Pathways toward the sustainable improvement of food security: Adopting the household food insecurity access scale in rural farming households in Bangladesh

Salman, Md.; Haque, Sadika; Hossain, Md. Emran; Zaman, Nafisa; Hira, Fatema Tuj Zohora Article Pathways toward the sustainable improvement of food security: Adopting the household food insecurity access scale in rural farming households in Bangladesh Research in Globalization Provided in Cooperation with: Elsevier Suggested Citation: Salman, Md.; Haque, Sadika; Hossain, Md. Emran; Zaman, Nafisa; Hira, Fatema Tuj Zohora (2023) : Pathways toward the sustainable improvement of food security: Adopting the household food insecurity access scale in rural farming households in Bangladesh, Research in Globalization, ISSN 2590-051X, Elsevier, Amsterdam, Vol. 7, pp. 1-10, https://doi.org/10.1016/j.resglo.2023.100172 This Version is available at: https://hdl.handle.net/10419/331096 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-nc-nd/4.0/ Research in Globalization 7 (2023) 100172 Available online 4 November 2023 2590-051X/© 2023 The Author(s). Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/bync-nd/4.0/). Pathways toward the sustainable improvement of food security: Adopting the household food insecurity access scale in rural farming households in Bangladesh Md. Salman a , Sadika Haque a , Md. Emran Hossain b , * , Nafisa Zaman c , Fatema Tuj Zohora Hira d a Department of Agricultural Economics, Bangladesh Agricultural University, Mymensingh-2202, Bangladesh b Department of Agricultural Finance and Banking, Bangladesh Agricultural University, Mymensingh-2202, Bangladesh c Department of Agricultural Economics, Bangladesh Agricultural University, Mymensingh 2202, Bangladesh d Faculty of Agricultural Economic and Rural Sociology, Bangladesh Agricultural University, Mymensingh-2202, Bangladesh ARTICLE INFO Keywords: Food security Rural farming households Bangladesh Household food insecurity access scale Migrated members ABSTRACT Achieving food security is a global concern that constitutes a major challenge, particularly for the least developed countries, such as Bangladesh. In the context of globalization, the nation continues to have ongoing food insecurity, particularly in rural areas, despite its overall economic growth and development. This has become a constraint in achieving the Sustainable Development Goals (SDGs) within the established time scale, particularly SDG2 (Zero Hunger). With this consideration in mind, the present study assesses the prevalence of household food in(security) and identifies the factors that influence this among rural farming households in Bangladesh. A sample of 350 farming households was surveyed randomly from the four villages in Mymensingh, Bangladesh. The household food insecurity access scale (HFIAS) was utilized to explore household food security. The results reveal that only 18% of rural farming households were food secure while the remainder were food insecure to some extent. Using a binomial logit regression model, we found that the household head’s educational level, as well as whether the household has a savings account, owns land, receives financial or other forms of support from household members abroad, has larger farm sizes, and practices homestead gardening significantly reduce household food insecurity, whereas a higher number of members in the household increases it. The findings of this study establish a foundational understanding of food security in rural areas by employing contemporary measurement tools and techniques. This addition to the existing knowledge base will assist in the design and implementation of a comprehensive and multifaceted policy outline not only for the rural areas of Bangladesh but also for sustainable development globally. Introduction In response to the two world food crises in 1972–2074 and 2006–2008, food (in)security has become a global policy concern (Gezimu Gebre, 2012; Sasson, 2012). Since the introduction of the term “food security” in 1974, several definitions have been posited (Smith et al., 1996). One of the earliest definitions of food security is “the access by all people at all times to enough food for an active healthy life” (World Bank, 1986). Another popular definition is “a situation that exists when all people, at all times, have physical, social and economic access to sufficient, safe and nutritious food that meets their dietary needs and food preferences for an active and healthy life” (FAO, 1996). Thus, the first pillar of food security is the sustainable production of food (Quisumbing et al., 1996). In this vein, agriculture plays a vital role (Arimond et al., 2010). However, global agricultural production is now under serious threat due to the depletion of natural resources and climate change (Hossain et al., 2022; Roberts, 2008). As a result of global warming, food output may be down anywhere from 3 % to 16 % worldwide by the year 2080 (Cline, 2008). Thus, there are several risks associated with both climate change and food instability that may endanger future generations (Wheeler & Von Braun, 2013). Nevertheless, several international organizations and the national governments of different countries are collaborating to comprehensively eliminate food insecurity throughout every aspect of society. Despite significant efforts from both the public and private sectors, it is evident that a substantial proportion of families * Corresponding author. E-mail addresses: [email protected] (S. Haque), [email protected] (Md.E. Hossain). Contents lists available at ScienceDirect Research in Globalization journal homepage: www.sciencedirect.com/journal/research-in-globalization https://doi.org/10.1016/j.resglo.2023.100172 Received 26 July 2023; Received in revised form 1 November 2023; Accepted 2 November 2023 Research in Globalization 7 (2023) 100172 2 worldwide—around one in 10—continue to face challenges in achieving food security (Nord, 2010). In the context of Bangladesh, food security is commonly understood as the ability to attain self-sufficiency in rice and wheat production (Mainuddin & Kirby, 2015). However, the country’s food production has encountered challenges due to its expanding population, thereby leading to the emergence of food insecurity (Abdullah et al., 2019). Bangladesh’s population exceeds 160 million individuals, while its land area spans 148 million hectares. Consequently, the matter of food security has assumed a heightened level of importance since its limited availability of land poses challenges in adequately nourishing its substantial populace. At the same time, the country is among the most vulnerable to climate change (Jalal et al., 2021), which threatens to lower its agricultural production in the coming years (Ruane et al., 2013). On the other hand, agricultural development is widely known as not only a driver of economic growth but also a tool for poverty alleviation (Myrdal, 1984; Victor Bekun & Akadiri, 2019). The advancement of the agricultural sector is therefore of the utmost importance, as it is correlated with countries’ gross domestic product and assists in maintaining sustainable economic growth both in the short and the long run (Gokmenoglu et al., 2016; Serto˘ glu et al., 2017). Such economic growth also broadens the pathways toward achieving food security (Torero, 2015). However, limited mechanization in agricultural production results in lower productivity and can represent a barrier in achieving food security (Olasehinde-Williams et al., 2022). Due to these factors, Bangladesh’s food security is now in jeopardy despite the government’s efforts to increase agricultural productivity, optimize the use of available land, and develop new crop varieties, among other measures. Despite the fact that undernutrition has decreased globally (Black et al., 2008), about 24 % of women in Bangladesh are still underweight (Khan & Talukder, 2013). Food insecurity is one of the underlying reasons for the persistently high rate of undernutrition (Black et al., 2008). Among Asian nations, Bangladesh has the lowest per capita vailability of calories for consumption (Ericksen et al., 2011). Thus, a comprehensive understanding of the determinants of the country’s household food (in)security is necessary (Harris-Fry et al., 2015). Food security is also influenced by gender equality (Duflo, 2012), as women play a primary role in the food value chain (Olumakaiye & Ajayi, 2006). However, most women are deprived of equal asset distribution when compared to their male counterparts. Despite significant advancements in terms of women’s emancipation, women in Bangladeshi society remain perpetually constrained. Hence, food insecurity prevails in Bangladesh even though new technologies have been adopted to improve agricultural productivity (Sraboni et al., 2014). On the other hand, food security comprises multiple dimensions that range from the global to the individual level (Abdullah et al., 2019). As having an impoverished and undernourished population is considered a hindrance to human and economic development (Mango et al., 2014), exploring the determinants of household food insecurity has become an important issue for Bangladesh, especially in the context of rural farming households, where women are heavily involved with farming. However, to address the abovementioned constraints and challenges, it is necessary to first understand the determinants of food security. This study has important practical implications and makes a novel contribution to the existing body of literature. First, as Bangladesh is a nation driven by its agricultural labor force, it is challenging to address its food insecurity problem, particularly among female-dependent rural farming households, who are already plagued by a variety of social, political, and health-related challenges. To identify the incidence and risk factors of food insecurity, this study assesses the household food security status of women who participate in farming activities. In doing so, it will deepen our understanding of Bangladesh’s food security status. The data-driven findings directly contribute to policy formulation by enabling an assessment of the country’s progress toward achieving the Sustainable Development Goals (SDGs) (Mota et al., 2019). Second, the determinants of food (in)security, as well as its underlying mechanisms and reasons for its persistence among rural farming households, are determined. Moreover, the findings reveal the inequalities in food access and availability in rural areas, which can improve the equitability of food distribution (Kennedy et al., 2004; Mota et al., 2019). Last, the results send actionable information to the global agricultural markets and investors in the agricultural sector (Kennedy et al., 2004). Third, this study employs the household food insecurity access scale (HFIAS), which allows us to examine food security from several perspectives given that previous studies on this issue have failed to do so. The HFIAS can be applied to not only distinguish food-insecure from food-secure households but also to track changes in food security over time and across cultural contexts (Coates et al., 2007). Fourth, this study provides suggestions for policy formulation to sustainably improve food security among agrarian households based on the empirical findings. These suggestions will help stakeholders’ plan and execute targeted developmental strategies to alleviate food insecurity and achieve the Fig. 1. Map of the research region. Md. Salman et al. Research in Globalization 7 (2023) 100172 3 SDGs (Kennedy et al., 2004). Finally, this research analyzes household food security from a rural perspective and thus enables global policy adoption given that the findings are applicable to many other countries that are similarly agrarian in nature. Accordingly, the insights gained through this research will facilitate global cooperation and provide a basis for global food system development from the local to the global levels. Data and methods Study area The research was primarily conducted in the newly formed Mymensingh division of Bangladesh, which we selected because of its agrarian characteristics. Mymensingh was declared a city corporation on November 20, 2017 and subsequently became the capital of the division (bdnews24.com, 2017). Therefore, it is necessary to evaluate how the new administrative authority and its functions will affect each economic sector, particularly rural agriculture. As a result, the specific research location was randomly selected from among the rural areas close to Mymensingh city. Four villages—Char Nilakshmia, Sutiakhali, Bhabkhali, and Parail—were chosen for this study (Fig. 1). All four villages are adjacent to Mymensingh city and have good transport and communication facilities. Furthermore, the agricultural practices and use of arable land are similar across all four villages. Sample design and sampling procedure After selecting the locations, we chose a sample from among the populace. Our study concentrated on agricultural households in which female members participate in agricultural production to fully depict the multidimensional nature of food insecurity in rural agrarian households. According to the Report of Agricultural and Rural Statistics (2018), there are 19,312 households in the Mymensingh region in which women are either directly or indirectly involved in agricultural production (BBS, 2018). Given that the population was limited, this study used the following formula to determine the representative sample size (Rusli Nordin et al., 2006): n=N× Z2×p×(1−p) e2 N−1×Z2×p×(1−p) e2 where n is the sample size, N is the population size (19,312), Z is the precession level at the 95 % confidence interval (1.96), p is the proportion of the population (35 %), and e is the error margin (5 %). The results revealed using the aforementioned method indicated that at least 343 households would serve as a representative sample for the entire population. The sample selection method accounted for 35 % of the population in light of budgeting and scheduling restrictions. Finally, 350 households were surveyed through face-to-face interviews using a random sampling technique. We included only households in which women were engaged in agricultural production and primarily responsible for food preparation. In this manner, primary data were collected from the respondents during February–March of 2022 using a wellstructured, pretested questionnaire. Analytical techniques Two types of analytical techniques were applied to achieve the study objectives: descriptive and inferential. In the descriptive analysis, the summary statistics of different variables, percentage distribution of the data, graphical representations, etc., were considered. The results of this analysis clarified the nature and distribution of the study variables. Household food insecurity access prevalence was employed in the inferential analysis as the outcome variable, and a binomial logit regression model was utilized to explore the significant predictors. Household food insecurity access scale We used the HFIAS to assess the level of household food insecurity (Coates et al., 2007). The individual components of the HFIAS yield information on food insecurity (access) at the household level. Household food insecurity was measured based on nine questions along with their frequency of occurrence. The respondents were asked whether they ever: 1. worry about food, 2. are unable to eat their preferred foods, 3. eat just a few kinds of foods, 4. eat foods they would rather not eat, 5. eat smaller meals than his/her aspiration, 6. eat fewer meals in a day (less than 3 meals), 7. have no food of any kind in the household, 8. go to sleep hungry, and 9. go a full 24 h without eating. From the results, four indicators were calculated, through which changes in household food insecurity (access) were assessed. These indicators provide summary information on household food insecurity access (a) conditions, (b) domains, (c) scale score, and (d) prevalence. All of the data were quantified in such a way that if a household responded “not experienced” to a particular condition it received a value of 0 and the frequency of that condition was also denoted as 0. However, if a household did experience a particular condition it received a value of 1 and the frequency was assigned a value of 1 for “rarely,” 2 for “sometimes,” and 3 for “often.” The detailed calculation process for the four indicators can be found in the Appendix. Binomial logit regression The binomial logit regression was used to explore the determinants of household food insecurity using the binary dependent variable for food insecurity (y) (Long and Freese, 2006). In this model, y takes the value of one when a household is food secure and zero when it is not. Table 1 presents the variables used in the regression model, which is expressed as: Ln[p/(1-p)] ¼β0+β1X where p is the probability that event Y occurs (p(Y =1)) [range =0 to Table 1 Description of variables used in the binomial logit regression. Variable Unit Description Age of household head years The age of the household head at the time of the survey Years of schooling of household head years How many years of formal schooling has received by the household head Household’s farming experience years How long has the household been engaged in producing crops Occupation of household head Dichotomous Whether a household head was engaged with crop farming or other types of occupation along with crops. Household size Discrete Number of persons living in a household Household has a savings account Dichotomous Whether the household has a saving account in a formal or informal financial institution or not Land ownership Dichotomous Whether the household has its own agricultural and non-agricultural land from where it receives income. Migrated member Dichotomous Represent any of the household member who has been migrated to any other place within the country or abroad to earn and assist the household financially. Farm size Decimal Area managed for cultivation by the household in the current year Homestead gardening Dichotomous Whether a household had a vegetable/ fruit garden in the yard of the home Md. Salman et al. Research in Globalization 7 (2023) 100172 4 1]. The odds ratio (OR) is defined as follows: p/(1-p) [range =0 to ∞] Ln[p/(1-p)] is the log OR, or “logit” [range =-∞ to +∞] The OR is used to explain the results with 95 % confidence intervals. The empirical model is as follows: log[P (1−P)]=β 0+ β 1 X Age of household head +β 2 X Years of schooling of household head +β 3 X Household farming experience +β 4 X Occupation of household head +β 5 X Household size +β 6 X Household have savings account +β 7 X Land ownership +β 8 X Migrated members + β 9 X Farm size +β 10 X Homestead gardening + ε i . (1) The confusion matrix (or classification test), model accuracy, precision, and recall were estimated in the post-estimation of the binomial logit regression model. The linktest command was utilized to test whether the regression equation was correctly specified and that no additional independent variables were significant except by chance. To find the link error, we let: y=f(Xβ) be the model and  β be the parameter estimates. Then, linktest calculated: _hat =X  β and, _hatsq =_hat 2 We can predict the model accuracy based on the significance of _hatsq, as suggested by Pregibon (1980) based on Tukey (1949). As linktest is universally applicable, straightforward, and a good secondorder approximation, it can be applied to any single-equation estimation technique (StataCorp, 2017). All analyses were undertaken using Stata V17 and the Python 3.10 integrated development environment. Ethical consideration In this study, ethical issues were considered during the field survey. Any kind of biases in area and sample selection were avoided. During the interview with the respondents, at first, their verbal consent was taken. The purpose of the study was clearly explained to them initially and requested to participate in the interview. If they agreed, then the interview was undertaken. The data handling and analysis process was done with intensive awareness. Any kind of people, animals, or natural resources were not harmed or destroyed while conducting the different stages of the study. Results Household information The results indicate that the average age of the household head was 46 years, where the age of the youngest head was 20 years and that of the oldest head was 75 years (Table 2). The average number of formal years of schooling was found to be only five, while nearly 34 % of the heads of household had no formal education. Of those who attended school, 25 % completed primary education and 13 % completed secondary education (Fig. 2). The average farming experience of the households was 18 years (Table 2), with a minimum of one year, a maximum of 60 years, and a nonnormal distribution (Fig. 2). The average household size was 4.92 members (Table 2), while the average farm size was 58 decimals (100 decimals =1 acre) (Table 5). Fig. 2 presents the distribution of farm sizes and shows that most of the households had less than 100 decimals of land. The occupations of the heads of households were classified into two categories: those who only produced crops and those who produced crops and engaged in other activities. The results revealed that around one-third of the heads of household were engaged in cropping activities, while the remainder were engaged with other activities such as fishing, rearing livestock or poultry, operating a small business, wage laboring, etc., in addition to cropping. More than 30 % of the households had a savings account. More than 80 % of the households owned agricultural land, but not all of that land was used for production in the current year. Around 8 % of the households had one or more members who migrated to other regions within Bangladesh or abroad to engage in incomegenerating activities. Almost 50 % of the households practiced homestead gardening at their dwelling. Food In(security) This study measured household food insecurity, which comprises household food insecurity access-related conditions, food insecurity access-related domains, food insecurity access scale scores, and food insecurity access prevalence. The results jointly define the food insecurity status of the studied households. Household food insecurity access-related conditions Table 3 presents the proportion of households that experienced food insecurity in a particular condition, as well as the frequency of experiencing that condition. The results revealed that almost 83 % of the households were worried about food insecurity and that 24 % of them had experienced this condition more than 10 times during the last four weeks as of the survey period. Around 81 % of the households were unable to eat their preferred foods, and 36 % frequently experienced this condition. Limited varieties of foods were consumed by 79 % of the households, and this condition was frequently experienced by 22 % of the households. Nearly 77 % of the households consumed foods that they would prefer not to eat, and almost 73 % had consumed smaller meals during the last four weeks as of the study period. Fewer than three meals per day were eaten by 65 % of the households, and 17 % of them experienced this condition regularly. Less than 1 % of the households frequently had no food to eat at home, and less than 1 % went to sleep hungry or regularly went a full day without eating. Food access-related conditions worsen as we proceed through the conditions. However, both the severity of these conditions as well as their frequency declined throughout the study period. Household food insecurity access-related domains We assessed three household food insecurity access-related domains. The results indicate that around 83 % of the households experienced Table 2 Summary of the study variables. Variable Mean ± Std. Dev. Min, Max Frequency (%) Age of the household heads 46.009 ± 12.244 20, 75 Years of schooling of the household heads 5.311 ±4.786 0, 16 Farming experience 18.411 ± 11.92 1, 60 Household size 4.92 ±1.764 2, 15 Farm size 58.357 ± 48.827 2.5, 300 Occupation of the household heads Crop only 114 (32.57) Crop & others 236 (67.43) Households’ savings account No 228 (65.14) Yes 122 (34.86) Land ownership No 47 (13.43) Yes 303 (86.57) Migrated members No 319 (91.14) Yes 31 (8.86) Homestead gardening No 178 (50.86) Yes 172 (49.14) Md. Salman et al. Research in Globalization 7 (2023) 100172 5 anxiety and uncertainty related to food insecurity. Approximately 81 % of the households had insufficient or poor-quality foods and almost 73 % had inadequate food intake and, as a result, suffered physical consequences (Table 4). Household food insecurity access scale score Table 5 shows the food insecurity access scale score of the studied Fig. 2. Summary of the studied variables. Table 3 Food insecurity access-related condition of the studied households. HFIAS questions Households experienced the condition at any time during the recall period (%) Households experienced the condition (often) at a given frequency (%) Worry about food 82.57 23.71 Unable to eat preferred foods 81.42 35.85 Eat a limited variety of foods 78.85 22.28 Eat foods that they really did not want to eat 77.42 21.14 Eat a smaller meal 72.57 18.57 Eat fewer meals in a day 64.85 16.57 No food to eat of any kind in the household 27.71 0.57 Went to sleep at night hungry 18.28 0.28 Went a whole day and night without eating anything 18.28 0.57 Table 4 Food insecurity access-related domains of the studied households. Anxiety and uncertainty (% of households) Insufficient Quality (% of households) Insufficient food intake and its physical consequences (% of households) 82.57 81.42 72.85 Table 5 Food insecurity access scale score. Variable Total score Mean Standard deviation Minimum Maximum HFIAS 3542 10.12 6.312 0 24 Md. Salman et al. Research in Globalization 7 (2023) 100172 6 households. The total score was 3542, with a mean of 10.12 and a standard deviation of 6.31. The minimum score was 0 and the maximum was 24 on a scale measured from 0 to 27. Household food insecurity access prevalence The prevalence of household food insecurity access was categorized into four groups. The results show that only 18 % of the households were food secure. Among the food-insecure households, 5 % were mildly food insecure, 35 % were moderately food insecure, and 42 % were severely food insecure (Fig. 3). Determinants of the risk factors of food insecurity After estimating the prevalence of household food insecurity access, we explored the risk factors of households’ food insecurity status. Mildly, moderately, and severely food-insecure households were treated as the insecure group, and the full sample of households was divided into two groups: food secure (18 % of the households) and food insecure (82 %) (Fig. 4). Due to the binary nature of the exploratory variable, we utilized the binomial logit regression model to explore the risk factors of food insecurity in the studied households. Furthermore, the OR was reported to explain the results of the regression model. Table 6 shows the results of the binomial logit regression model, which indicate that each one-year increment of formal education of the household head improved household food security by 0.921 times. The addition of one member to the household raised the risk of household food insecurity by 0.18 times; if the household had a savings account, food security improved by 0.291 times. Households who owned land were 0.308 times more food secure than those that did not. The presence of migrated household members reduced the risk of household food insecurity by 0.401 times. Similarly, a one-decimal increase in households’ farm area reduced food insecurity by 0.989 times. Finally, households that practiced homestead gardening were 0.269 times more food secure compared to those that did not. Each effect of the explanatory variable was valid when holding the effects of other variables constant. In the post-estimation section of the model, the specification test Fig. 3. Food insecurity access prevalence. Fig. 4. Distribution of the exploratory variable (food insecurity). Table 6 Binomial logistic regression on household food insecurity status. Variables Odds ratio St. Err. pvalue Age of the household head 1.007 0.02 0.717 Years of schooling of the household head 0.921** 0.033 0.022 Household’s farming experience 0.994 0.021 0.765 Occupation of the household head (crop & other) 1.038 0.418 0.926 Household size 1.18* 0.115 0.089 Household has a savings account 0.291*** 0.098 0.000 Land ownership (yes) 0.308* 0.205 0.076 Migrated member(s) (yes) 0.401* 0.196 0.062 Farm size 0.989*** 0.003 0.000 Homestead gardening (yes) 0.269*** 0.093 0.000 Constant 65.556*** 73.785 0.000 Model diagnosis (Linktest) _hat 0.900 0.269 0.001 _hat square 0.095 0.095 0.665 Others diagnostic test Mean dependent var 0.820 SD dependent var 0.385 Pseudo r-squared 0.194 Chi-square 64.161[0.000] Akaike information criteria (AIC) 287.814 Bayesian information criteria (BIC) 330.252 Accuracy 0.84 Precision 0.84 Recall 0.99 *** p <.01, ** p <.05, * p <.1. Table 7 Confusion/ classification matrix of the food insecurity status model. Classification D ~D Total +279 47 326 – 8 16 24 Total 287 63 350 Md. Salman et al. Research in Globalization 7 (2023) 100172 7 shows that the squares of the coefficients were statistically nonsignificant, which indicates that the model was correctly specified (Table 6). The confusion or classification matrix showed that the model correctly classified 279 cases as true positive and 16 cases as true negative (Table 7), which implies that these cases were correctly predicted and that the model accuracy was 84 % (Table 6). The precision was 0.84, which implies that the model predicted 84 % of the food insecurity status of the studied households (Table 6). Moreover, the recall was 0.99, which implies that the logistic regression model identified 99 % of the food insecurity status of a given household (Table 6). Discussion The main contribution of this study lies in its in-depth exploration of household food (in)security. The occurrence of severe food-related issues was found to be lower in the food insecurity access-related condition. This may indicate that significant progress has been made in terms of food security in Bangladesh in recent years (IPC, 2022). Mitu et al. (2022) found that rural areas of Bangladesh experience low dietary diversity, which is linked to food availability and thus anxiety and uncertainty. The results of this study indicate that the greater levels of food-insecure access in the two domains we considered were associated with increased physical repercussions resulting from inadequate food consumption. The mean of the food insecurity access score was low, which is likely a result of the continuous growth of food production in Bangladesh (BER, 2021). The findings relating to the frequency of food insecurity in rural households vary at the national and regional levels (IPC, 2022). The results indicate that only 18 % of rural farming households are food secure. Panezai et al. (2022) reported similar findings, claiming that 72 % of marginal farmers and 81 % of landless farmers in coastal Bangladesh experience food insecurity. These findings are comparable since both studies focused on farming activities in rural Bangladesh, although they were conducted in two different agro-climatic regions. The results further indicate that increased education can significantly reduce household food insecurity. Evidence from field observations revealed that heads of household were the main decision-makers with regard to food choices and food availability in rural Bangladesh. Therefore, their education plays a role in achieving household-level food security. When the individual responsible for overseeing domestic affairs is not present, women assume a leadership role; otherwise, they remain in a secondary role. Furthermore, those in positions of authority are responsible for making financial decisions when purchasing food items, and their choices are influenced by their understanding of food and nutrition. Several previous studies also showed that the educational level of the household head is negatively related to household food insecurity (e.g., Amaza et al., 2009; Bashir et al., 2017; Gezimu Gebre, 2012; Idris & Gwary, 2008; Mango et al., 2014). A larger household size was found to be negatively associated with household food security. Field observations revealed that households cannot increase their food production or the availability of food without increases in household income. At the same time, food prices and production costs are continuously rising in both rural and urban areas (Alexander et al., 2023). Thus, households with a fixed income must distribute the available food among household members. Moreover, managing the food supply for all members of a household becomes more difficult when an additional member is introduced into the family while its income remains fixed. Jacobs (2009) found that larger households consume more food and thus need to increase their food expenditure and compete for scarce resources (Ndobo & Sekhampu, 2013), which makes them more likely to be food insecure compared to smaller or more nuclear households (Babatunde et al., 2007). According to our findings, households with savings accounts are less likely to experience food insecurity. Having savings also encourages investment in agricultural and home matters (Bevan & Mammo, 2002), which may increase food production and thus improve household food security. Field observations showed that some families were very much concerned about their financial well-being and tried to save so as to mitigate future shocks, depositing a portion of their income in banks or with Non-government Organizations (NGOs). By utilizing their accumulated savings, individuals occasionally leased additional land, thereby increasing their overall production and availability of food. These savings were also used to purchase food for household members when there were no other sources of income or during economically challenging periods. Demeke et al. (2011) found that households with savings accounts had a higher food security score and thus a better food security status. In addition, the findings of this study are consistent with those of the World Food Program (WFP, 2009) report, which indicated that savings accounts are a common factor in determining household food security. Our results show that owning land may reduce the risk of household food insecurity in rural areas. In this study, agricultural land included all of the land owned by households, including that which was currently leased out, used for production, or fallow for any reason. Ownership of land also indicates households’ socioeconomic status in Bangladesh to some extent, as it enables food production for family sustenance (Molony & Smith, 2010) and increases the household’s income-generating potential (Adebayo & Ojo, 2012). Farmers may receive cash or in-kind payments when they lease land out. Household food security and income can be improved by lengthening the land-lease tenure and property rights of farming households (World Bank, 2017), which is consistent with the findings of this study. We found that an increase in farm size may significantly reduce household food insecurity. Rahman and Islam (2013) showed that a positive relationship exists between households’ food intake and farm size. Mannaf and Uddin (2012) also showed that large-farm owners are more likely to be food secure than small farm owners in Bangladesh. This implies that large-farm owners are able to consume more food. Furthermore, the increased revenues of large-farm owners notably increase overall levels of food consumption in Bangladesh. A similar result was found in Nigeria, where it was indicated that each unit increase in farm size increased household food security (Adebayo Oyefunke Olayemi, 2012). People that migrate from rural Bangladesh primarily go to metropolitan regions or overseas for work or to improve their living conditions (Sikdar et al., 2014). The results of this study show that having migrant members in a household may increase its food security. Field observations indicated that the migrated members of households regularly send money back home, through which the households’ food access increases in terms of quality, quantity, and diversity (Ndobo & Sekhampu, 2013). Tiwari and Joshi (2012) found that food security is largely dependent on the availability of employment for members who have migrated to other regions. Some previous studies have also found that migration is positively linked with household food security (e.g., Crush, 2013; Kuuire et al., 2013), which is also in line with the findings of this study. Homestead gardening—defined as growing vegetables and/or fruits in the homestead area in this study—was found to improve household food security. Although home gardens vary in terms of their size, biodiversity, seasonal cropping patterns, adaptation to locally available resources, and cultural preferences, they are common in Bangladesh (Talukder et al., 2000). Households may gain access to several types of foods from their home gardens. Homestead gardening is also important for good human health and well-being (Lal, 2020) while enhancing access to food in rural areas (Chadha & Oluoch, 2003). Taken together, homestead gardening ensures household food security (Galhena et al., 2013) by increasing the supply and availability of food to households. Conclusions and policy recommendations Most of the population in Bangladesh still lives in rural areas and is therefore directly or indirectly dependent on farming. Hence, conducting an assessment of the food security status of rural farming Md. Salman et al. Research in Globalization 7 (2023) 100172 8 households will yield valuable insights for both policy-makers and researchers. With this consideration in mind, the present study evaluated the food security status of, and identified the related risk factors in, rural farming households characterized by active female participation in agricultural production. From a methodological perspective, we employed the HFIAS to gauge the food (in)security status of households and the binomial logit model to evaluate the risk factors associated with it in Bangladesh. The results revealed that a high prevalence of food insecurity exists in rural farming households and that its physical consequences are noticeable. Although the incidence of severe food-related issues has fallen, the seriousness of the existing conditions should not be underestimated. The findings of this study indicate that a mere 18 % of rural farming households exhibited food security, while the remaining homes experienced varying degrees of food insecurity. The results also uncovered the determinants of food insecurity in farming households in rural areas. The educational level of the household head, as well as whether the household has a savings account, owns land, has members who have migrated to other regions for work, has larger farm sizes, and practices homestead gardening, significantly reduce the risk of household food insecurity, which may mean that farming households need to have diversified sources of income, as farming alone may not be sufficient to ensure food availability throughout the year. Moreover, larger households had a higher risk of food insecurity than the smaller households in rural areas. The results of this study can serve as a reference for policy formulation at the local and national levels. Recommendations and policy guidelines are derived based on the empirical results. First, we discovered that education has the potential to significantly reduce household food insecurity. In rural areas of Bangladesh, girls’ education is lagging despite the government attempting to prioritize this issue. At the household level, girls are disadvantaged in terms of receiving a quality education in science, English, and mathematics. While this trend is slowly reversing, there is still cause for concern. To ensure that women and heads of household are educated, formal and informal educational programs should be introduced, which may include nighttime schooling, stipend programs, agricultural incentives (e.g., the provision of seed or fertilizer), and skills training. Second, as smaller household sizes may reduce food insecurity, governments and NGOs may implement measures to promote smaller families and family planning, which is in accordance with Bangladesh’s developmental aims. In addition to food security, other benefits of having a small family (e.g., better education, living standards, care, and management) should be emphasized. Third, we found that in rural farming households, holding a savings account alleviates food insecurity. Therefore, both public and private banks, as well as governmental and nongovernmental financial organizations, should take care to give rural residents access to practical credit and savings options by easing the complex existing procedures for obtaining financial services. Fourth, total farm area (i.e., total agricultural and nonagricultural land ownership) is a crucial determinant in overcoming the challenges associated with food insecurity. As the amount of land is limited in Bangladesh, land intensification, precision agricultural practices, land development, integrated farming, etc., can significantly increase agricultural productivity. Fifth, migration promotes food security, but the target area of that migration should be taken into consideration. The implementation of effective security measures in areas that absorb migration can ensure that the influx of revenue it provides to rural regions is sustainable. Finally, adequate training programs, agricultural inputs, management practices, and extension services should be introduced in rural areas to promote year-round homestead gardening given that it improves household food security. In this regard, the Bangladesh Agricultural Development Corporation, as the leading seed-producing and -distributing governmental organization in Bangladesh, can play a pivotal role. Limitations and scope of the study With the addition of the new area of knowledge, this study also has some limitations. These limitations are: the study considered only the crop (rice/ paddy) farmers where the main crop was rice. So, the results, discussion, conclusion, and recommendation could not be implicated or justified directly to other agricultural enterprises, such as livestock, fisheries, or forestry. The study chose a sample size that is small but representative. More samples could be added if the time length and budget were widened. Also, the study used a cross-sectional study design which is not comparable with time series analysis. A large sample with different farming practices like crops, livestock, fisheries, etc. could be included and compared among the groups. A time series analysis could be conducted to trace the trend of food security over time. More knowledge in this field could help to sustainable growth and development of rural agriculture and food security through interventions and time-demanding policy implications. Funding This study was supported by a grant (project no.: 2021/131/BAU) from the Bangladesh Agricultural University Research System (BAURES) to carry out the research activities. However, there was no fund to publish this article. Authors’ contribution Md. Salman did the conceptualization, analyzed and prepared the initial draft; Sadika Haque supervise the whole work process, Writing – original draft, Writingreview and editing; Md. Emran Hossain managed the data, Writing – original draft, Writingrevise the paper; Nafisa Zaman was responsible for data collection, questionnaire preparation, and Writing – original draft; Fatema Tuj Zohora Hira did the literature review and write the conclusion section. Declaration of Competing Interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. Table A1 Measurement of household food insecurity access-related conditions. Indicators Calculation procedure Households experiencing the condition at any time during the recall period. Percent of households that responded, “yes” to a specific occurrence question. For example: “Percent of households that ran out of food.” Example:(Number of households with response =(“1 ″ to Q7) / (Total number of households responding to Q7) X 100 Households experiencing the condition at a given frequency Percent of households that responded “often” to a specific frequency-ofoccurrence question. For example: “Percent of households that ran out of food often.” Example:(Number of households with response =(“3 ″ to Q7a) / (Total number of households responding to Q7) / X 100 Source: Coates et al. (2007). Note: Q# denotes the particular question related to the condition and Q#a represents the frequency of occurrence of a particular condition. Md. Salman et al.