Determinants of poverty in rural cocoa farming communities in Ghana: unidimensional and multidimensional analysis

Agyeman-Boaten, Solomon Yaw Article Determinants of poverty in rural cocoa farming communities in Ghana: unidimensional and multidimensional analysis Cogent Economics & Finance Provided in Cooperation with: Taylor & Francis Group Suggested Citation: Agyeman-Boaten, Solomon Yaw (2024) : Determinants of poverty in rural cocoa farming communities in Ghana: unidimensional and multidimensional analysis, Cogent Economics & Finance, ISSN 2332-2039, Taylor & Francis, Abingdon, Vol. 12, Iss. 1, pp. 1-21, https://doi.org/10.1080/23322039.2024.2397808 This Version is available at: https://hdl.handle.net/10419/321583 Standard-Nutzungsbedingungen: Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden. Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen. Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Terms of use: Documents in EconStor may be saved and copied for your personal and scholarly purposes. You are not to copy documents for public or commercial purposes, to exhibit the documents publicly, to make them publicly available on the internet, or to distribute or otherwise use the documents in public. If the documents have been made available under an Open Content Licence (especially Creative Commons Licences), you may exercise further usage rights as specified in the indicated licence. https://creativecommons.org/licenses/by/4.0/ Cogent Economics & Finance ISSN: 2332-2039 (Online) Journal homepage: www.tandfonline.com/journals/oaef20 Determinants of poverty in rural cocoa farming communities in Ghana: unidimensional and multidimensional analysis Solomon Yaw Agyeman-Boaten To cite this article: Solomon Yaw Agyeman-Boaten (2024) Determinants of poverty in rural cocoa farming communities in Ghana: unidimensional and multidimensional analysis, Cogent Economics & Finance, 12:1, 2397808, DOI: 10.1080/23322039.2024.2397808 To link to this article: https://doi.org/10.1080/23322039.2024.2397808 © 2024 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group View supplementary material Published online: 01 Sep 2024. Submit your article to this journal Article views: 1078 View related articles View Crossmark data Citing articles: 1 View citing articles Full Terms & Conditions of access and use can be found at https://www.tandfonline.com/action/journalInformation?journalCode=oaef20 DEVELOPMENT ECONOMICS | RESEARCH ARTICLE Determinants of poverty in rural cocoa farming communities in Ghana: unidimensional and multidimensional analysis Solomon Yaw Agyeman-Boaten a,b a Department of Economics, University of Ghana, Legon, Ghana b School of Humanities and Education, Knutsford University College, East Legon, Ghana ABSTRACT Poverty is a rural phenomenon due to the dominance of subsistence farming in rural communities. This study compares the estimates of unidimensional and multidimensional methodologies to analyse the factors that influence the poverty levels of cocoa farming households predominantly in rural Ghana. A census was conducted in the Chorichori community in Ghana using a structured questionnaire to gather information from 386 cocoa farming households. The multidimensional poverty index and expenditure-based poverty measures were used to estimate a bivariate probit regression to find the determinants of cocoa farming households’poverty. The study’s outcome indicates that poverty among the cocoa farmers is jointly determined, unidimensionally and multidimensionally, by the access to healthcare, household child deaths, household’s school-age child not in school, access to farm inputs, and the age of the household head. Whereas the education level of the household head, frequency of ill-health, use of external labour, migration status, and relationship to the household head were significant in determining multidimensional poverty, the number of household members, cooperative union membership, access to farm water, occupational diversity, household access to financial credit, and the marital status of the household head significantly determined unidimensional poverty among the cocoa farmers. Even though both poverty measures produced fairly different results, the study’s findings showed the mutual and exclusive importance of the unidimensional and multidimensional poverty approaches in determining poverty and formulating good developmental policies for cocoa farmers. Therefore, selecting an approach should be based on prevailing circumstances, such as differences across locations and within households or entities. IMPACT STATEMENT This research provides critical insights into the complex nature of poverty among cocoa farming households in rural Ghana. By comparing unidimensional and multidimensional poverty measures, the study reveals significant disparities in how poverty is assessed, offering a more comprehensive understanding of the socio-economic challenges faced by rural communities. This dual approach not only uncovers the socioeconomic factors that contribute to poverty but also reveals discrepancies and complementarities between different poverty measurement methods, offering a nuanced perspective that can enhance the accuracy of poverty assessments. The findings highlight the need for policymakers, development agencies, and social planners to target interventions that address the specific factors contributing to poverty in rural areas, thereby improving the effectiveness of poverty alleviation programs. By informing more holistic and inclusive policies, this study supports efforts to reduce poverty among rural cocoa farmers, ultimately contributing to broader economic development and social equity in Ghana and similar contexts across developing countries. ARTICLE HISTORY Received 22 April 2024 Revised 1 August 2024 Accepted 23 August 2024 KEYWORDS Bivariate probit regression; cocoa farmers; determinants; multidimensional poverty; rural areas; unidimensional poverty SUBJECTS Development Economics; Agricultural Economics, Economics and Development, Microeconomics, Econometrics, Economics, Finance JEL CLASSIFICATIONS C31; I32; O13; Q12; Q14 CONTACT Solomon Yaw Agyeman-Boaten [email protected],[email protected] Department of Economics, University of Ghana, Legon, Ghana Supplemental data for this article can be accessed online at https://doi.org/10.1080/23322039.2024.2397808. ß2024 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. The terms on which this article has been published allow the posting of the Accepted Manuscript in a repository by the author(s) or with their consent. COGENT ECONOMICS & FINANCE 2024, VOL. 12, NO. 1, 2397808 https://doi.org/10.1080/23322039.2024.2397808 Introduction Poverty is a major concern for all governments globally. In the 1990s, development literature focused on how economic growth can be used as a strategy to reduce poverty and noted progress in many countries’poverty reduction with rapid and high rates of growth (Dollar & Kraay, 2000; World Bank, 2000). Subsequent literature argued that poverty reduction was not achieved primarily through the structure of growth and income inequality but specifically through growth rates (Asogwa et al., 2012; Mellor, 1999; Ravallion & Datt, 1996). Recent literature advocates for using innovation systems to reduce poverty (eg Onumah et al., 2023). Dzanku (2015) noted that ‘growth in agriculture is reported to be more pro-poor than growth in other sectors’. Poverty is a rural phenomenon (De Janvry & Sadoulet, 2000; GSS, 2014, 2018). According to Ghana Living Standard Survey (GLSS) 5–7 reports 1 , rural areas contributed 85.3%, 78% and 83.2% of total poverty in Ghana, despite the different degrees of poverty among rural dwellers. In Sub-Saharan Africa, rural dwellers are predominantly farmers and distinctions are classified by the ecological zones of rural areas (forest, savannah and coastal). Forest areas are suitable for cocoa cultivation, and hence cocoa farmers dominate the forest zones in Ghana. The trend of poverty incidence for all the rural zones and national poverty in Ghana increased from GLSS 1–2 and then decreased in GLSS 3 reports. However, except for the GLSS 3 report in which rural forest areas had a higher poverty incidence (26.5%) than coastal areas (22.6%), the earlier rounds of GLSS reports on poverty incidence by locality in Ghana (GLSS 1–3) show that rural forest areas had a lower incidence of poverty than other zones and the national poverty incidence (Coulombe & McKay, 1995). Likewise, Hern andez-N u~ nez et al. (2022) found among Colombian rural households that Cocoa farmers have the highest capital endowment and level of welfare compared to diversified farmers and new cocoa farmers. The current estimates of poverty incidence in Ghana by locality are measured by dividing total household consumption by the number of adult equivalents in the household, which is different from calculating the total household expenditure, per capita, in constant prices, adjusted for the effects of differential recall error between GLSS 1, 2, and GLSS 3. Despite the differences in the methodologies, the current trends of GLSS results are not different from earlier ones (Coulombe & McKay, 1995; GSS, 2014,2018). GLSS 5–7 report that rural forest areas have the lowest poverty incidence among the different rural zones, rural areas, and the national poverty level. Even with extreme poverty, only the rural coastal zone recorded a lower incidence of 9.6% compared to 12.6% in the rural forest. Nonetheless, the rural forest zone has recently followed the general trend of reduction in Ghana’s poverty incidence. Although the contribution of rural forest zones to the total poverty of Ghana has been lower than that of rural savannah zones, it has been far above that of rural coastal zones. Rural forest areas, on the other hand, increased their contribution to total poverty from 29.1% to 30.3% from GLSS 5 to 6, before falling to 25.3% in GLSS 7 (GSS, 2018). Furthermore, evidence suggests that most of the world’s impoverished live and work in rural areas (IFAD, 2001; World Bank, 1999); hence, reducing overall poverty depends largely on reducing rural poverty. In Africa, programmes and policies implemented to serve poverty reduction strategies included the Structural Adjustment Programme and Poverty Reduction Strategy Papers (PRSPs). In Ghana, additional strategies such as LEAP (Livelihood Empowerment Against Poverty), the Capitation Grant, the School Feeding Programme, the free distribution of school uniforms, exercise books, and textbooks, the elimination of schools under trees, and the free senior high school education are just a few of the programmes that have been implemented to alleviate poverty among the vulnerable population. Consequently, it is critical to adopt a poverty alleviation strategy by primarily assessing the socioeconomic characteristics to determine the factors that influence the poverty levels of rural dwellers who are predominantly engaged in the agricultural sector. Based on the capabilities approach by Sen (1979,1985), many scholars have argued that poverty is multidimensional and includes dimensions of basic needs: education, health, and other indicators of standard of living (eg De Janvry & Sadoulet, 2000; Fergusson et al., 2001). Current studies also take into account intangible factors such as the satisfaction of being employed, empowerment, community ties, legal and human rights, and political freedoms (World Bank, 2000). As such, the World Bank and other international bodies in recent years have revised their earlier reliance on measuring poverty using a unidimensional approach, which views welfare according to income or expenditure (Ningaye et al., 2011). Consequently, recent studies favour the multidimensional poverty measure over the traditional 2 S. Y. AGYEMAN-BOATEN expenditure-based measure for analysis and policy formulation due to its inclusion of many indicators of life. Moreover, cocoa farmers, predominantly in rural forest areas, face many challenges, including crop diseases (cocoa swollen shoot virus disease (CSSVD), black pods), lack of social amenities, and inadequate extension officers. For instance, the cocoa swollen shoot virus disease has caused a massive decline in the living standards of many cocoa farmers in Ghana in recent years (Agyeman-Boaten & Fumey, 2021). Although recent records have shown a decline in poverty in Ghana, the poverty phenomenon persists in rural areas dominated by farmers. Despite all these challenges, it is difficult to find current studies that comprehensively investigate the factors determining cocoa farmers’poverty and the differences in the estimates of the two poverty measuring techniques (money-metric and non-money metric). Determining the socioeconomic well-being of Ghana’s cocoa farmers is difficult without empirical evidence. Given the critical role of cocoa farming in the rural economy and the persistent poverty in these regions, more accurate assessments to inform effective policy interventions are needed. This study aims to assess the factors that influence the poverty levels of cocoa farming households in rural Ghana by comparing the findings of both unidimensional and multidimensional approaches. The specific objectives include (1) identifying and examining the socioeconomic characteristics of rural cocoa farmers that influence their poverty levels, and (2) comparing the findings of the multidimensional poverty measure to the unidimensional (incomeor expenditure-based) measure. To achieve this, quantitative data for the study were collected through a survey from Chorichori (a cocoa farming community in Ghana). The bivariate probit regression analysis was employed for the study estimates, with multidimensional poverty index (MPI) and expenditure-based approaches as the measures for the dependent variables (multidimensional poverty and unidimensional poverty), respectively. This study contributes significantly to literature and policy-making, focusing on poverty measurement and alleviation among rural cocoa farmers in developing countries. Firstly, it contributes to understanding the dynamics between the unidimensional and multidimensional poverty methods using data from rural cocoa farmers. Through empirical research, this study uniquely juxtaposes monetary and non-monetary welfare indicators from a rural population in a developing country in Africa. This approach highlights the discrepancies and complementarities between the two poverty methods. Secondly, it examines the factors determining the poverty levels of rural cocoa farmers, which are under-explored in existing literature. The findings are intended to inform governments, poverty analysts, and social planners to target the most important factors for alleviating poverty in the rural sector. The remainder of the paper is organised as follows. The next section provides an overview of poverty in Ghana, followed by a brief review of the literature. The methodology and data collection procedure are presented in section three. The study’s findings, analysis, and discussions are presented in the fourth section. The final section summarises the findings and makes policy recommendations for poverty alleviation. Poverty in Ghana Poverty in Ghana experienced a significant decline between 1988/1989 and 1991/1992 after increasing between 1987/88 and 1988/89 (GLSS 1–3). It rose from 36.8% in 1987/88 to 41.8% in 1988/89 before falling to 27.9% in 1991/92 (Coulombe & McKay, 1995). According to the Ghana Statistical Service’s GLSS report, new poverty lines and price deflators have demonstrated a drop in national poverty rates across all poverty indicators from 1992 to 2017 (GLSS 3–7) (GSS, 2018). Between 1987 and 1992, poverty decreased dramatically between 1988/89 and 1991/92, more than offsetting the increases in all rural areas between 1987/88 and 1988/89. Ghana achieved the first Millennium Development Goal (MDG) of halving poverty from 51.7% of the population in 1992 to 24.2% of the population in 2013. It is poised to achieve Sustainable Development Goal (SDG) 1, which calls for an end to poverty in all its forms everywhere. From 1992 to 2013, the headcount ratio declined from 56.5% to 24.2%. In recent GLSS reports, poverty incidence has been reduced slightly to 23.4%. Additionally, extreme poverty decreased from 33.2% in 1992 to 8.2% in 2017. As a result, nearly a quarter of Ghana’s population is poor, with less than a tenth of the population living in extreme poverty. Generally, the dynamics of Ghana’s current poverty condition, from GLSS 5 to GLSS 7 (2005/06–2016/17), show that poverty is still predominantly a rural phenomenon, with rural savannah having the highest incidence as the poorest locality. Since 1991/92, COGENT ECONOMICS & FINANCE 3 poverty has been more widespread in other urban areas than in Accra, but poverty levels in urban areas are generally lower than in any of the three rural localities (Coulombe & McKay, 1995; GSS, 2014,2018). The GLSS 7 report indicated that self-employed household heads engaged in the agricultural sector contribute the highest to poverty in Ghana, with their highest poverty incidence. Meanwhile, private employees and self-employed household heads working in industries other than agriculture are less likely to be poor than those who work in agriculture (GSS, 2018). Literature review There are many theories of poverty discussed and classified differently in the literature. A common classification in recent literature is based on the root causes of poverty. Well-known classifications include theories based on the cause of poverty in individual deficiencies (conservative) and those that attribute poverty to broader social phenomena (liberal or progressive). The classification can be extended into five causes of poverty, which are the theories that originate from individual deficiencies, cultural belief systems, political or economic distortions, geographical disparities, and cumulative and circumstantial origins (Bradshaw, 2007). This study deliberates on two phenomena of poverty:cultural/behavioural and structural/economic. The cultural thesis linked to the individual theory of poverty, argues that poverty is rooted in the behaviour of the poor and deficient character that undermines success and economic well-being (Aikaeli, 2010). As a result of the poor’s dysfunctional values concerning mainstream social norms on family, education, and work, poverty is passed down from generation to generation, resulting in rising rates of violence, crime, substance abuse, and family breakdown (Aikaeli, 2010). This argument is criticised for blaming and stigmatising the poor, as it merely outlines the symptoms of poverty without addressing systematic causes. The ‘structural or economic’school of thought also argues that most poverty results from favours for a certain group over others, which ensues from structural factors in economies and institutional environments. Holzer (1991) noted that an individual’s poverty can be ascribed not just to personal attributes, but also to the conditions in which they live. Economic opportunities, such as productive employment and infrastructure, and social amenities, may be among these variables. Marxists endorse this viewpoint that poverty in the developing world highlights the developed world’s past exploitation (Aikaeli, 2010). However, because some present prosperous countries were once colonies, this theory has been criticized as lacking adequate proof. Notwithstanding these two main arguments, people, cultures and institutions in certain geographical regions face disparities in essential resources necessary to generate welfare and income. Furthermore, they are limited in their ability to advocate for equitable redistribution of resources. The structural approach and the theory of poverty due to geographical disparities are emphasised in this study. It ties rural income to driving production factors such as ecological conditions, land development, and the size and education of the labour force (Aikaeli, 2010; Samuelson & Nordhaus, 1992). Empirically, some studies have employed different methods to analyse poverty. Mekonnen (2024) used a new vulnerability to income and multidimensional poverty estimation index to measure the likelihood of individuals falling into and remaining in poverty. The study found that in Ghana vulnerability to income and multidimensional poverty were 37.9% and 56%, respectively with 23.4% of the 37.9% at risk of falling into poverty and 14.57% at risk of remaining in poverty. Costa (2003) also employed rank correlation analysis to determine how the unidimensional and multidimensional techniques reveal the presence of two distinct groups of poor households. He concluded that income-based evaluation only delivers a partial picture of poverty. Comparing multidimensional and unidimensional poverty estimates for the Mandi Bahuddin District in Pakistan for 2010 and 2014, Khan et al. (2020) identified mixed results in comparing unidimensional estimates with multidimensional poverty in Asia and Africa. They recommended a holistic strategy to identify multidimensional poverty in the social sector because the relative proportion of educational and health poverty towards MPI remained larger. Bersisa and Heshmati (2021) used the logit model to investigate the determinants of household poverty status in Ethiopia, employing both unidimensional and multidimensional measurements. They found that intensity, severity, and depth of poverty differ significantly between the two measurements. Meanwhile, Belhadj and Limam (2012) revisited two issues raised by the use of fuzzy measures rather than traditional poor/non-poor 4 S. Y. AGYEMAN-BOATEN dichotomous measures, proposing the use of fuzzy monetary and non-monetary measures with poor and non-poor membership functions. Besides, many studies have tried to investigate the determinants of welfare or poverty among different groups of population with the expenditure-based approach to welfare (eg Asogwa et al., 2012; Babatunde et al., 2008; Eyasu, 2020; Owuor et al., 2007; Salami et al., 2017; Woldie et al., 2020). In the context of multidimensional poverty, Huluka (2024) utilized an ordered probit model and found the location, household’s head demographics (sex, literacy and age), family size, land area and region of residence as the key influencing factors of multidimensional poverty in Ethiopia, which is predominantly rural. However, limited literature has empirically highlighted the differences between the unidimensional and the multidimensional methods, especially with a dataset from rural areas and developing countries such as Ghana. Costa (2003) believes that comparing unidimensional and multidimensional methodologies is an appropriate step in poverty analyses and that if the differences are minor, unidimensional analysis should continue to play a significant role in poverty research. When the two groups of poor, on the other hand, show significant differences, an appropriate framework must be chosen. Methodology Research design Surveys serve as a primary methods for gathering data from respondents for quantitative studies, enabling the analysis of large samples before generalising the findings to the broader population (HesseBiber, 2010)). Consequently, the study utilises quantitative methods and data to draw conclusions on the numerical values. The study is part of a broader study to investigate the impact of cocoa swollen shoot virus disease (CSSVD) on the livelihood of cocoa farmers in Ghana (Agyeman-Boaten & Fumey, 2021). The study relied mainly on primary data; however, secondary information was also sourced from the Ghana Statistical Service, Ghana Cocoa Board and the Sefwi Akontombra District Assembly. 3.2. Study area The research area, Chorichori, is in the Sefwi Akontombra District of the Western North Region, near the Sui Forest Reserve. Figure 1 illustrates the district map of Sefwi Akontombra District, delineating the specific position of Chorichori, while Figure 2 shows the spatial layout of the Chorichori community. The community is located 281 km from Accra, Ghana’s capital city. From 1964 to 2001, the community received an average of 1461 mm of annual rainfall, and the climate is tropical, with an average temperature of 22 C–27 C (Boni et al., 2004). During the cocoa season, workers and traders from francophone Figure 1. Akontombra district population projection map. Source: Akontobra District Assembly. COGENT ECONOMICS & FINANCE 5 countries and other parts of Ghana, primarily from the north, used to do business in the study area. Unfortunately, the emergence of the cocoa swollen shoot virus disease in recent years has caused a massive decline in the economic activities of these migrants in and around the study area. Over 75% of the people in the area work in agriculture, farming cocoa and food crops including plantain, cassava, vegetables, and rice, among others (Schulte-Herbruggen, 2012). The climate, soil type, topography, and other characteristics of the study location make it ideal for cocoa cultivation. The community is connected to the national electricity grid. Other social amenities of the community include a basic school up to junior high level, a community centre, two boreholes, a river, two public pit-latrine toilets, a community health planning and services (CHPS) compound, nine churches and a mosque. The main cooking fuel used by the community dwellers is wood and the primary building materials are metal roofing sheets, cement floors, and mud or clay for the outer walls. About 84 percent of the people in the community grow cocoa, making it a fair representative for the rural cocoa farming communities. However, the survey does not claim to be completely representative of Ghana but anticipates that the findings based on this sample will serve as a useful estimate for rural cocoa farmers when broadly interpreted. Sampling procedure and technique Data collection The study data were collected primarily from a primary source in four weeks from January to February 2018. A structured questionnaire was administered to gather the primary data. The questionnaire aimed to capture various aspects of the respondents’socioeconomic conditions, particularly focusing on their engagement in cocoa farming. Sampling technique Mart ınez-Mesa et al. (2016) noted that sampling results, especially in studies aiming to estimate the prevalence of disease, can be affected by random or sampling error. To mitigate these errors and ensure the reliability of the data, an exhaustive approach within the cocoa farming community was adopted. Given the practical limitations such as time restrictions, ethical concerns, budgetary constraints, and logistical challenges, a non-probabilistic sampling method, specifically purposive sampling technique, was adopted. Target population The target population of the study comprised households within a rural cocoa farming community that was significantly affected by the cocoa swollen shoot virus disease. The goal was to obtain comprehensive data from this specific population segment. To achieve a thorough and representative data set, a census of the cocoa farming community was conducted. It involved interviewing all households in the community, but the survey was terminated if a household was found not to engage in cocoa farming. This approach ensured that only relevant respondents (cocoa farmers) were included in the survey. Figure 2. Chorichori: study site. Source: Google map. 6 S. Y. AGYEMAN-BOATEN Sample size and data collection process Data were collected from 386 rural cocoa farmers who were identified through a community survey conducted in Chorichori and its surrounding hamlets. The survey was conducted as a one-time exercise through face-to-face interviews. The method facilitated comprehensive data collection and allowed for clarifications and additional information to be gathered during the interviews. The objective of obtaining a total enumeration of households in the community was successfully achieved through this method. Ethical compliance To achieve full participation from respondents as well as meet research ethical requirements, the questionnaire was meticulously designed to assure respondents of confidentiality and anonymity. Moreover, respondents’participation in the interview was entirely voluntary, with each respondent providing verbal consent following a thorough explanation of the questionnaire’s purpose and procedures. This study was part of a broader study to investigate the impact of cocoa swollen shoot virus disease (CSSVD) on the livelihood of cocoa farmers in Ghana, and hence, ethical clearance was secured through the University of Ghana, Legon, Department of Economics Ethics Committee in adherence to the guidelines set forth by the Ethics Committee for Humanities. Consequently, the department issued an introductory letter to the stakeholder institutions and respondents, outlining the ethical considerations and objectives of the study. Instruments Quantitative data was collected with a paper-based structured questionnaire administered to households in the cocoa farming community. Access to the questionnaire and study data is facilitated through the data availability statement. The survey collected indicators on the cocoa farmers’household demographic or personal characteristics, capability (education and health), CSSVD and farm characteristics, economic resources (Food and Non-food expenditure), housing/dwelling-related information, and household asset ownership (living standard indicators). The questionnaire was designed following Ghana living standard surveys of the Ghana statistical service prior to 2018 and the new (acute) multidimensional poverty index for developing countries proposed by Alkire and Santos (2011). Analytical technique: estimation approach To ensure accuracy, consistency, and reliability, the primary data collected were screened. Codes were assigned to the responses extracted from the questionnaire before data entry. Descriptive analyses of the variables were conducted to analyse the socioeconomic characteristics of the cocoa farmers. The Multidimensional Poverty Index (MPI) and the unidimensional (expenditure-based) approaches were used as dependent variables to measure the poverty variable. Finally, the research objective to determine the predictors of the poverty of rural cocoa farmers was achieved with the estimation of a bivariate probit regression model. Afterwards, an analysis was made to compare the multidimensional and the unidimensional determinants of poverty among rural cocoa farmers. Study variables The dependent variable 1: the multidimensional poverty index (MPI) Typically, the conventional way of measuring poverty, welfare, or living standards of a household is likely to be unidimensional, preferably total household expenditure. Meanwhile, current studies have opposed using unidimensional approaches to measure poverty for multidimensional methods to capture certain quality-of-life attributes (Ningaye et al., 2011). Bourguignon and Chakravarty (2002)noted that the idea that all non-monetary aspects of welfare have no markets, and when they do, they are imperfect, is a key underlying feature of the multidimensional approach. Thus, there is no guarantee that they will be supplied with goods. Because some authors believe the money-metric approach is unsuitable because it focuses solely on monetary indicators, the Multidimensional Poverty Index (MPI) was used in this study to estimate the welfare or poverty levels of cocoa farmers in the study area and compared to expenditure-based unidimensional estimates. Initially, the global Multidimensional Poverty Index (MPI) was aligned with indicators used to track the Millennium Development Goals COGENT ECONOMICS & FINANCE 7 At the 1% level, the coefficient for the number of household members was also significant and indicated a negative relationship with unidimensional poverty, which is confirmed by Babatunde et al. (2008) and Salami et al. (2017). It indicates an increase in the number of members in a cocoa farming household by one person is associated with higher income and a decreased likelihood of being unidimensionally poor by about 0.309 units. Some studies argue that with a larger number of children and fewer active adult members of a household (unemployed and elderly), the burden of meeting the Table 6. Regression estimates of the rural cocoa farmer’s poverty status. Model_1a Model_1b Model_2a Model_2b eWelf MPI eWelf MPI Household Head Marital status 0.393 −0.008 0.451 0.073 (0.09) (0.097) (0.101) (0.092) Age of Household Head −0.020.014−0.025 0.018 (0.011) (0.008) (0.011) (0.009) Access to Financial Credit −4.12 0.126 −3.894 0.667 (0.523) (0.563) (0.571) (0.608) Migration Status of the Household −0.216 0.838 −0.144 0.757 (0.282) (0.355) (0.327) (0.365) Member of Cooperative Union −0.917 0.078 −1.214 −0.009 (0.446) (0.236) (0.431) (0.245) Access to Farm Inputs −0.922 0.331−1.113 0.377 (0.239) (0.188) (0.29) (0.185) Cocoa Land Size −0.029 0.022 −0.01 −0.024 (0.125) (0.11) (0.123) (0.109) Access to Farm Water 2.003 0.274 2.003 0.244 (0.549) (0.192) (0.56) (0.202) Number of Household Members −0.309 −0.019 −0.349 −0.008 (0.061) (0.041) (0.062) (0.047) Occupation Diversification 0.619 −0.267 0.509 −0.361 (0.218) (0.203) (0.239) (0.203) Use of External Labour Force 0.36 0.5320.245 0.512 (0.315) (0.31) (0.35) (0.321) Access to Healthcare 3.553 −6.241 3.084 −7.143 (0.484) (0.361) (0.406) (0.612) Ill-Health Frequency −0.09 −0.684 −0.04 −0.761 (0.154) (0.16) (0.167) (0.164) Child Death 0.711 7.722 1.576 7.523 (0.359) (0.448) (0.511) (0.473) Household’s child not in school 0.936 2.966 0.781 2.956 (0.293) (0.311) (0.314) (0.321) Household Head Education level −0.114 0.304 (0.14) (0.112) Relationship to the household head −0.048 −0.252 (0.092) (0.091) 1bn. Household Head Education level 2.Preschool/Primary −0.702−0.046 (0.361) (0.341) 3.Middle/JSS/JHS −0.503 0.563 (0.312) (0.278) 4.SSS/SHS/Voc/Tech/Com 0.516 1.281 (0.584) (0.456) 5.Tertiary 0.9 0.232 (0.691) (0.923) 1bn. Relationship to the household head 2.Spouse 0.06 −0.331 (0.377) (0.296) 3.Son or Daughter 0.224 −0.366 (0.317) (0.236) 4.Grandchild 0.123 −1.944 (0.41) (0.608) 5.Other Relative −1.117 −0.463 (0.816) (0.483) 6.Non Relative 0.125 −4.895 (0.759) (0.285) _cons −4.508 3.243 −3.874 4.233 (1.364) (1.24) (1.253) (1.132) athrho:_cons 0.333*0.292 (0.201) (0.202) Observations 381 381 Pseudo R 2 .z .z Robust standard errors are in parentheses.  p<.01,  p<.05, p<.1. 14 S. Y. AGYEMAN-BOATEN minimum cost of a household’s nourishment becomes heavier, resulting in higher levels of poverty (Asogwa et al., 2012; Omonona, 2001; World Bank, 1996). This study conversely suggests that households with more members were less likely to be unidimensionally poor on average if these members are more active. Contrary to the study of Huluka (2024), the negative relationship of the multidimensional model was not significant. Contrary to earlier studies by Adenuga et al. (2013) and Babatunde et al. (2008), the age of the household head was found to be significant in both models but at a marginal influence of 10% in the main model and 5% in the second model for both poverty measures. Nonetheless, Adeoti (2014) and Huluka (2024) found a significant relationship between age and multidimensional poverty. At the 1% level, the coefficient of education level of household heads showed a positive and statistically significant relationship with multidimensional poverty but was not significant with unidimensional poverty. The positive relationship indicates an increase in the level of education of cocoa farmers’household heads is associated with a 0.304 unit probability of being multidimensionally poor. However, Adenuga et al. (2013) found education level to be positive but not significant in influencing multidimensional poverty. Adeoti (2014) and Huluka (2024), on the other hand, found that education level significantly decreases the probability of being multidimensionally poor. Although the relationship between the education level of the household head and unidimensional poverty was not significant, World Bank (1996), Babatunde et al. (2008), Asogwa et al. (2012), Salami et al. (2017) and Eyasu (2020) found an increase in expenditure welfare to associate with a rise in education level; thus, a negative effect between education levels and unidimensional poverty suggests a rise in education level leads to a reduction in poverty level. The ambiguous results regarding the relationship between education levels and both MPI and unidimensional poverty prompted further investigation, which required the second model. A further examination indicated that some education up to the primary level (about 6–8 years, including preschool) was marginally significant (at the 10% level) in lowering the cocoa farmers’unidimensional poverty. Higher levels were not significant. On the other hand, middle school or junior high (secondary) school and senior high (secondary) school, vocational or technical school leavers who are rural cocoa farmers are likely to be poor multidimensionally at a 5% and 1% significance level. Other levels were not significant. Farmers’understanding of new technologies is believed to improve with education, enhancing their knowledge of cocoa production and management practices. Some education level of the cocoa farmer household heads builds their capabilities to appropriately follow the best practices in cocoa farming, such as the right application of fertilizer, weedicide, and pesticides, which reflects high crop yield and their ability to spend more. However, higher education does not influence the unidimensional poverty of rural cocoa farmers. On the multidimensional front, attaining a secondary level education increases the probability of a rural cocoa farmer being poor. The results indicate that it would benefit an individual with higher education to engage in large-scale cocoa production with improved technologies and management practices rather than engaging in subsistence farming which is predominant in rural cocoa farming communities. Household members’relationship with the household head was also significant and negatively influenced only multidimensional poverty at the 1% level. The coefficient suggests that the relationship of the household members to the household head is associated with a lower household’s multidimensional deprivation by approximately 0.252 units. For further understanding of the relationship, the results based on the second model showed that being a grandchild or non-relative to the cocoa farming household head were the only significant indicators to determine multidimensional deprivation. The negative coefficients (−1.944 and −4.895) signify the low probability of the grandchild or non-relatives being multidimensionally poor relative to the household head. No other relationships to the household head were found to be associated with multidimensional poverty or expenditure-based poverty. Even though the coefficient of the marital status of household heads (0.393) was highly significant in the expenditure model at the 1% level, it was not significant in the multidimensional model. It implies that a household head’s marital status is associated with unidimensional poverty through expenditure but does not influence multidimensional poverty. This finding is consistent with the studies of Adenuga et al. (2013) and Bersisa and Heshmati (2021). Likewise, Adeoti (2014) also confirmed the insignificance of the household head’s marital status in a multidimensional model. COGENT ECONOMICS & FINANCE 15 The coefficient for households that migrated to the study area was significant at a 5% level and negatively related to the multidimensional poverty but not the expenditure-based poverty. The coefficient (0.838) indicates that settlers are more likely to be multidimensionally poor than indigenous households. This finding could be partly attributed to the fact that while indigenous households accumulate and maintain assets such as comfortable homes and improved living standards in the study area, settlers may prioritise these aspects elsewhere, hence deprived of those indicators. This finding is supported by Amfo et al. (2022,2023). The results showed that households’access to financial credit negatively and significantly influenced expenditure poverty at the 1% level. However, it did not significantly influence the depth of multidimensional poverty, as confirmed by Adenuga et al. (2013). The results signify that households that households with access to loans experience lower unidimensional poverty by approximately 4.12 units. The ease with which such funds can be utilized in numerous household activities may explain the large influence that the households’access to credit has on their expenditure welfare (Apata et al., 2010). Some authors argue that access to credit enables households to acquire more productive resources for their household enterprises and meet consumption expenditures such as medical costs, school fees, food, and social emergencies, thereby enhancing their income-generating capacity and overall welfare (Apata et al., 2010; Asogwa et al., 2012; FOS, 1999; Owuor et al., 2007; Omonona, 2001). The coefficient for the cocoa farmer households’membership in a cooperative union was negative and significantly influenced unidimensional poverty at a 5% level but was not significant multidimensionally. It deviates from Adenuga et al. (2013) findings that membership in an association was negative and significant in influencing the depth of multidimensional poverty. The results suggest that a cocoa farming household that joins a cooperative union is associated with higher expenditure wellbeing of approximately 0.917 units. This indicates that households that join a cooperative union are more likely to improve their money-metric welfare than those that do not join. Various benefits through credit facilities, access to improved production inputs, and access to information, among others, extended to members could be the reason for their improved well-being. As a form of social capital, Hern andez-N u~ nez et al. (2022) advocated for the participation of rural household members in local organisations. In the expenditure-based poverty model, the coefficient for the households’access to farm water was positive and significant at a 1% level, but not significant in the multidimensional poverty model. The coefficient indicates that a cocoa farmer’s household with access to farm water is associated with expenditure-based poverty by approximately 2.003 units. Thus, households with access to water for cocoa farming or near river bodies are more likely to experience unidimensionally poverty than those without. A simple explanation could be that some farmers believe that cocoa farms near water bodies suffer specific challenges such as reduced sunlight in the farm, causing darkness on cocoa farms, which leads to rotten cocoa pods, thereby reducing cocoa yield and depriving the households of income or revenue. The findings showed that occupational diversity has a significant (1% level) and positive influence on unidimensional poverty. The coefficient shows that a cocoa farmer household that diversifies occupationally is associated with a higher unidimensional poverty of about 0.619 units. It suggests that a household’s unidimensional poverty increases with occupational diversification and confirms the findings of Hern andez-N u~ nez et al. (2022) in the cocoa sector. Conversely, some studies have found that occupationally diverse households enhance living standards unidimensionally and have been confirmed by Asmah (2011), Schulte-Herbruggen (2012), Guatem and Andersen (2016), Ampaw et al. (2017), and Eyasu (2020). This study could not confirm the claim that diversification ensures income stability as the farmer engages in other economic activities (Asogwa et al., 2012). Even though there was a negative and marginal association at a 10% level in the second model, occupational diversity was not significant in influencing multidimensional poverty in the main model. However, Dagunga et al. (2020) found that the extent and dimension of diversification reduce multidimensional poverty among farming households. The coefficient of cocoa farming households with access to healthcare was significant for both dimensions in the two models at a 1% level. Access to healthcare negatively influences and is associated with a lower probability of multidimensional poor cocoa farming households by approximately 6.241 units. Also, it positively influences and is associated with higher expenditure-based poverty by approximately 3.553 units. While access to healthcare is likely to reduce the multidimensional poverty of cocoa farmers, 16 S. Y. AGYEMAN-BOATEN it is likely to increase expenditure-based poverty. Furthermore, a household that recorded a child’s death significantly influenced both the multidimensional and unidimensional models at 1% and 5% levels, respectively. The positive coefficient of both dimensions indicates that a household with a record of child death in the last 12 months is associated with a higher probability of being deprived multidimensional by 7.722 units and is likely to be unidimensionally poor by 0.711 units. A child’s death may be attributed to malnutrition, bad hygiene, and financial difficulty, resulting from the household’s deprivation rendering it incapable of saving the lives of children living in the household. Meanwhile, the frequency of ill-health in the cocoa farming household was negative for both dimensions in the two models but significant at the 1% level for multidimensional poverty and not significant in expenditurebased poverty. The coefficient (0.684) indicates that a decrease in the frequency of illness among cocoa farming households is associated with a low probability of being multidimensional poor. This suggests that good health remains a vital indicator of the multidimensional well-being of the cocoa farmer. The position of this finding has been confirmed by Adenuga et al. (2013) and Lawal et al. (2018). A Household with a school-age child not attending school was positive and significantly influenced both unidimensional and multidimensional poverty at the 1% level. The coefficients indicate that on average a cocoa farming household with a school-age child not attending school is associated with multidimensional poverty by approximately 2.966 units and unidimensional poverty by 0.936 units. Children who are not able to attend school may be due to the deprived status of the household and low household income, hence, unable to cater for the child through school. The study’s findings also showed that the multidimensional poverty of cocoa farmers is positively influenced by the use of an external labour force at a 10% significance level. However, it showed no significance for unidimensional poverty. The marginal significance indicates that the engagement of the external labour force by the cocoa farming households is associated with the households’multidimensional poverty by approximately 0.532 units. The household’s access to inputs was positive and significant in influencing multidimensional poverty at the 10% level. It was also found to be negative and significant in the unidimensional model. It indicates that households with access to farm inputs are associated with a higher depth of multidimensional poverty by approximately 0.019 units. Additionally, the results confirm the study by Woldie et al. (2020), who found the negative and significant influences of input utilization on expenditure-based poverty. Finally, the cocoa land size of the household had no significant influence on both unidimensional and multidimensional poverty, contradicting the findings of Huluka (2024). Ravallion (1996) argues that a model’s usefulness depends on its ability to capture and determine what it ought to, not on the degree of its sophistication. According to Khan et al. (2020), multiple methods for identifying and designing poverty alleviation programs should be promoted because the severity and depth of poverty differ throughout villages and within households. Similar to the findings of earlier scholars (eg Costa, 2003; Khan et al., 2020), different results were realised for both monetary and nonmonetary poverty or welfare measures. Costa (2003) contends that because the two approaches describe two distinct groups of poor households, any socioeconomic policy to decrease poverty based on income data is unlikely to meet its stated objectives because it is directed at socioeconomic units that are, in reality, non-poor. As a result, it is only within the context of a multidimensional approach that it is feasible to appropriately identify the poor and create poverty-reduction strategies. Conclusion and policy implications In Ghana, traditional poverty measurement and analysis have primarily been based on a single-dimension monetary or expenditure-based approach, with little attention devoted to multidimensional poverty assessment. However, many authors assert that the multidimensional poverty approach is more suitable to measure poverty than a unidimensional approach and have highlighted the critical limitations of the monetary approaches to measuring poverty. Comparing the unidimensional and multidimensional approaches using the bivariate probit model, this study examines the factors that influence the poverty of rural cocoa farmers in Ghana. The result demonstrates some differences drawn from the multidimensional and unidimensional estimates. It is evident from the result that the influence of the variables on either the expenditure-based approach or the multidimensional poverty approach depends on several COGENT ECONOMICS & FINANCE 17 factors, which include the differences in cocoa farming household characteristics, the household head characteristics, and the setting of the study. For instance, unidimensional poverty was significantly influenced by household characteristics rather than multidimensional poverty. However, many of the variables simultaneously influenced both multidimensional poverty and unidimensional poverty. Another potential factor could be attributed to recent prevalence of the cocoa swollen shoot virus disease in the study area, ravaging cocoa production and their income-earning ability. Since multidimensional welfare is attained over time, these households were considered multidimensionally sufficient before the incidence of the disease, hence the less multidimensional impact. Although the directions and significance levels may differ, the two approaches estimated five variables that simultaneously influenced poverty significantly with coefficient from 0.014 to 7.722 units. They include access to healthcare, household child death, household’s school-age child not in school, access to farm inputs and the age of the household head. In addition, the two approaches produced differing significant associations for the remaining variables. While the education level of the household head, ill-health frequency, use of external labour force, migration status, and the relationship to the household head were significant to determining multidimensional poverty, the number of household members, cooperative union, access to farm water, occupational diversity, household access to financial credit, and marital status of the household head significantly determined unidimensional poverty of the cocoa farmers. The two approaches diverged on many variables in terms of direction and significance because poverty indicators cannot be directly acquired through the income of cocoa farmer households. The study’s findings suggest that to improve the living standards of cocoa farmers who mostly depend on the cocoa output for their livelihood, a measure to move from subsistence to large-scale farming is crucial, especially for higher-educated farmers. The availability of improved and subsidised farm inputs and access to financial credit are among the measures to move many cocoa farmers from income poverty and hence, subsistence to a larger scale. The formation of effective farmers’unions and farmers’organisations to promote knowledge transfer, savings mobilisation, inputs, and output marketing and distributions, and farm credit sourcing and supply should be encouraged among cocoa farming households to improve economic well-being. The health status of the cocoa farming household largely determines their multidimensional welfare. Therefore, governments, policymakers, and stakeholders of the cocoa industry and the agricultural sector should ensure that access to healthcare services, and measures to reduce the households’frequency of ill health, and child death, among others, are maintained. Larger households were found to reduce unidimensionally poverty among the cocoa farming households. It indicates that a large labour force in the household is required to significantly make a fortune out of cocoa farming. Since the result showed that cocoa farmers require some level of education of about eight years to reduce unidimensional poverty, prioritising comprehensive social policy, which guarantees access to quality education at the basic level is essential. This study concludes that even though multidimensional methods of poverty measurement provide wide coverage of deprivations determining poverty, analysis based on both the unidimensional and the multidimensional methods provides a better understanding of the poverty situation of a population for the formulation of good developmental policies. Hence, an approach should be adopted based on circumstances at stake, such as the differences across locations and within households or entities. Notes 1. The Ghana Living Standard Surveys are round of surveys initiated in the 1980s and conducted by the Ghana Statistical Service to provide benchmark data on living standard of the population, monitor, and evaluate progress made by planners and policymakers in improving and sustaining those standards. The rounds include GLSS 1 (1987/88), GLSS 2 (1988/1989), GLSS 3 (1991/1992), GLSS 4 (1998/1999), GLSS 5 (2005/06), GLSS 6 (2012/ 2013) GLSS 7 (2016/17). 2. MPI deprivation rates and cutoffs of indicators analysis based on OPHI Country Briefing 2017, on Ghana (OPHI, 2017); Average MPI ¼0.26. 18 S. Y. AGYEMAN-BOATEN Authors contributions The corresponding author solely contributed to the conception and design, analysis and interpretation of the data, and drafting and revision of the manuscript for intellectual content to be published. He agrees to be accountable for all aspects of the work. Disclosure statement No potential conflict of interest was reported by the author(s). Funding This research received no direct funding. About the author Solomon Yaw Agyeman-Boaten holds a postgraduate degree (Master of Philosophy) and a bachelor’s degree in Economics from the Department of Economics, University of Ghana, Legon. Yaw lectures at Knutsford University College, East Legon, Ghana. He is also pursuing an MSc in Quantitative Finance in Kiel, Germany. Yaw has published extensively in renowned journals, focusing on Development Economics, Monetary Economics, Entrepreneurial Economics, Applied Econometrics, and Quantitative Finance. He serves as a peer reviewer for several academic journals, including SN Business & Economics (Springer Nature) and the Journal of Social and Economic Development (Springer). ORCID Solomon Yaw Agyeman-Boaten http://orcid.org/0000-0002-4299-5419 Data availability statement The data supporting the findings of the study are openly available in Mendeley Data at 10.17632/hpk3ddmjr3.1 and the questionnaire is included as a supplementary document. References Adenuga, A. H., Omotesho, O. A., Ojehomon, V. E. T., Diagne, A., Olorunsanya, E. O., & Adenuga, O. M. (2013). Poverty analysis of rice farming households: A multidimensional approach. Albanian Journal of Agricultural Sciences,12(4), 641. Adeoti, A. I. (2014). Trend and determinants of multidimensional poverty in rural Nigeria. Journal of Development and Agricultural Economics,6(5), 220–231. https://doi.org/10.5897/JDAE2013.0535 Agyeman-Boaten, S. Y., & Fumey, A. (2021). Effects of cocoa swollen shoot virus disease (CSSVD) on the welfare of cocoa farmers in Ghana: evidence from Chorichori community of the Sefwi Akontombra district. SN Business & Economics,1(11), 1–31. https://doi.org/10.1007/s43546-021-00152-8 Agyeman-Boaten, S. Y., Fumey, A., Norman, S.& ,B . (2022). Impact evaluation of households participation in agriculture on welfare in Ghana. African Social Science and Humanities Journal,3(3), 1–21. 10.57040/asshj.v3i3.123 Aikaeli, J. (2010). Determinants of rural income in Tanzania: An empirical approach. Research on Poverty Alleviation. Alkire, S., & Foster, J. (2011). Understandings and misunderstandings of multidimensional poverty measurement. The Journal of Economic Inequality,9(2), 289–314. https://doi.org/10.1007/s10888-011-9181-4 Alkire, S., & Jahan, S. (2018). The new global MPI 2018: Aligning with the sustainable development goals. OPHI Working Paper No. 121, Oxford Poverty and Human Development Initiative (OPHI). University of Oxford. Alkire, S., & Robles, G. (2017). Multidimensional poverty index summer 2017: Brief methodological note and results. OPHI Methodological Notes, 45. University of Oxford. Alkire, S., & Santos, M. E. (2011). Acute multidimensional poverty: A new index for developing countries. Oxford Poverty & Human Development Initiative (OPHI) Working Paper No. 38, United Nations Development Programme Human Development Report Office Background Paper No. 2010/11. Amfo, B., Aidoo, R., Mensah, J. O., & Maanikuu, P. M. I. (2023). Linkage between working conditions and wellbeing: Insight from migrant and native farmhands on Ghana’s cocoa farms. Heliyon,9(2), e13383. https://doi.org/10.1016/ j.heliyon.2023.e13383 COGENT ECONOMICS & FINANCE 19 Amfo, B., Osei Mensah, J., & Aidoo, R. (2022). Migrants and non-migrants’welfare on cocoa farms in Ghana: Multidimensional poverty index approach. International Journal of Social Economics,49(3), 389–410. https://doi. org/10.1108/IJSE-07-2021-0386 Ampaw, S., Nketiah-Amponsah, E., & Senadza, B. (2017). Urban farm-nonfarm diversification, household income and food expenditure in Ghana. Studies in Business and Economics,12(2), 6–19. https://doi.org/10.1515/sbe-2017-0017 Apata, T. G., Apata, O. M., Igbalajobi, O. A., & Awoniyi, S. M. O. (2010). Determinants of rural poverty in Nigeria: Evidence from small holder farmers in South-western, Nigeria. International Journal of Science and Technology Education Research,1(4), 85–91. Asmah, E. E. (2011). Rural livelihood diversification and agricultural household welfare in Ghana. Journal of Development and Agricultural Economics,3(7), 325–334. Asogwa, B. C., Okwoche, V. A., & Umeh, J. C. (2012). Analysing the determinants of poverty severity among rural farmers in Nigeria: A censored regression model approach. American International Journal of Contemporary Research,2(5), 166–176. Babatunde, R. O., Olorunsanya, E. O., & Adejola, A. D. (2008). Assessment of rural household poverty: Evidence from South-western Nigeria. American-Eurasian Journal of Agricultural and Environmental Science,3(6), 900–905. Belhadj, B., & Limam, M. (2012). Unidimensional and multidimensional fuzzy poverty measures: New approach. Economic Modelling,29(4), 995–1002. https://doi.org/10.1016/j.econmod.2012.03.009 Bersisa, M., & Heshmati, A. (2021). A distributional analysis of uni-and multidimensional poverty and inequalities in Ethiopia. Social Indicators Research,155(3), 805–835. https://doi.org/10.1007/s11205-021-02606-w Boni, S., Nuhu, R. I., Reuter, F., & Da Re, G. (2004). Anthropological, environmental and soils assessment of the Sefwi Wiawso District, Ghana. Ricerca e Cooperazione. Bourguignon, F., & Chakravarty, S. R. (2002). Multi-dimensional poverty orderings. Delta. Bradshaw, T. K. (2007). Theories of poverty and anti-poverty programs in community development. Community Development,38(1), 7–25. https://doi.org/10.1080/15575330709490182 Costa, M. (2003). A comparison between unidimensional and multidimensional approaches to the measurement of poverty. (IRISS Working Papers. No. 2003-02). CEPS/INSTEAD. Coulombe, H., & McKay, A. (1995). An assessment of trends in poverty in Ghana 1988-1992 (No. 31097 (p. 1). The World Bank. Dagunga, G., Ayamga, M., & Danso-Abbeam, G. (2020). To what extent should farm households diversify? Implications on multidimensional poverty in Ghana. World Development Perspectives,20, 100264. https://doi.org/ 10.1016/j.wdp.2020.100264 Datt, G., & Jolliffe, D. (1999). Determinants of poverty in Egypt: 1997. Food consumption and nutrition division. International Food Policy and Research Institute. De Janvry, A., & Sadoulet, E. (2000). Rural poverty in Latin America: Determinants and exit paths. Food Policy,25(4), 389–409. https://doi.org/10.1016/S0306-9192(00)00023-3 Deutsch, J., & Silber, J. (2008). The order of acquisition of durable goods and the measurement of multidimensional poverty. In Quantitative approaches to multidimensional poverty measurement (pp. 226–243). Palgrave Macmillan. Dollar, D., & Kraay, A. (2000). Growth is good for the poor. World Bank. Dzanku, F. M. (2015). Transient rural livelihoods and poverty in Ghana. Journal of Rural Studies,40, 102–110. https:// doi.org/10.1016/j.jrurstud.2015.06.009 Eyasu, A. M. (2020). Determinants of poverty in rural households: Evidence from North-Western Ethiopia. Cogent Food & Agriculture,6(1), 1823652. https://doi.org/10.1080/23311932.2020.1823652 Fergusson, D. M., Hong, B., Horwood, J., Jensen, J., & Travers, P. D. (2001). Living standards of older New Zealanders: A summary. Ministry of Social Policy. FOS. (1999). Poverty assessment profile for Nigeria 1980–1996. Federal Office of Statistics. Greene, W. (1998). Gender economics courses in liberal arts colleges: further results. The Journal of Economic Education,29(4), 291–300. https://doi.org/10.1080/00220489809595921 Greene, W. H. (2003). Econometric analysis. Prentice-Hall. Grootaert, C., & Braithwaite, J. (1998). Poverty correlates and indicator-based targeting in Eastern Europe and the former Soviet Union. World Bank Publications. GSS. (2014). Ghana living standards survey round 6 (GLSS 6), Main Report. Ghana Statistical Service. Accra GSS. (2018). Ghana living standard survey round 7 (GLSS 7). 2018 Poverty trends in Ghana 2005-2017. Ghana Statistical Service. GSS. (2020). Ghana Statistical Service Multidimensional Poverty-Ghana. Retrieved March 07, 2024, from https://ophi. org.uk/wp-content/uploads/Ghana_MPI_report_2020.pdf Guatem, Y., & Andersen, P. (2016). Rural livelihood diversification and household wellbeing: Insights from Humla, Nepal. Journal of Rural Studies,44, 239–249. Hern andez-N u~ nez, H. E., Guti errez-Montes, I., Bernal-N u~ nez, A. P., Guti errez-Garc ıa, G. A., Su arez, J. C., Casanoves, F., & Flora, C. B. (2022). Cacao cultivation as a livelihood strategy: Contributions to the well-being of Colombian rural households. Agriculture and Human Values,39(1), 201–216. https://doi.org/10.1007/s10460-021-10240-y Hesse-Biber, S. N. (2010). Mixed methods research: Merging theory with practice. Guilford Press. 20 S. Y. AGYEMAN-BOATEN Holzer, H. J. (1991). The spatial mismatch hypothesis: What has the evidence shown? Urban Studies,28(1), 105–122. https://doi.org/10.1080/00420989120080071 Huluka, A. T. (2024). How is the multidimensional poverty changing in Ethiopia? An empirical examination using demographic and health survey data. Cogent Economics & Finance,12(1), 2364359. https://doi.org/10.1080/ 23322039.2024.2364359 IFAD. (2001). The rural poverty report 2001. International Fund for Agricultural Development. http://www.ifad.org/poverty/index.htm Khan, M., Saboor, A., Rizwan, M., & Ahmad, T. (2020). An empirical analysis of monetary and multidimensional poverty: evidence from a household survey in Pakistan. Asia Pacific Journal of Social Work and Development,30(2), 106–121. https://doi.org/10.1080/02185385.2020.1712663 Lawal, J. O., Famuyiwa, B. S., & Taiwo, O. (2018). Shocks and factors influencing welfare among cocoa farming households in Cross-Rivers State, Nigeria. In International Symposium on Cocoa Research (ISCR), Lima, Peru, 13-17 November 2017. International Cocoa Organization (ICCO). Mart ınez-Mesa, J., Gonz alez-Chica, D. A., Duquia, R. P., Bonamigo, R. R., & Bastos, J. L. (2016). Sampling: how to select participants in my research study? Anais Brasileiros de Dermatologia,91(3), 326–330. https://doi.org/10.1590/ abd1806-4841.20165254 Mekonnen, A. G. (2024). Estimating vulnerability to income and multidimensional poverty: a new methodological approach. International Journal of Development Issues. ahead-of-print. https://doi.org/10.1108/IJDI-05-2024-0145 Mellor, J. B. (1999). The structure of growth and poverty reduction. World Bank. Nawar, A. H. (2014). Multidimensional poverty index and tackling interlocking deprivations in the Arab States. Working Paper No. 125. International Policy Centre for Inclusive Growth (IPC-IG). Ningaye, P., Ndjanyou, L., & Saakou, G. M. (2011). Multidimensional poverty in Cameroon: Determinants and spatial distribution. AERC. Okunmadewa, F. Y., Yusuf, S. A., & Omonona, B. T. (2005). Social capital and poverty reduction in Nigeria. Africa Economic Research Consortium (AERC). Olaniyan, O. (2000). The role of household endowments in determining poverty in Nigeria. Department of Economics. Olaniyan, O., & Abiodun, S. (2005). Human capital, capabilities and poverty in rural Nigeria research report submitted to the African Economic Research Consortium (AERC), for the second phase collaborative poverty research project. African Economic Research Consortium. Omonona, B. T. (2001). Poverty and its correlates among rural farming households in Kogi State, Nigeria [PhD. thesis]. University of Ibadan. Omonona, B. T. (2009). Quantitative analysis of rural poverty in Nigeria (No. 9). International Food Policy Research Institute (IFPRI). Oni, O. A., & Yusuf, S. A. (2007). Determinants of expected poverty among rural households in Nigeria [PhD Thesis]. Department of Agricultural Economics, University of Ibadan. Onumah, J. A., Osei, R. D., Martey, E., & Asante, F. A. (2023). Welfare dynamics of innovations among agricultural households in Ghana: Implication for poverty reduction. Heliyon,9(7), e18066. https://doi.org/10.1016/j.heliyon. 2023.e18066 OPHI. (2017). Oxford Poverty and Human Development Initiative (June 2017). “Ghana Country Briefing”. University of Oxford. https://ophi.org.uk/wp-content/uploads/GHA_2017.pdf Owuor, G., Ngigi, M., Ouma, A. S., & Birachi, E. A. (2007). Determinants of rural poverty in Africa: The case of small holder farmers in Kenya. Journal of Applied Sciences,7(17), 2539–2543. https://doi.org/10.3923/jas.2007.2539.2543 Ravallion, M. (1996). Issues in measuring and modelling poverty. The Economic Journal,106(438), 1328–1343. https:// doi.org/10.2307/2235525 Ravallion, M., & Datt, G. (1996). How Important to India’s poor is the sectoral composition of economic growth? The World Bank Economic Review,10(1), 1–25. https://doi.org/10.1093/wber/10.1.1 Rogan, M. (2016). Gender and multidimensional poverty in South Africa: Applying the global multidimensional poverty index (MPI). Social Indicators Research,126(3), 987–1006. https://doi.org/10.1007/s11205-015-0937-2 Salami, M. F., Babatunde, R. O., Ayinde, O. E., & Adeoti, E. I. (2017). Determinants of poverty among local rice processors in Kwara state, Nigeria. Trakia Journal of Science,15(4), 386–391. https://doi.org/10.15547/tjs.2017.04.022 Samuelson, P. A., & Nordhaus, W. D. (1992). Economics (14th ed.). McGraw-Hill. Inc. Schulte-Herbruggen, B. (2012). The importance of bushmeat in the livelihoods of cocoa farmers living in a wildlife depleted farm-forest landscape, SW Ghana [Doctoral dissertation]. UCL (University College London). Sen, A. K. (1979). Personal utilities and public judgment: Or what’s wrong with welfare economics? Economic Journal, 89(355), 53758. Sen, A. K. (1985). Commodities and capabilities. North-Holland. Woldie, D. T., Haji, J., & Mehare, A. (2020). Intensity and determinants of rural poverty in Banja District of Awi Zone, Amhara National Regional State, Ethiopia. International Journal of Agricultural Economics,5(3), 49. https://doi.org/ 10.11648/j.ijae.20200503.11 World Bank. (1996). Nigeria—Poverty in the midst of plenty. The challenges of growth with inclusion. Report 14733. World Bank. (1999). World development indicators 1999. World Bank. (2000). Attacking poverty, world development report 2001/2002. Oxford University Press. COGENT ECONOMICS & FINANCE 21