Entrepreneurial intention is a catalyst for entrepreneurial behaviour (Kautonen, Van Gelderen & Fink 2015:655–647). Malebana (2012:35) argues that entrepreneurial behaviour, which includes identifying gaps in the market and starting businesses, is intentional and a planned behaviour that can be predicted by intention.

The current study used the theory of planned behaviour (TPB) model to discuss the phenomenon of entrepreneurial intention among commerce matric students. According to Jain, Khan and Mishra (2017:6) and Oni and Mavuyangwa (2019:7), the TPB model illustrates the dependence of intention on attitudes, subjective norms and perceived behavioural control. In addition, the intention predicts the willingness to update behaviour, which in turn is influenced by the individual disposition regarding behaviour, subjective norms and perceived behavioural control (Alonso & Krajsic 2015:97; Phau et al. 2014:248). Previous studies that tested and subsequently validated the TPB model include Otuya et al. (2013:132), Muofhe and Du Toit (2011:9), Mueller (2011:55–74), Iakovleva, Kolvereid and Stephan (2011:353–370), Engle et al. (2010:35–57), Gird and Bagraim (2008:711–724) and Wiklund and Shepherd (2003:1919–1941). The TPB model is therefore suitable for the current study, because attitudes, subjective norms and perceived behavioural control are connected with entrepreneurial intentions and behaviours of matric commerce students.

Many definitions of entrepreneurial intention have been suggested by various researchers since the 17th century. For example, Thompson (2009:669) defined entrepreneurial intention as the self-declared beliefs of individuals that they intend to start new business ventures in the future. This definition is supported by Davey, Plewa and Struwig (2011:337), who defined entrepreneurial intention as the intention to start a business at some point in life with no specific time constraints, and by Adam (2016:24), who defined entrepreneurial intention as the desire to strive for entrepreneurial action in the future. For Bird (1988:442–453) and Lin et al. (2013:1), entrepreneurial intention is defined as the key element in understanding the process of starting a new company.

Entrepreneurial intentions are influenced by factors such as scarce employment opportunities, entrepreneurial education, mindset and skills, according to Oni and Mavuyangwa (2019:7). Malebana and Swanepoel (2015:1–26) discovered that demographic factors such as gender, work experience, previous business start-up experience and entrepreneurial family background all influenced entrepreneurial intention. Sesen (2013:624) discovered that locus of control and entrepreneurial self-efficacy influence entrepreneurial intention. According to Nasip et al. (2017:825), entrepreneurial intention is positively influenced by innovativeness, self-confidence, willingness to take risks and ambiguity.

Seth (2020:38) observed that there is a lack of a precise definition of entrepreneurship education. Jones and English (2004), cited in Gautam (2015:22), defined entrepreneurship education as a process by which a person with commercial opportunities is empowered to recognise the know-how and the insights and skills to use them. Hood and Young (1993), cited in Seth (2020:38), defined entrepreneurship education as an educational process for profitable business start-ups that contributes to economic development. Fayolle et al. (2006:702) defined entrepreneurship education as any educational process or pedagogical course that aims to impart entrepreneurial attitudes and know-how.

In the early research work of Lüthje and Franke (2003:135–147), a positive relationship was found between entrepreneurship education and entrepreneurial intention. Wu et al. (2022:1) discovered a statistically significant relationship between entrepreneurship education and entrepreneurial intention in their recent study. Further evidence can be found in the study of Mahlaole and Malebana (2021:10), which found that entrepreneurship education is positively related to increased entrepreneurial intention. Nabi et al. (2018:452–467) discovered that students who are exposed to entrepreneurship education are more inspired to pursue entrepreneurship than their counterparts who are not exposed to entrepreneurship education.

Despite this evidence, there is a scarcity of literature on the status of matric commerce students in rural areas. Because of a number of important circumstances unique to this group, the current study identifies this as a gap. Rural schools are distinguished by a slew of negative factors that undermine educational quality (Du Plessis 2014:1109). The negative factors range from a lack of basic infrastructure to poor socio-economic conditions and information and communications technology (Du Plessis 2014:1109). Rural schools also lack proper governance structures as well as technologically advanced learning and teaching tools. As a result, the study aims to investigate students’ entrepreneurial intentions in rural areas.

In this study, a deductive, positivistic research design was chosen because the deductive approach makes it possible to explain causal relationships between concepts and variables, to measure concepts quantitatively and to generalise research results to a certain extent (Van Wyk 2012:15). Regarding the survey design, a cross-sectional design was used in this study in order to investigate possible relationships between the study variables without experimental manipulation (Munyanyi, Munongo & Pooe 2021:10). In addition, a deductive, positivistic research design was found to be appropriate for this study because this study examines the influence of independent variables on a dependent variable and is quantitative in nature. The study examined factors influencing the entrepreneurial intentions of business math students in rural areas of KwaZulu-Natal, South Africa.

The population for this study included matric commerce students from rural schools in 11 districts of the KwaZulu-Natal province of South Africa. The target population constituted 11 200 people on average. The sample size was calculated using the Raosoft online sample calculator with a 5% error rate, a 95% confidence level and a 50% response distribution. A sample size of 372 students was recommended as the bare minimum. Table 1 shows the demographic characteristics of the participants.

The instrument for data collection was a questionnaire. The questionnaire consisted of 29 questions divided into six sections: Section A, biographical and company data (three questions); Section B, entrepreneurial intention (eight questions); Section C, attitude towards entrepreneurship (six questions); Section D, solicited information on innovativeness (four questions); Section E, solicited information on entrepreneurial self-efficacy (15 questions); and Section F, entrepreneurship education (three questions). Sections A and B used a nominal scale (‘yes or no’ answer types), whereas sections B–F used a five-point Likert scale (1 = strongly disagree to 5 = strongly agree). All questions in sections B, C, E and F were adapted from the Lin and Chen (2006, 2009) Entrepreneurial Intention Questionnaire, which Malebana (2012) used, with minor changes. The questions in Section D were adopted with minor changes from the Entrepreneurial Intention Questionnaire developed by McMillan and Schumacher (2001). The use of these questions is justified because they have been validated in other studies, increasing the reliability of the designed questionnaire. Furthermore, pilot testing was conducted at KwaPata Secondary School, where seven students were randomly selected to participate in the pilot testing.

All seven participants returned the questionnaire. Data obtained from the pilot testing revealed that respondents understood the questions and the questions were free of ambiguity. Table 2 depicts the characteristics of the sample.

A total of 433 matric commerce students registered for the 2021 academic year were the respondents. Of those polled, 52.9% were women, while 47.1% were men. The majority of respondents (55.9%) were 18 years old, 25.9% were 19 years old, 17.1% were 17 years old and just over 1% were 20 years old. The majority of respondents (98.4%) were black students, with mixed race and white students accounting for the remaining 1.4%. One per cent of respondents did not disclose their ethnicity.

The main method of collecting data was face to face. This method was adopted because it guarantees the greatest response rate to the researcher. In addition, literature offers several advantages associated with face-to-face administration of the questionnaire. For example, when in doubt about a question, participants could seek explanation or clarity from the researcher. As mentioned in the data collection process, the researchers explained the purpose of the research to the respondents and ensured they understood the research and signed the informed consent form before completing the questionnaire.

The researchers applied to the Department of Education for permission to carry out this study at the selected schools. The researchers asked the school principals of the respective schools for permission to distribute the questionnaires. After permission was given, the researchers made appointments to administer the questionnaire. This was done to ensure that school authorities were aware of the data collection days so that classes were disrupted as little as possible. In addition, it was also necessary to set dates, as the researchers sometimes also had to rely on physical and logistical support from the school authorities, for example, distribution of the questionnaires to the learners and collection from them once completed. Before the research participants started completing the questionnaire, the researchers went through the questionnaire with them to clarify each question on the questionnaire so that in the end everyone had the same understanding of the questions.

The collected data were cleaned and coded using the Statistical Package for the Social Sciences (SPSS) version 22, and structural equations were performed using the Analysis of Moment Structures (AMOS) version 22. The data were subjected to descriptive statistics, confirmatory factor analysis (CFA) and structural equation modelling, with the results used to answer the research questions. Pearson’s correlation coefficient was used to determine the strength of the relationships between the study variables.

Ethical approval for the study and distribution of the research questionnaire was granted by the Durban University of Technology’s Ethics Committee (reference number: IREC 136/18).

To establish the reliability and validity of the research instruments, Cronbach’s alpha, composite reliability (CR) and average variance extracted (AVE) were assessed (Fornell & Larcker 1981; Hair et al. 2010). Instrument reliability indicates the internal consistency of a set of questions designed to measure a specific variable. Good reliability is assured with alpha and CR values of 0.7 and above and an AVE of 0.5 or higher (Hair et al. 2010). The reliability results of this study are in line with Hair et al.’s (2010) recommendations as shown in Table 3. Convergent validity measures how well a set of designed questions measures or reflects the specific variable that it was meant to measure (Hair et al. 2010). Factor loadings of 0.5 or higher, CR of 0.7 or higher and AVE of 0.5 or higher depict adequate convergent validity and internal consistency (Fornell & Larcker 1981).

Discriminant validity distinguishes a study variable from other related variables using statistical information (such as the square root of the AVE) to depict the unique nature of such a latent variable. Discriminant validity is achieved if the value of the square root of the AVE for every latent variable in a study exceeds the value of the correlations with other constructs (Hair et al. 2010). Table 4 shows the values obtained for this study, thus establishing convergent validity and discriminant validity.

The following indices were used to assess the model’s goodness of fit: a statistically nonsignificant chi-square–degree of freedom (χ²/[df]) value ≤ 3, the comparative fit index (CFI) value ≥ 0.9, the Tucker–Lewis Index (TLI) value ≥ 0.9, the incremental index of fit (IFI) value ≥ 0.9 and the root mean square error of approximation (RMSEA) value ≤ 0.08 (Hair et al. 2010). Table 5 with initial results of the CFA shows some of the indices were adequate. To ensure all the indices meet the threshold values, modification indices (MI) as suggested by AMOS were applied. Three error terms were correlated on the entrepreneurship intention variable, and five error terms were correlated on the entrepreneurship self-esteem variable. Subsequent evaluation (Table 5) revealed a good model fit based on the values of all indices.

A structural model path analysis was conducted to evaluate the hypothesised relationships among the constructs of interest. The goodness-of-fit of the structural model (Table 6) was assessed with the following indices: chi-square goodness-of-fit, RMSEA), CFI, TLI and SRMR.

At least three-quarters of those polled felt that the curriculum had influenced their desire to start a business positively. Furthermore, 70% were ready to start a business after finishing their studies and believed that their professional goal was to be an entrepreneur. The highest rated mean score indicated that the curriculum had positively contributed to their interest in starting a business, which was followed by a strong intention to start a business in the future.

Attitude towards entrepreneurship has predictive power with regard to the likelihood of entrepreneurial behaviour (Karali 2013). The results of this study are consistent with other studies, which provide evidence that attitude is a much stronger predictor of entrepreneurial intention. The results indicate that the attitude towards entrepreneurship had a positive and significant relationship with entrepreneurship intention (β = 0.725, p < 0.001). The more positive an attitude one has towards entrepreneurship, the more likely one is to have the intention to engage in a small-scale business. These results are supported by other researchers, for instance, Ayalew and Zeleke (2018) and Bilgiseven (2019), who argued that staying positive, even when setbacks arise, is key to moving forward and pushing through especially in entrepreneurship. However, business disruptions because of coronavirus disease 2019 (COVID-19) are signalling a strong decline in revenue and profitability of small businesses (McKinsey 2020), probably reducing the entrepreneur’s confidence and creating a negative attitude towards entrepreneurship.

Many small businesses do not have the financial, operational or strategic structures that larger corporations do. Research results suggest that if an individual has a high sense of self-efficacy, he or she will have higher entrepreneurial success (Al-Ghazali & Afsar 2021). Furthermore, individuals with higher entrepreneurial self-efficacy are more confident in their ability to run their own business with high performance (Santos & Liguori 2019). The results of this study are consistent with the literature and indicate that entrepreneurship self-efficacy has a positive and significant relationship with entrepreneurship intention (β = 0.157, p < 0.001). An increase in self-efficacy is likely to lead to more people intending to engage in small-scale businesses. According to the McKinsey (2020) report, breaking free from a restrictive owner mindset and assuming a more strategic role is important. However, because of pessimism about the economy and future outlook, there could be some hesitancy to invest in small-scale business unless there are mitigating factors that would cushion the finances if their revenues fell.

Entrepreneurship education has a positive and significant relationship with entrepreneurship intention (β = 0.110, p < 0.05). These results suggest that increasing entrepreneurship education was catalytic to increase the intention to engage in small-scale business. Because of a lack of knowledge of available funding options as well as knowledge in managing cash flows and earnings, insufficient or lack of knowledge regarding entrepreneurship can lead to small-scale businesspeople failing to secure small-scale funding. Even when funding is available, Small Medium Enterprises (SMEs) face significant barriers to accessing the necessary support (McKinsey 2020), emphasising the importance of entrepreneurship education. Other than not qualifying, the top two reasons for not receiving financial assistance were that entrepreneurs were unaware of the opportunities or did not know where to find the information needed to apply, emphasising the importance of entrepreneurship education in entrepreneurship intention.

In the aftermath of the COVID-19 pandemic’s market competition, small-scale entrepreneurs with a better understanding of shifting demand, potential new client bases and local substitutes for products can shift their focus to new target markets in order to sustain demand (Zizile & Tendai 2018). This necessitates being inventive and making use of available technology. This study’s findings contradict this evidence, as respondents’ innovativeness had nothing to do with their intention to start a business. These findings are concerning in light of the COVID-19 pandemic, which highlights the importance of creativity in order for the company to grow. Accessing the right markets to sell products can be difficult if entrepreneurs are not sufficiently creative. To connect with potential buyers during the lockdown, they could sell their products through WhatsApp groups or through other social networking sites.