East African Journal of Arts and Social Sciences

The study was conducted in the Wanging’ombe district council in the Njombe region, Southern highlands of Tanzania. The objective of the study was to assess the determinants of women’s control over their earnings. The study employed a cross-section research design. Data were collected from 150 women who engaged in various economic activities. The structured interview method was used to collect data using a structured questionnaire. Data were processed and analysed using IBM SPSS statistics version 20 and Stata version 11. The analysis involved descriptive analysis in exploring the characteristics of women in the study area, a chi-square test for association


ABSTRACT
The study was conducted in the Wanging'ombe district council in the Njombe region, Southern highlands of Tanzania. The objective of the study was to assess the determinants of women's control over their earnings. The study employed a cross-section research design. Data were collected from 150 women who engaged in various economic activities. The structured interview method was used to collect data using a structured questionnaire. Data were processed and analysed using IBM SPSS statistics version 20 and Stata version 11. The analysis involved descriptive analysis in exploring the characteristics of women in the study area, a chi-square test for association, and a binary logistic regression to explore the predictors. The results revealed that more than half (54.7%) of women reported that their husbands/partners decide for them on the use of money earned. Three factors, namely ownership of the house, kind of earning activity (source of income), and amounts of money earned per month are predictors of the control over women's cash earnings in the household. The study concluded that, although husbands have less chance to decide on income earned by women who engage in economic activities other than farming and those who earn a higher income, household resources including women's earnings are controlled by husbands to a large extent. The local government authorities in Wanging'ombe and nongovernment organisations interested in women's empowerment should intensify efforts to address violence against women, especially economic abuse. This will enable women to engage in different economic activities and be employed in informal and formal sectors.

INTRODUCTION
Women's control over resources is a vital element in achieving women's empowerment agenda. Empowerment of women has been defined by different scholars to generally mean women gaining a significant share of control over resources including financial resources like money, knowledge, information, and ability to make decisions in the household, community, society, and at the national level (Menon et al., 2020). Empowerment is all about changing an individual from being powerless to a position of power that entails enabling someone to have control over his or her own life. Economic empowerment of women has particularly been portrayed as more foundational to foster more empowerment in many other aspects of the lives of women. Economic empowerment is seen as a cornerstone in achieving sustainable goals, especially Sustainable Development Goal 5 which focuses on achieving gender equality and empowering all women and girls (UN Secretary 2016, as cited in del Campo andSteinert, 2020). Several benefits have been associated with women's economic empowerment in different parts of the world both in developed and less developed countries. The benefits include a reduction of gender-based violence, an increase in women's ability to make decisions, improved household livelihood, and reduced poverty among women and the whole society (Ndossi et al., 2022;Opata et al., 2020).
Women's control over financial resources including their earnings is associated with economic independence. It depends on how the decisionmaking process goes, which reflects the resource control management (Abate, 2019). This means that women have control over them if they have access to and can make decisions as was observed in the study of the empowerment of women through cooperatives in India (Bharti, 2021). The global efforts stem from the fact that more than one billion people in the world today, particularly in developing countries, live in multidimensional poverty, and the majority of them are women (OPHI, 2018;Women UN, 2022). Poverty affects the household as a whole because of the gender division of labour and responsibilities for household welfare. Women bear a disproportionate burden of poverty, attempting to manage household consumption and production conditions in persistent inequality to access and control productive resources, especially in agriculture where the majority are employed (Anderson et al., 2021). The efforts to step up women's empowerment in Africa have been a big agenda all over the continent. The African Women's Development Fund (AWDF), which involves itself in fundraising and grant-making initiatives to support the work of the African Women's movement, is one example.
The government of Tanzania, like other developing countries, recognises the importance of women's empowerment and has been formulating policies and programs to support women in their socioeconomic activities in order to improve the livelihood of women in Tanzania (Mayoux, 2020). Since 2015, it has initiated the women in development approach to reducing women's workload and coordinating women towards development through different economic activities. Tanzania's government has been emphasising empowering women by providing more chances in decision-making, education, property ownership and access to loans to improve their livelihood status (Mayoux, 2020). The Tanzania government's efforts include the establishment of the Women Development Fund (WDF) through Local Government Authorities (LGAs) through district community development officers (DCDOs) that help to link women groups to get soft loans in order to run their own activities that help them to get their income.

Benefits of Women's Access and Control over Resources
Generally, evidence from low and middle-income countries (LMICs) have shown the association between women's access to and control over household resource and proper allocation of the resources to household needs. For example, among female-headed households, a large proportion of household income was spent on household needs, particularly food and education, while in maleheaded households, some spent household income on personal needs such as alcohol (Doss, 2006;Anderson et al., 2021). According to a study that was conducted in Nigeria, women's earnings were used mostly to meet household food, children's clothes, pay school fees, and energy fuels (Opata et al., 2020). The study in northern Tanzania in the Mara region confirmed that women who owned more agricultural land felt more power than women who had no land (Ndossi et al., 2022).

Status of Women's Control Over their Earnings
Control over resources including money earned by women and other household resources is of great concern in the world as far as women's empowerment is concerned. Studies in various places in the world show the difference in the degree of control over resources and decision-making power of women (Abate, 2019). According to the study in Ethiopia, Abate shows that major sources of income are controlled by husbands while wives control small income sources. This implies that women in this society have little chance concerning access to resources, decision-making, and control over them. However, at the domestic level, they enjoy limited power to make some decisions. In India, the study by Bharti (2021) also reveals that gender inequality exists due to a lack of access to productive resources and low or lack of education among many women. Similar findings were found in Bangladesh, where control over income by women increased as the level of education increased (Roy et al., 2017).
In Tanzania, according to TDHS -MIS 2015 -16 only 36% of married women who received cash earnings decided on how the income should be used, while 55% made the decisions jointly with their husbands and 9% mainly made by husbands (United Republic of Tanzania [URT], 2016). Also, women are often faced with discrimination in terms of wages and advancement opportunities. A study that was conducted in Mwanza, North-West Tanzania on the understanding of economic abuse against women revealed that women's earnings are controlled by their husbands in some ways; for example, when a husband directs his wife on how to use the money she has. Furthermore, tracking how women spend household income also reflects the less power women have in decision-making (Serpa et al., 2021). Such economic abuse limiting control of women over household resources including their earnings has far-reaching consequences such as household necessities and poverty among women, their families, and society at large. Unfortunately, in Tanzania, studies conducted have focused on women empowerment in general, especially in various productive activities such as agriculture (Ndossi et al., 2022) and engaging in women's groups but not specifically focusing on the state of women's control over their earnings and how they decide on what to do with them. Therefore, this study aimed to assess the determinants of control over women's cash earnings in the household.

Study Area
The study was conducted at the Igosi ward in Wanging'ombe District Council. Igosi is a rural ward in the Wanging'ombe district in the Njombe region of the Tanzanian Southern Highlands. According to the 2012 Population and Housing Census report, there were 7,204 people in the ward, where 3,900 were females and 3,304 were males (URT, 2013). The ward is found in the latitudes 9.2537503 degree south and longitudes 34.504685 degrees east. Furthermore, the ward is located on an undulating plateau intersected by deep valleys, and it is bordered by the villages of Ujindile in the north, Mdasi and Makoga in the east, Mwilamba and Utelewe in the south, and Mafinga and Kipengere in the West. Rainfall varies from year to year. Low temperatures and frequent cloudiness ensure sufficient rainfall to support a variety of cash and food crops and to give a green appearance all year round. The ward is located on the undulating Makambako Mwanging'ombe plateau, intersecting by relatively deep river valleys such as Fukulwa, Nyamadoda, and Chalove. The ward has few social and economic infrastructures. The services available are a secondary school, primary schools, a health centre, and telecommunication. The people in the area produce paddy, maise, and beans which are the staple food for the community (URT, 2020).

Study Design, Sampling Frame, and Procedures
The study used a cross-sectional design that collected data at one point at a time. The sampling frame of this study was a list of all households with women in the Igosi ward in the Wanging'ombe district council. The sampling unit was a single woman. To avoid bias, a simple random sampling was used to obtain respondents.

Data Types, Sources, and Collection Methods
The study used quantitative data collected from primary and secondary sources. Primary data were collected from women at the household level. A structured questionnaire was employed to get the required information and administered through face-to-face interviews. Secondary data were obtained from different documents including various reports, records, books, and journals.

Sample Size
The sample size of 150 women was determined by using the formula by Kothari (2004); the critical value (Zα/2) of 1.96 at a 95% confidence level, the proportion of factors that tend to affect women assumed (P) to be 50% accounting for a maximum level of variability in the attributes and decided marginal error (e) of 8% which is the allowable error,

Study Variables
The dependent variable of the study is a decision on the use of money earned by women (wife) in the household, which is measured in dichotomous such that the decision can be made either by a woman (wife) and both or by husband/partner. If the decision is made by a woman (wife) and both then the dependent variable (Y = 0) and if the decision is made by a husband/partner then the dependent variable (Y = 1). On the other hand, the independent variables of the study were age, education level, partners' level of education, household size, occupation, partners' occupation, listening to the radio at least once a week, watching TV at least once a week, reading newspapers at least once a week, ownership of the house, income earning activity, records of cash received, amount of income earned per month.

Data Processing, Analysis, and Presentation
The collected data was verified, entered, edited, and coded to remove discrepancies. The analysis of data was done by IBM SPSS statistics version 20 and STATA 11. The analysis involved descriptive such as frequency and percentages to explore the characteristics of women in the study area and the results were presented by tables and figures. Moreover, the study involved inferential analysis such as the chi-square test, which was used to explore the association between dependent and independent variables. Also, a binary logistic regression was used to measure the determinants of women's control over the usage of their income. Table 1 shows the characteristics of the respondents; the results revealed that 30.7% of the respondents were aged from 28 to 37, the majority (41.3%) had secondary education, and most of their partners had secondary education. Also, more than half (63.3%) of the respondents came from households with 5 to 9 members; 57% were selfemployed, while 51.3% of their partners were selfemployed. Regarding media exposure at least once a week, the majority (95%) listened to the radio, 92.7% watched TV, and 76.7% read the newspaper; also, a majority (96%) recorded their cash earnings.

Socio-Demographic Characteristics of the Respondents
Moreover, results in Figure 1 depicted that 45.3% of the respondents engage in farming activity, and the results in Figure 2 revealed that 32.7% % earn income per month from 100,000/= to 199,999/= and 31.3% earn income from 50,000/= to 99,999/=.

Decision on the Use of Money Earned by Women in the Household
The results in Figure 3 show the control over women's cash earnings in the households and who decides on the use of money earned by women in the households from different sources. The results revealed that more than half (54.7%) of women reported that their husbands/partners decide for them on the use of money earned, 27.3% reported that they decide jointly (both) with their husbands on the use of money and 18% of women reported that they decide themselves on the use of the money they earn.

Socio-Demographic Characteristics versus Decision on the Use of the Money Earned by Women in the Household
The cross-tabulation results in Table 2 revealed that husbands make decisions of income earned by their wives regardless of the age of the wife. Likewise, regardless of the level of education of wives, husbands make a decision based on their earnings. However, results have indicated that as the level of education increases, there is an increase in a joint decision on the income earned by a wife in the household, which may imply that education gives some sort of power to a woman in this case. Findings in Table 2 continue to show that partners or husbands who have higher education levels have even more control over women's earnings compared to husbands of low education levels. Other variables as seen in Table 2, such as household size, wife, or husband occupation, listening to the radio, or reading newspapers, do not make any difference as in all cases, husbands to a large extent decide on women's earnings in the household. A little difference however can be noticed in terms of house ownership and level of income. A woman who owns a house has an increased power to decide on her earning by 10% close to 12% by husband or partner, with increased bargaining power shown by joint decision (23%). On the other hand, women who earn low income (64%) of those who earn below 50,000/-Tanzanian shillings decide more on their earnings compared to those who earn higher incomes. The results show that as the income level earned by a woman increases the decision on how to use is made by the husband, but the level of joint decision increases, which may mean that the woman's power is increasing.
The chi-square test was used to assess the association between the dependent and independent variables. The results in Table 2 showed that the age ( 2 = 12.18, = 0.058 < 0.1), partners' occupation ( 2 = 11.82, = 0.019 < 0.05), listening to the radio at least once a week ( 2 = 8.44, = 0.015 < 0.05), an education level ( 2 = 21.81, = 0.001 < 0.05), kind of income earning activity ( 2 = 22.38, = 0.004 < 0.05), ownership of the house ( 2 = 14.5, = 0.001 < 0.05), reading newspapers at least once a week ( 2 = 5.78, = 0.056 < 0.1) and the amount of income earned per month ( 2 = 12.33, = 0.056 < 0.1) are statistically significantly associated with the decision on the use of income earned by women in the households. On the other hand, household size, occupation of a woman, watching TV at least once a week, and records of cash earned or received are not statistically significantly associated with the decision for the use of income earned by women in the households. Therefore, independent variables which were statistically significantly associated with the dependent variable (decision on the use of income earned by women in the households) were used in binary logistic regression to explore the factors that influence the control over the women's cash earnings which portrays who decides on the use of income earned by the women in the households.

Factors Influencing the Control over Women's Cash Earnings in the Households
The binary logistic regression analysis in Table 3 indicated that three (3) factors of ownership of the house, kind of earning activity (source of income), and amount of money earned per month were found to be statistically significant in influencing the control over women's cash earnings in the household. For odds ratios shows that a house whose owner is a husband/partner (male) is more likely the decision of income earned by a woman (wife) to be made by a husband/partner (P = 0.008, OR = 3.573, 95% CI = 1.405 -9.088), for the source of income; a woman (wife) whose source of income is pastoralism the decision on income earned to be made by husband/partner was less likely (P = 0.001, OR = 0.119, 95% CI = 0.035 -0.409) compared to a woman whose source of income is from farming. Again, for a woman (wife) whose source of income is from Village Community Bank (VICOBA) is found to be less likely the decision on the income, she earned to be made by her husband/ partner (P = 0.017, OR = 0.055, 95% CI = 0.005 -0.598) compared to a woman whose source of income is from farming. And a woman (wife) whose earnings are from entrepreneurship activity is less likely to make a decision on income earned to be made by a husband/partner compared to a woman whose earnings are from farming, although it is not statistically significant.
On the other hand, the level of income shows significance in decision-making regarding women's earnings. Results in Table 3 show that for a woman (wife) whose earning per month ranges from 50,000/= to 99,999/= is more likely that the decision on income earned will be made by a husband /partner (P = 0.091, OR = 4.219, 95% CI = 0.796 -22.374) compared to a woman whose earnings per month is below 50,000/=. For a woman whose earnings per month range from 100,000/= to 199,999/= is more likely the decision of income earned to be made by a husband/partner (P = 0.07, OR = 4.498, 95% CI = 0.885 -22.871) compared to a woman whose earnings per month is below 50,000/= and a woman (wife) whose earnings per month equal or greater than 200,000/= is more likely the decision to be made by a husband or partner (P = 0.095, OR = 4.602, 95% CI = 0.766 -27.649) compared to a woman whose earnings per month is below 50,000/=. Age, level of education, partners' occupation, listening to the radio, and reading newspapers were not statistically significant in influencing the women's control over the income in the households.

DISCUSSION
From the findings presented in Tables (from Table  2 to Table 3), it is clear that in most cases, control over women's earnings and household resources, in general, is under men or husbands. Just as in similar studies, control over resources is shadowed by economic empowerment, which is manifested in the access to productive resources and the ability to decide on how to use those sources (Anderson et al., 2021, Menon et al., 2020. For example, results show that where household assets such as a house are owned by a man (husband), the decision over a woman's earnings will be more often made by a husband. This ownership of a house may represent ownership of all other assets in the same way that if a woman does not own or share ownership, she is powerless over those resources and will not decide on them. Findings of a similar study in Ethiopia conform to the results of this study whereby the major sources of income are controlled by husbands while wives only control minor sources of income.
The study in Ethiopia further revealed that at the domestic level, women have some limited decisionmaking power and are in a disadvantaged position when it comes to decision-making (Abate, 2019).
The source of income in this study was a significant determinant of women's earning control in the household. According to the result, for women whose major source of income was farming the decision on their income was more likely to be made by husbands compared to other sources of income.
The findings further show that for women who own other income-generating activities such as smallscale entrepreneurship and engaging in VICOBA, decisions on their earnings are less likely to be made by husbands. Such results may imply that such women are more economically independent than those who engage in farming. The results are similar to findings (Bharti, 2021) in India on the role of cooperatives in economic empowerment, which revealed that the cooperatives increased access to and control over resources by women and increased their economic independence.
The level of income earned by a woman is an important factor in influencing a husband to make decisions on the usage of income earned. Results show that for a woman whose earning per month ranges from 50,000/= to 99,999/= is more likely that the decision on income earned will be made by a husband /partner compared to a woman whose earnings per month is below 50,000/=. For a woman whose earnings per month range from 100,000/= to 199,999/= is more likely the decision of income earned to be made by a husband/partner compared to a woman whose earnings per month is below 50,000/= and a woman whose earnings per month are equal or greater than 200,000/= is more likely the decision to be made by a husband or partner compared to a woman whose earnings per month is below 50,000/=. The results are contrary to (Eggers del Campo and Steinert, 2020), who revealed that women who earn more income have increased bargaining power and hence they can decide on their earnings and other household resources. Table 3 have shown that sociodemographic characteristics such as age, education, and occupation of partners are not statistically significant in determining the control over women's earnings. These findings are contrary to some similar studies which found that control over resources and the ability to decide is associated with the age and education level of women (Roy et al., 2017;Ndossi et al., 2022). The findings in the study area may imply that the root of the problem of inequality in control over resources and decisionmaking may be in cultural and social norms, as was observed in Ethiopia.

CONCLUSIONS
Generally, household resources including women's earnings are controlled by husbands. The findings of the study have demonstrated that the key predictors of women's control over their earnings are ownership of assets, particularly a house, a major source of income (kind of earning activity), and the number of earnings or level of income. The study has also revealed that husbands have less chance to decide on income earned by women who engage in economic activities other than farming and those who earn a higher income.

Recommendations
Based on the findings the inequality among men and women in access and control over resources is evident. Worse enough, women do not even have control over their earnings. As per the literature reviewed in this study, several benefits are accompanied by women's economic empowerment for women themselves, their families, community, society, and the nation at large. Henceforth the study recommends that; deliberate efforts be made by local government authorities to create awareness of the benefits of equal access to productive resources at the household level by both men and women. In this case, that strategy should be the abolition of customs and norms that put women in a marginal position. Also, local government authorities in Wanging'ombe and non-government organisations interested in women's empowerment should intensify efforts to address violence against women, especially economic abuse. This will give women to get more freedom to engage in different economic activities and be employed both in informal and formal sectors.