THE ENTREPRENEURIAL ORIENTATION – PERFORMANCE RELATIONSHIP: A SOUTH AFRICAN SMALL BUSINESS PERSPECTIVE

The high failure rate among small businesses in South Africa has created an urgent need to identify strategies that will improve their levels of performance. The purpose of this study was to investigate the relationship between entrepreneurially orientated strategies implemented by small businesses in the Eastern Cape and the influence of these strategies on business performance. Entrepreneurial orientated strategies were assessed in terms of the five dimensions of entrepreneurial orientation, namely innovativeness, proactiveness, competitive aggressiveness, risk-taking and autonomy. Business performance was assessed in terms of profitability and growth. A measuring instrument was developed based on valid and reliable items. Statistical techniques including descriptive statistics, Pearson’s product moment correlations and structural equation modelling, were performed on data gathered from 317 small business enterprises. The results of this study showed that the more small businesses implement the strategies of proactive innovativeness, competitive aggressiveness, and autonomy, and the less risk-taking their strategies are, the more likely their businesses are to be successful.


INTRODUCTION AND BACKGROUND TO THE STUDY
Small businesses play an important role in stimulating economic growth, innovation and competitiveness, as well as in reducing unemployment and alleviating poverty (Abedian, Blottnitz, Coovadia, Davel, Masilela and Rees, 2008:23).They contribute to 30 percent of gross domestic product and account for between 70 and 80 percent of employment in South Africa (Mahembe, 2012:13-14).Despite their importance, the failure rate of small businesses is high (Maswangayi, 2012).
Various reasons are given for this high failure rate, including a lack of finance and knowledge, poor strategic management, a lack of access to finance and poor cash flow management (Short, Payne, Brigham, Lumpkin and Broberg, 2009;Junehed and Davidson, 1998;Sexton and Bowman-Upton, 1991).Abedian et al. (2008) suggest that failure to anticipate or react to competition, new technology, or other changes in the marketplace are also common reasons why small businesses fail.This failure to react to or anticipate change occurs when the business does not act entrepreneurially (Casillas, Monero and Barbero, 2010:30).
The driving force behind the pursuit of entrepreneurial activities is an entrepreneurial orientation (Covin and Wales, 2011:677).According to Miller (1983), "entrepreneurial orientation" (EO) refers to a business that is geared towards innovation in the product-market field by carrying out risky initiatives, and which is the first to develop innovations in a proactive way in an attempt to defeat its competitors.Similarly, Lumpkin and Dess (1996) describe EO as the processes, practices and decisionmaking styles of firms that act entrepreneurially.More specifically an entrepreneurial firm is defined as one that exhibits five entrepreneurial behaviours, namely autonomy, competitive aggressiveness, innovativeness, proactiveness and risktaking (Short et al., 2009;Lumpkin and Dess, 1996;Miller, 1983).
In recent times there has been an increase in research devoted to the field of entrepreneurship as well as a growing interest in the EO of small businesses, particularly in developing countries (Chye, 2012:8;Casillas et al., 2010;Melia, Boulard and Peinado, 2007:67).However, conflicting views exist with regard to the relationship between EO and business performance.Empirical evidence supporting the view that EO has a positive influence on business performance has started to grow (Chye, 2012;Lotz and Van der Merwe, 2010;Junehed and Davidsson, 1998;Brown, 1996) and various authors (Fatoki, 2012;Gurbuz and Aykol, 2009;Wiklund, 1998;Zahra and Covin, 1995) have reported a positive relationship between EO and performance.However, Covin and Slevin (1991), as well as Sexton and Bowman-Upton (1991), contend that a lack of systematic empirical evidence exists proving that EO actually leads to improved firm performance.Similarly, Hughes and Morgan (2007:651) contend that EO sometimes, but not always, contributes to improved business performance.Hart (1992), for example, suggests possible negative consequences as a result of EO and hypothesises that entrepreneurial strategymaking is likely to lead to lower rather than higher performance because of role imbalances between top management and lower-level workers.Simmons (2010:46-48) also suggests that EO does not always result in improved performance because of the possible strain that would be placed on a firm to allocate its scarce resources to risky projects.Oswald (2008:317-333) believes that a limited understanding exists of why entrepreneurial activities vary from business to business.Furthermore, Casillas et al. (2010:29-33) assert that EO literature needs to provide more knowledge concerning the conditions under which EO as a whole is related to business performance, as well as how the dimensions of EO influence performance separately.This study attempts to address this need.

PROBLEM STATEMENT AND RESEARCH OBJECTIVES
Given the high failure rate of small business in South Africa, the need to identify strategies to improve their performance is clearly evident.According to Lotz and Van der Merwe (2010:131), EO is an important path to competitive advantage and improved performance for all types of businesses.Several studies have shown a positive relationship between the implementation of entrepreneurially orientated strategies and business performance (Chye, 2012;Fatoki, 2012;Short et al., 2009;Wang, 2008;Wiklund, 1998;Zahra and Covin, 1995).This implies that the more small businesses implement entrepreneurially orientated strategies and behave in an entrepreneurially orientated manner, the more successful they are likely to be.Very few small businesses do, however, undertake entrepreneurially orientated activities (Fairoz, Hiobumi and Tanaka, 2010:134-140).
The purpose of this study is to investigate the relationship between the EO of small businesses in the Eastern Cape and their level of business performance.EO will be assessed in terms of five dimensions, namely proactiveness, innovativeness, competitive aggressiveness, autonomy and risk-taking, while business performance will be assessed in terms of profitability, growth and goal achievement.The primary objectives of this study are to establish the level of EO of small businesses in the Eastern Cape and to establish the influence of this orientation on business performance.

ENTREPRENEURIAL ORIENTATION
EO should be differentiated from entrepreneurship.Entrepreneurship relates to new business formation, and is concerned primarily with the questions "What business do we enter?" and "How do we make the new business succeed?"(Richard, Barnett, Dwyer and Chadwick, 2004:258).EO relates to a process that concerns the "methods, practices and decision-making styles that businesses use" (Lumpkin and Dess, 1996:136).EO is taken from a strategic management perspective, and is concerned with the intentions and actions of the various stakeholders "functioning in a dynamic generative process" in a business (Lumpkin and Dess, 1996:136).Being entrepreneurially orientated encourages the involvement of "multiple management levels" in the design and execution of entrepreneurial strategies (Callaghan and Venter, 2011:37).EO is primarily a firm-level construct that is closely linked to strategic management and the strategic decision-making process (Richard et al., 2004:257;Covin and Slevin, 1991).
According to Miller (1983:771), a business's level of EO can be seen by the extent to which the enterprise innovates, takes risks and acts proactively.Miller (1983) specifically identified three dimensions, namely "innovativeness", "risk-taking", and "proactiveness" to characterise EO.His original conceptualisation of the threedimensional entrepreneurial construct received much support from Covin and Slevin (1991), and Lumpkin and Dess (1996).Covin and Slevin (1989) and Lumpkin and Dess (1996), subsequently extended and refined Miller's concept of EO.Lumpkin and Dess (1996) define EO as a firm that exhibits five entrepreneurial behaviours, namely innovativeness, proactiveness, risk-taking, competitive aggressiveness and autonomy.
Innovativeness is concerned with supporting and encouraging new ideas as well as experimentation and creativity which are likely to result in new products, services or processes (Lumpkin and Dess, 1996;Covin and Slevin, 1989;Miller, 1983).
Proactiveness is concerned with "first-mover" and other actions aimed at seeking to secure and protect market share, as well as with a forward-looking perspective reflected in actions taken in anticipation of future demand (Lumpkin and Dess, 1996;Covin and Slevin, 1989;Miller, 1983).A proactive approach implies taking the initiative in an attempt to shape the environment to gain a competitive advantage, and to anticipate competitors' movements and market needs (Lumpkin and Dess 1996).A risk-taking propensity denotes the willingness to make investments in projects that have uncertain outcomes (Lumpkin and Dess, 1996)."Competitive aggressiveness" refers to a business's tendency to "directly and intensely challenge its competitors to achieve entry or improve position to outperform industry rivals in the market place" (Lumpkin and Dess, 1996:138).Autonomy refers to "the independent action of an individual or a team in bringing forth an idea or a vision and carrying it through to completion" (Lumpkin and Dess, 1996:136).This dimension of EO is instrumental in allowing the other four dimensions to have an impact on the performance of the business; however, it is often very difficult to measure (Gurbuz and Aykol, 2009).
EO is said to consist of the five dimensions described above, which may vary independently, with each having a different influence on business performance (Simmons, 2010:16-18;Lumpkin and Dess, 1996).An enterprise can exhibit relatively high levels of one or more dimensions and, at the same time, relatively low levels of other dimensions.
Based on the above, EO can be defined as "the dimensions of entrepreneurial behaviour along which an opportunity is pursued as measured through its level of innovativeness, proactiveness, competitive aggressiveness, autonomy and risktaking" (Callaghan and Venter, 2009:31), these being the key dimensions of EO.

HYPOTHESIS DEVELOPMENT
In this study the five dimensions of EO identified in the literature (Lumpkin and Dess, 1996;Miller, 1983) serve as the independent variables, while Business performance serves as the dependent variable.It is hypothesised that the existence of these five dimensions of EO in small businesses has a positive influence on their business performance.Evidence to support the hypothesised relationships will be presented in the paragraphs that follow.

Dependent variable
According to Acs, Glaeser, Litan and Fleming (2008:11-12), consensus does not exist on appropriate measures for small business performance.Mayer-Haug, Read, Brinckmann, Dew andGrinchnik (2013:1255) assert that identifying the true nature of business performance is a challenging assignment and that "the choice of performance measures is a critical issue in research".Earlier research has mainly focused on variables for which information is easy to gather (Cooper, 1995).Several researchers advocate growth as the most important performance measure for small businesses (Wiklund and Shepherd, 2005:80;Brown, 1996;Tsai, MacMillan and Low, 1991).Wiklund and Shepherd (2005:80) contend that growth as a measure of performance may be more accurate and accessible than accounting measures of financial performance.D 'Souza and McDougal (1989) believe that sales growth is the best measure of growth.Sales growth reflects both short-and long-term changes in firms, and is easily obtainable.These authors, as well as Barkham, Gudgin, Hart and Hanvey (1996), maintain that entrepreneurs consider sales growth to be the most common indicator of good performance.Hosseini and Eskandari (2013:206) state that firm growth in terms of number of employees, as well as the number of offices is a good measure of business performance.Mayer-Haug et al. (2013:1255-1256) define performance in terms of different categories, namely the growth of the business, the number of employees, the profitability of the business, other financial performance measures as well as several qualitative measures of performance.These qualitative measures of performance include adhering to the budget, firm survival, market share, human resource management knowledge acquired and overall performance versus competitors.
An alternative view considers performance as being multi-dimensional in nature, and suggests that it is advantageous to integrate different dimensions of performance in empirical studies (Wiklund and Shepherd, 2005:80;Lumpkin and Dess, 1996;Cameron, 1978).It is possible to regard financial performance and growth performance as different aspects of performance, as each will reveal important information (Zahra, 1995).Therefore it can be inferred that taken together, growth and financial performance will give a richer description of the actual performance of a firm than each does separately.
For the purpose of this study, Business performance will be measured in terms of both growth and financial indicators and refers to the business experiencing growth in employee numbers, profits and turnover, as being profitable and financially secure, as well as achieving its planned financial goals and growth rate.Selfreported measures of performance were considered acceptable for this study as it is in line with other studies on EO (Hosseini and Eskandari, 2013;Covin and Lumpkin, 2011;Wiklund and Shepherd, 2005;

Independent variables
Several studies (Mahmood and Hanafi, 2013;Lotz and Van der Merwe, 2010;Gurbuz and Aykol, 2009;Short et al., 2009) have investigated the relationship between EO and business performance.The seminal studies (Lumpkin and Dess, 1996;Covin and Slevin, 1989;Miller, 1983) investigating EO made use of an overall measure of EO to show a relationship between EO and business performance.Raunch, Wiklund, Lumpkin, and Frese (2009:767) found 37 other studies that considered EO a "one-dimensional" construct, while 14 more recent studies have viewed the concept of EO as a multi-dimensional construct.
Viewing the concept of EO as one-dimensional means that the various dimensions of EO will affect business performance in the same way, while viewing the concept as multi-dimensional involves analysing how the different dimensions individually relate to business performance (Lim, 2009:3921;Raunch et al., 2009:764,767).The majority of studies on EO focus on the relationship between EO as an integrated one-dimensional construct and overall business performance, while few studies focus on the individual dimensions of EO and their individual influence on business performance.Lotz and Van der Merwe (2010:131) report a positive relationship between EO and business performance, and conclude that EO plays an important part in the improved performance of a business.The thrust of the argument for the positive influence of EO on business performance is related to the first-mover advantages, and the tendency to take advantage of emerging opportunities implied by EO (Fairoz et al., 2010:138).Mahmood and Hanafi (2013:86) also report that EO has a positive influence on small business performance.They conclude that EO impacts the firm's ability to gain a competitive advantage, which in turn enhances its business performance.
According to Zahra and Covin (1995), businesses with a high level of EO can target niche market segments and set the trend in the market ahead of their competitors.
These businesses monitor market changes and respond quickly to take advantage of emerging opportunities.Innovation keeps them ahead of their competitors, and as a result of this competitive advantage, financial results improve.Proactiveness gives firms the ability to present new products and/or services to the market ahead of competitors, which also gives them a competitive advantage (Gurbuz and Aykol, 2009: 321-336).Tsai et al. (1991) suggest that the influence of EO on performance is long-term, rather than short-term in nature.Proactive firms can introduce new goods and services ahead of their competitors.As first-movers they can control access to the market by dominating distribution channels.By introducing their products or services ahead of competitors, businesses can, if successful, establish industry standards.These actions should help first-movers to acquire sustained rather than temporary high performances (Zahra and Covin, 1995).Chye (2012:167-169) reports that the association between EO and business performance is "significant in strength and positive in direction".This leads the author to conclude that the EO-performance relationship is not only confirmed as being strongly significant, but is also tested as being sustainable over an extended period.Fatoki (2012:129) also observes a positive association between EO and the business performance of firms in South Africa and states that the adoption of entrepreneurially orientated activities can provide a method to reduce the weak performance and high failure rate of small businesses.
Empirical evidence exists supporting a positive relationship between entrepreneurial activities in firms and financial performance (Morris and Sexton, 1996:8), and this relationship appears to continuously strengthen over time (Zahra, 1995:242).To cope with business challenges, organisations are increasingly turning to entrepreneurship as a means of innovation, growth and strategic renewal (Bhardwaj, Agrawal and Momaya, 2007:131).Continuous innovation and an ability to compete effectively in international markets are two skills that are expected to increasingly influence performance in the 21st century's global economy (Kuratko and Welsch, 2001: 347).Chye (2012:77) argues that most literature on EO and firm performance is meant for larger enterprises or corporate entrepreneurship research.The author adds that there are "inadequate studies on the relationship between EO and the performance of small businesses", particularly research on the impact of the individual dimensions such as innovativeness, proactiveness and risk-taking on small business performance.Most studies view EO as a composite construct consisting of different independent but related dimensions (Casillas et al., 2010:28;Lumpkin and Dess, 1996).This implies that a business can show high levels of EO in some dimensions but not necessarily in all of them (Casillas et al., 2010:28) and each dimension of EO can be related to performance in a different way (Casillas et al., 2010:29).According to Casillas et al. (2010:29), this independence of the EO dimensions suggests the need to differentiate the dimensions and investigate them individually.
When examining the results of studies done on the influence of individual dimensions of EO on business performance, differences can be seen regarding which variables are found to be statistically significant by different researchers.Hughes and Morgan (2007:636) found that of the five dimensions, namely innovativeness, proactiveness, competitive aggressiveness, risk-taking and autonomy, only innovativeness and proactiveness had a significant influence on business performance.This finding led the authors to suggest that not all the EO dimensions will lead to an improvement in performance.Based on the results of their study among 200 small and medium-sized Finnish businesses, Soininen, Puumalainen, Sjögrén and Syrjä (2010:3-20) conclude that the more innovative and proactive the business is, the more successful it will become; while the more risktaking activities it adopts, the lower the profitability of the business will be, except in a time of recession.
Studies conducted by Lim (2009:3925-3926) find competitive aggressiveness to be the most significant dimension positively influencing business performance.Wang (2008:12) consider innovativeness to be the most significant positive influential dimension on perceived business performance relative to proactiveness and competitive aggressiveness.Casillas et al. (2010:38) also report innovativeness to have a significantly positive influence on the growth performance of the businesses participating in their study.Simmons's (2010) results differ from those of Casillas et al. (2010), Lim (2009), Wang (2008) and Hughes and Morgan (2007).He finds risktaking to be the most significant dimension of EO influencing business performance, while innovativeness, proactiveness and competitive aggressiveness are found to be statistically insignificant.
These differences in findings have led researchers (Hyunjoong, 2012:253;Simmons, 2010;Lim, 2009;Wang, 2008;Hughes and Morgan, 2007;Lumpkin and Dess, 1996) to believe that the influence of the individual dimensions of EO on business performance cannot be universalistic and that a multi-dimensional approach appears to be more realistic.It is for this reason that in the present study, EO is viewed as a multi-dimensional construct, and the influence of each dimension on the dependent variable is individually investigated.
Against this background, the following directional hypotheses have been formulated and will be subjected to empirical testing: H1: There is a positive relationship between the level of Innovativeness and

H2:
There is a positive relationship between the level of Proactiveness and

H3:
There is a positive relationship between the level of Risk-taking and Business performance.

H4:
There is a positive relationship between the level of Competitive aggressiveness and Business performance.

H5:
There is a positive relationship between the level of Autonomy and Business performance.

Development of the measuring instrument
A survey was used to collect the raw data on the dimensions of EO and on Business performance.A measuring instrument was developed for this purpose.The independent and dependent variables investigated in this study were operationalised using reliable and valid items sourced from previous empirical studies (Eybers, 2010;Lotz and Van der Merwe, 2010;Farrington, 2009;Short et al., 2009;Stam and Elfering, 2008;Quince and Whitaker, 2003;Lumpkin and Dess, 1996).Where necessary, the items were rephrased to make them more suitable for the present study.The measuring instrument consisted of three sections.Sections A and B contained questions aimed at obtaining demographic information about the respondents as well as information about the small business.Section C contained 46 randomly sequenced statements assessing the five dimensions of EO and Business performance.Using a five-point Likert-type scale ranging from Strongly disagree (1) to Strongly agree (5), respondents were requested to indicate the extent of their agreement with each statement.

Sampling and data collection
The population for this study consisted of small businesses in the Eastern Cape.For the purpose of this study a small business is a business which has been in operation for at least one year and does not employ more than 50 full-time employees.To date, no national database or list of small businesses in South Africa or in the Eastern Cape exists.As a result, a convenience sampling technique was employed.
Questionnaires were distributed by field workers and collected upon completion.
Each questionnaire was accompanied by a covering letter, and respondents were guaranteed that their response would be treated with confidentiality.The survey yielded 317 usable questionnaires.

Statistical analyses
The data collected from the 317 usable questionnaires were subjected to various statistical analyses using the computer programmes Statistica version 10 and AMOS version 19.An exploratory factor analysis was undertaken and Cronbach's alpha coefficients were calculated to assess the discriminant validity and reliability of the measuring instrument respectively.Descriptive statistics were calculated to summarise the sample data and correlation coefficients to establish the relationships between the factors under investigation.The hypothesised relationships were assessed by means of structural equation modelling (SEM).

Sample description
The majority of the respondents who participated in the study were male (76.97%), while 23.03% were female.Most respondents were aged between 40 and 49 years (36.28%),followed by respondents between the ages of 50 and 59 years (25.24%) and between the ages of 30 and 39 years (24.59%).The majority of respondents were either white (58.36%) or black (24.29%).Most of the respondents (66.25%) indicated that they were in possession of a post-matric qualification.
The respondents' small businesses operated predominantly in the service (45.74%) and wholesale/retail (22.71%) industries.Most respondents indicated employing between 5 and 10 employees (37.85%) or between 1 and 4 employees (25.55%) in their businesses.The majority indicated that their business had been running for 10 years or less (60.57%).

Discriminant validity and reliability results
In order to determine the construct validity of the measuring instrument used in this study, an exploratory factor analysis (EFA) was conducted before conducting the SEM analysis.In identifying the factors to extract for the model, the percentage of variance explained and the individual factor loadings were considered.
The factor structure resulting from the EFA is reported in Annexure 2. for each of the aforementioned dimensions are reported in Table 1.
According to the literature (Miller, 2011;Lumpkin and Dess, 1996;Covin and Slevin, 1989;Miller, 1983:771), Innovativeness and Proactiveness are considered two separate dimensions of EO.These constructs were measured using eight and six items respectively.However, the results of the EFA revealed that several of the items originally intended to measure these two constructs, loaded together onto one factor.The factor was named Proactive innovativeness, and hypotheses H1 and H2 were modified to reflect this change.Five of the six items measuring Proactiveness and four of the eight items measuring Innovativeness loaded onto Proactive innovativeness.Other studies (Piirala, 2012:91-92) have also reported that items measuring these two different constructs have loaded together.Eight of the nine items intended to measure Autonomy and six of the seven items intended to measure Competitive aggressiveness loaded as expected.The item RISK3 also loaded onto the Competitive aggressiveness construct.Of the seven items originally intended to measure Risk-taking, three items loaded together as expected.
For the purpose of this study Cronbach's alpha (CA) coefficients of .7 (Lehman, 2005:145;Nunnally and Bernstein, 1994) indicate a scale to be reliable.Table 1 summarises the operational definitions of the factors and provides details concerning the validity and reliability of the scales measuring these factors.From

H2:
There is a positive relationship between the level of Risk-taking and

H3:
There is a positive relationship between the level of Competitive aggressiveness and Business performance.

H4:
There is a positive relationship between the level of Autonomy and Business performance.

Descriptive statistics and correlation coefficients
Descriptive statistics relating to the sample data were calculated and are summarised in Table 2.For the sake of brevity and for discussion purposes, response categories on the 5-point Likert type scale were categorised as Disagree (1.0-2.6),Neutral (2.7-3.4) and Agree (3.5-5.0).With regard to the dependent variable Business performance, a mean score of 4.054 was observed.Most of the respondents (63.41%) agreed that their businesses had experienced growth in profits and turnover, were profitable and financially secure, and were achieving their planned financial goals and growth rate.
Proactive innovativeness returned a mean score of 4.118.The majority of the respondents (63.09%) agreed that they emphasised innovation and continuous improvement in their businesses; they regularly sought out, made changes to and introduced new processes, products and services; they encouraged creativity and experimentation, and they continuously sought and pursued new opportunities.
For Risk-taking a mean score of 3.035 was reported.Most of the respondents disagreed with (40.38%) or were neutral (34.07%) about the statements measuring Risk-taking.This means that the majority of small business owners participating in this study were not concerned with having a preference for or a willingness to commit to high-risk, high-return projects or to encourage risk-taking with new ideas.
With regard to the factor Competitive aggressiveness, a mean score of 3.868 was reported.Most of the respondents agreed (49.53%) or were neutral (42.59%) when responding to the statements measuring Competitive aggressiveness.Only half of the respondents agreed that their businesses could be described as aggressive and intensely competitive, offensive in overcoming threats posed by competitors, initiating actions to which competitors responded, striving for first-mover advantage, and being bold when faced with potential opportunities.
Autonomy returned a mean score of 3.716.Most respondents (44.80%) agreed with the statements measuring Autonomy.Slightly fewer (41.01%) respondents were neutral with regard to these statements.This finding suggests that the majority of respondents (55%) either did not allow or were neutral regarding allowing employees to work independently and without continual supervision, to make decisions or to be flexible and creative in finding solutions.
In addition to the descriptive results, Pearson's product moment correlations were used to assess the associations between the variables under investigation in this study (see Table 2).The independent variables Proactive innovativeness, Autonomy Competitive aggressiveness were all significantly and positively correlated (moderate to weak associations) with the dependent variable Business performance.
No significant correlation was reported between Risk-taking and Business performance.Furthermore, the independent variables were all positively and significantly correlated with each other.The highest correlation was observed between Proactive innovativeness and Competitive aggressiveness.

Structural equation modelling
Structural equation modelling (SEM) was the major statistical technique used to assess the hypothesised relationships between the dimensions of EO investigated in this study and Business performance.The SEM analysis produced the structural model (Figure 1) with 5 parameters, 39 estimate variances and 6 covariances as well as 38 point estimates.In order to identify the goodness-of-fit indices, the model was examined to determine whether the measurement and structural model indicated an acceptable approximation of the data.The goodness-of-fit indices showed that a CMIN/DF value of less than 3 (2.473) was reported, which suggests that there is an acceptable fit between the data and the model.The RMSEA figure of 0.068 reported was between 0.05 and 0.08 (Hair et al., 2006:748), which suggests a relatively good fit between the data and the model.Although the CFI value of 0.841 reported was lower than the recommended value of 0.9 (Hair et al., 2006:753), it was close to this value.Therefore although the model does not fit the data perfectly, based on the CMIN/DF and RMSEA fit indices it can be described as having an acceptable or reasonable fit.
Although the results of the EFA have already proved the scales measuring the factors under investigation to be valid, this analysis was confirmed by means of the CFA component of SEM.The results of the CFA produced by the SEM analysis are summarised in Annexure 3, where it can be seen that the items loaded onto the various factors as expected, and all reported factor loadings of greater than 0.5.The results of the CFA thus confirm the results of the EFA reported.The validity of the scales were thus again confirmed.

STRUCTURAL MODEL
The results of the SEM analysis are summarised in Table 3 (the parameter estimates and p-values).

DISCUSSIONS AND MANAGERIAL IMPLICATIONS
The primary objective of the study was to establish the level of EO of small businesses in the Eastern Cape, and to establish the influence of this orientation on business performance.Based on these findings, several managerial implications are suggested.
When assessing the levels of EO of small businesses in the Eastern Cape, it can be concluded that the levels of Competitive aggressiveness, Risk-taking and Autonomy are low, but higher for Proactive innovativeness.Although 63% of respondents agreed that their businesses had experienced growth in profits and turnover, were profitable and financially secure, and were achieving their planned financial goals and growth rate, the findings suggest that if they were to increase their levels of Competitive aggressiveness and Autonomy, this percentage could be higher.
Proactive innovativeness is reported as having a significant positive influence on Business performance.This finding implies that the more a business emphasises innovation and continuous improvement; regularly searches for and makes changes; introduces new processes, products and services; encourages creativity and experimentation, and continuously searches for and pursues new opportunities, the more likely it is to be successful.This finding corresponds with that reported in previous studies (Fairoz et al., 2010;Lotz and Van der Merwe, 2010;Lumpkin et al., 1996) that a positive relationship exists between both innovativeness and proactiveness, and business performance.According to Li, Tang, Tang, Marino and Zhang (2006), proactiveness and innovativeness are particularly crucial for new small businesses to succeed, because they are at the start-up phase -the phase when firms are most limited by resources.
Small business owners need to identify and understand how they can be more innovative and proactive in their particular business environments.This will require them to continuously seek out changes and opportunities in business contexts; to take calculated risks in order to take advantage of these changes and opportunities; and to audit themselves to identify their strengths and weaknesses.Therefore, small businesses need to formulate innovative ways to use those scarce resources, and proactively seek opportunities to gain a foothold in the market.Small businesses who implement Proactive innovativeness are opportunity-seeking and forward-looking, and tend to introduce new products and services ahead of their competitors, acting in anticipation of future demand.
Of the four dimensions of EO investigated in this study, Competitive aggressiveness reported the greatest significant positive influence on Business performance.In other words, the more a small business acts in an aggressive, intensely competitive manner, is offensive in overcoming threats posed by competitors, initiates actions to which competitors respond, strives for first-mover advantage and is bold when facing potential opportunities, the more likely it is to be successful.This positive influence corresponds with the findings reported in other studies (Lotz and Van der Merwe, 2010;Short et al., 2009;Gurbuz and Aykol, 2009).Lim (2009:3925-3926) also find Competitive aggressiveness to be the most significant dimension positively influencing business performance.
Small businesses should not avoid competitive encounters with other businesses, but should rather actively assume competitive stances through aggressive advertising and low-cost leadership, so as to outperform those businesses with a low level of competitive aggressiveness (Lim, 2009:3926).By adopting this approach, small businesses will be able to take advantage of emerging opportunities and to actively respond to the actions of competitors.Small businesses need to adopt an aggressive mind-set to gain a competitive advantage over their competitors.This could be achieved by implementing strategies that promote Competitive aggressiveness, such as keeping prices as low as possible and sacrificing profitability to gain an increased market share.Enterprises could also spend aggressively to obtain manufacturing capacity, in order to enable them to fully take advantage of any opportunities that may arise from serving new market needs.
In this study the relationship observed between Risk-taking and Business performance was found to be significant but negative.This finding implies that the less small businesses have a preference for and a willingness to commit to high-risk, high-return projects, and the less they encourage taking risks with new ideas, the more likely the business is to be successful.This finding concurs with that of Simmons (2010) who also found a negative relationship between risk-taking and business performance.Soininen et al. (2010:3-20) conclude that the more risk-taking activities a business adopts, the lower the profitability of the business will be, except in a time of recession.The finding also partly agrees with that of Lumpkin, Wales and Ensley (2006), who argue that pursuing "continuously high levels of risk-taking beyond the venture's early youth will become detrimental to the venture's performance."The findings of this study and the aforementioned studies, however, contradict many others who report a positive relationship between risk-taking and performance (Lotz and Van der Merwe, 2010;Short et al., 2009;Gurbuz and Aykol, 2009).
According to Lumpkin and Dess (1996), in order for a small business to be competitively aggressive and proactively innovative, it must exhibit some risk-taking behaviour.Small businesses that are bold and aggressive in pursuing opportunities and which make large resource commitments to risky projects to obtain high returns, are said to exhibit Risk-taking.Therefore, calculated risks rather than excessive risktaking should be taken by small businesses.By taking calculated risks, opportunities in the business environment can be exploited, even when their outcomes are uncertain.
Autonomy reported the lowest influence on Business performance.Despite this low influence on performance, the finding is still significant, and implies that the more a business allows its employees to make decisions, work independently and without continual supervision, and to be flexible and creative in finding solutions, the more likely the business is to be successful.This finding corresponds with the findings of several studies (Raunch et al., 2009;Lumpkin and Dess, 2005:150) who also report Autonomy as key to allowing the other dimensions of EO to have an impact on the performance of the business.The findings did, however, contradict Hughes and Morgan (2007:636), as well as Soininen et al. (2010:15), who do not find Autonomy to have a significant influence on Business performance.

It is argued that businesses cannot function entrepreneurially without facilitating
Autonomy in their structures (Coulthard, 2007:36).Small businesses should promote and encourage independent thought and allow their employees to make decisions and proceed with actions independently, without any restrictions.

CONTRIBUTIONS OF THE STUDY
This study has added to the theoretical and empirical body of entrepreneurship and EO literature by investigating the relationships between the dimensions of EO and performance in the context of the South African small business sector.The study has therefore broadened the knowledge of EO as well as the relationship between EO and performance in the small business context.Furthermore, it has addressed some of the gaps in the current literature in terms of the impact of EO on firm performance in developing countries, particularly in South Africa.
By developing a measuring instrument suitable for measuring the individual dimensions of EO in the South African context, this study has added to the discussion and research on suitable scales to measure the various dimensions of EO.Different studies have made use of different scales, resulting in different results.
This study has shown that the scales and their interpretation may be influenced by the context in which they are administered.
Most studies on the relationship between EO and performance have investigated EO as a one-dimensional construct.This study has investigated the influence of each dimension of EO individually on the performance of a small business.The study has thus added to the body of knowledge of EO as a multi-dimensional concept.Most studies focus on Miller's (1983) three dimensions, namely innovativeness, proactiveness and risk-taking, whereas this study has focused on Lumpkin and Dress's (1996) five dimensions.
In this study, the items used to measure the dimensions Proactiveness and Innovativeness were perceived by respondents as measuring the same thing.As such, these dimensions of EO could not be subjected to further testing separately, and a new variable (Proactive innovativeness) was formulated.This could imply that small businesses are unable to differentiate between being innovative and being proactive, and consider the separate constructs to mean the same thing.This finding contributes to the debate on whether Proactiveness and Innovativeness are in essence separate constructs or not.
As far as can be established, no other studies have made use of SEM to investigate the relationships between the individual dimensions of EO and performance in the South African context.This study has made a contribution to this field of study in that a more sophisticated multivariate statistical technique has been adopted than has been used to date.
Establishing which dimensions of EO influence the performance of small businesses has provided small business owners, policy-makers and researchers with greater insights into the role of entrepreneurial behaviour in small business performance.
From these insights, steps and measures can be taken by small business owners to adapt and improve their processes, practices and decision-making styles in order to improve their chances of success and long-term survival.

LIMITATIONS AND FUTURE RESEARCH
Despite the contributions of this study, several limitations should be noted.Firstly, the use of convenience sampling introduces a source of potential bias into the study.
The findings can thus not be generalised to the entire small business population.
Furthermore, this study is limited to small businesses in the Eastern Cape, and generalising the results to all South African small business may not be appropriate.
Future studies should attempt to identify a database from which probability samples can be drawn, and include small businesses throughout South Africa.
The demographic profile of the respondents was also a factor to consider in this study.The majority of the respondents were from a single ethnic group, and were therefore not representative of all ethnic groups in the country.Future studies investigating the influence of EO should attempt to obtain a more balanced representation of the different ethnic groups.Possibly a comparison could be done to observe the differences in the levels of entrepreneurial orientation among small business owners from different ethnic groups.
The responses in this study were based on the individual responses of small business owners, and were thus based on personal perceptions and on one-time self-report measures.Self-reporting does not necessarily lead to the problem of common method bias, and in many cases the bias may be so small that it does not jeopardise the validity of the results (Meade, Watson and Kroustalis, 2007).Common method bias could, however, be a factor that has influenced the results of this study.
This study investigated the entrepreneurial orientation of small businesses by applying the model of Lumpkin and Dess (1996).The level of innovativeness, proactiveness, risk-taking, competitive aggressiveness and autonomy exhibited by small businesses was investigated.Existing items found valid and reliable in previous studies, were used for this purpose.However, the items used to measure the variables Proactiveness and Innovativeness were perceived by respondents as measuring the same construct.In future studies, researchers should develop scales that more accurately measure Proactiveness and Innovativeness as individual constructs, so that these two constructs are clearly distinguishable from each other.
Several internal and external factors (both moderating and mediating variables) have been shown to influence the relationship between EO and performance (Covin and Lumpkin, 2011;Casillas, 2010:29).These moderating and mediating variables have not been considered in this study.Differences in EO dimensions could be explained by other aspects such as firm size and industry, and even environmental characteristics (Short et al., 2009:18).These characteristics have also not been accounted for in this study.Future studies should include a wider range of business types and sizes.
Despite several limitations, this study has provided insights into the EO of small businesses in the Eastern Cape, as well as the influence of implementing entrepreneurially orientated strategies on business performance.This study adds to the body of entrepreneurship knowledge by providing a greater understanding of the EO-performance relationship.

SUCC1
My small business has experienced growth in turnover in the past three years.

SUCC3
My small business has experienced growth in profits in the past three years.

SUCC4
My small business is achieving its planned growth rate.

SUCC5
My small business can be regarded as successful.

SUCC6
My small business is profitable.

SUCC7
My small business is financially secure.

SUCC8
My small business is achieving the financial goals that have been set for it.Employees in my small business do their job without continual supervision.

PROACTIVE INNOVATIVENESS
AUTO 5 Employees in my small business are allowed to be creative and try different methods to complete their job.

AUTO 7
Employees in my small business are allowed to make decisions without going through elaborate justification and approval procedures.
AUTO 8 Employees in my small business have the ability to work independently when acting on an opportunity.

AUTO 9
Employees in my small business often independently bring an opportunity from the idea stage to completion.

RISK 2
Employees in my small business are often encouraged to take calculated risks concerning new ideas.

RISK 4
My small business has a strong preference for high-risk projects (with chances of very high return).

RISK 7
My small business is willing to commit a relatively large portion of assets to pursue a high-risk high-return project.

COMP 1
In dealing with competitors, my small business typically adopts a very competitive "outdo-the-competitor" approach.COMP 2 My small business is aggressive and intensely competitive.
COMP 4 My small business effectively assumes an aggressive posture to combat industry trends that may threaten its survival or competitive position.
COMP 5 My small business assumes an offensive combative posture to overcome threats posed by competitors.COMP 6 My small business devises strategies aimed at defending its market position.
COMP 7 My small business strives to obtain the "first-mover" advantage.

RISK 3
When confronted with uncertain decisions, my small business typically adopts a bold posture in order to maximise the probability of exploiting opportunities.

Business
Five factors were extracted, explaining 46.46% of the variance in the data.Items with factor loadings of ≥ 0.5(Hair, Black, Babin, Anderson and Tatham, 2006) and those only loading onto a single factor were considered significant and thus considered for further statistical analysis.The factors extracted could be identified as the theoretical dimensions of Business performance, Proactive innovativeness, Autonomy, Risktaking and Competitive aggressiveness.The minimum and maximum factor loadings Table 1 it can been seen that the scales measuring Proactive innovativeness, Autonomy, Competitive aggressiveness, Risk-taking and Business performance all report factors loading of greater ≥ 0.5 and CA coefficients of greater than .7.Evidence of acceptable validity and reliability is thus provided.

TABLE 1 OPERATIONAL DEFINITIONS AND VALIDITY AND RELIABILITY OF RESULTS
* See Annexure 1 for a full list of items retained for the statistical analysis As a result of the EFA, the five dimensions (independent variables) of EO identified in the literature were reduced to four.The dependent variable Business performance remained unchanged.The revised directional hypotheses to test the proposed relationships are listed below: H1:There is a positive relationship between the level of Proactive innovativeness and Business performance.
In the past few years, my small business has introduced many new lines of products and/or services.INNO 8Experimentation and creativity to continuously come up with new products and/or processes is encouraged in my small business Employees in my small business are encouraged to manage their own work.