective measures

9; 1342 . Figure 5 does show a significant difference between objective measures 0.07; & (0.16/0.29 and subjective measures 0.44; & 0.33/0.53 . Which implies that using objective measures to measure export performance results in a weaker entrepreneurial – export performance relationship. However, the mixed measures confidence interval overlaps both the objective and the subjective confidence intervals &

0.24/0.65 , which results in an overall non-significant result 0.127 . Figure 4:

(33)

27 Within the sample of this meta-analysis, most of the studies used multiple measures

32; 7251 for export performance as opposed to a single measure 9; 1363 . The confidence interval of single measure studies & 0.32/0.51 and multiple measure studies & (0.01/0.48 largely overlap as shown in figure 6. Thus, the moderator analysis reveals that the number of measures does not have a significant influence on the entrepreneurial orientation – export performance relationship 0.513 .

Figure 5:

Measure type moderator confidence intervals

Figure 6:

(34)

28 Finally, the control variable year of data collected split the sample into studies who collected their data before 2000 3; 686 , studies who collected data between 2000 and 2009

16; 3169 , and those who collected data between 2010 and 2019 22; 4759 . The confidence intervals of the data collection time frames (& (0.12/0.58, & 0.18/0.47, and & 0.3/0.56 respectively) mostly overlap, as is shown in figure 7,

indicating that the time frame in which the data was collected did not significantly influence the results 0.295 .

Figure 7:

(35)

29

Table 4:

Moderator analysis results

k N rc SD(ρ) 95% CI LL UL Q test p(Q) I² p Region 38 8042 0.38 0.33 0.25 0.49 881.63 0.000 95.80 0.253 Africa 4 756 0.51 0.19 0.15 0.75 23.2 0.000 87.07 Asia 13 2530 0.41 0.42 0.22 0.58 396.63 0.000 96.97 Europe 18 4175 0.37 0.31 0.21 0.51 375.74 0.000 95.48 North America 3 581 0.19 0.19 -0.35 0.63 15.65 0.000 87.22 Firm size 41 8614 0.39 0.32 0.26 0.5 892.04 0.000 95.52 0.461 SME 17 3455 0.31 0.33 0.14 0.47 357.27 0.000 92.52 Mixed 24 5159 0.44 0.31 0.33 0.54 482.52 0.000 95.23 Measure type 41 8614 0.34 0.32 0.07 0.56 892.04 0.000 95.52 0.127 Objective 7 1203 0.07 0.22 -0.16 0.29 54.83 0.000 89.06 Subjective 25 6069 0.44 0.3 0.33 0.53 517.62 0.000 95.36 Mixed 9 1342 0.47 0.37 0.24 0.65 163.21 0.000 95.10 Number of measures 41 8614 0.36 0.32 0.19 0.51 892.04 0.000 95.52 0.513 Single measure 9 1363 0.25 0.36 -0.01 0.48 158.62 0.000 94.96 Multiple measures 32 7251 0.42 0.31 0.32 0.51 706.94 0.000 95.61

Year data collected 41 8614 0.35 0.32 0.25 0.45 892.04 0.000 95.52 0.295

1990-1999 3 686 0.27 0.15 -0.12 0.58 11.45 0.003 82.53

2000-2009 16 3169 0.34 0.26 0.18 0.47 200.64 0.000 95.52

(36)

30

6. DISCUSSION

This research has three main goals. First, determining the strength and direction of the relationship between entrepreneurial orientation and export performance. Second, testing the influence of culture on the relationship between entrepreneurial orientation and export performance. And third, exploring factors other than culture that may influence the

entrepreneurial orientation – export performance relationship. By investigating the data from 41 studies 8614) interesting findings were made which are relevant for both theory and practice. Additionally, throughout this research, some potential avenues for future research have been made clear. Therefore, this chapter contains the implications on a theoretical and practical level, in addition to having a paragraph dedicated to the limitations of this research and the potential for future research. Below, Table 5 shows the summary results of the hypotheses used in this thesis.

Table 5:

Hypotheses summary result

Hypothesis Result rc p H1 Significant 0.39 <0.01 H2 N.s. 0.38 0.253 H3 N.s. 0.39 0.461 H4 N.s. 0.34 0.127 H5 N.s. 0.36 0.513 N.s. = Non-significant

6.1 Theoretical contributions and implications

The first theoretical contribution made by this research is the meta-analytic estimation of the relationship between entrepreneurial orientation and export performance. Which, as far as this author is aware, is the first meta-analysis specifically conducted on the entrepreneurial orientation – export performance relationship. This research set out to discover the strength and direction of this relationship and found with H1 that it is a significant positive

relationship 0.39) with a significant level of 0.01. This result is consistent with the expectations and empirical arguments. While entrepreneurial orientation had been expected to have a positive relationship with export performance, in theory, it had not been tested on a wide scale and some individual studies showed mixed or negative results. Furthermore, the relationship found in this thesis is stronger compared to the relationship between

(37)

31 relationship of entrepreneurial orientation and export performance indicates that

entrepreneurial orientation is more effective in an export environment. This result is a novel contribution to the literature as it implies that entrepreneurial orientation plays a larger role in export performance compared to regular performance. Based on this result, several

interesting performance related questions can be raised. While entrepreneurial orientation is well suited for export based on its characteristics, are there any underlying factors that contribute to the higher performance in export compared to regular performance? Or

inversely, are there indicators within firm performance that score worse with entrepreneurial orientation, holding down the average performance? Using the resource-based view, the findings of this thesis suggest that the resource bundles commonly used in export are more compatible with entrepreneurial orientation, compared to those used for other business activities. This perspective opens up an interesting avenue to research which resources are compatible with entrepreneurial orientation. Additionally, based on the findings of this thesis, it can also be suggested that other business activities, benefiting from one or more of the entrepreneurial orientation characteristics, may see an increase in performance with entrepreneurial orientation.

The second contribution is in the moderator analysis conducted in this thesis. While none of the moderators included in the moderator analysis was significant, they still offer some valuable insights. The first moderator, which concerns H2, was the cultural moderator which separated the data based on the region it was collected in. The regions that were included were Africa, Asia, Europe, and North America. Of these regions, not enough data was collected from Africa ( = 4) and North America ( = 3) for the results of these regions to be representative. The data from Europe and Asia showed that these regions have almost identical results, which is contrary to the empirical assumption that the cultures who are better at producing firms with entrepreneurial orientation will also be better at leveraging it for increased export performance (Mueller & Thomas, 2001). This outcome implies two options, either the culture does not impact the relationship between entrepreneurial

orientation and export performance or, the way this study measured cultural influence was ineffective. The second option stems from the fact that one needs to keep in mind that cultures are rarely uniform, for example, Hofstede cultural dimensions of a country are averages and are generalised. Hofstede himself found significant intra-country differences in his sample (Williamson, 2002). Therefore, if countries are already experiencing intra-country differences that need to be averaged, this would make regions even more prone to this

(38)

32 Asia makes them very similar to the point that comparisons are pointless. Therefore, the contribution of this thesis on the influence of culture is limited to the finding that a meta-analysis is not effective at measuring the influence of culture, and another type of research is necessary.

The second moderator concerning H3 was the firm size moderator which separated the sample into SME sample studies and mixed sample studies. The third hypothesis was based on empirical evidence which suggested that SMEs would gain more benefit from entrepreneurial orientation. The moderator analysis showed an insignificant result which implies that firm size does not influence the effectiveness that entrepreneurial orientation has on export performance. This can be explained by the fact being a larger firm also has

advantages. For example, the resources of large firms allow them to be more effective in innovation, and their larger amount of resources gives them more space to take risks

(Balabanis & Katsikea, 2003). Since the impact of a failed project will not harm a larger firm as much, the risk tolerance of larger firms may be higher compared to small firms.

Additionally, the information networks of a larger firm are undoubtedly more impressive than those of a small firm which allows them to be aware of opportunities ahead of smaller firms. Based on the results of the moderator analysis it seems that the advantages of SME sized firms and those of larger firms cancel each other out, resulting in a similar strength

relationship between entrepreneurial orientation and export performance. The novelty of this finding is that the influence of firm size on the entrepreneurial orientation – export

performance relationship has not been researched before. Thus, finding that firm size does not influence the entrepreneurial orientation – export performance relationship is a

contribution. Furthermore, the insignificant result might not necessarily be the end of the road for firm related factors, as much remains unknown. For example, the difference between high-tech and low-tech firms, as well as the difference between service and manufacturing industry could potentially influence the entrepreneurial orientation – export performance relationship.

The third moderator concerning H4 was the type of measurement used by the studies included in the meta-analysis. The overall result was insignificant, however, there was a significant result between objective and subjective measures. Furthermore, instead of being in the middle of the objective and subjective subgroups, the mixed measure group finds a

(39)

33 that the confidence interval of the mixed measure study group is large and as it is a non-significant result, it can not be taken as evidence. Studies using subjective measures found a stronger relationship between entrepreneurial orientation and export performance compared to objective measures. The reason for this significant difference might be that managers subjectively overestimate their export performance compared to objective financial data. The variation in the results based on the measure types reaffirms the issues of export performance as an outcome variable addressed in the empirical part of this paper. There is a need for a consensus in terms of how to measure export performance to make it easier to compare studies. While it is not possible to reach a consensus based on this thesis alone, when judging by accuracy, mixed measure studies seem to be the way forward.

(40)

34 6.2 Practical implications

This meta-analysis has resulted in some general findings that may be useful for management in using entrepreneurial orientation to increase their export performance. While several primary research efforts had already suggested the positive influence of entrepreneurial orientation on export performance, this meta-analysis confirms the positive relationship. With this confirmation, exporters are recommended to adopt an entrepreneurial orientation through hiring managers and employees with the innovativeness, risk-taking, and proactiveness capabilities.

Second, this thesis found that the effectiveness that entrepreneurial orientation has on export performance does not change by region. Although this research cannot make

assumptions on the ease of implementing entrepreneurial orientation in firms with different geographical locations, once it has been implemented the effectiveness will be the same. For management, this means that they do not need to be located in or need to seek people from specific areas to properly reap the benefits of entrepreneurial orientation with regards to export performance. The result also implies that culture does not change the effectiveness of entrepreneurial orientation. This allows firms to implement entrepreneurial orientation regardless of which national culture the firm is established.

(41)

35 6.3 Limitations

Before recommendations are made for future research, several limitations need to be addressed and kept in mind when the findings of this meta-analysis are interpreted.

First, a meta-analysis will always be limited by the primary research included in it. This is because the limits of the primary research will become the limit of the meta-analysis. The moderator analysis included in this thesis was limited by the number of relationships examined, as well as the description of the research settings in the primary studies. As a result, the moderator analysis could only examine the relationships and the contextual and methodological moderators which were in the primary studies. For example, the primary studies included in this meta-analysis rarely used an industry-specific sample. This made it impossible to include industry as a moderator, which could explain some of the heterogeneity in the sample.

Second, all the studies used for this meta-analysis were cross-sectional, which is a type of study that analyses the data of a sample at a certain point in time. The cross-sectional data limits the ability of this thesis to make a causal reference between entrepreneurial orientation and export performance. Meta-analyses are known to be insensitive to causal directions which makes longitudinal studies, where data is repeatedly collected from the same sample, necessary to establish a causal relationship (Aguinis, Pierce, Bosco, Dalton, &

Dalton, 2011; Cooper, Hedges, & Valentine, 2009).

Third, some of the conclusions drawn from the moderator analysis are based on a small number of studies. Especially the region moderator suffers from this, as two out the four regions had four or fewer studies to represent their region, calling into question the representativity of these regions. The limited amount of studies should be kept in mind when considering the results and conclusions drawn in this thesis.

(42)

36 6.4 Future research

The results of this meta-analysis indicate several avenues suitable for future research. The first recommendation is to research the effect of entrepreneurial orientation as part of a resource-bundle. Since the empirical argument of this thesis is that entrepreneurial orientation is a firm resource, it would further the understanding of this topic if it can be determined in which interrelationships entrepreneurial orientation is most effective as part of a resource bundle. This approach would shed light on which other performance indicators, or business activities aside from export, entrepreneurial orientation may have a positive effect on. This can be done through investigating how different capabilities complement entrepreneurial orientation in the pursuit of higher export performance, or vice versa, through specific

statistical techniques like the fuzzy-set qualitative analysis (fsQCA) (Kraus, Ribeiro-Soriano, & Schüssler, 2018). Studies of this kind can explore successful bundles of capabilities that increase different types of performance. This approach goes beyond the findings of

entrepreneurial orientation in an isolated analysis, and the results might have high practical value for management. Another approach to finding other business activities that benefit from entrepreneurial orientation is to research empirically how much these activities would benefit from the characteristics of entrepreneurial orientation, after which it can be tested.

The second recommendation is regarding the influence of culture on the

entrepreneurial orientation – export performance relationship. While this thesis contributed little to the knowledge on the effect of culture, it has been deduced that a meta-analysis is not the correct way to measure this effect. To determine the effects of culture, a primary study needs to be conducted on the relationship between entrepreneurial orientation and export performance in two different countries, one with low uncertainty avoidance and high individualism (more suitable) and the other with high uncertainty avoidance and low individualism (less suitable). To cover up for a big limitation of the meta-analysis, these studies would preferably be longitudinal studies to also confirm a causal relationship. Comparing longitudinal studies of different countries may grow the understanding of the effect that culture has on the entrepreneurial orientation – export performance relationship.

Master Thesis draft for MSc IB&M