The study involved three out of 55 FDCs in Tanzania, namely: Mamtukuna (Kilimanjaro Region), Monduli (Arusha Region) and Chisale (Dodoma Region). These FDCs were selected for this study because one of their major objectives of training is to equip the learners with the knowledge and skills that would enable them to be self-employed and self-reliant based on local situations. The three colleges were selected purposively because of the similarity in the nature of the agricultural courses which were blended with an entrepreneurship course. The study population was all final-year certificate students pursuing agriculture courses.

This study employed a cross-sectional design, which was appropriate for this study because the data were collected from three colleges which are located in three different regions at one point in time. A sample size of 300 students was formed from an estimated population of 1200 from the three colleges using the formula developed by Israel (2013):

n = N/ (1 + N (e2))[Eqn 1]

In Equation 1, n is the sample size, N population size and e is the level of precision. The formula assumes that p = 0.05 (maximum variability). The desired confidence level is 95% and the degree of precision or sampling error accepted is ± 5%. Therefore n = 1200/ (1 + 1200(0.05)) = 300.

Every element in the sample was selected using simple random sampling, as this procedure considers the sampling elements to have homogenous characteristics (all are final year and their courses are blended with an entrepreneurship course). The sample was drawn from FDCs’ admission records and directories. The selection of sample from each institution is shown in Table 1.

The study employed questionnaires, focus group discussions and an interview guide as data collection tools. Pre-testing of questionnaires was done before being administered. The questionnaire forms were distributed to 12 respondents, equivalent to 4% of the sample size during pre-testing. A few unfamiliar terms were noted, and the researcher made adjustment to those terms by replacing them with more familiar terms. Three hundred questionnaire forms were administered and its development was guided by the Theory of Planned Behaviour as retested by Liñán and Chen (2006). The completed questionnaires forms were 294 (98%) which were used in the analysis. Six focus groups each consisting of seven students were formed through nomination strategy. Six college staff members (two per college) and two Ministry of Health, Community Development, Gender, Elderly and Children (ministry responsible for FDCs) officials were purposively selected and involved in interviews based on their role, knowledge and experience.

The data collected include: respondents’ socio-demographic characteristics, types of courses studied, types of teaching methodology applied, existence of intention and expected learning outcomes. Six items that assess the expected learning outcomes (independent variable) were developed from the content of the courses studied and the teaching methods applied. The six items were developed under the guidance of the Damian and Wallace (2015) and Roomi and Redman (2016) studies. The six items were measured by a five-level Likert scale labelled 1 = strongly disagree to 5 = strongly agree. Likewise, the intention (dependent variable) was assessed by nine items developed under the guidance of Liñán and Chen (2006) and Malebana (2012) and measured on a five-level Likert scale labelled 1 = strongly disagree to 5 = strongly agree.

The data supporting the two objectives of this study were analysed by using descriptive non-parametric inferential statistics and content analysis. Specifically, respondents’ socio-demographic characteristics, types of courses studied, types of teaching methodology applied, existence of youth farm entrepreneurial intention and perception of the expected learning outcomes were analysed by using frequencies, percentages, means and standard deviations. The relationship between expected learning outcomes and youth farm entrepreneurial intentions were analysed by using Somers’ D non-parametric model.

Somers’ D of Y with respect to X is defined as:

D(Y/X)= T(X, Y)/T(X, X)[Eqn 2]

In Equation 2, Somers’ D coefficient of association for asymmetrical variables, X is the independent variable pair which include expected learning outcomes assessed by six indicators or items and Y is the dependent variable pair which is the intention variable assessed by nine indicators or items. If Somers’ D coefficient > 0 ≤ + v ₁, the variable is regarded to have impact on intention. The choice of Somers’ D is based on the central role it plays in rank statistics for non-parametric (Newson 2013). Qualitative information collected from key informants and focused group discussion were conceptually analysed by coding the themes systematically and drawing the key that emerged from them.

Internal reliability of items for the self-administered questionnaire was measured by Cronbach’s alpha as defined by Fami (2000):

α=K/K−1×ST2−∑​SI2[Eqn 3]

In Equation 3, α is the alpha coefficient, K is the number of items, S T2 is the total variance of the sum of the item and the S I2 variance of each individual item. A positive alpha coefficient ranging from 0.7 to 1 was taken into consideration as most recommended internal consistency (Gliem & Gliem 2003). Pairwise deletion method was applied in performing the reliability analysis. To obtain the required alpha results, some of the items that were in the questionnaire were deleted. Nine items measuring expected learning outcomes were presented to the respondents but three items were deleted after reliability test results with their overall Cronbach’s alpha coefficient reading 0.707. No item was deleted for the entrepreneurial intention item after the reliability test and their respective coefficient read 0.870.

To ensure that the instrument covered all the components of information, content validity was determined through reviewing previous studies in assessing the adequacy and the accuracy of what it measures. The questionnaire items that measured farm entrepreneurial intention were adopted and modified and fixed to the context from work of Liñán and Chen (2006), Ajzen (1991) and Malebana (2012). The development of topics list, entrepreneurial teaching methodology and expected learning outcomes were guided by the following studies: Damian and Wallace (2015), Gibb and Price (2014), Roomi and Redman (2016), Vesala and Jarkko (2008), Adeyemo (2009), Klein and Bullock (2006) and Vandenbosch (2006).

The analysis of the data shows that the mean age of the respondents was 20.6 years, the lowest being 15 years, and the highest age was 31 years with a standard deviation of 2.439. The average age falls within the age criterion definition of youth by the United Nations. It is also in line with the operational definition of youth used in this study. The distribution by sex shows that women were 11.6% more than men as shown in Table 2. The respondents involved in the study were in two main groups. The first group was those who specialised in animal husbandry and the second group was those who studied general agriculture (both animal and crops husbandry). The second group did not specialise because they were not sitting for VETA exams which have enrolment limitations as per Form Four National Examination results. In the analysis, the two groups were combined since they are taught using FDC and VETA curricula.

Studies by Assane (2015) and Sumra and Katabaro (2014) have noted mismatched, narrow and outdated agricultural education curricula in developing countries. Following the inadequacy in the curriculum, competencies that focus on preparing the youth to be farm entrepreneurs were assessed. Also, studies by Roomi and Redman (2016), and Vesala and Jarkko (2008) have recommended skills training and attribute development for context-specific entrepreneurship education. Therefore, the competencies and skills assessed in this study have been categorised according to the aspects that guide farm enterprise development or growth (agriculture and general entrepreneurship competencies). The assessment of competencies and skills was conducted in the form of topics studied at this level, as shown in Table 3. The list presented was cross-checked against the existing FDC curriculum.

The findings (Table 3) show that for the agricultural courses, animal husbandry is the most studied topic, followed by crop production; the least studied topic was agro-mechanics, followed by agro-economics. The topics assessed are mostly found in the curriculum of FDCs, except the value addition and value chain. However, content-wise, some topics are taught at a very elementary stage (lower levels of cognitive domain; action verbs such as define and mention) and some important topics are missing. Following the inadequate content in the FDC curriculum, in 2013 the government through the Ministry of Health, Community Development, Gender, Elderly and Children allowed the parallel use of VETA curriculum to supplement the FDC curriculum as explained by the ministry director coordinating Community Development Training Institutes and FDCs:

This was further confirmed during the interviews with FDC principals and Ministry of Health, Community Development, Gender, Elderly and Children officials.

On the other hand, the review shows that although the entrepreneurship course was not in the FDC curriculum, the students were taught using the VETA curriculum. Students have studied all courses that teach basic entrepreneurship knowledge and skills, except the Human Resource Management course. The Human Resource Management course is not an entrepreneurship course per se but it is important because it helps an entrepreneur to manage their employees properly during the expansion stage of an enterprise. However, it has not been critically adapted to the agricultural context as it lacks a practical facet. General implication of the score is that learners are expected to possess the right entrepreneurship competencies.

Based on these findings, the majority of the respondents have basic agricultural and entrepreneurship competencies that can help them to establish farm enterprise. Although some basic topics such as value addition, value chain and human resource management are missing in both the FDC and VETA curricula, with such competencies, the youth can at least manage to start up an enterprise. This is because the ideal requirement is awareness or exposure and basic applied skills that are enterprise-specific in order to launch an enterprise.

The question of adequacy of syllabus in terms of content was asked to the tutors and the response was that with the introduction of the VETA-curriculum, their syllabus was deemed adequate; when the follow-up question was asked about why the graduates find it difficult to start or establish an enterprise, the Mamtukuna tutors explained:

This also emerged during the focus group discussion as one of the discussants explained:

What is taught is one thing but how it is taught to obtain the best outcome in learning is a different matter. In this case, the teaching methods that were used to deliver the agricultural and entrepreneurship competencies to FDC students were assessed as shown in Table 4.

The findings show that six methods of teaching are commonly applied in the FDCs, with lecture, research and guest speakers less common, as shown in Table 4. As far as developing entrepreneurship traits among students is concerned, the last two techniques of research and guest speakers are crucial as noted by Adeyemo (2009). Research techniques help to build innovation, creativity and analytical capabilities, traits that are important for an entrepreneur. In addition, inviting successful entrepreneurs helps to develop motivation, networking and the development of a good attitude concerning the field in which they are expected to work. Thus, less regular use of those methods could have affected the learners’ entrepreneurial traits.

During focus group discussions, respondents were asked if they are happy with the teaching methods. They were positive about those courses in which the tutors have good experience and provide practical details in their respective subjects. However, they complained that many of practical skills-oriented topics are only theoretically taught. The focus group discussant explained:

The discussion shows that lack of qualified tutors and appropriate infrastructure for student learning are a constraint to the experiential teaching method. The findings concur with Prathima et al. (2008) who noted that training is limited to traditional classrooms.

The expected learning outcomes in relation to farm entrepreneurship were assessed as shown in Table 5. The findings show that students agreed and strongly agreed with items 1–6 intended to measure the skills, knowledge and any other lesson factors. Students gained farm entrepreneurship knowledge and were satisfied with teaching methodology applied compared to skills-related outcomes as evidenced by scores in Table 5. These findings concur with those of Rammolai (2009) who noted that time was the limiting factor for entrepreneurship studies in Zimbabwe. However, the actual time allotted is 6 months per term which exceeds university time allotted per semester by 2 months. Moreover, during the interviews with principals of colleges and during the focus group discussions, it was noted that the majority of tutors work part time. This arose during the focus group discussion; for example, one discussant mentioned that:

In general, the respondents have a good perception of the expected learning outcomes measured in relation to farm entrepreneurship. Also, the curriculum used is relevant to the existing environment and industry as no mismatch has been noted at this perception level. However, practical classroom learning is lacking as reported during focus group discussions; this may have implications in terms of a mismatch with industry requirements for farm entrepreneurship. This finding is consistent with Gemma and Manipol (2015) and Prathima et al. (2008) who found much theoretical emphasis in teaching. In addition, the content of the two curricula that are currently used in parallel was reported as sufficient but some of the students who do not sit for VETA exams are still using FDC curricula and implementers are not mandated as per evaluation to use the VETA curriculum. Furthermore, since the FDC curriculum lacks addressing some of the skills and since the objectives of the two curricula are not the same, there is a need to harmonise the use of the two curricula.

An index was developed to determine the overall level of attainment of the expected learning outcomes of the respondents which was then analysed by descriptive statistics as shown in Table 6. Table 6 was developed from Table 5 where the Likert scale consists of six items and five response options with their respective weights – strongly disagree (1), disagree (2), unsure (3), agree (4) and strongly agree (5) – were transformed into three level responses. With regard to respondents’ responses, the total minimum score for the six items was 6, the total neutral or unsure scores for the six items was 18 and total maximum score for the six items was 30. In developing the index, the researcher grouped the strongly disagree and disagree scores and labelled them as ‘learning has no impact’, unsure was labelled as ‘undecided’ and agree and strongly agree were grouped as ‘learning has impact’.

Generally, the descriptive statistics in Table 6 show that the courses studied have impact. The implication for this finding is that the youth who studied agri-entrepreneurship courses believe they possess the basic farm entrepreneurial knowledge and skills in the related affective learning domains. They are aware of the concept of farm entrepreneurship, familiar with enterprise start-up and its management and they believe they possess basic agriculture field skills. This finding concurs with Nabi et al. (2018) who found the type of entrepreneurial learning (basic knowledge about entrepreneurial process and associated skills) influences individual entrepreneurial intention.

Farm entrepreneurial intention as a key aspect in this study was assessed by nine items measured on a five-point Likert scale (strongly disagree, disagree, unsure, agree and strongly disagree). The determinants for farm entrepreneurship intention that were assessed include readiness, determination, interest and internal drives (internal locus of control), measured by nine items as shown in Table 7. The findings show that the higher ratings of the respondents fall under the agree and strongly agree measurements. There is a small discrepancy between prior intention in relation to item 8. This implies that after students’ enrolment in the college, they received more exposure, skills, knowledge and confidence which led to an increase (scores) in intention to pursue farm entrepreneurship. By considering the eight item scores measuring intention, the agricultural education blended with entrepreneurship course contributes to the increase in farm entrepreneurship intention of the final-year FDC students. This finding is supported by the studies done by Malebana (2012) and Liñán et al. (2010) who found an increase in intention after studying. Also, the scores for prior intention imply that college agriculture education is not the only contributing factor for intention in farm entrepreneurship; there are other factors that are not explained by this study.

An index was developed to determine the overall level of youth farm entrepreneurial intention of the respondents after exposure to studying agricultural courses. As shown in Table 8, the Likert scale consists of nine items and five response options: strongly disagree (1), disagree (2), unsure (3), agree (4) and strongly agree (5). With regard to respondents’ responses, the total minimum score for the nine items was 9, the total neutral or unsure scores for nine items was 27 and total maximum score for the nine items was 45. In developing the index, the researcher grouped the strongly disagree and disagree scores and labelled them as ‘no intention’, unsure was labelled as ‘undecided’ and agree and strongly agree were grouped as ‘presence of intention’. Generally, the descriptive statistics in Table 8 show that the majority of youth have acquired farm entrepreneurial intention.

The relation between farm entrepreneurial intention and expected learning outcome was assessed. Farm entrepreneurship intention was first assessed based on the fact that they have studied the required courses. However, to obtain a direct relation, the learning outcome factors were established and computed against nine factors of farm entrepreneurship intention.

The findings (Table 9) show that, all eight factors measuring farm entrepreneurship intention, excluding factor number 9, have significant relation for the majority of expected learning outcome determinants. However, the Somers’s D coefficient ranges from below 0.02 to above 0.2 indicating very weak to moderately weak strength of relationship. The relationship between youth farm entrepreneurship intention and expected learning outcome was tested at the 5% level of significance.

The findings further show that six factors measuring farm entrepreneurial intention have a moderate weak significant relationship with the first determinant of the expected learning outcomes. Since the first expected learning outcome determinant focused on the possession of farm entrepreneurial skills, it is clear that the respondents’ farm entrepreneurial intention is influenced by studying agriculture and entrepreneurships courses. Also, significant relationships exist between four factors measuring farm entrepreneurial intention and the remaining five expected learning outcome determinants. The implication is that the respondents have acquired farm entrepreneurial knowledge and behavioural competencies which consequently impacted their farm entrepreneurial intention.

This finding concurs with Emanuel’s (2012) and Remeikiene, Startiene and Dumciuviene’s (2013) studies which found the existence of a farm entrepreneurial intention after studying agriculture courses blended with entrepreneurship courses. This finding is supported by the Theory of Planned Behaviour which asserts that intention can be changed directly or indirectly by the formation of beliefs resulting from the environment (Ajzen 1991). Beliefs in this case were developed by studying agri-entrepreneurship courses. The impact of the training is manifested through a change in the youth’s attitude, perception and perceived behavioural control with regard to farm entrepreneurship career. The sum of these behavioural antecedents subsequently changes the youth farm entrepreneurial intention.