Entrepreneurial Characteristics Among Small Business Owners and Non-Business Owners in the Kumasi and Accra Areas

1. INTRODUCTION

In the last few decades, the role of small- and medium-sized enterprises in the economy became widely recognised (OECD 2017). Their importance is still growing both in developed and in developing countries. Indeed, SMEs are often seen as the answer to the problems of economic development and economic recovery (Ayandibu and Houghton 2017; Abor and Quartey 2010). SMEs are the result of the entrepreneurial effort of many people; an effort that can bring about the economic development of an area.

It is recognised that various factors are necessary for a development that is rooted in the local society and has a good perspective to be sustained in the longer run. One of the most important among them is the resourcefulness of the local people that translates into proactive entrepreneurial attitude. Such an entrepreneurial spirit is essential for starting and spreading a process of development that is vital and locally rooted (Hoselitz 1952).

The debate on the essence of entrepreneurship has engaged the attention of scholars for a long time. The literature maintains that the agent of entrepreneurship is the entrepreneur. Consequently, entrepreneurship is analysed referring to this actor: entrepreneurship is the essence of what the entrepreneur does. In order to start an entrepreneurial activity (to see entrepreneurship in action) the existence is required of entrepreneurial opportunities, which must be discovered and exploited (Shane 2002, I:6). However, the ability to discover and exploit entrepreneurial opportunities is not widespread; only a small part of a population succeeds in becoming an entrepreneur. Many factors contribute to explain this fact (Hindle 2004), among them the existence of personal characteristics that contribute to making a successful entrepreneur.

Many factors influence the existence, and the proportion, of individuals with entrepreneurial traits within a population, not least the culture of the population itself. The extent of the diffusion of these characteristics among a given population is a gauge of the entrepreneurial potential of that population. The level of entrepreneurial potential may decide whether a certain process of development can take place, whether certain policies or support measures can be successful in the longer run.

Defining who is an entrepreneur has been subject of a long debate (Ray 1993). One of the ways proposed to define an entrepreneur takes into consideration the personal characteristics of an individual and states that certain characteristics are typical of an entrepreneurial personality, concluding that individuals who strongly present those traits are potential entrepreneurs (Low and MacMillan 1988). Although a number of researchers considered the trait approach as unsatisfactory (Gartner 1988), a sizeable branch of entrepreneurial studies has focused on it.

Many authors since the beginning of the economics studies have mentioned the peculiarities of the entrepreneurs and their functions, assigning to them various degrees of importance^[1]. After a period of eclipse due to the prevalence given in the classical economics authors to the function of capital in economic development^[2], the role of the entrepreneur came again to the fore in the last part of the 19^th century (Baumol 1968; Hebert and Link 2009).

Some authors came to consider entrepreneurship as a fundamental factor of growth in a capitalistic system. For example, Pareto, while considering capital the indispensable factor for the development of the economic system, sees entrepreneurship as the agent that implements such development. Moreover, because of this centrality in the process of economic development, the entrepreneur takes on, additionally, a social function (Pareto 1906). The growing attention to the role of entrepreneurship and the entrepreneur in the economic system and, particularly, in the process of economic development reaches its climax with Schumpeter. According to him, entrepreneurship is THE essential and indispensable factor of economic development. He maintains that the growth of an economy (due to an increase of population and wealth) is not real development. Development comes with qualitative shifts due to innovation (in any aspects of the economic activity) and agent of these shifts is the entrepreneur. According to Schumpeter, the entrepreneur implements innovation, and through innovation introduces development factors in the economic system. The ability to innovate and thereby cause upheavals in the system (the “creative destruction” that moves progress and reshapes the world) is not common and the entrepreneur in a Schumpeterian sense is quite rare.

This understanding of the entrepreneur as a world-changer has caused a long-running debate (Carland et al. 1984), not yet settled, about whether any or some type only of business owner can be defined as an entrepreneur. For various reasons, not least that in the author’s first language the term “imprenditore” means any person who professionally runs an economic activity aimed at the production or trade of goods and services, here the term ‘entrepreneur’ will be used with the twin meaning of ‘business owner’.^[3]

2. THE CHARACTERISTICS OF THE ENTREPRENEUR AND INFLUENCE OF CULTURE

Following McClelland’s seminal work (McClelland 1961), numerous efforts were made to explain the term ‘entrepreneur’ on the basis of certain psychological, personal and social characteristics, maintaining that, to be able to carry out his activity, an entrepreneur needs to have some specific personal traits. Some of these traits are widespread among a population while some others are much less common. Individuals who possess the right combination of these traits are more likely than the rest of the population to become entrepreneurs and to develop their activity successfully (Low and MacMillan 1988).

Reviewing the literature, one finds long lists of widely different characteristics associated with entrepreneurs and constituting a sort of ‘entrepreneur’s profile’. It is possible to divide these characteristics into three groups: the first relates to the entrepreneur’s personality, the second to the entrepreneur’s skills and the third to the entrepreneur’s background and experience (Ray 1993). In consideration of the fact that many of the factors mentioned in those lists are different facets of the same few main characteristics, they can be consolidated into a much smaller set of profile dimensions (Kuratko and Hodgetts 1992; Zeelie et al. 2007; Kuratko 2013). Indeed, at the price of a certain degree of simplification, it is possible to whittle down the number of traits, which should characterize an individual, in order for him/her to be deemed ‘entrepreneur grade’. Limiting ourselves to traits related to the entrepreneur’s personality and basically following Timmons (Timmons 1990), one identifies: attitude towards risks; commitment and determination; creativity; independence and leadership; motivation to progress; and obsession with opportunity.

Most authors agree in naming these six traits among the distinctive characteristics of the entrepreneur. They should belong to any entrepreneur, wherever he lives and whatever his origin. The lack of one of them would cause a fatal weakness in an individual ability to act entrepreneurially, because there is a close connection between personal characteristics of an individual, his being able to become an entrepreneur and the economic success of the firm, as measured by growth and profitability (Casson 1982).

However, individuals with those characteristics are not evenly present in all groups. Many factors influence the existence, and the proportion, of individuals with entrepreneurial traits within a population. Among these factors, the culture of the population is very important, because it shapes values, norms and behaviour of people, which shape the social environment wherein people live. Indeed, this connection between the cultural background of an individual and the propensity to economic activity was seen by economics authors already a long time ago. Weber must be mentioned here, because he ascribes great relevance to cultural (religious) factors in determining the development of the society and the motivation of the entrepreneur as agent of change and development (Weber [1904] 2001). However, we can consider Sombart as the forerunner both of the modern line of enquiry that seeks to understand the entrepreneur on the basis of the entrepreneurial traits and of the connection between entrepreneurship and culture. In fact, he asserted that the presence and the manifestation of the traits he identified as typical of the entrepreneur is helped (or hindered) by the “national psychology”, which shapes the personality of the entrepreneur as that of any individual of a given nation (Sombart [1911] 2001).

The relationship between culture and entrepreneurship has received particular attention from the last decade of the 20^th century, when various studies were carried out to empirically assess it (Hayton, George, and Zahra 2002). Indeed, the culture of a population determines its predominant characteristics in terms of behaviour, aspirations, and ways to tackle problems and so on (Ronen 1986). It was observed that national culture moulds individuals’ perceptions, disposition, and behaviours from childhood, through the process of encouragements and discouragements reinforced in social institutions like workplace and family (Triandis 1989). Culture, then, can influence the type and frequency of the personal qualities found within a society. Therefore, culture influences the attitude towards entrepreneurship and entrepreneurs, affecting the social status given to them. The social environment shaped by culture may create obstacles to the diffusion of entrepreneurial activity, because an individual with entrepreneurial proclivities cannot move in a culturally unsympathetic environment as ‘a fish in its water’ but finds obstruction and hostility. This leads to the assumption that the ‘entrepreneurial’ traits are more likely to be found in certain cultures. Indeed, it has been verified that a supportive environment influences the relationship between psychological traits and entrepreneurial propensity (Okhomina 2010).

Thus, the cultural characteristics of a given nation or ethnic group help to cause a stronger or weaker diffusion of ‘grassroots entrepreneurship’ (Trevisan and Matuella 1994) within a given population. Stronger, if a group’s culture predisposes the individuals belonging to it towards starting their own independent activity when the occasion presents itself (or to create such an occasion); weaker, if a group’s culture creates obstacles to the development of entrepreneurial activities of its members. A culture can be more or less ‘entrepreneurial’ and the diffusion of entrepreneurial traits (or lack thereof) among a population could point to a higher or lower inclination towards entrepreneurship. Therefore, based on the frequency of these traits among the individuals of given cultures, it should be possible to measure the higher or lower ‘entrepreneurial potential’ of different ethnic groups (Trevisan 1997).

3. THE RESEARCH

This paper reports the findings of a research carried out in the Kumasi area of Ghana, aimed at collecting some indications to assess the diffusion of entrepreneurial traits, comparing small business owners (E/SBOs) and people not involved in business activities as owners of the business. Furthermore, a comparison is made with the findings of a previous research that was carried out in the Accra area.

To ascertain the presence of entrepreneurial traits (and, a priori, that the selected traits actually do belong to entrepreneurs/small business owners), the following hypothesis was tested:

H1 selected personal characteristics, defined as ‘entrepreneurial traits’, are actually attributes of entrepreneurs/small business owners anywhere, irrespective of their cultural background.

To ascertain that these personal characteristics are peculiar of the E/SBOs and therefore are more strongly present among them than among non-businesspeople, the following hypothesis was tested:

H2 these characteristics are typical of entrepreneurs/small business owners (E/SBOs), i.e., within a given culture E/SBOs will possess these traits to a larger extent than non-E/SBOs.

In order to test these hypotheses, we decided to look at four of the six characteristics, mentioned earlier, that are most often mentioned in the literature and in other research projects carried out in other places (Trevisan and de Jager 2000). To them a trait was added, related to the ability of looking at the long term. The conceptual framework is inclusive, since it elaborates on the Timmons model that had great importance in the theory of entrepreneurial traits (Timmons 1990).

The five selected traits are then: attitude towards risks, commitment and determination, independence and leadership, creativity and long-term orientation. Attitude towards risks is the disposition to face the possibility of loss, failure and danger, the aptitude to start a new activity without knowing the results. Commitment and determination is the firmness of individuals trying to reach their aims, their total dedication to succeed in their enterprise. Independence /internal locus of control is the ability to think and act autonomously, to be self-reliant; it is the belief that each person can guide one’s own destiny. Creativity is the capability of developing new ideas or combining old and/or new ideas to create new arrangements that can better satisfy needs. Long-term orientation is the understanding that entrepreneurial activity requires forethought (and thus planning) and that, often, decisions cannot bear immediate fruit but their effects can take some time to appear (and thus aiming at continuity).

The fieldwork was carried out in 2017 in the area of Kumasi, Ghana – in particular in the city of Kumasi, in the town of Offinso and in the rural areas in between – using a questionnaire developed in English (the same used is similar fieldwork in other countries). Since it was expected that English would not be commonly understood, making it difficult possibly to convey the correct meaning of the questions to some interviewees, local collaboration was recruited. In Ghana, the help was sought of A Ghanaian collaborator helped in this phase, who could translate the questions in some of the local languages, when English was not well understood.

In this paper, the answers to a specific group of questions will be analysed. These are dichotomous questions, aimed at ascertaining the presence of entrepreneurial traits in the respondents. For each question one answer is “entrepreneurial”, i.e., shows that the specific trait is present, and one is not. The five entrepreneurial traits mentioned above were considered and four questions for each trait asked. The more “entrepreneurial” answers, the stronger is the entrepreneurial propensity of the respondent. Each respondent can be graded on a range from zero to four for each trait and on a range from zero to 20 for entrepreneurial propensity.

The sampling was non-probabilistic and the samples are convenience samples. This, of course, makes it impossible to generalise the results reported later on in this paper. This is inevitable given the characteristic of the research as an exploratory research. Therefore, any conclusion drawn here can only be seen as a starting hypothesis for further research.

4. THE KUMASI SAMPLE

Two hundred questionnaires were collected, 100 from non-E/SBO and 100 from businesspeople, 50 in the manufacturing industries and 50 in services. 40% of the questionnaires were collected in Kumasi city, 48% in Offinso town and the remaining 12% in Offinso rural. Out of the 100 non-E/SBO questionnaires, 34 were collected in Kumasi, 42 in Offinso urban and 24 in rural areas. The distribution by gender follows that of the Kumasi district, i.e., 49 % female and 51% males. The most numerous age group is the 36-50 years old (39%), followed by 25-35 years old (24%), up to 24 years old (21%), and more than 50 years old (16%).

Forty-five of the 100 questionnaires from E/SBO were collected in Kumasi city and 45 in Offinso town. The distribution by gender is clearly skewed with 83% males and 17% females. The female E/SBO are concentrated in the service sector, where they make up 26% of the sub-sample. The most numerous age group is again the 36-50 years old (44%), followed by 25-35 years old (36%), and more than 50 years old (16%); the youngest age group (up to 24 years old) is inevitably almost absent from the business owners (4%).

The educational level of the non-E/SBO in the sample is relatively high: 32% completed secondary school, 27% have a diploma and 12% have a university degree. Only a very small proportion of the remaining 29% who has less than full secondary education did not do at least some years of further schooling beyond primary (only 2% has only primary education or lower). The educational level of businesspeople is not so high: 31% completed secondary school, 10% have a diploma and 12% have a university degree. The largest group (38%) did not finish secondary school and 9% has only primary education (2%) or only a few years of schooling.

The most common occupation of the people in the sample is unskilled worker (16%), followed by skilled worker (14%), student (11%), manager (10%), and teacher (9%). Only 3% are farmers but 16% answer other. The most common occupation of the businesspeople before they started their business was student (37%), followed by unskilled worker, unemployed and entrepreneur at the same level (9%). Teacher, skilled worker and self-employed artisans (all with 4%) are also among the previous occupations. A sizeable proportion (10%) said other.

5. ENTREPRENEURIAL TRAITS

In following pages, the two assumptions previously mentioned shall be discussed.

i) selected personal characteristics, defined as ‘entrepreneurial traits’, are actually attributes of entrepreneurs/small business owners anywhere, irrespective of their cultural background, and

ii) these characteristics are typical of E/SBOs, i.e., within a given culture E/SBOs will possess these traits to a larger extent than non-E/SBOs.

To test assumption

i) selected personal characteristics, defined as ‘entrepreneurial traits’, are actually attributes of entrepreneurs/small business owners anywhere, irrespective of their cultural background, the average score of the two E/SBO groups for each of the selected entrepreneurial traits and for the cumulative entrepreneurial propensity shall be presented and discussed as well as the distribution of the “entrepreneurial” answers of each respondent. Considering that each respondent could score between 0 and 4 in each trait, a trait will be considered strongly present in the group if the average score is from 3 upwards (meaning that the group as a whole has given on average at least three “entrepreneurial” answers out of four). If the average score is from 2 downwards (meaning two or less “entrepreneurial” answers) the trait will be considered weakly present or absent; if the average score is between 2 and 3, the trait will be considered present but not strongly.

The same will be presented for the non-E/SBOs and the comparison between them and the former will allow the testing of assumption

ii) these characteristics are typical of E/SBOs, i.e., within a given culture E/SBOs will possess these traits to a larger extent than non-E/SBOs.

In table 3, the average score for entrepreneurial traits, as well as the median, are presented. Grouping together the E/SBOs, it appears that one trait, long-term orientation, is strongly present (mean 3.25, median 4); three are present but not strongly (one, commitment, scores just below 3), while one, creativity, is weakly present. Entrepreneurial propensity is particularly (and unexpectedly) low, at a normalised mean of 2.48. (the discussion about entrepreneurial propensity shall come later).

When splitting the E/SBOs in the two sub-groups, it appears that among manufacturers, long term orientation is definitely strongly present, and commitment is also strongly present (although not much above the threshold: mean 3.06; median 3). Attitude to risk and independence are present but not too strongly (mean 2.77 and 2.27, respectively; median 3 for the former and 2 for the latter). However, the trait creativity (mean 1.04, median 1) is almost absent among them. Those who operate in the service sector, show a partially different pattern: while, as is the case of manufacturers, long-term orientation is strongly present, for them also attitude to risk, scores just enough to be considered strongly present. Commitment and independence are present but not too strongly (mean 2.82 and 2.20, respectively; median 3 for the former and 2 for the latter). Creativity is barely present (mean 1.28, median 1).

In general, industrial E/SBOs show a slightly stronger presence of the traits than the traders, the exception being attitude to risk (also creativity is lower but it is so low for both that it makes no difference). At a first reading, then, industrial E/SBOs show clearly the presence of four traits (two of which strongly), while the fifth one is almost absent. The services E/SBOs show, although less intensely, the presence of four traits (again two, but not the same two, strongly enough), while the fifth one (the same as in manufacturers) is very weakly present.

The low score of independence in both sub-samples is particularly relevant, since in the literature it is emphasised that being able to take decisions while others remain sceptical is an important characteristic of entrepreneurs. It can be said that the low score of independence among the E/SBOs sample in Kumasi confirms the findings in other African countries – such as Mozambique and Senegal, as well as of a sample in Accra – where again independence scored rather low (Trevisan 2016). It is then possible to surmise that some common cultural trait might make it more difficult for African people, including the entrepreneurs, to show a large degree of independence. The much lower score of attitude to risk among manufacturers is an unexpected result. Unexpected, too, is the practical absence of creativity in both samples (and in the non-E/SBOs, too).

Assumption i), for the time being, is not fully proved. The entrepreneurial traits are present in the businesspeople of the sample, but not as strongly as it would be expected in consideration of the fact that they are a defining factor of entrepreneurship. Contrary to what was found in other samples, the differences between manufacturing and services businesspeople, although present, are not particularly strong, at least at this level of analysis.

When confronting the score of non-E/SBOs with that of the businesspeople, it is possible to see a difference between the two groups. Their score is lower in all the traits bar creativity, which in any case scores so low that has to be considered as almost absent. Moreover, for long term they, too, show a strong presence of the trait. The differences do not appear very big, however. Therefore, for the time being, assumption ii) appears proved but not conclusively.

To look at the issue in a different way and to give a concise view of the intensity of entrepreneurial traits among the respondents, an intensity index is used (Trevisan 2009), built on the proportion of “entrepreneurial” answers given to the various questions related to each trait. The index ranges from 0 to 1, where 0 means that no “entrepreneurial” answer was given to any of the four questions and 1 that all the answers of all respondents were “entrepreneurial”. The actual score will always be somewhere in between and the interpretation given to the index is the following: an index higher than 0.750 means the trait is strongly present; from 0.501 to 0.750 that trait is present; from 0.251 to 0.500 the trait is weakly present, 0.250 and below it is absent.

The intensity index confirms that four traits are presents among the E/SBOs (see Table 4). However, only one, long-term, is strongly present among them, when they are grouped together. It is present in both the sub-groups of businesspeople together with another one (different in the two). In both cases, independence is the trait less strongly present. Creativity is weakly present in the services and absent in the manufacturers. Services E/SBOs show a less strong presence of some traits and stronger in attitude to risk and creativity. The service E/SBOs appear to have a marginally weaker presence of the traits, but in comparison with previous research, they are closer to manufacturers than expected.

A χ² test on the figures underlying this index shows that the differences between manufacturers and traders are significant at 99% level for attitude to risk, long term, and creativity, and not significant for the other two traits. Therefore, the two sub-groups are rather different, but it is less clear than in other research that they should be considered as two populations apart. In this analysis, too, assumption i) is proved but not yet satisfactorily: the traits are present but with lower intensity than expected. Only long term is strongly present in both sub-samples, while attitude to risk and commitment are strongly present alternatively in one and not in the other of the sub-groups.

Among non-businesspeople the traits are present, with long term in the “strongly present” range and only creativity in the “weakly present” range. For all traits, excluding creativity, non-E/SBOs show a lower intensity than the businesspeople do. The χ² test shows that the differences between E/SBOs and non- ESBOs are significant at the 99% level for commitment and at the 95% level for attitude to risk. Therefore, the difference between these two groups is less marked than the difference between the two groups of businesspeople.

Thus, assumption ii) cannot be proved, being validated only for two out of the five traits analysed. The case is further muddied by the diverse pattern of manufacturers and traders, where the differences of the former with non-E/SBOs are significant for four of the traits (three at the 99% and one at the 95% level) and those of the latter significant at 99% level for two traits.

The last analysis presented is related to the entrepreneurial propensity, i.e., the presence of the entrepreneurial traits as a whole set and not taken separately. Again, the average score and the intensity index are taken into consideration (see table 5). This index is built on the basis of the number of “entrepreneurial” answers to the 20 dichotomous questions and from it measures the entrepreneurial propensity of respondents. A score from 0 to 5 would mean lack of any entrepreneurial propensity; from 6 to 10, a low propensity; between 11 and 14, a moderate entrepreneurial propensity and from 15 upwards a high entrepreneurial propensity.

The average score of the E/SBOs is not high, and the intensity index shows that the propensity is present but not strongly. Again, assumption i) is not satisfactorily proved: the entrepreneurial traits are present in the businesspeople of the sample but not as strongly as would be expected, considering that they are thought of as a peculiar characteristic of entrepreneurs. A surprising result (in the light of previous research) is that manufacturing and services E/SBOs score almost identically and even more surprising, the score is (slightly) higher for the latter.

Assumption ii) appears to be proved. When the figures are shown for the E/SBOs grouped together, the difference between E/SBOs and non-E/SBOs now appears clearly (and, again, it is significant at the 99% level). Both average score and intensity index of the former are clearly higher, although, again, not as much as could be expected. When looking at the two business groups separately, both show higher average score and intensity index than the non-E/SBOs and the difference is significant at the 99% level. The difference in the intensity index between the two sub-groups of E/SBOs is small and not significant.

Two additional indices were developed to analyse entrepreneurial propensity: the index of entrepreneurial propensity, for low and high propensity. The index EP1 gives the proportion of respondents that show a high degree of entrepreneurial propensity, while the index EP2 gives the proportion of respondents that show high and moderate entrepreneurial propensity (see table 6).

Analysing the EP index, it appears that the businesspeople in this Kumasi sample are moderately entrepreneurial. Less than one fifth of them show a high entrepreneurial propensity. When putting together high and moderate entrepreneurial propensity, however, more than four fifths of them show it (EP2 index = 82.0). While for EP2 the two E/SBOs sub-groups are very similar, the difference between manufacturing and services E/SBOs shows very clearly in the EP1 index, which in the latter is double that in the former. From this analysis, it is possible to conclude that assumption i) is again verified but not fully satisfactorily. There is a clear presence of entrepreneurial propensity (which shows that the traits are present among E/SBOs) but it is not very strong. The EP2 index is good, but the EP1 is poor, showing that most of the E/SBOs have only a moderate entrepreneurial propensity (and therefore a moderate strength in the presence of the entrepreneurial traits).

From this analysis, assumption ii), that E/SBOs possess entrepreneurial traits to a larger extent than non-E/SBOs is verified: EP1 is 18.0 and 7.0 respectively (but 18.0 is really low for E/SBOs) and EP2 is 82.0 and 69.0 (but 69.0 for EP2 of non-SBOs is quite high).

6. THE ACCRA SAMPLE

To put the findings from the Kumasi sample into perspective, a comparison is made with the findings of a similar research carried out some ten years previously in the greater Accra area. In that case, the entrepreneurial traits that were taken into consideration were four because creativity was not included. However, excluding the four dichotomous questions related to creativity, the questionnaires were identical.

As the sample is different and the time lapse is large, there is no pretension of significance in this comparison; however, presenting these findings can give an idea of how differences in the intensity of entrepreneurial traits can be noticed in different areas of the same country. The fieldwork was carried out in the metropolitan area of Accra, Ghana, using the same questionnaire in English. In this case, too, the researcher sought the help of a Ghanaian collaborator, who could translate the questions in some of the local languages, when it appeared that the correct meaning of the questions was not clearly transmitted to the interviewees.

The valid questionnaires collected were 230, of which 102 (44.3%) from general population (non-E/SBOs) and 128 (55.7%) from business owners (E/SBOs). This latter group was composed industrial business owners (88 questionnaires, 38.3% of the total) and commercial business owners (40, 17.4%).

Although the majority of the business owner were males (58%), this however implies a rather high proportion of female business owners. This can be explained mostly by the structure of the business of the interviewees: most businesses, particularly in the textile and garment industry are rather small and can be considered not more than largish dressmaker’s workshops, typically owned by women. In table 7, the average score for entrepreneurial traits, as well as the median, are presented. Grouping together the E/SBOs, it appears that three of the traits are present but not strongly (only one, attitude to risk, scores, just, above 3), while one, independence, is weakly present.

When splitting the E/SBOs in the two sub-groups, it appears that among manufacturers, attitude to risk is definitely strongly present (mean 3.52; median 4). Commitment and determination and long-term orientation are definitely present, albeit not too strongly (mean 2.64 and 2.74, respectively; median 3 for both). Independence (mean 1.62, median 2), however is barely present among. The traders show a pattern only partially different: while, as is the case of manufacturers, commitment and long-term orientation are definitely, though not too strongly, present (with a slightly lower score and clearly lower median), the weakly present trait is attitude to risk (mean 1.90 median 2). Conversely, independence (mean 2.25, median 2.5) is not strong among them but it is clearly stronger than among manufacturers.

In general, industrial E/SBOs show a stronger presence of the traits than the traders, the exception being independence. At a first reading, then, industrial E/SBOs of this sample show clearly the presence of three traits (one of which very strong), while the fourth one is quite weak. The commercial E/SBOs show, although not too clearly, the presence of three traits, while the other one (different from the one in manufacturers) is just below the limit, mentioned earlier, of the weak presence.

The low score of independence in both sub-samples is particularly important. As mentioned above, this is a very important trait for entrepreneurs. The low score in this trait underlines a common feature with the Kumasi sample (as well as with other African samples). The much lower score of attitude to risk among traders is also an unexpected result. It also is in full contrast with the findings in the Kumasi area, where the traders scored significantly higher than the manufacturers for this trait.

When looking at the score of non-E/SBOs, it is possible to see a clear difference in comparison with the E/SBOs. Their score is clearly lower in three of the four traits (in two of them their score is in the ‘weakly present or absent’ range or minimally above it), the odd one out being independence, although even this trait is weakly present (mean 2,00; median 2). In this case, too, despite offering a clearer picture than in Kumasi, in the first overview the two assumptions are not conclusively proved. Therefore, also for the Accra sample further analysis was carried out, developing first the intensity index and thereafter the two indices of Entrepreneurial Propensity.

The intensity index (presented in Table 8) confirms that all four traits are presents among the E/SBOs. However, only one, attitude to risk, is strongly present among them and then only among the manufacturers. The other three traits are also present among manufacturers, although one of them, independence, in the weakly present. Among the traders, three traits are present, albeit with lesser intensity than among manufacturers, while for one, long-term, the index is in the “weakly present” range. A χ² test on the figures underlying this index shows that the differences between manufacturers and traders are significant at 99% level for attitude to risk and independence and not significant for the other two traits.

In this analysis, too, assumption i) is not fully proved: the traits are present but with lower intensity than expected; only the industrial E/SBOs show a clear presence of one of them (attitude to risk), which is only weakly present in the traders. On the other hand, the trait independence is weakly present among manufacturers. And in both cases the trait that is weakly present among one group of business owners is, in this group, weaker than among non-businesspeople.

Among non-businesspeople the traits are present, although commitment is in the weakly present range and independence and long-term are borderline cases. For all traits, excluding independence, non-E/SBOs show a lower intensity than the businesspeople. The significance of this difference between the two groups is borne out by the χ² test, which shows that the difference is significant at the 99% level for all three traits. However, in the case of attitude to risk, while the trait is significantly weaker in the non-E/SBOs than in businesspeople, when the traders are isolated, they show a significantly weaker presence than the non-E/SBOs. Therefore, although evidence increases that assumption ii) can be proved, being validated for three out of the four traits analysed, the case is weakened by the diverse pattern of manufacturers and traders, each showing weaker presence of one of the traits (a different one) than the non-E/SBOs.

The next step of the analysis is the assessment of the entrepreneurial propensity (see Table 9). The intensity index shows that the propensity is present but not strongly, because the score for the e/SBO as a group is lower range of presence (and (for the traders it is rather close to the lower limit of this range). Assumption i) is not satisfactorily proved: the entrepreneurial traits are present in the businesspeople of the sample but not as strongly as would be expected, considering that they are thought of as a peculiar characteristic of entrepreneurs. Moreover, the difference between industrial and commercial businesspeople is so large (and significant at the 99% level) that the two must be considered as something clearly different and cannot be lumped together in the same group of businesspeople.

When looking at the difference between E/SBOs and non-E/SBOs, assumption ii) appears to be proved. Looking at the score for the E/SBOs grouped together, the difference between E/SBOs and non-E/SBOs now appears clearly (and, again, it is significant at the 99% level). Both average score and intensity index of the former are clearly higher, although not as much as could be expected. When looking at the two business groups separately, both show higher average score and intensity index than the non-E/SBOs. In particular, the difference between the latter and the industrial E/SBOs is clearly large. On the contrary, the difference in the intensity index between them and the traders is not significant, again pointing to the conclusion that in the Accra sample, traders and manufacturers are two different populations.

According to the EP index, the businesspeople in this Accra sample are moderately entrepreneurial (see Table 10). Less than 30% of them show a high entrepreneurial propensity and although when putting together high and moderate entrepreneurial propensity almost three quarters of them show it (EP2 index = 74.2), more than one quarter of the businesspeople interviewed show low entrepreneurial propensity or even a lack of it. The difference between manufacturers and traders becomes glaring: the EP1 for the former is 81.9, for the latter is 5.0. For the EP2, which tends to smooth the distance, the difference is also very high (EP2 is 81.9 and 57.5 respectively). From this analysis, it is possible to conclude that assumption i) is verified for the industrial E/SBOs but not for the commercial ones.

At a first reading, assumption ii), that E/SBOs possess entrepreneurial traits to a larger extent than non-E/SBOs is clearly and definitely verified: EP1 is 28.9 and 3.9 respectively and EP2 is 74.2 vs. 46. However, the huge difference between manufacturers and traders influences also this aspect. The difference between industrial business owners and non-business people is so high to warrant a definitely positive answer to assumption ii) (EP1: 39.8 and 3.9; EP2 81.9 and 46, respectively); but the difference between traders and non-business people is quite small (EP1 5 vs 3.9; EP2 57.5 vs 46) and not significant.

7. CONCLUSION

From the findings presented it is possible to assert that in both samples there is a difference between E/SBOs and non-E/SBOs that is statistically significant. The entrepreneurial traits taken into consideration are present in the former and basically absent in the latter. However, the two samples differ in many ways and this difference clouds the outcome. The sample in Accra shows more clearly the difference between the two groups, but also shows more strongly the difference between manufacturers and traders. Indeed, in this sample the real difference is between manufacturers and non-E/SBOs, while the difference between the latter and the trades, although present is not statistically significant. The Kumasi sample shows a less strong (although still statistically significant) difference between E/SBOs and non-E/SBOs. It also shows very little difference between manufacturers and traders. There might be various explanations for this difference, that require a more detailed analysis of the influence of some demographic variables to ascertain, for example, whether the reduced difference between E/SBOs and non-E/SBOs shown in the Kumasi sample might be ascribed to a higher degree of schooling among the non-E/SBOs interviewed. In the end, one of the assumptions tested assumption ii), that E/SBOs possess entrepreneurial traits to a larger extent than non-E/SBOs is verified, while the other, assumption i), that the entrepreneurial traits taken into consideration are indeed traits common to all E/SBOs is verified only for a few of them.