This study was conducted in Nakuru County, Kenya. The county boarders Baringo to the North, Kajiado to the South, Laikipia to the north-east, Bomet to the West, Nyandarua to the East and Narok to the south-west. It lies in longitude 0.3031˚S, and latitude 36.0800˚E. The county has a population of about 2.1 million with youth constituting 33% of the total population [28] . Nakuru county cover an area of 7495.1 km² and has 11 sub-counties which include Nakuru West, Nakuru East, Njoro, Kuresoi North, Kuresoi South, Molo, Gilgil, Subukia, Naivasha, Bahati and Rongai. The main economic activity in the county is agriculture. Main crops grown are wheat, maize, potatoes and horticultural crops like vegetables and fruits. According to Quarterly Labour Force Survey, business sector in Kenya recorded the highest increase in employment with 111,100 new jobs created in the second quarter of 2021. This trend is also reflected in Nakuru county, especially after corona virus outbreak which caused job losses and which made many people to consume fruits to boost their immunity. The study was conducted in Nakuru East and Nakuru West because they have more youth who sell fruits compared to the other sub-counties (Figure 1).

Multistage sampling was employed when conducting the study. First, selection of Nakuru county was purposively done because it is one of the counties where youth practice agripreneurship. Secondly, Nakuru town West and Nakuru town East sub-counties we selected purposively they have highest population in the county and therefore attract agripreneurship. Similarly, they have major fruit business markets. In the third stage, three wards, Biashara, Nakuru East and Kaptembwo were selected because of the high number of youths who sell fruits. Agripreneurs were then selected from the chosen wards using proportionate random sampling to obtain the sample size.

Semi-structured questionnaires were used to collect data for the study. The questionnaires were composed of both open and close-ended questions and were used as a survey instrument to obtain primary data from the sample. The primary data collected included the motivational factors behind the fruit business venture, the agripreneurs social capital and psychological capital. The fruit business performance in terms of financial and non-financial indicators was also be collected. To ensure the validity of the questionnaire, items in the questionnaire were carefully considered to ensure the collection of relevant information.

The unit of sampling were youth who sell fruits while the sample size was made of youth between 18 - 35 year in Nakuru West and Nakuru East who have sold fruits for a year or more. To obtain the sample size, Yamane [29] was used as the population of the study area was known with certainty. The applied formula was

n = N 1 + N ( e 2 )

n = 739 1 + 739 ( 0.05 ) 2 = 259.52

n = 260

where; n = Sample size; N = the total population of interest l 2 = 0.05-margin of error (qualitative variables). This gave a sample size of 260 respondents. The proportionate was obtained by adding the total population of the three wards then diving the target population of each ward with the total population as shown in Table 1 and then multiplying with the target sample size to obtain the sample for each ward.

PLS-SEM was used to determine the effects of agripreneurial motivation on social and psychological capital. Structural equation model (SEM) was used as it allows both unobserved and observed variables to the incorporated in the same

Source: County government of Nakuru.

model [30] . It also handles measurement errors within exogenous variables which have several indicators using Confirmatory Factor Analysis (CFA). Similarly, SEM allows multiple linear regression between independent variables to be analysed simultaneously, allow multiple path analysis, asses the fitness of the overall model and also direct and indirect effects which cannot be done by traditional regression method of analysis. In addition, the model offers measures of fit to assess the entire model [31] . The general question model used consisted of measurement and structural models:

Y = v + Λ n + ε (1)

η = α + B η + ξ (2)

where Y is the vector of p observed variables in a considered study (p > 1), ν is the vector of observed variable mean intercepts the p × 1 vector of observed variable mean intercepts, Λ is the p × q matrix of factor loadings, η is q × 1 latent factors assumed in it (q > 0), ε the vector of p pertinent residuals (error terms), α is the q × 1 vector of latent variable intercepts, B is a q × q matrix of latent regression coefficients and ξ is the q × 1 vector of corresponding latent disturbance terms.

From the above general Equations (1) and (2), the structural equation model for the two factors; necessity motivation (ξ₁), opportunity motivation (ξ₂), with manifest endogenous variables social capital and psychological capital (Y₁, Y₂) was modelled as follows:

Y 1 = α 1 + β 1 ξ 1 + β 2 ξ 2 + e 1 (3)

Y 2 = α 2 + β 1 ξ 1 + β 2 ξ 2 + e 1 (4)

The general matrix expression is given in the following equation:

Y 1 Y 2 = α 1 + ℜ 1 ξ 1 + e 1 (5)

where;

ℜ 1 = β 1 , β 2 and ξ 1 = ξ , 1 ξ 2 (Table 2).

In Equation 5 above, Y₁, Y₂ is the manifest endogenous variables (psychological capital and social capital), α₁ is the latent intercepts, ℜ 1 are the coefficient vectors for the linear effects of n latent predictors, ξ₁ are the latent factors and finally e 1 is the latent disturbance. The direct effects of motivation on psychological and social capital are shown in Figure 2.

It is assumed that agripreneurial motivation (opportunity and necessity) directly affect the items of psychological capital (self-confidence, hope, resilience and optimism) and social capital (cognitive, relational and structural) of the youths who sell fruits as shown in Figure 3.

Convergent validity is realized when a group of indicators of a construct represents a single underlying construct or converge Hair et al. [31] . Cronbach’s alpha (CA), rho A and the composite reliability (CR) were used to test the relia

bility of the constructs. CA values ranged between 0.528 - 0.811, meeting the threshold of 0.5. The CR values ranges 0.761 and 0.877, exceeding the minimum threshold of 0.70 [31] . rho A values ranged between 0.577 and 1. Therefore, reliability and internal consistency was achieved as evidenced by the results in Table 3.

To assess the convergent validity of the constructs, average variance extracted was used. The values were more than the threshold of 0.4 [31] . Similarly, according to Fornell and Larcker [32] , even though the AVE values are less than 0.5, but the CR is more than 0.6, the constructs still achieve convergent validity. From these results, the convergent validity of all the constructs used in the SEM were acceptable. The VIF values of all the variables were less than 3.3 showing that multicollinearity was not a problem [33] [34] .

Average full collinearity VIF (AFVIF) was used to assess the model’s goodness of fit. The VIF values was 1.643 VIF (AFVIF) = 1.643, acceptable if ≤ 5, ideally ≤ 3.3 [35] . Based on these results, the SEM fitted well for estimating the effects of motivation on social capital and psychological capital. The generated coefficients, therefore, yielded suitable estimation on the role of agripreneurial motivation, social capital and psychological capital on performance youth-based agrienterprises.

This study used the heterotrait-monotrait ratio of correlations (HTMT) ratio and cross loading to test discriminant validity. For cross loading to meet discriminant validity, no indicator should have higher punctuation than the construct being measured [31] . In Table 4, all the constructs of individual items’ loadings had greater values for their respective constructs than other constructs as shown by the numbers in bold.

The HTMT ratios of the constructs used to examine the effect of entrepreneurial motivation on psychological and social capital of youth-based fruit agri enterprises are shown in Table 5.

Henseler et al. [36] stated that HTMT is a better criterion to assess discriminant validity. Similarly, according to Monte Carlo, HTMT are better than measures like Fornell and Lacker [32] [37] . In this study, there was no discriminant validity problem as the HTMT ratios were less than 0.85 [38] (Table 5). HTMT values of less than 0.90 are still good according to Gold et al. [39] .

Motivation, psychological and social capital had discriminant validity with their constructs, hence good reliability and validity as the HTMT ratios were between 0.229 to 0.878 shown by bolded values. The constructs had no potential measurement bias and were suitable to estimate effect of agripreneurial motivation on social and psychological capital of youth-based fruit agrienterprise.

It is hypothesized that necessity motivation influences the constructs of social and psychological capital. The path coefficients of direct effects of entrepreneurial necessity motivation on psychological capital and social capital of youth-based fruit Agri enterprises are in Table 6. Necessity motivation and cognitive social capital had a negative and significant relationship at 10% significance level. From these results, it can be said youth who chose entrepreneurship because they were not employed or because they are poor are less likely to communicate well with their customers and are also not sure with the quality of the products they sell. The probable explanation for this is that push type entrepreneurs do not have passion for entrepreneurship, therefore, lack entrepreneurial skills like good communication skills which are required to have good customer relationship. There was a positive and significant relationship between necessity motivation and relational social capital at 5% significance level, therefore, youth who choose entrepreneurship due to poverty respect their customers because they desire to retain and increase their customers to make more sales and increase their profits. Necessity motivation and self-confidence had a positive and significant relationship at 5% significance level. This means push type

*NM- Necessity motivation.

entrepreneurs or those who want to provide for their families are likely to trust their ideas. Such entrepreneurs make sure they start an enterprise they trust and are confident with because they do not want any challenging opportunities/business environment and also want to avoid failure. The results also indicate a positive relationship between necessity motivation and resilience at 1% significant level. This means that youth who started their agrienterprise due unemployment or because they lost their jobs are able to adapt to the challenges they face in their enterprises. This is because they do not have any option other than entrepreneurship. Such entrepreneurs do not want to go back to their original situation, thus have no option but to look for ways to overcome their challenges. These results collaborate with those of Ephrem et al. [7] which indicates youth who are necessity driven are resilient.

It is hypothesized that opportunity motivation influences the constructs of social and psychological capital. The results in Table 7 indicate the path coefficients of direct effects of entrepreneurial opportunity motivation on psychological and social capital of youth-based fruit Agri enterprises. From the results, there was a positive and significant relationship between opportunity motivation and cognitive social capital at 1% significance level. Therefore, youth who sell fruits because they saw an opportunity on fruit agrienterprise are more likely to have shared values and communicate effectively with their customers. This is because such youth possess entrepreneurial skills, have passion and understand their customer needs since they conduct research before deciding on the enterprise to venture. Opportunity motivation and structural social capital had a positive and significant relationship at 5% significance level. Youth who have passion for entrepreneurship, therefore, interact frequently and maintain their relationship with their customers. The probable explanation for this is that such youth wants to understand their customers better so as to meet their needs. A positive and significant relationship was also experienced between opportunity motivation and relational social capital at 1% significance level meaning youth who have perceived entrepreneurial opportunity trust and relate well with their

customers. The reason for this positive relationship is because youth who are opportunity driven possess social networks, trust and respect their customers. Increased interactions with their customers also lead to spread of trust. This result is in line with that of [40] who stated that trust paves the ground for relations and conversation.

Youth whose entrepreneurial motivation was of opportunity-type had a positive and significant relationship with hope at 1% significance level. This imply that those who venture into agripreneurship because they love it are more likely to overcome challenges they face in their enterprises and also to achieve their goals. This is because they are endowed with entrepreneurial ingredients that drive their choice of an entrepreneurial career, hence, set realistic goals about their opportunities. Self-confidence and opportunity motivation had a positive and significant relationship at 1% significance level, meaning youth whose reason to venture into entrepreneurship was out of passion is more likely to trust their business ideas and abilities. This can be explained by the fact that opportunity entrepreneurs do market research prior to venturing into the business hence believe in their entrepreneurial competencies and have high perceived feasibility which influence their agripreneurial intentions [41] . Opportunity motivation and optimism had a positive and significant relationship at 1% significance. Those who start business because they have passion, therefore, keep expecting positive outcome even when things are uncertain. The probable explanation is that such youth understand that there are low and high seasons in business, therefore, foresee a better future despite the uncertainties. Resilience and opportunity motivation had a positive and significant relationship at 1% significance level, showing that youth who enjoy entrepreneurship are likely to adapt easily to the challenges they face. Such youth would have many ways to overcome difficulties and can adapt to the stressful life events as they have strong desirability for self-realization unlike entrepreneurs by necessity. Opportunity entrepreneurs enter into businesses because they already believe in their entrepreneurial competencies and have more means to overcome any drawback compared to necessity entrepreneurs who often startup businesses with limited funds and poor entrepreneurial skills. This result supports that of Ephrem et al. [7] which states that opportunity driven entrepreneurs are highly resilient.