Determinants of Takaful Patronage Among Entrepreneurs in Sabon Gari Market, Kano State, Nigeria

Objective – This paper examined the determinants of Takaful patronage among the entrepreneurs in Sabon Gari market of Kano metropolis, Nigeria. Design/methodology – The paper adopted survey research approach through the use of questionnaire. Respondents were selected using a purposive sampling technique. Multivariate regression was adopted to test the hypothesized model of the study using variance-based SEM. The data was analyzed using Smart-PLS 3. Results – The results show that religiosity, trust and confidence and awareness most often positive and are found to significantly influence patronage of Takaful services at significant level of 0.05 while the three explanatory variables were able to explain 58% variation in the dependent variable.


Introduction
Islamic finance is seen as one of viable vehicle for sustainable economic growth, financial inclusion and poverty alleviation across the globe; as it has been stated by Central Bank of Nigeria in its 2020's national financial inclusion strategy (Sanusi, 2011), the potential for its growth and greater penetration of the market is very high in Nigeria especially in the predominantly Muslim northern part of the country where the entrepreneur's experiences frequent risk and disasters, such as Sabon Gari market fire inferno and several incidences in Kano state.
Islamic insurance (Takaful) as one of the most important areas of Islamic finance can serve as conscientious component to mitigate the lost engulf by such disasters, never the less; risk is a natural phenomenon in any society that everybody is exposed to in daily life. These risks and perils affect the lives of many individuals in the society and could be devastating especially for the vulnerable and helpless. Therefore, Takaful has enormous potential, largely untapped, among the vast Muslim population.
Takaful company in Nigeria was first introduced in 2003 with registration of three insurer companies, the African Alliance Insurance Plc, the Corner Stone Insurance Plc, through its division, Halal Takaful Nigeria Limited and the Niger insurance Plc. The development was further strengthened by issuance of the Takaful Regulation Policy in 2013 by National Insurance Commission of Nigeria, this and also paved way for the 2016 licenses of the two full-pledged Takaful companies, namely Jaiz and Noor Takaful (Ayuba, 2014). Despite the mandatory policy requirement as well as the voluntarily insurance policy, the scheme of insurance yet to be accomplished the desire anticipated results due to low level of patronage (Darazo, 2011). According to Nigerian Bureau of Statistics (NBS 2017) the contribution of insurance industry in Nigeria to the Gross Domestic Product (GDP) is below 1%. This is however, as a result of the low performance of both Islamic and conventional insurance in Nigeria. Darazo (2011) pointed out other

Awareness
Awareness derived from the word "aware" which means cognizant, conscious, sensible or awake, it means having knowledge or discernment of something. Aware implies knowledge gained through one's own perceptions or by means of outside information. Consumer awareness refers to the combination of the knowledge of the product purchased by the consumers in terms of its quality (Ishak & Zabil, 2012). Awareness can be seen as knowledge that something exists, or understanding of a situation or subject at the present time based on information or experience (Cambridge dictionary). Lateh, Ismail, & Ariffin (2009) have asserted that, awareness is referred to the knowledge about 'Consumer Rights. This means that, the consumer must know that he/she has the right to get the right kind of product or services. Awareness in marketing, measure of how well known a brand, firm, or product is. Raising Takaful awareness among individual consumers remains major challenge. It is about general knowledge of Takaful and specific familiarity with Takaful provider. The level of awareness can be found in regard to basic of concept, difference between Takaful and insurance, terms and others. However, it has been argued that the level of awareness about Takaful concept and principles is relatively low and unpopular in Nigeria due to its nature of infancy.

Religiosity
Religiosity is defined as the degree to which a person adheres to the religious values, beliefs, and practices that are used in their daily life (Worthington Jr. et al., 2003). Islam is a religion that urges its followers to do good and noble deeds; it guides the acts. of every follower or believer in all cases including good relations not only with God but also with other human beings. McDaniel & Burnett (1990), defines religion as a belief in God accompanied by a commitment to follow principles believed to be set forth by JAROE VOL. 3(3) God. Several studies have attempted to measure the impact of the religiosity factor on the selection of Islamic financial institution products. The study by Arifin, Yazid, & Hussin (2014) revealed that religious values stand as the most influential factor among academicians in public institutions of higher learning to patronize Islamic financial institution in Malaysia.

Trust and Confidence
Trust and confidence repose on Takaful operators refer to as willingness of consumer to buy the services and feel comfortable that his investment is in competent, credible, honest and safe hand. A study by Bello & Ayuba (2014) revealed that trust and confidence is a factor or determinant that positively and significantly influence Takaful patronage.

Patronage
According to International Institute of Sustainable Development (IISB) (1995), patronage is described as the use of goods and services that respond to the human basic needs. Patronage is an attitude of consumers guided by their mind set with some level of like and dislikes and attitude pre-empt consumers intended behavior, individual's dispositions, situations and context are controlled by their attitudinal actions and inactions. Human behaviors are guided by Religious, Traditional and cultural beliefs which can also change one's attitude for consuming goods and services such as Takaful products, so also other non-economic constructs with human behavioral element such as information and Trust.
Empirical Literature Review Hassanuddin, Muda, & Karim (2016) in Malaysia conducted a survey to obtain peoples' response towards Takaful insurance. The study investigated the factors influencing customer's intentions in participating in Takaful, using a judgmental Sampling method to obtain the sample. Also, a personally-administered questionnaire was distributed randomly among respondents and data is analyzed using structural equation modelling (SEM). From the analysis, it has been found that religion and awareness were among the factors that contribute significantly towards employee's participation to Takaful. The findings of this study proposed that Takaful companies should concentrate in their marketing and promotional matters. However similar study can be conducted in different environment. Adamu (2018) studied acceptance and patronage of insurance services in northern Nigeria. The researcher used theory of planned behavior (TPB) to describe northern Nigerians' acceptance and patronage of insurance services. By applying the theory with additional variable as a contribution, to see what precisely account for relatively poor patronage of insurance services in northern Nigeria, particularly now that Islamic insurance is established. Furthermore, the study examined the Interrelationship among the variables (attitude, subjective norms, perceived behavioral control, awareness and religion) using multiple regression to see the extent of influence each has on customers' acceptance and patronage. A survey is conducted on 288 members of the public, (both insured and non-insured) using SPSS software as a tool of analysis. The research found out that the attitudes and awareness are the most often negative cause of low patronage of insurance services. And religious values, subjective norms and behavioral control factors does not account for poor acceptability and patronage. Maiyaki & Ayuba (2015) investigate consumers' attitude toward Islamic insurance services (Takaful) patronage in Kano State. The study examined the extent to which consumers' awareness, perception of Takaful services and the trust and confidence they reposed on Takaful Operators have significant relationship with their attitude towards Takaful services patronage. A structured questionnaire was administered to 384 regarded as the sample selected on judgmental sampling technique, the data was analyzed using multiple-regression the researchers found that awareness, perception, trust and confidence are significantly related to the consumers' attitude toward Takaful services patronage. However, a more suitable robust technique of analysis or model such as Structural Equation Modelling (SEM) can be used to further test the relationship between the variables that are lantern in nature.
Similarly, Bello & Ayuba (2014) test two (2) models representing different constructs explaining consumers' perception using survey of 266 respondents. Two dimensions: awareness and income earners' trust and confidence; were tested in Model A, using multiple regression. It was discovered that consumers' awareness of Takaful services showed least effect in influencing favorable perception of Takaful consumption. The trust and confidence reposed on the Takaful operators by consumers made average impact on determining favorable perception of Takaful services consumptions. While Model B' dimensions consist of age, marital status, gender, qualification, occupation, and income of consumer; were found to be insignificant and cannot make any capable impact in influencing consumers' favorable perception of Islamic services consumption in Kano Metropolis, Nigeria. These findings are unique as many studies have discovered factors such as income of consumers as capable of determining demand of Takaful services. However, a more suitable robust technique of analysis or model such as Structural Equation Modelling (SEM) can be used to further test the relationship between the variables that are lantern in nature.
A study by Maiyaki & Ayuba (2015) on the Consumers' Attitude Toward Islamic Insurance Services (Takaful) Patronage in Kano Metropolis, Nigeria, examining the extent to which consumers' awareness, perception of Takaful services and the trust and confidence they reposed on Takaful Operators have significant relationship with their attitude towards Takaful services patronage. A structured questionnaire was the only instrument used in collecting the data. Although, the population of the study is indefinite, 384 subjects are administered with the questionnaires using judgmental sampling technique and the employment status of the respondents was the basis for the selection and inclusion of the respondents in the study. The data was analyzed using multipleregression with SSPS version 17. The paper found that awareness, perception, trust and confidence are significantly related to the consumers' attitude toward Takaful services patronage. However, a more robust technique i.e., multivariate regression can be employed to test the relationship between the lantern variables.

Theoretical Literature Review
The theory of planned behavior (TPB) was developed by Ajzen in 1985 as an extension of Theory of Reasoned Action (TRA) whereby perceived behavioral control (PBC) was included. TPB seeks to study the linkages between beliefs, attitude and subject norms, perceived behavioral control, intention and behavior. TPB has been widely used in predicting an individual's behavior intention. The model of TPB has been extended to various studies and has the flexibility of integrating some variables for better understanding within the theory, and has been applied in many studies of intention, comments that TPB is well accepted and modified in various study of behavioral intention. Some researchers have argued that TPB model it lacking some determinant variables in facilitating behavioral intention a phenomenal; that there are other dimensions that can interfere when measuring behavioral intention towards a phenomenal (Echchabi & Aziz, 2012). Also, Jaffar & Musa (2014) used TPB in investigating and determining factors that influence the attitude towards Islamic financing.
However, in this paper, awareness, religiosity, and trust and confidence will be developed as a foundation of beliefs that serve as the component of TPB model to determine their significance on behavioral intention of entrepreneurs of Sabon Gari market towards patronizing Takaful product and services.

JAROE VOL. 3(3)
Source: Adapted from (Husin & Rahman, 2013) The figure 1 model presents the hypothetical framework about the arguments in this research paper. It consists of three dimensions as the independent variables: Awareness, religiosity and trust and confidence. While Takaful patronage in Sabon Gari market is the dependent variable. The model will be used to determine significant relationship between independent and dependent variables.

Research Method
Sabon Gari Market is a popular and one of the largest markets in Kano state, Nigeria which attracts customers from all over the Northern and other part of Nigeria. Sabon Gari market was built in 1914, is a one-stop, all-inclusive market for commodities located right at the heart center of the city and covers an area of approximately 22 hectares with over 13,000 shops. The population of the study comprises 35,000 registered entrepreneurs and 25,355 unregistered entrepreneurs constituted the total population of 60,355 (Yakasai, 2019). Three hundred and seventy (370) were selected as the sample size (Barlett, Kotrlik, & Higgins, 2001). The paper relies heavily on cross sectional data collected from the questionnaires distributed. A judgmental sampling technique was adopted to capture the target respondents of the study, resulting in 261 questionnaires were retrieved and a response rate of 70.5% was achieved. Data is analyzed using structural equation model (SEM) in which variance-based SEM is used with the help of Smat-Pls3 the choice of this technique is line with (Bello & Ayuba, 2015;Hassanuddin et al., 2016). The following model will be used in the study:

Findings and Analysis
This section presents and analyses the data generated for the study using Partial Least Square (PLS) path modelling. It begins by reporting the results of the preliminary analysis i.e., normality and multicollinearity test, followed by the descriptive statistics of demographic features of the respondents Thereafter the measurement model was assessed to determine the individual item reliability, internal consistency reliability, convergent validity and discriminant validity. Finally, results of structural model are reported (i.e., level of the R-squared values, effect size (F 2 ), and predictive relevance of the model (Q 2 ).

Preliminary Analysis
Initial data screening is very crucial in any multivariate analysis because it helps researchers identify any possible violations of the key assumptions regarding the application of multivariate techniques of data analysis (Hair, Money, Samouel, & Page, 2007). Additionally, initial data screening helps researchers to better understand the data collected for further analysis, the following preliminary data analyses were performed: (1) Normality test, and Multicollinearity test.

Normality Test
This is a pre-requisite and parametric statistical testing of data to ensure an underlying assumption and objective judgement. of normality aims toward Previous research has traditionally assumed that, PLS-SEM provides accurate model estimations in situations with extreme non-normal (Wetzels, Odekerken-Schröder, & van Oppen, 2009 From table 1, the Kolmogorov and Shapiro Test of normality shows that the variables of the study namely religiosity, trust & confidence, awareness and patronage were statistically insignificant i.e., P>0.05, hence all the variable are statistically normality distributed and therefore further analysis can be conducted.

Multicollinearity Test
Multicollinearity refers to a situation in which two or more independent variables become highly correlated. The presence of multicollinearity among the independent variables can substantially distort the estimates of regression coefficients and their statistical significance tests. In particular, multicollinearity increases the standard errors of the coefficients, which in turn render the coefficients statistically non-significant (Tabachnick & Fidell, 2007). In this study Variance Inflated Factor (VIF) and Tolerance Value were examined to detect multicollinearity problem. Hair, Ringle, & Sarstedt (2011) suggested that; multicollinearity is a concern if VIF value is higher than 5 and Tolerance Value is less than 0.20  Table 2 indicates that multicollinearity did not exist among the exogenous latent constructs as all VIF values were less than 5 and tolerance values exceeded 0.20, as suggested by Hair et al. (2011). Thus, none of the variables is highly correlated as such the data is free from multicollinearity.

Demographic Profile of the Respondents
This section analyses the demographic profile of the respondents of the study. The demographic characteristics are gender, age group, marital status, type of business and Annual level of Income.

Frequency
Percentage ( Table 3 above demonstrates that 258 of the respondents are males which is equivalent to (98.8%), and only 3 females' respondents are found to be only (1.2 %), this signifies the dominant of males in the study and that was due to the socio-cultural sitting of the respondents that businesses within the market's premises are dominated by males.
Table 3 above shown that 16 of the respondents which is equivalent to (6.1%) are within (18-25) age group, while (26-35) age bracket consists of 28(10.7%) of the respondents and 157(60.2%) respondents fall within (36-45), also 33(12.6%) are within (46-55) age bracket and lastly 27(10.4%) of the respondents fall within (56-Above). This entails that majority of the respondents are young entrepreneurs that fall within (36-45) age bracket then follow by (46-55). Table 3 above reveals that 32(12.3%) of the respondents are single, while 168 which is equivalent to (64.4%) are married and lastly 61 equivalents to (23.3%) of the respondents are divorced. This signifies that majority of the respondents are married.  Table 3 above shown that 73 which is equivalent to (27.9%) of the respondents are selling different forms of shoes, while 61(23.4%) of the respondents engaged in selling cosmetics, 54(20.6%) of the respondents are selling pharmaceuticals products and only 33(12.6) of the respondents sells electronics/electrical appliances and lastly 40(15.5%) of them fall under other category i.e., Jewries and sport wares. This signifies that majority of the respondents are selling cosmetics then follow by pharmaceuticals. Table 3 above reveals that 7(2.6%) have annual income of below N500, 000, while 74(28.4%) of the respondents have an annual income of N500, 000 -100,000,000 and 109 of the respondents which is equivalents to (41.8%) have the annual income of N1, 000,001-5,000,000 and 47(18%) of the respondents earn annual income of 5,000,001-10,000,000 and lastly 24(9.2%) of the respondent's annual income range from 10,000,001 and above. This indicates that majority of the target respondents are small scale entrepreneurs with annual income ranging from 1,000,000 to 5,000,000.

Assessment of PLS-SEM Path Model Results
This section presents a two-step process to evaluate and report the results of PLS-SEM path, as suggested by (Henseler, Ringle, & Sinkovics, 2009). This two-step process comprises of (1) the assessment of a measurement model, and (2)

Assessment of Measurement Model
An assessment of a measurement model involves determining individual item reliability, internal consistency reliability, content validity, convergent validity and discriminant validity (Hair et al., 2011;Hair et al., 2014;Henseler et al., 2009). Source: Computed from Field Survey (2019) using Smart-Pls 3

Individual Item Reliability
Individual item reliability was assessed by examining the outer loadings of each construct's measure (Duarte & Roposo, 2010;Hair et al., 2014). Following the rule of thumb for retaining items with loadings between 0.40 and 0.70 (Hair et al., 2014), it was discovered that out of 25 items, three were deleted because they presented loadings below the threshold of 0.40., however, the deletion should not be more than 25% of items measuring a lantern variable as suggested by (Hair et al., 2014). Internal consistency reliability refers to the extent to which all items on a particular (sub) scale are measuring the same concept (Sun et al., 2007). Cronbach's alpha coefficient and composite reliability coefficient are the most commonly used estimators of the internal consistency reliability of an instrument. In this study, composite reliability coefficient was chosen to ascertain the internal consistency reliability of measures adapted. Two main reasons justified the use of composite reliability coefficient. Firstly, composite reliability coefficient provides a much less biased estimate of reliability than Cronbach's alpha coefficient because the later assumes that all items contribute equally to its construct without considering the actual contribution of individual loadings (Götz, Liehr-Gobbers, & Krafft, 2010).

Convergent Validity
According to Hulland (1999), individual reliability is the process of assessing loading of the multiple items in relations to their respective construct of the study. The items of a particular construct are expected to be consistent in measuring the proposed construct (Hair, Ringle, & Sarstedt, 2013). Achieving reliability and validity explained that the items are free from random errors and systematic errors. The individual item reliability was assessed using their individual loadings, obtained from PLS algorithm result. Researchers argued that latent variable should explain at least 50% of the variance in the observed variable shared with the construct (Henseler et al., 2009). There exist various positions in the literature (Hulland, 1999;Iacobucci & Churchill, 2018), on the threshold or rule of thumb regarding individual item reliability. However, majority of scholars agreed on loadings between 0.4 and 0.7 (Hair et al., 2013). Following this suggestion, all items with less than 0.4 were subsequently deleted, meanwhile the deletion should not be more than 25% of items measuring the lantern construct as suggested by (Hair et al., 2013).   Table 4 show that the entire items factor loadings, whereas, factor loadings of less the yardstick of 0.4 were deleted (i.e. PTRG6= -0.056, TAC3= 0.029, RLG3= 0.143 and AWN6=0.343), however the deletion should not be greater than 25% of items measuring a constructs as requested by (Hair et al., 2013), hence RLG1=0.287, RLG2=0.329, TAC1=0.253 and TAC2=0.242 item will be used in the model because the deletion within each lantern variable in up to 25%, in case of Composite reliability a lantern variable should explain at least 0.5 of the variance in the observed variable shared with the construct, all the lantern variable of the study revealed reliability of >0.5, therefore the reliability is achieved, while the average variance extracted value of the lantern variable have exceed the threshold of >0.4. Cronbach alpha value of all variables of the study exceeds the yardstick of 0.6.

Discriminant Validity
Discriminant validity refers to the extent to which a particular latent construct is different from another latent variable (Duarte & Roposo, 2010). In the present study Discriminant validity was determined following Chin (1998)  In table 5, the correlations among the latent variable were compared with the square root of the average variances extracted (values in bold face). Table 5 also shows that the square root of the average variances extracted were all greater than the correlations among latent constructs, suggesting adequate discriminant validity (Fornell & Larcker, 1981).

Structural Model Assessment
The validity of the outer model (measurement model) gives room for evaluating the inner (structural) model (Henseler et al., 2009). The structural model is concerned about R 2 , coefficient and P-value and T-value by way bootstrapping (Hair et al., 2013), Additionally, the model predictive relevance (Q 2 ), as well as the effect size (F 2 ) of each variable, was explained by the structural model. The inner model is to evaluate the significance of loadings and paths coefficient that exists between variables (Barclay,  Thompson dan Higgins, 1995). Specifically, the structural model is aimed at model evaluation, as well as, an examination of the regression of the study. Source: Computed from Field Survey (2019) using Smart-Pls 3 Direct Relationship Note: ***Significant at 0.01 (two-tailed) Source: Computed from Field Survey (2019) using Smart-Pls 3

Test of Hypotheses
In testing a hypothesis, the coefficient is significance if T-value is equal to or greater than 1.96 at 0.05 significance level using two-tail test and also if the P-value is less than 0.05 (Hair, Black, Babin, & Anderson, 2010).
Null hypothesis 1 postulates that religiosity does not significantly affect patronage of Takaful among entrepreneurs in Sabon Gari, Kano State. The result revealed a coefficient of (β = 0.478 i.e., 47.8%) and statistically significant (P-value= 0.000 and T-value= 7.798), hence we reject the null hypothesis because the P-value is less than 0.05 and T-value is greater than 1.96. This finding is consistent with the previous findings of (Hassanuddin et al., 2016;Maiyaki & Ayuba, 2015).
Null hypothesis 2 postulates that trust and confidence not significantly influence patronage of Takaful among entrepreneurs in Sabon Gari, Kano State. The result revealed a coefficient of (β = 0.354 i.e., 35.4%) and statistically significant (P-value= 0.000 and T-value=5.541), hence we reject the null hypothesis because the P-value is less than 0.05 and T-value is greater than 1.96. This finding is consistent with the previous findings of (Hassanuddin et al., 2016;Maiyaki & Ayuba, 2015).
Null hypothesis 3 postulates that awareness does not significantly affect patronage of takaful among entrepreneurs in Sabon Gari, Kano State. The result revealed a coefficient of (β = 0.212 i.e., 21.2%) and statistically significant (P-value= 0.000 and T-value= 4.478), hence we reject the null hypothesis because the P-value is less than 0.05    /www.jurnal.unsyiah.ac.id/JAROE and T-value is greater than 1.96. This finding is consistent with the previous findings of (Hassanuddin et al., 2016;Maiyaki & Ayuba, 2015).

Assessment of Variance Explained in the Dependent Latent Variables
Another important criterion for assessing the structural model in PLS-SEM is the R-squared value, which is also known as the coefficient of determination (Hair et al., 2012). The R-squared value represents the proportion of variation in the dependent variable(s) that can be explained by one or more independent variable (Hair et al., 2010). Chin (1998) suggests that the R-squared values of 0.67, 0.33, and 0.19 in PLS-SEM can be considered as substantial, moderate, and weak, respectively. Table 7 presents the R-squared values.

Variable
Variance Explained (R 2 ) Patronage 0.579 Source: Computed from Field Survey (2019) using Smart-Pls 3 The results in table 7 of the R squared indicate that it is estimated that the predictors of the patronage of takaful explain (0.579 = 57.9 percent) of its variance. In other words, the error variance of patronage of takaful is approximately (0.421= 42.1percent), this signifies that religiosity, trust and confidence and awareness influences the patronage of takaful by 57.9 percent while other 42.1 percent changes in patronage takaful is course by other variables that are not stated in the model which are represented by the error term.

Assessment of Effect Size (F 2 )
Effect size indicates the relative effect of a particular independents variable on dependent variable by means of changes in the R-squared (Chin, 1998). It is calculated as the changes in R-squared of the variable to which the path is connected (Chin, 1998). Cohen (1988) describes F 2 values of 0.02, 0.15 and 0.35 as having weak, moderate and strong effects respectively. Table 8 shows the respective effect sizes of the independent latent variables of the structural model. From the result displayed in table 8, it could be deduced that all the variables have some exploratory power towards the dependents variable. Specifically trust and confidence and awareness have a moderate effect on patronage, while, religiosity, has a strong effect on patronage.

Assessment of Predictive Relevance
The study also applied Stone-Geisser Test of predictive relevance of the research model using blindfolding procedures (Geisser, 1974;Stone, 1974). Hence, because the dependent variable was reflective in nature, a blindfolding procedure was applied mainly to this dependent variable. In particular, a cross-validated redundancy measure (Q²) was applied to assess the predictive relevance of the research model (Geisser, 1974;Hair et al., 2013). According to Henseler et al., (2009), a research model with Q 2 statistic (s) greater than zero is considered to have predictive relevance. Additionally, a research model with higher positive Q 2 values suggests more predictive relevance. Table 9 presents the results of the cross-validated redundancy Q 2 test.   As shown in table 9, the cross-validation redundancy measure Q² for the dependent variable is above zero, hence, suggesting predictive relevance of the model. This finding further reaffirms and support the research model of the study that revealed the causal relationship between the variable of the study (i.e., awareness, religiosity and Trust and Confidence are the predictors of Patronage). Furthermore, the result testifies the model of the study as valid and reliable to be use by relevant stakeholders i.e., regulators and policy makers for predictions, decisions and analysis. Meanwhile, this finding is consistent with the findings of (Hassanuddin et al., 2016;Maiyaki & Ayuba, 2015).
The primary aim of this study is to identify the determinants of Takaful patronage among the entrepreneurs of Sabon Gari Markets. The specific objectives of the study are to assess whether, religiosity, trust & confidence and awareness have significant effects on Takaful patronage among entrepreneurs of Sabon Gari market. To achieve this, data is collected and analyzed, hence, the following conclusions were drawn. Religiosity positively influence Takaful patronage among entrepreneurs in Sabon Gari, market, therefore, this study concludes that religiosity is the most influencing determinant of Takaful patronage among entrepreneurs of Sabon Gari market, Kano state.
The result of this study also revealed that trust and confidence positively influence Takaful patronage among entrepreneurs of Sabon Gari market. Therefore, this study concludes trust and confidence reposed on Takaful companies is the second most determinant that improve the level of Takaful patronage among entrepreneurs in Sabon Gari market, Kano state. Furthermore, the result of this study revealed that awareness influence Takaful patronage among of entrepreneurs of Sabon Gari market. Therefore, this study concludes that awareness is the third most determinant factor that improve the level of Takaful patronage among entrepreneurs in Sabon Gari markets, Kano state.
The recommendations were scrutinized based on the findings of the study, hence, the following recommendations are made in order to provide adequate measure to complement the proper Takaful patronage among the entrepreneurs in Sabon Gari market, Kano state: (i) Strategic awareness of benefits: The Takaful providers have to create not just awareness of its existent of the products, but aggressive enlightens campaigns by all stake holders, on door-to-door basis, to sensitize the business community on the enormous potentials of Takaful to the growth and survival of business enterprises, also there is need for publicity via all outlets of media (radio programs, television, social media, bill board, etc.).
(ii) There is need to involve the religious scholars (Ulama) for the needful patronage of the Takaful, by enlightens the Umma about the benefit of participating in Takaful and also reward to obtain by engaging in Shari'ah compliant mitigating risk, this could be achieved through Friday sermons.
(iii) There is need for creating business trust and confidence between the entrepreneurs and the operators by providing efficient and effective service delivery through sound disclosure of their record and ensuring sound and easy claim settlement of losses which can be even made public. The stakeholders in the industry should provide the needed capital investments which can enable the Takaful business to be well established and be capable in winning more customers.
(iv)There is the need for operators of Takaful in conjunction with traders' union ensure joint organization of regular customer awareness forum, which should be doorto-door at their markets in reaching out to the entrepreneurs.
This study was able to assess the determinants of Takaful patronage among entrepreneurs in Sabon Gari market, Kano state. However, there is need to widen the scope of the study to different markets places within Kano state and beyond so that proper

Conclusions
Takaful Patronage, Entrepreneur http://www.jurnal.unsyiah.ac.id/JAROE generalization can be made. There is also need to introducing new independent, mediating or moderating variables by testing it effect to the used model through analyzing it reliability and validity. Takaful Patronage, Entrepreneur http: //www.jurnal.unsyiah.ac.id/JAROE