Open Access © The Author(s) 2025. Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://​creat​iveco​mmons.​org/​licen​ses/​by-​nc-​nd/4.​0/. RESEARCH Zenbaba et al. Journal of Innovation and Entrepreneurship (2025) 14:48 https://doi.org/10.1186/s13731-025-00495-8 Journal of Innovation and Entrepreneurship Impact of wheat production technology packages adoption on smallholder farmers’ food security and income in Horo Guduru Wollega Zone, Ethiopia Oliyad Sori Zenbaba1,2*   , Mengistu Ketema3, Moti Jaleta4 and Kedir Jemal1  Abstract  Adoption of agricultural technology packages got considerable attention in enhanc- ing smallholders’ food security and farm income in Ethiopia. However, the impact evaluation of such technology packages’ contributions to households’ food security and income are limited. This study is aimed at identifying factors affecting households’ decisions in adopting wheat technology packages and its ex-post adoption impact on their food security and wheat production income. The food consumption score and households’ dietary diversity score were used as proxy measures of households’ food security. The study employed a multinomial endogenous switching regression model to account for selection bias. Analysis results show that household decisions to adopt combinations of wheat technology packages influenced by age, sex, edu- cation level of household head, distance from markets, plots and services, phone ownership, access to credit services, agricultural cooperative membership, farm size, livestock, and farm training services . Moreover, the study shows that adoption of  full wheat technology packages has a greater positive impact on households’ food security and wheat production income than adopting these packages in a few combina- tions or in isolation. The findings of the study suggest strengthening the provision of improved agricultural services to support farmers in adopting packages of technolo- gies for a better food security and livelihood outcomes. Keywords:  Technology packages adoption, Impact, Multinomial logit model, Multinomial endogenous switching model, Ethiopia Introduction Majority of population in Sub-Saharan countries rely on agriculture for their subsist- ence and livelihoods (Abay et al., 2021; Setsoafa et al., 2022; Ssozi et al., 2019). Besides, agriculture contributes to economic growth by sustaining smallholders’ well-being and reducing poverty in the region. In Ethiopia, for example, agriculture, national develop- ment goals, and the overall livelihood of smallholder farmers are highly interlinked. It dominantly serves as a source of economy, income, employment, exports, and revenue generation, and contributes to food insecurity and poverty reduction in the country *Correspondence: oliyadsorizen@gmail.com 1 School of Agricultural Economics and Agribusiness, Haramaya University, Haramaya, Ethiopia 2 Department of Agricultural Economics, Wollega University, Nekemte, Ethiopia 3 Ethiopian Economics Association, Addis Ababa, Ethiopia 4 International Maize and Wheat Improvement Center (CIMMYT), Addis Ababa, Ethiopia http://creativecommons.org/licenses/by-nc-nd/4.0/ http://crossmark.crossref.org/dialog/?doi=10.1186/s13731-025-00495-8&domain=pdf http://orcid.org/0000-0002-2590-2460 Page 2 of 26Zenbaba et al. Journal of Innovation and Entrepreneurship (2025) 14:48 (ATA, 2018; Welteji, 2018; Zerssa et al., 2021). By considering agricultural development and its importance for national economic development, the government has formu- lated and developed different strategies and policies, but due to the insufficient efforts in implementing these policies, agricultural productivity is still low in Ethiopia (Shikur, 2020). Despite its crucial importance for household welfare, agriculture in Ethiopia faces pro- duction constraints, such as dependence on rainfall, low rates of technology adoption, soil degradation, limited irrigation, lack of investment infrastructure, gender difference in access to technologies, vulnerability to climate change, subsistence production, and lack of value addition (Abro et al., 2014; Shimeles et al., 2018; Zerssa et al., 2021). As a result, the performance of agricultural productivity in Ethiopia is extremely low com- pared to the rest of the world, and it is difficult to meet food demand, improving small- holders’ welfare and boosting economic development (Ssozi et al., 2017). Therefore, the problems of food insecurity, hunger, and malnutrition continue to be more influential challenges than ever, and people are chronically undernourished (FAO, ECA, & AUC, 2021; Shimeles et  al., 2018). Wheat is the main staple food crop and is considered as strategic crop to realize food self-sufficiency, serve to improve household income, nutritional value, and welfare in Ethiopia. It is ranked second by providing 14% of the total caloric intake next to maize (2018b; Anteneh & Asrat, 2020; Brasesco et al., 2019; Tadesse et al., 2018a). Food insecurity is among the top-listed global challenges worsening the livelihood of human beings (Arouna et al., 2017). It is a lack of securing sufficient, safe, and nutritious food, which is a consequence of the non-sustainability of normal growth, development, and an active and healthy life (FAO, 2021). When it once happens, eradicating it from its root is complex and challenging (Cordero-Ahiman et al., 2020). Among the indica- tors used for identifying the status of nations’ well-being, food security measurement is top-listed (Hoddinott & Yohannes, 2002). In Ethiopia, food insecurity and malnutri- tion problems are growing due to population growth, drought, inflation, rising prices of goods and services, political instability, internal conflict and displacement, and unem- ployment (Alemu & Mengistu, 2019; WFP, 2023). This issue chronically exists in Ethio- pia, where many people are highly food insecure and depend on Safety Net Program (SNP) aid for survival (FSIN, 2017). Agricultural technology adoption significantly boosts agricultural productivity, food security, consumption, income, and poverty reduction (Kaliba et al., 2018; Mark et al., 2014), but it also requires the technology to be fully disseminated and utilized in society (Glover et al., 2019). According to Hailu et al., (2021), smallholder farmers’ preference, willingness, and ability to jointly adopt interdependent agricultural technology packages determine the technology uptake and utilization. Jointly adopting recommended tech- nology packages can realize expected outcomes (Teka & Lee, 2020). However, most pre- vious studies, such as Yirga and Alemu (2016), Solomon et al. (2016), Mekonnen (2017), Wordofa et  al. (2021), and Assaye et  al. (2022) and Merga et  al.(2023), failed to focus on the impact of the adoption of multiple technology packages but rather discussed the impact of single agricultural technology adoption on household welfare, income, and agricultural productivity. Though wheat productivity in Ethiopia is low, demand and consumption for wheat and wheat products have increased due to an increased Page 3 of 26Zenbaba et al. Journal of Innovation and Entrepreneurship (2025) 14:48 population, increased urbanization, and awareness of value addition (2018b; Brasesco et  al., 2019; Gedefe & Bekele, 2019; Tadesse et  al., 2018a). To meet the demand gap along with smallholders’ food security and improve farm income, the adoption of wheat technology packages is profoundly considered criteria. According to Tadesse et  al., (2018a, 2018b) and Adhikari et al. (2021), these production technology packages include improved wheat varieties, seeding rates, row planting, better irrigation, weed manage- ment, recommended fertilizer, etc. Adopting these technology packages contributes to improving wheat productivity, food security, and the welfare of households (Atinafu et al., 2022; Giller et al., 2017; Nirgude & Sonawane, 2017). Agricultural technology adoption analysis got considerable attention to influence the welfare and livelihood of agriculture dependent smallholder farmers (Kaliba et al., 2018). There is a large empirical literature regarding the impact of wheat production technol- ogy adoption on household food security and income. However, most previous impact studies of wheat technology adoption (Mulugeta & Hundie, 2012; Shiferaw et al., 2014; Solomon et al., 2016; Wake & Goshu, 2019) did not focus on the impact of joint tech- nology package adoption, which could be resulted in a better outcome but were con- ducted using a single or few technology package(s). However, evaluating the impact of a single agricultural technology adoption and its causal effect may result in a biased estimate (Biru et al., 2020; Ward et al., 2018). Although the adoption of wheat technol- ogy packages plays a key role in achieving food security and farm income, their status and packages’ selection remained challenging and limited to relying on a few technolo- gies. Similarly, its impact studies on smallholder farmers’ income improvement and food security did not get the expected attention, and there exist literature gaps for such an impact analysis. Therefore, this study focuses on a key research questions such as: (1) what factors determine adoption of both technology packages combinations and a single technology; (2) does adoption of technology packages improve households’ food secu- rity (food consumption score and dietary diversity score) and wheat production income; (3) if impact exists, which technology package/s adoption improve households’ food security and income and which not? and (4) what impact policy and strategy should be applied to improve adoption of technology packages? Overall, identifying the impact of adoption of wheat production technology packages on households’ food security and wheat production income has a vital role for imple- menting and intervening agricultural policies and strategies. Thus, this study attempts to address the gaps in the literature in twofold. First, it identifies factors that affect farmers’ decisions to adopt  full wheat technology packages. This method emphasizes the impor- tance of adopting multiple wheat technology packages rather than using a single or few packages. Second, it analyzes the impact of recommended wheat technology packages on households’ food security and income. Besides, to mitigate selection bias originated from both observed and unobserved heterogeneity, the study employed the multino- mial endogenous switching regression technique, which depends on a multinomial logit model. The rest of the paper is organized as follows: the section "Empirical review" rep- resents reviewed empirical literatures. The methodology used for adoption and impact analysis is presented by the section  "Research methodology". The 4th section presents estimations of empirical results and the 5th section draws conclusion and implications of the study. Page 4 of 26Zenbaba et al. Journal of Innovation and Entrepreneurship (2025) 14:48 Empirical review Impact evaluation brings positive feedback for routinely undertaken product develop- ment. So far, many studies on the impact analysis of agricultural technology adoption have been extensively undertaken in different areas. However, among these studies, many were focused on analyzing the impact of a single or a few technologies, which might not achieve the real potential expected outcomes. Several studies analyzed impact evaluation using the adoption of a single technology (Ahmed et al., 2016; Khonje et al., 2015; Oyetunde-Usman et al., 2021; Richard et al., 2020; Shiferaw et al., 2014; Solomon et al., 2016; Wake & Goshu, 2019), while some others focused on multiple or fully rec- ommended agricultural technology adoption (Ahmed, 2022; Biru et  al., 2020; Gebre- mariam & Wünscher, 2016; Kassie et al., 2015; Khonje et al., 2018; Manda et al., 2016; Marenya et al., 2020). The impact of improved agricultural technologies on household food security was studied by Hailu et al. (2021) and revealed that adoption of improved agricultural tech- nologies significantly increases dietary energy supply, dietary diversity, and food con- sumption score. Further, they concluded that impact is improved when technologies are adopted jointly rather than separately. Wordofa et al. (2021) discussed that adoption of agricultural technologies profoundly increases the income of smallholder farmers and this can be improved when awareness about agricultural technology adoption is created for farmers by the university extension wing, the district bureau of agriculture and natu- ral resources, non-governmental organizations (NGOs), and model farmers. Biru et al. (2020) revealed that the impact evaluation of agricultural technology adoption is less understandable with the adoption of a single or few agricultural technologies, but the adoption impact on household welfare becomes high as households adopt several agri- cultural technologies. Overall, the study identified that the adoption of different combi- nations of agricultural technologies had a greater impact on consumption, poverty, and vulnerability among smallholders than the adoption of single technologies. Zegeye et al. (2022) studied the impact of agricultural technology adoption on house- hold food consumption expenditure and identified that adopting agricultural technology significantly increases household food consumption expenditure per adult equivalent, while actual and counterfactual scenario differences in food consumption expenditure per adult equivalent of households are observed. Ahmed (2022) analyzed the impact of improved seed and inorganic fertilizer on maize yield and consumption expenditure and discussed that adoption of agricultural technology combinations (improved varie- ties and inorganic fertilizer) boosts maize yield and consumption expenditure more than adopting single agricultural technology. Khonje et al. (2018) revealed that adopting full or  multiple agricultural technology packages  improves household welfare (poverty), yields, and household income more than adopting each technology in isolation. The on- farm impacts of adopting a combination of improved agronomic practices (IAPs) on net crop income and agrochemical use in Malawi were conducted by Kassie et  al. (2015). Their findings revealed that the adoption of a combination of technologies had a positive impact on net crop income and reduced other costs of purchasing technologies com- pared to single technology adoption. Oyetunde-Usman et  al. (2021) discussed the welfare impact of organic fertilizer adoption in Nigeria and revealed that its adoption influenced per-capita total house Page 5 of 26Zenbaba et al. Journal of Innovation and Entrepreneurship (2025) 14:48 expenditure, per-capita asset value, and per-capita food expenditure. Jaleta et al. (2018) studied the impact of improved maize variety adoption on household food security and identified that adoption of improved maize varieties affected food security in maize growing areas of Ethiopia. Moreover, adopting improved maize varieties improves per- capita food consumption and increases the probability of a smallholder having a food surplus. Wake and Goshu (2019) showed that the adoption of high yielding wheat varie- ties had a greater impact on the farm income of households than non-adopters. Shiferaw et al. (2014) found that access to modern wheat varieties and their usage increase food security more than households that did not adopt improved wheat varieties. Khonje et al. (2015) conducted a study on the adoption and welfare impacts of improved maize varieties in eastern Zambia and revealed that adoption of improved maize improves gains in crop incomes, consumption expenditures, and food security. Solomon et  al. (2016) conducted a study on the impact of improved wheat technology adoption on pro- ductivity and income and discussed that improved wheat variety adoption increases the income of adopters more than that of non-adopters. Research methodology Description of the study area The study was conducted in Horo Guduru Wollega Zone, Oromia Region, Ethiopia, which is located 324  km away from the capital city of the country, Addis Ababa. The zone has a population of 576,737 inhabitants, while the proportion of men is 50.1% and the remaining women (CSA, 2007). The study area is characterized by different agro- ecologies, such as lowland, midland, and highland, and it has a high potential for wheat production. Wheat production serves as a primary source of food and serves the liveli- hood of smallholder farmers by generating income from selling the crop in the study area. Moreover, the area is blessed with a diversity of different agricultural crops, such as teff, barely, wheat, oat, nug, groundnut, sesame, maize, fruits, and vegetables. Rar- ing animals are well known in the study area (HGWZAO, 2023). Specifically, the study was undertaken in three randomly selected districts of the zone, namely, Horo, Ababo Guduru, and Abe Dongoro districts (Fig. 1). Methods of data collection The study used quantitative data collected from different sources. To appropriately col- lect a cross-sectional data from randomly selected farm households, enumerators were trained on the methods of data collection and household survey method was employed. These primary data were collected from households’ socio-demographic, socio-eco- nomic, institutional, resource factors, technological factors, income, and food secu- rity status using a structured questionnaire. Translating questionnaires to the farmers’ mother tongue (from English to Oromo) were also undertaken to avoid any misunder- standings of the questions. Accordingly, the required data were successfully collected from the head of the household (male or female). Secondary data were also collected from secondary data sources, such as published and unpublished papers, working papers, and agricultural office the zone and selected districts of the study area. Page 6 of 26Zenbaba et al. Journal of Innovation and Entrepreneurship (2025) 14:48 Sampling techniques and sample size To select a representative sample in the study area, a combination of purposive and prob- ability sampling techniques were employed. Horo Guduru Wollega zone was purposely selected as it is one of the zones in the western Ethiopia where potentially producing rain- fed wheat for home consumption and marketing is undertaken. Before selecting the repre- sentative districts in the zone, districts were stratified into lowland, midland, and highland agro-ecologies. Accordingly, at the first stage, three potential districts and one (1) district from each agro-ecology (one from highland, one from midland, and one from lowland areas) were selected purposefully. In each selected district, the list of kebeles was taken from agricultural offices, and through comparing the differences between groups, two kebe- les were randomly selected. Then, the list of wheat producers was taken from each kebeles development workers, and finally, a sample representative of smallholders was selected ran- domly. The sample size determination was undertaken using Kothari (2004) by comparing differences among groups at each agro-ecology. The allocation of sample size to each kebele was undertaken proportional to the household size of each kebele within their respective districts, as represented by Table 1 (1)n = Z2pqN e2(N− 1)+ Z2pq Fig. 1  Administrative map of the study area. Source: Own sketch Page 7 of 26Zenbaba et al. Journal of Innovation and Entrepreneurship (2025) 14:48 where n = is the sample size drawn from total population, N = is the total population of wheat producers from which sample size will be drawn, e is acceptable error, Z (1.96) = is the standard cumulative distribution that corresponds to the level of confidence, and p represents the proportion of an adopter that is present in the population, while q is the population proportion of non-adopter in the population when q = 1− p. Estimation strategy Making the decision to adopt full  technology packages could enable farmers to respond to external factors such as diseases, drought, pests, erosion, and soil infertility. Addi- tionally, their decision to select one or more interdependent agricultural technology packages could be affected by observed factors, such as demographic, institutional, and economic factors and/or unobserved factors that represent the innovation and ability of farmers (Ehiakpor et  al., 2021; Setsoafa et  al., 2022). Adoption of agricultural tech- nologies is exposed to self-selection problems by their nature, especially when farmers are categorized as non-adopters, single technology adopters, and two or more full pack- age adopters. In this case, if appropriate methods are not carried out, the end result of the impact analysis would be biased and inconsistent (Setsoafa et  al., 2022). However, some empirical studies, such as Jambo et al. (2021), Aweke et al. (2021), Wordofa et al. (2021), Wang et al. (2020), Ahmed et al. (2016), Kassa et al. (2014), and John et al. (2020), employed different econometric models such as PSM, OLS, and logit models, which may ignore observable selection bias. Besides, some other studies (Assaye et al., 2022; Aye- new et al., 2020; Hailu et al., 2021; Jaleta et al., 2018; Ngomi et al., 2020) used endog- enous switching regression model. However, most empirical studies employed multinomial endogenous switching regres- sion (MESR) if smallholder farmers had the probability of adopting two or more tech- nology packages. The model estimates the average treatment effects of adoption of technology packages on outcome variables. By applying this model, the issues of farmers’ self-selection bias among the technology packages could be addressed as observable and (1.96)2(0.5)(0.5)N (0.05)2(N− 1)+ (1.96)2(0.5)(0.5) = (1.96)2(0.5)(0.5)13,221 (0.05)2(13,221− 1)+ (1.96)2(0.5)(0.5) = 373, Table 1  Sample size of selected kebeles. Source: HGWZAO, 2023 Districts/Kebeles Household size of selected districts and kebeles Sample size of selected districts and kebeles Horo 4,253 120 Laku Igu 569 69 Gitilo Dale 417 51 Ababo Guduru 6,451 182 Lalistu Loya 671 97 Loya Malole 589 85 Abe Dongoro 2,517 71 Wirtu Senxa 261 39 Tulu Moti 219 32 Total 13,221 373 373 Page 8 of 26Zenbaba et al. Journal of Innovation and Entrepreneurship (2025) 14:48 unobservable factors of outcome variables might affect farmers’ interest. Accordingly, Ahmed (2022), Setsoafa et  al. (2022), Sisay (2024), Ngango et  al. (2022), Kassie et  al. (2015), Zegeye (2021), Ali et al. (2022), and Danso-Abbeam and Baiyegunhi (2018) used the MESR model. The model considers both observed and unobserved factors of impact analysis and comparatively solves the issue of selection bias when alternative technology adoption options exist. Multinomial logit selection model At this stage, MNL is employed to analyze factors affecting smallholder farmers’ deci- sions to adopt technology packages or combinations. The estimation of multinomial logit (MNL) models accounts for unobserved heterogeneity by generating inverse Mills ratios. At the second stage, using computed inverse Mills ratios as an independent vari- able to address selection bias problems, the outcome equations are estimated. The expected utility, U∗ im , that the household derives from the adoption m technology package is specified as follows: where U∗ im is a latent variable, Xi is observed exogenous variables, and εim is an error term that accounts for unobserved characteristics. Let M represent an index that denotes the farmer’s choice of package, such that M is given as follows: Smallholder farmers adopt a single or more combinations of technology packages M if they expect more utility from the adoption of the technology package(s) M than any other package j = M (Bourguignon et  al., 2007).According to Mc-Fadden (1973), the probability that farmer i , select technology package m given exogenous variable xi, by assuming that the error terms are identical and independently Gumbel distributed is specified as follow: where Prob (m|xi) is the probability that a given farmer i selects to adopt technology packages m, σm is constant term of technology packages m, βm is vector of parameters to be estimated, and Xi is a set of observed exogenous variables. Multinomial endogenous switching regression  Multinomial endogenous switching regression is employed to analyze different combinations of farmers’ technology pack- age combinations and their adoption impact on household food security (HFSC and HDDS) and wheat production income. This study used wheat technology packages that are widely applied in the study area, such as improved wheat varieties, row plant- ing, and the application of fertilizers. Farmers could have the possibility of using these technologies in eight (23) possible combinations, depending on their selection inter- (2)U∗ im = Xiβm + εim, (3) (4)Prob (m|Xi) = exp(σm + Xiβm) ∑m n=1 exp(σn + Xβn) , Page 9 of 26Zenbaba et al. Journal of Innovation and Entrepreneurship (2025) 14:48 ests. These possible combinations are, none of the packages adopted (I0R0F0) which is used as base category, improved seed only (I1R0F0), row planting only (I0R1F0), ferti- lizer only (I0R0F1), improved seed and row planting only (I1R1F0), improved seed and fertilizer only (I1R0F1), row planting and fertilizer only (I0R1F1), and all improved seed, row planting, and fertilizer (I1R1F1). In the second stage, the model equation, which shows the relationship between out- come variables (food security and farm income) and a set of endogenous variables given as A for the selected technology package(s), is separately estimated for both non-adopters and adopters of the technology packages. Accordingly, adoption of combinations of technology packages is represented as ( m = 2 . . . n ) and while non- adoption is given as m = 1 which serves as a reference category and the overall equa- tion is specified as follows: where Qi is ith farmer’s outcome variables with and without adoption technology pack- ages M; I is an index which represent farmer i’s selection of technology package/s; Ai is a set of exogenous variables; α1 and αM are parameters to be estimated; ui1 and uiM are error terms. The MESR framework estimates and calculates the selectivity correction terms under Eq. 2 and to mitigate unobserved selection bias included the corrections under Eqs.  5 and 6. The observed selection bias problems given by these equations were solved using vector of observed covariates given under Ai . However, if farmers’ deci- sion to adopt technology packages and outcome variables is simultaneously affected by the same unobserved factors, the error terms given under Eq. (2) and Eq. (5) would be correlated and unobserved selection bias occurs which resulted in biased estimate. Using these assumptions, Eq. (5) could be rewritten as follows: where Qi are ith farmer’s outcome variables with and without adoption technology pack- ages M; Ai is a set of exogenous variables; �1 & �M are selectivity correction terms applied for solving selection bias issues computed from the estimated probabilities in Eq.  (4) where, in this model, it is given as M-1 for each technology package; σ1 & σM are covari- ance between error terms given in Eq. (2), Eq. (5). To account for the heteroscedasticity arising from the generated repressor, the standard errors in Eq. (6) are bootstrapped. (5) (6) Page 10 of 26Zenbaba et al. Journal of Innovation and Entrepreneurship (2025) 14:48 According to Kassie et al. (2015), it is possible to identify the parameters of a mul- tinomial ESR model, even though selection and outcome equations have the same regressors. This means that to identify multinomial treatment effects’ parameter model, inclusion of instrumental variables in the selection equation is not strictly necessary. However, following Ahmed (2022), Kassie et  al. (2018) included the selection instrument to improve the consistency of the model but excluded it from outcome equations. For the validity of the MESR model, proper and consistent esti- mation of appropriate selection instruments was included in exogenous variables but not in outcome equations (Ai). Market distance and distance from the training center (measured in minutes) were used as selection instruments, and these vari- ables were not expected to directly influence outcome variables but rather through package combination adoption. A simple falsification test was performed to con- firm the admissibility of these instruments, following Di Falco et al. (2011). Accord- ingly, the selected instruments were valid and jointly affected wheat technology package combinations adoption decisions but not affected non-adopter outcome variables such as food security (FCS and HDDS) and wheat production income (Appendix Table 7). Estimation of average treatment effects  To analyze impact of wheat production technol- ogy package adoption on households’ food security and wheat production income, the study used observational cross-sectional data. Using such data for impact analysis of non- excremental observations might create a challenge with the estimation of counterfactual outcome (if they had adopted the technology) as the outcome of technology adopters is not observed. However to solve these issues, the study employed multinomial endoge- nous switching regression (MESR) method which computes average treatment effects on treated and untreated (ATTs and ATUs). Following studies undertaken by Setsoafa et al. (2022); Ahmed (2022); Kassie et al. (2018),  the average treatment effect on the treated (ATT) estimation for actual and counterfactual scenarios is given using equations. Adop- ters with adoption (actual) are given by Eq. 7a & 7b. The equations represent the expected outcomes of food security and wheat production income for adopters and non-adopters of technology packages.  Adopters had they decided not to adopt (counterfactual) is given by Eq. 8a & 8b.  Therefore, ATT is calculated using the differences of the above actual and counterfactual Eqs. (6a & 8a or 6b & 8b). By using 7a & 8a, ATT is presented as follows: (7) (8) Page 11 of 26Zenbaba et al. Journal of Innovation and Entrepreneurship (2025) 14:48 Measurement of outcome variables In this study, the outcome variables represent household food security status measured by Food Consumption Score (FCS) and Household Dietary Diversity Scores (HDDS) and smallholder farmers’ annual wheat production income (in ETB). Food security measure- ment is not easy, and no fixed method is employed to identify its status and impact, as it might be calculated through methods with different unobservable characteristics. The challenges with food security measurement were along its level of measurement, with its ability to assess, quantify, and qualify food security (Jones et al., 2013; Webb et al., 2006). However, according to Maxwell (1996), Izraelov and Silber (2019), Ogundari (2017), Maxwell et al. (2014), and Cafiero et al. (2014), different indicators of food secu- rity measurements are developed and being used to measure the status of food security in households. This study used the household food consumption score and household dietary diversity scores developed by WFP and are mostly used as proxies for the main food access indicators as proxies for food security (WFP, 2008). The Food Consumption Score (FCS) is an index that aggregates dietary diversity, food frequency, and the nutritional importance of food groups consumed. The family mem- bers who make food were asked yes-or-no questions about how many days at least one member of the family consumed at least one food group during the preceding 7 days of the interview. It is calculated by multiplying the frequency of food groups consumed over the previous seven days by the weighted relative nutritional value (WFP, 2008). The weighting value of each food type varies depending on the nutritional density of the food group classified by WFP (INDDEX Project, 2018). The eight types of food items include: main staples, pulse, vegetables, fruit, meat/fish, milk, sugar, and oil. Household Dietary Diversity Score (HDDS) measures the quantity and quality of food access for 12 food groups consumed by a household over a given reference period. It relies on food or food items consumed within the last 24 h and households’ ability to access food (Hoddinott & Yohannes, 2002; Kennedy et al., 2011). DDS is aimed at iden- tifying whether a family member consumed at least one food group from each of the 12 food categories over the preceding 24 h at interview time (Swindale & Bilinsky, 2006). The procedure follows mostly by asking a household member who prepares food using ‘yes’ or ‘no’ questions and counting down the list of food items consumed, which may range from 0 to 12, with 0 for none of the food groups consumed and 12 for all food groups consumed within the last 24 h. By the two methods, this study did not consider food consumed occasionally; rather, it depends on food consumed regularly to reduce estimation bias. These 12 food groups included in this study are: cereals, roots/tubers, vegetables, fruits, meat/poultry, eggs, fish, pulses/legumes/nuts, milk/milk products, oil/fats, sugar/honey, and miscellaneous. Wheat production income represents small- holders’ wheat production annual income and measured in Ethiopian Birr (ETB). It is obtained by valuing wheat product (quintal) at market price and deducting the costs of all variable inputs during last production seasons of when survey was undertaken. (9) Page 12 of 26Zenbaba et al. Journal of Innovation and Entrepreneurship (2025) 14:48 Results and discussion Descriptive statistics result The study depends on different categories of wheat technology packages adoption, such as wheat improved seed, row planting, and the application of recommended quantities of chemi- cal fertilizer. The selected three packages generated eight possible combinations of the pack- ages in a single or combinations of the packages. In this study, a package or combinations of packages of 25 and above observations were selected, and ineligible package of zero obser- vation was dropped. Accordingly, among the combinations of the packages, 19.57% (73) of the sampled households did not adopt any technology package (I0R0F0). About, 6.70% (25) sampled households adopted improved seed and row planting only (I1R1F0), about 10.99% (41) farmers adopted improved seed and chemical fertilizers only (I1R0F1), and none of them adopted wheat row planting only (I0R1F0). Among the packages, 9.38% (35) of households adopted improved seed only (I1R0F0) and 15.28% (57) adopted improved chemical fertilizer only (I0R0F1). About 28.15% (105) farmers adopted full wheat technology packages, such as improved seed, row planting, and the recommended amount of chemical fertilizer (I1R1F1). Therefore, for this study, among the total combinations of technology package(s), such as no adoption (I0R0F0), adoption of all packages (I1R1F1), adoption of improved wheat seed and row planting only (I1R1F0), adoption of improved seed and chemical fertilizers only (I1R0F1), adoption of row planting and chemical fertilizers only (I0R1F1), adoption of improved seed only (I1R0F0), and adoption of chemical fertilizer only (I0R0F1) package combinations, were selected, while no adoption technology package (I0R0F0) is used as a reference category (Table 2). Descriptive statistics of explanatory variables, outcome variables, and their compar- ison with respect to different categories of wheat technology packages are given by appendix Table 5. Empirical results Factors affecting adoption decision of technology packages The first stage of the multinomial endogenous treatment effects model is estimated using the multinomial logit model, which is represented by Table 3. To compare the technology package or combination, non-adoption of any technology package (I0R0F0) is used as a base category. The model fits the data substantially well with the Wald chi2(90) = 3117.59; χ2 = 0.000 which shows that the rejection of the null hypothesis that assumed all the regression coefficients is jointly equal to zero. The model has been checked for multi- collinearity problems using VIF, and no multicollinearity problems are seen, as given in Table 2  Different wheat technology packages’ combinations Source: Own survey result, 2023 Packages Packages descriptions Frequency Percentage I0R0F0 No adoption (base category) 73 19.57 I1R1F1 Improved seed, row planting, and fertilizer 105 28.15 I1R1F0 Improved seed and row planting only 25 6.70 I1R0F1 Improved seed and fertilizer only 41 10.99 I0R1F1 Row planting and fertilizer only 37 9.92 I1R0F0 Improved seed only 35 9.38 I0R1F0 Row planting only 0 0.00 I0R0F1 Fertilizer only 57 15.28 Page 13 of 26Zenbaba et al. Journal of Innovation and Entrepreneurship (2025) 14:48 Appendix Table 6. As shown in Table 3, the age of the household head positively influ- enced the likelihood of adoption of recommended fertilizers combination (I0R0F1)only, at 5% significance level. However, the age of household farmers  had no significant effect on other technology packages adoption  which implies that older farmers may lack the knowledge of importance of adoption of  full technology packages but adopt single and incomplete technology package/s.Younger farmers may be educated, trained, decision and idea makers, risk takers, and eagers to adopt recommended and timely diffused tech- nologies. Sex of household positively influenced the probability of adopting improved seed and row planting technology packages (I1R1F0) only at 1% significance level, respec- tively. Compared with female farmers, male households mostly engage in income gener- ating activities and get higher income which may help them to purchase improved wheat Table 3  Determinants of  smallholder farmers’adoption of wheat  technology package combinations. Source: Own survey result, 2023 Where ***, **, & * represent the significance at 1%, 5%, & 10% probability levels, respectively. Independence of Irrelevant Alternatives of MNL assumption is checked by the Stata command “modern hausman mlogtest iia” and confirms the validity of the model Variables Categories of technology package combinations I0R0F1 I1R0F0 I0R1F1 I1R0F1 I1R1F0 I1R1F1 Age 0.046** (0.022) 0.030 (0.029) − 0.026 (0.031) 0.023 (0.028) 0.019 (0.026) 0.003 (0.023) Sex 1.141 (0.701) –0.059 (0.605) –0.337 (0.631) 1.217 (0.846) 13.561*** (0.493) -0.309 (0.532) Education 0.131 (0.102) 0.171 (0.109) –0.016 (0.130) 0.015 (0.108) 0.157 (0.124) 0.217** (0.086) Market dis- tance –0.039*** (0.014) –0.034** (0.015) –0.033** (0.016) –0.035** (0.017) –0.035* (0.021) –0.053*** (0.014) Training distance –0.039** (0.019) –0.056*** (0.020) –0.048** (0.021) –0.059*** (0.022) –0.043** (0.021) –0.064*** (0.020) Farm distance –0.007 (0.021) –0.011 (0.026) –0.019 (0.024) –0.025 (0.027) 0.012 (0.024) –0.058** (0.024) Mobile owner- ship 0.539 (0.416) 0.227 (0.478) 0.447 (0.478) 1.051** (0.443) 0.985* (0.509) 1.034*** (0.382) Radio owner- ship 0.238 (0.408) 0.139 (0.476) –0.558 (0.483) –0.331 (0.473) –0.159 (0.601) 0.267 (0.387) Credit services 0.717* (0.407) –0.219 (0.450) –0.394 (0.456) 0.089 (0.440) –0.296 (0.500) 0.518 (0.372) Extension services –0.573 (0.429) –0.497 (0.503) 0.038 (0.460) –0.635 (0.464) –0.633 (0.537) –0.505 (0.378) Cooperative membership 0.459 (0.386) 0.062 (0.460) 0.741* (0.445) 0.348 (0.428) 0.660 (0.519) 0.103 (0.369) Farm size –0.068 (0.232) –0.581* (0.308) –0.273 (0.249) –0.460 (0.282) –0.318 (0.331) –0.071 (0.215) Livestock 0.100 (0.114) –0.014 (0.125) 0.098 (0.135) 0.204* (0.119) 0.137 (0.154) 0.238** (0.108) Farm training 0.635 (0.403) 0.249 (0.464) 0.050 (0.438) 0.599 (0.441) 1.174** (0.534) 0.787** (0.370) Off/non-farm income 0.480 (0.391) 0.024 (0.448) 0.450 (0.436) 0.460 (0.432) 0.062 (0.513) 0.259 (0.354) Constant –2.903** (1.443) 1.297 (1.551) 2.681* (1.617) –0.830 (1.788) –14.86*** (1.657) 1.227 (1.287) Wald chi2(90) = 3117.59 Log likelihood = –604.414 Pseudo R2 = 0.1185 Number of observation = 373 Prob > chi2 = 0.0000 Page 14 of 26Zenbaba et al. Journal of Innovation and Entrepreneurship (2025) 14:48 seed and chemical fertilizer. Male farmers may withstand the risks of finding improved seed and chemical at far distance when the availability of these inputs is not sufficient, and which influence female famers using the recommended amount of these inputs. The result is in line with the research findings of (Hailu et al., 2021). The educational level of households has a significant effect on the adoption of combi- nations of full technology packages, such as improved seed, row planting, and chemical fertilizer (I1R1F1) at 5% significance level, while having no significant effect on other sin- gle or combinations of technology package(s). This implies that educated farmers have knowledge and awareness about the importance of new agricultural technologies and the effect of adopting all the recommended technologies for improving their welfare rather than adopting a single or few packages. Educated farmers easily access informa- tion through collecting, analyzing, and using it for the adoption of all the recommended improved technology packages compared to those who were less educated. The result is consistent with the research findings of Manda et al. (2016). Distance from the nearest market significantly and negatively influenced the proba- bility of households adopting I1R1F1 (improved seed, row planting, chemical fertilizer) technology combinations, I1R1F0 (improved seed and row planting) technology combi- nations only, I1R0F1 (improved seed and chemical fertilizer) technology combinations only, I0R1F1 (row planting and chemical fertilizer) technology combinations only, and I1R0F0 (improved seed) technology combinations only and I0R0F1 (chemical fertilizer) technology combinations only. This shows that farmer households who are located far from input markets are constrained to get and purchase agricultural inputs, which might be due to a lack of information, a high cost of transportation, a lack of roads, and being forced to purchase easily transportable inputs such as chemicals. The result is in line with the research findings of Ali et  al. (2022). Similarly, distance from training cent- ers like farmers’ training center (FTC) and households’ location distance from their farm site negatively and significantly influenced all farmers’ adoption of wheat pack- age combinations, such as I1R1F1, I1R1F0, I1R0F1, I0R1F1, I1R0F0, and I0R0F1. This shows that as households are located far from training centers, they might lack understanding about technology adoption, and their probability of adopting technology packages has decreased. Moreover, farm distance negatively and significantly influenced households’ adoption decision of full wheat production technology package combinations (I1R1F1) at 5% significance level. . As farms are far from their homestead, they may face the prob- lems of regularly supervising their land and the higher cost of production, which can hinder their decision to adopt wheat technology packages. Mobile phones are found to have a positive and significant influence on households’ decisions to adopt wheat technology packages (I1R1F1, I1R1F0, and I1R0F1,). Mobile phones are a key resource for farmers to exchange information about the importance, methods, availability, price, and sources of agricultural inputs, which further contributes to farm- ers’ adoption of all or nearly all recommended technology packages. Access to credit also had a significant and direct effect on farmers’ adoption of chemical fertilizer technol- ogy packages (I0R0F1). Farmers who had access credit service for agricultural production have adopted recommended amount of chemical fertilizer technology package (I0R0F1) which enable them to be able to purchase fertilizer at right time and quantity. Farmers’ agricultural cooperative membership positively and significantly influenced adoption of Page 15 of 26Zenbaba et al. Journal of Innovation and Entrepreneurship (2025) 14:48 combination of recommended row planting and fertilizer application (I0R1F1) only. Agri- cultural cooperative membership enables farmers to easily access to agricultural produc- tion inputs, access to credit, share knowledge and experience which could improve their adoption decision of combination of wheat row planting and application of fertilizer. The farmers’ land size negatively and significantly influenced the probability of adopting improved seed and chemical fertilizer (I1R0F1) technology package combinations only. Larger land size needs efficient management and farmers less efficient in applying recom- mended improved wheat seed as they own large size of land. Farm training obtained through advisory services had a significant and positive influ- ence on the likelihood of farmers’ adoption of I1R1F1 and I1R1F0. This suggests that farm households that get training services are more likely to adopt technology packages than their counterparts. Training helps households as an instrument in acquiring knowledge on newly implemented agricultural technologies and getting agricultural input price infor- mation from different sources and their availabilities. Overall, training centers train farm- ers about technology package adoption through advising, teaching, raising awareness, and creating practical work for minimizing households’ welfare constraints. The result is con- sistent with the research findings of Hailu et al. (2021) and Zegeye et al. (2022). The presence of a larger number of livestock owners has positively and significantly affected the probability of adopting I1R1F1 (improved seed, row planting, and chemical ferti- lizer) and I1R0F1 (improved seed and chemical fertilizer) technology package combinations only. This shows that farmers who have more livestock adopted full technology packages than those who have less. Livestock helps farmers generate income for purchasing agricul- tural inputs and improves soil fertility through the provision of manure and the preparation of compost. The result is consistent with Zegeye et al. (2022) and Ali et al. (2022), who dis- cussed that livestock had a positive effect on the recommended technology package. Average treatment effects of wheat technology packages adoption The second stages of the multinomial endogenous switching regression model are employed to estimate determinants of food consumption score (Table  8), household dietary diversity score (Table  9), and wheat production income are given in appendi- ces (Table 10). After identifying determinants of adoption of wheat technology package combinations, the treatment effects of wheat technology package adoption on household food security (measured by household food consumption score and household diversity dietary score) and wheat production income are estimated and represented by Table 4. The findings of the study identified that adoption of full technology packages improves outcome variables (such as household food security and wheat production income) more than adopting a single or a few combinations of technology packages. According to ATT estimation, adoption of full technology packages, such as improved seed, row planting, and application of recommended fertilizer (I1R1F1), increase house- hold food consumption score by 13.29%. This implies that implies that jointly adopting technology packages improves HFCS than adopting them in a few combinations or in isolation. In support to this finding, Aweke et  al. (2021), Hailu et  al. (2021) discussed that joint adoption of improved agricultural technology packages resulted in highest food consumption score than when adopted in isolation. Similarly, adoption of full tech- nology packages had a significance impact on the probability of households to diversify Page 16 of 26Zenbaba et al. Journal of Innovation and Entrepreneurship (2025) 14:48 their diets. Households can improve their diversity of food by 8.88% as they jointly adopt all the three wheat production technology packages (I0R0F1). The result is in line with the research findings of Hailu et al. (2021), Setsoafia et al. (2022), Aweke et al. (2021). Similarly, the results revealed that adoption full technology packages, such as improved seed, row planting, and application of recommended fertilizer (I1R1F1), increase wheat production income (Birr/hectare) by 4.45%. This implies that adoption of these packages improves the households’ wheat production income than adopting the technologies in isolation. Previous studies, such as Marenya et al. (2020), Kassie et al. (2018), Setsoafia et al. (2022), Wordofa et al. (2021), and Manda et al. (2016), showed that adoption of  full technology packages improves farm income than adopting a few or single technology. Moreover, the findings of the study show that households’ adoption of improved seed only (I1R0F0) resulted in a significant and positive impact on household wheat produc- tion income and it increases wheat production income (Birr/hectare) by 1.43%. The study also revealed that households who adopted only fertilizer (I0R0F1) increase their income from wheat production by 1.44% (Birr/hectare). The result of the study also showed that, adoption of the combination of improved seed and row planting (I1R1F0) increases the wheat production income (Birr/hectare) by 1.53%. Households who adopted by com- bining row planting with chemical fertilizer (I0R1F1) could improve HDDS, HFCS and wheat production income (Birr/hectare) by 8.88%, 13.29%, and 1.84%, respectively. A farm household that adopt the combination of improved seed and chemical fertilizers (I1R0F1) improves their level of DDS, FCS, and income obtained from wheat production in ETB by 9.22%, 15.54%, and 2.25%, respectively. Table 4  Treatment effect estimation of impact of wheat technology packages adoption on FCS, HDDS, and income. Source: Own survey result, 2023 Where ***, **, & * represent the significance at 1%, 5%, & 10% probability levels, respectively. Outcome variables Technology packages Adopting (actual) (A1) Non-adopting (counterfactual) (A0) ATT (B) (A1-A0) ( B A0 )*100 FCS I1R1F1 50.54(1.75) 43.35(1.29) 7.19(2.17)*** 16.58 I1R1F0 50.52(4.26) 44.97(1.09) 5.54(4.40) 12.31 I1R0F1 51.51(3.74) 44.58(1.09) 6.93(3.90)* 15.54 I0R1F1 50.70(2.66) 44.75(1.14) 5.94(2.89)** 13.27 I1R0F0 46.30(2.89) 45.24(1.13) 1.05(3.10) 2.32 I0R0F1 45.58(3.24) 45.29(1.11) 0.28(3.43) 0.61 HDD I1R1F1 6.80 (0.16) 6.18(0.12) 0.62(0.20)*** 10.03 I1R1F0 6.40(0.46) 6.35(0.10) 0.05 (0.47) 0.78 I1R0F1 6.87(0.32) 6.29(0.10) 0.58(0.34)* 9.22 I0R1F1 6.86(0.30) 6.30(0.11) 0.56(0.32)* 8.88 I1R0F0 6.45(0.29) 6.34(0.11) 0.11(0.31) 1.73 I0R0F1 6.47(0.27) 6.33(0.11) 0.14(0.29) 2.21 Wheat production income I1R1F1 10.08(0.05) 9.65(0.04) 0.42(0.06)*** 4.35 I1R1F0 9.91(0.07) 9.76(0.03) 0.15(0.08)* 1.53 I1R0F1 9.97(0.10) 9.75(0.03) 0.22(0.11)** 2.25 I0R1F1 9.93(0.09) 9.75(0.03) 0.17(0.17)* 1.84 I1R0F0 9.90(0.07) 9.76(0.03) 0.14(0.08)* 1.43 I0R0F1 9.89(0.06) 9.75(0.04) 0.14(0.07)* 1.44 Page 17 of 26Zenbaba et al. Journal of Innovation and Entrepreneurship (2025) 14:48 Conclusions and policy suggestions Various impact studies of agricultural technologies on households’ food security and income rely on a single or few technology package combinations, which mostly face the problem of incomplete and inconsistent outcomes. This study was initiated to identify the impact of wheat technology package adoption on households’ food security (food consumption score and dietary diversity score) and wheat production income among smallholder farm- ers in western Ethiopia. To do so, it employed a multinomial endogenous switching regres- sion (MESR) model to analyze the determinants of smallholder farmers’ adoption decisions of wheat technology packages and estimate the influence of package(s) adoption on house- hold food security (FCS and HDD) and wheat income. Findings of the study identified that adoption of full  technology package combinations greatly contributes to the improvement of household food security and wheat income compared to packages adopted in isolation or in a few combinations. Therefore, to maximize the expected outcome of technology package adoption, farmers need to adopt full technology package combinations rather than adopting incomplete packages. Besides, the concerned bodies should design and set policies and strate- gies that aim to tackle wheat production technology package adoption constraints smallholder farmers’ wheat production and aware farmers about the joint adoption of all the packages to improve the expected returns of adoption effects. The multinomial logit model result of the study also showed that farmers’ decision to adopt wheat technology package/s combinations, such as I1R0F0, I0R0F1, I1R1F0, I1R0F1, I0R1F1, and I1R1F1, was significantly influenced by different socio-demographic and socio-economic fac- tors. Among these influential factors, age, sex, education level of the household head, distance to market, training centers and farm plots, mobile ownership, access to credit service, agri- cultural cooperative membership, farm size, livestock, and farm training services significantly influenced adoption decision of households’ technology package combinations. Focusing on opportunities and challenges of wheat production technology packages through providing credit services, training services, and intervening in other adoption constraints, profoundly improve adoption of wheat technology packages and achieve households’ expected maximum potential in wheat production productivity and production income. These achievements con- tribute for tackling households’ long-lasting food insecurity problems and enable them to expand their own sources of income generation which will play an important role in reducing the level of poverty in the region. The effect of all these factors is clearly identified when all households adopt full tech- nology package combinations (I1R1F1), which needs further improvement of all these sig- nificant factors for farmers’ to fully adopt these technology packages. Hence, the study recommends strengthening the provision of adoption enhancing services and encour- aging smallholders to adopt packages of technologies for better food security and live- lihood outcomes. Overall, future studies could undertake further studies as this study used cross-sectional data from specific locations which may fail to capture the complete impact of wheat production technology packages adoption on households’ food security and wheat production income over a long time at with wide area coverage. Appendix See Tables 5, 6, 7, 8, 9 and 10 Page 18 of 26Zenbaba et al. Journal of Innovation and Entrepreneurship (2025) 14:48 Ta bl e  5  Su m m ar y of d es cr ip tiv e st at is tic v ar ia bl es . So ur ce : O w n su rv ey re su lt, 2 02 3 I 1R 1F 1 r ep re se nt s ad op tio n al l o f i m pr ov ed s ee d, ro w p la nt in g, a nd c he m ic al fe rt ili ze r; I 1R 1F 0 r ep re se nt s ad op tio n of im pr ov ed s ee d an d ro w p la nt in g on ly ; I 1R 0F 1 r ep re se nt s ad op tio n of im pr ov ed s ee d an d ch em ic al fe rt ili ze r o nl y; I 0R 1F 1 r ep re se nt s ad op tio n of ro w p la nt in g an d ch em ic al fe rt ili ze r; I 1R 0F 0 r ep re se nt s ad op tio n al l o f i m pr ov ed s ee d on ly ; I 0R 0F 1 r ep re se nt s ad op tio n of c he m ic al fe rt ili ze r o nl y Va ri ab le s I 1R 1F 1 I 1R 1F 0 I 1R 0F 1 I 0R 1F 1 I 1R 0F 0 I 0R 0F 1 I 0R 0F 0 M ea n St d. d ev M ea n St d. d ev M ea n St d. d ev M ea n St d. d ev M ea n St d. d ev M ea n St d. d ev M ea n St d. d ev A ge 38 .9 2 8. 00 41 .0 4 8. 38 40 .5 8 9. 72 38 .0 0 7. 89 41 .1 4 8. 98 42 .7 0 8. 71 39 .7 3 7. 74 Se x 0. 84 0. 36 1 0 0. 95 0. 21 0. 78 0. 41 0. 82 0. 38 0. 94 0. 22 0. 83 0. 37 Ed uc at io n 4. 43 2. 41 4. 04 2. 00 3. 58 2. 22 3. 21 2. 29 3. 94 1. 99 3. 85 2. 52 3. 09 1. 95 M ar ke t d is ta nc e 29 .8 5 12 .9 32 .4 13 .2 32 .5 6 13 .6 32 .4 3 13 .9 32 12 .3 32 .2 8 13 .9 38 .2 1 13 .8 Tr ai ni ng c en te r d is ta nc e 17 .8 5 8. 90 20 .0 0 7. 21 18 .1 7. 88 18 .2 8. 26 18 .7 6. 22 19 .9 9. 03 22 .8 7 13 .5 Fa rm d is ta nc e 11 .7 1 8. 02 16 .6 8. 0 13 .6 5 9. 81 14 .1 8 8. 12 14 .7 1 9. 31 15 .6 1 9. 86 16 .0 9 8. 17 M ob ile o w ne rs hi p 0. 48 0. 50 0. 52 0. 50 0. 51 0. 50 0. 29 0. 46 0. 31 0. 47 0. 42 0. 49 0. 23 0. 42 Ra di o ow ne rs hi p 0. 34 0. 47 0. 28 0. 45 0. 26 0. 44 0. 21 0. 41 0. 34 0. 48 0. 35 0. 48 0. 32 0. 47 C re di t s er vi ce 0. 64 0. 48 0. 44 0. 50 0. 56 0. 50 0. 40 0. 49 0. 45 0. 50 0. 66 0. 47 0. 46 0. 50 Ex te ns io n se rv ic e 0. 45 0. 50 0. 32 0. 47 0. 34 0. 48 0. 45 0. 50 0. 37 0. 49 0. 36 0. 48 0. 47 0. 50 Co op er at iv e m em be rs hi p 0. 44 0. 49 0. 60 0. 50 0. 53 0. 50 0. 59 0. 49 0. 45 0. 50 0. 56 0. 50 0. 42 0. 49 La nd s iz e 1. 97 0. 80 1. 90 1. 00 1. 81 0. 86 2. 00 0. 81 1. 73 0. 84 2. 02 0. 96 2. 17 0. 77 Li ve st oc k 6. 68 2. 04 6. 24 1. 72 6. 33 1. 61 5. 99 1. 61 5. 81 1. 24 6. 17 1. 80 5. 76 1. 58 Tr ai ni ng re ce iv ed 0. 53 0. 50 0. 60 0. 50 0. 48 0. 50 0. 37 0. 49 0. 40 0. 49 0. 47 0. 50 0. 35 0. 48 O ff- fa rm in co m e 0. 48 0. 50 0. 40 0. 50 0. 51 0. 50 0. 48 0. 50 0. 40 0. 49 0. 50 0. 50 0. 38 0. 48 FC S 50 .1 9 17 .8 50 .5 2 21 .7 51 .5 1 24 .3 50 .7 0 16 .4 46 .3 17 .3 45 .5 8 24 .7 29 .7 6 12 .8 H D D S 6. 75 1. 73 6. 40 2. 38 6. 87 2. 08 6. 86 1. 85 6. 45 1. 78 6. 47 2. 07 5. 08 1. 89 W he at in co m e 26 47 3. 81 12 31 8. 92 21 83 2. 00 89 97 16 25 98 0. 48 15 02 6. 63 24 11 4. 86 13 10 5. 75 22 05 4. 28 10 74 5. 18 21 99 7. 36 98 55 .0 4 97 78 .0 8 65 08 .9 0 Page 19 of 26Zenbaba et al. Journal of Innovation and Entrepreneurship (2025) 14:48 Table 6  Multicollinearity test. Source: Own survey result, 2023 Variables VIF 1/VIF Age of household head 1.08 0.922614 Sex of household head 1.05 0.954336 Education of household head 1.22 0.818627 Market distance 1.05 0.954728 Training center distance 1.08 0.931972 Farm distance 1.21 0.823485 Mobile ownership 1.06 0.939543 Radio ownership 1.07 0.938751 Access to credit services 1.09 0.918226 Access to extension services 1.17 0.854532 Cooperative membership 1.06 0.945857 Farm size 1.13 0.882179 Livestock 1.27 0.855561 Farm training 1.05 0.952526 Access to off/non-farm income 1.03 0.975956 Mean vif 1.10 Table 7  Validity test of the selection instrument based on non-adopter households Variables FCS HDDS lnIncome Age of household head 0.262(0.228)  -0.026(0.033) -0.019(0.011)* Sex of household head  -4.950(4.404)  -0.514(0.651) 0.282(0.212) Education of household head  -0.433(0.807) 0.053(0.119) 0.061(0.038) Market distance  -0.076(0.121)  -0.011(0.017) -0.003(0.005) Training center distance  -0.013(0.125)  -0.022(0.018) 0.006(0.006) Farm distance 0.143(0.217) 0.005(0.032) -0.011(0.010) Mobile ownership 1.776(3.891)  -0.188(0.575) -0.391(0.187)** Radio ownership -8.843(3.504)** 0.367(0.518) 0.613(0.169)*** Access to credit services 5.933(3.407)* 1.215(0.503)** -0.146(0.164) Access to extension services 2.696(3.216) -0.159(0.475) -0.1645(0.155) Cooperative membership  -2.186(3.174) 0.634(0.469) 0.536(0.153)*** Farm size 1.998(2.203) -0.058(0.325) 0.071(0.106) Livestock 1.187(1.123) 0.365(0.166)** -0.001(0.054) Farm training 3.272(3.194) 0.125(0.472) -0.137(0.154) Access to off/non-farm income  -5.690(3.423) -0.058(0.506) -0.017(0.165) Constant  14.945(12.474) 4.420(1.844)** 9.178(0.602)*** R-squared 0.2882 0.2834 0.4175 Page 20 of 26Zenbaba et al. Journal of Innovation and Entrepreneurship (2025) 14:48 Ta bl e  8  Se co nd s ta ge e st im at es o f M ES R fo r f oo d co ns um pt io n sc or e (F C S) W he re * ** , * *, & * re pr es en t t he s ig ni fic an ce a t 1 % , 5 % , & 1 0% p ro ba bi lit y le ve ls , r es pe ct iv el y Va ri ab le s I 0R 0F 0 I 1R 1F 1 I 1R 1F 0 I 1R 0F 1 I 0R 1F 1 I 1R 0F 0 I 0R 0F 1 Co ef St d. d ev Co ef St d. d ev Co ef St d. d ev Co ef St d. d ev Co ef St d. d ev Co ef St d. d ev Co ef St d. d ev A ge -0 .5 2 0. 41 -1 .0 3 1. 12 -4 .6 7 4. 66 -3 .3 4 5. 69 -1 .2 2 4. 47 -1 .0 9 2. 00 1. 80 * 1. 04 Se x -3 .7 6 15 .7 -3 0. 35 32 .6 4 -5 .7 8 9. 89 -5 4. 65 30 1. 8  2 .1 7 6. 60  4 6. 07 57 .1 0 81 .6 8* * 39 .6 6 Ed uc at io n  -3 .1 4* 1. 86 -1 .0 1 1. 96 11 .0 8 18 .6 8 -6 .1 2 21 4. 6  0 .9 1 10 .1 2 -1 0. 47 11 .9 9 -2 .7 5 5. 28 M ar ke t d is ta nc e 0 .0 33 0. 39 -0 .0 6 0. 42 2. 06 1. 96 -0 .3 0 55 .1 5  1 .5 3 1. 79  1 .3 7 3. 55  1 .3 6* ** 0. 44 Tr ai ni ng d is ta nc e  -0 .0 7 0. 29 -0 .0 8 0. 78  3 .4 7* ** 1. 12 -0 .0 08 19 .1 3  1 .8 9 5. 96  1 .4 9 2. 56  2 .2 1 1. 88 Fa rm d is ta nc e 0 .2 1 0. 38 -1 .6 3 1. 11 6. 40 6. 08 -0 .8 4 72 .8 3  1 .1 9 1. 48  2 .4 4 6. 21  3 .1 9* 1. 63 M ob ile o w ne rs hi p 8 .6 6 6. 08 -4 .7 7 5. 83 -6 5. 2* ** 18 .8 1. 62 98 7. 5 -2 9. 49 58 .8 2 -4 0. 42 37 .6 2 -3 9. 57 * 23 .3 9 Ra di o ow ne rs hi p -1 6. 9* ** 3. 18  1 .1 7 17 .8 1 -4 1. 33 11 6. 7 -3 2. 19 54 1. 3  1 0. 12 38 .1 9 -3 2. 82 79 .6 4 -1 3. 73 37 .9 3 C re di t s er vi ce s  -1 .8 2 5. 21  6 .9 2 4. 80 -1 62 .9 ** 79 .3 3 -3 8. 88 16 92 .2 -1 5. 98 77 .0 6 -1 6. 62 96 .5 5 -4 3. 1* ** 8. 63 Ex te ns io n se rv ic es 6 .6 7 5. 37  7 .2 3 6. 36  2 4. 34 86 .3 8 48 .5 4 46 9. 9 16 .1 4 11 .9 1  1 0. 79 15 .9 8  1 .9 8 17 .3 5 Co op er at iv e m em be rs hi p 2 .5 7 5. 14  3 .4 8 7. 90  5 2. 81 10 6. 23  4 5. 35 30 0. 1 -2 2. 64 92 .1 3  6 .8 3 98 .0 8  9 .4 4 18 .0 2 La nd s iz e 2 .6 5 1. 83 5 .1 7 4. 00 -1 7. 57 28 .7 4  9 .2 2 39 5. 5  4 .6 3 68 .4 3  2 .3 8 15 .2 0 -1 9. 57 16 .5 5 Li ve st oc k  1 .7 6 2. 77 2 .0 8 3. 42 -3 0. 6* * 12 .9 1  3 .8 2 10 6. 3 -5 .3 2 17 .6 9 -7 .4 0 39 .5 9 -1 1. 6* ** 4. 05 O ff/ no n- fa rm in co m e -3 .7 6 9. 14 -0 .4 6 10 .2 3 -9 7. 0* ** 25 .4 4  7 .4 4 22 9. 4 -6 .2 8 59 .2 6 -2 8. 39 22 .8 3 -1 9. 00 24 .0 5 An ci lla ry σ 2 99 57 .4 16 77 52 28 .7 62 89 51 63 ** * 16 89 14 12 11 78 17 57 24 00 75 14 57 55 99 95 ** * 36 42 0 � 0 0 .6 3  0 .4 1  0 .4 0 0. 91 -0 .2 8 0. 19  0 .7 5 0. 58  0 .5 6 0. 52  0 .9 1 0. 39 � 1 0 .0 4 0. 82 -0 .1 8 0. 81 -0 .0 4 0. 86 -0 .4 5 0. 90 -0 .7 3 0. 99 -1 .0 7* * 0. 42 � 2  -0 .0 5 0. 53 -1 .3 ** * 9. 47  0 .0 9 0. 44 -0 .5 5* * 0. 23  0 .9 2 0. 86 0. 73 0. 58 � 3  1 .1 3* ** 0. 37 -0 .1 0  0 .5 8 -1 .1 0* * 0. 51 -0 .1 2 0. 64 -0 .9 ** * 0. 34 -0 .0 9 0. 46 � 4  0 .4 4 0. 34  0 .8 9  1 .0 1  0 .4 0 0. 88  1 .4 0* 0. 81  0 .1 3 0. 23 -0 .3 7 0. 71 � 5 -0 .8 6 0. 66 -0 .4 1  0 .8 2  0 .7 8 0. 98 -0 .7 7 0. 88  1 .0 5* 0. 59  0 .7 ** * 0. 23 � 6 -0 .7 1* * 0. 36 -0 .0 9  0 .7 5 -0 .5 5 0. 41 -0 .2 9 0. 37 -0 .5 9 0. 86  0 .1 1 0. 53 N o of o bs er va tio n 73 10 5 25 41 37 35 57 Page 21 of 26Zenbaba et al. Journal of Innovation and Entrepreneurship (2025) 14:48 Ta bl e  9  Se co nd s ta ge e st im at es o f M ES R fo r h ou se ho ld d ie ta ry d iv er si ty s co re (H D D S) W he re * ** , * * & * re pr es en t t he s ig ni fic an ce a t 1 % , 5 % & 1 0% p ro ba bi lit y le ve ls , r es pe ct iv el y Va ri ab le s I 0R 0F 0 I 1R 1F 1 I 1R 1F 0 I 1R 0F 1 I 0R 1F 1 I 1R 0F 0 I 0R 0F 1 Co ef St d. D ev Co ef St d. D ev Co ef St d. D ev Co ef St d. D ev Co ef St d. D ev Co ef St d. D ev Co ef St d. D ev A ge -0 .1 ** * 0. 03 -0 .1 2 0. 08 -0 .5 2 0. 95 -0 .1 7 0. 39 -0 .1 2 0. 69 -0 .3 1 0. 34  0 .0 3 0. 08 Se x -0 .5 0 1. 93 -4 .3 5* ** 1. 24 -0 .9 8 1. 34 -1 0. 08 7. 15  1 .7 9 73 .3 9 -9 .1 3 18 .5 5  8 .0 8 5. 39 Ed uc at io n -0 .1 6 0. 37 -0 .0 5 0. 18 -0 .0 1 2. 02  0 .8 0 1. 82 -0 .7 2 2. 97  0 .0 7 3. 20 -0 .3 2 0. 31 M ar ke t d is ta nc e -0 .0 06 0. 02 -0 .0 2 0. 03 -0 .0 01 0. 08 -0 .1 0 0. 33  0 .1 2 0. 24 -0 .0 3 0. 42  0 .0 3 0. 08 Tr ai ni ng d is ta nc e -0 .0 1 0. 10 -0 .0 2 0. 04  0 .2 2 0. 30 -0 .0 8 0. 22  0 .0 3 0. 38  0 .2 5 0. 52  0 .0 6 0. 14 Fa rm d is ta nc e  0 .0 1 0. 06 -0 .0 9* * 0. 04  0 .1 5 0. 14 -0 .2 3 0. 48  0 .1 1 0. 88  0 -.1 9 0. 51  0 .1 1 0. 13 M ob ile o w ne rs hi p -0 .1 1 2. 21 -1 .3 8* ** 0. 38 -0 .8 2 1. 75 -2 .1 8 4. 55 -1 .8 9 14 .5 1 -3 .5 3 8. 18 -0 .3 0 2. 95 Ra di o ow ne rs hi p -0 .4 6 2. 26  0 .1 8 0. 64  4 .5 3 4. 19  1 .1 4 6. 86 -2 .5 9 23 .6 8  3 .3 5 3. 82 -1 .1 6 2. 31 C re di t s er vi ce s  0 .2 1 1. 80  0 .2 7 0. 60 -6 .3 6 5. 15 -1 .7 4 4. 17 -2 .9 10 .4 0  4 .6 1* * 2. 35 -2 .2 8 3. 76 Ex te ns io n se rv ic es  0 .6 3 0. 91  1 .0 0 0. 64  2 .3 7 5. 36  1 .4 8 4. 44  1 .7 2 20 .6 7  3 .7 7 4. 63  0 .1 6 2. 50 Co op er at iv e m em be rs hi p  1 .8 8 1. 57  1 .1 9* * 0. 47  3 .7 8 3. 30 -0 .9 2 3. 46 -1 .3 3 16 .8 3 -1 .7 8 8. 81  2 .3 7* * 1. 02 La nd s iz e  0 .1 5 0. 41  0 .5 7 0. 41  1 .3 8 3. 93 -0 .0 6 1. 60 -1 .8 1 8. 53  5 .3 6 4. 19 -0 .6 2 1. 33 Li ve st oc k  0 .3 6 0. 29  0 .2 0* * 0. 09 -0 .4 9 1. 77 -0 .4 0 1. 08 -0 .3 5 3. 00  0 .2 9 1. 04 -0 .2 7 0. 38 O ff- fa rm in co m e  0 .2 9 1. 50  0 .0 3 0. 27 -2 .3 9 5. 26 -1 .0 6 8. 11  0 .0 07 8. 38 -1 .0 9 5. 49 -0 .4 1 2. 24 An ci lla ry σ 2 91 .6 1* 50 .2 20 9. 06 ** * 28 .3 1 18 54 .7 32 19 84 1. 1 11 87 96 8. 77 14 63 26 91 ** 12 06 29 8. 4 41 3. 7 � 0 0. 05 0. 44  0 .3 5 0. 86 -0 .6 0 0. 54  0 .1 3 0. 25  0 .3 9 1. 07 -0 .1 0 0. 49 � 1 -0 .1 9 0. 89  0 .2 1 0. 60  0 .4 3 0. 79 -0 .5 8 0. 65  0 .5 8 0. 42 -0 .9 8 0. 72 � 2  0 .5 2 0. 97 -0 .6 6 0. 90 -0 .5 7 0. 49 -0 .7 3 1. 20  0 .0 5 0. 72  1 .2 3 0. 82 � 3 -0 .2 1 0. 57 -0 .8 6* ** 0. 23 -1 .1 5* * 0. 58  1 .1 3* ** 0. 41 -1 .3 6* 0. 70 -0 .2 7 0. 79 � 4  1 .1 4 0. 88  1 .2 9* ** 0. 21  0 .9 1 0. 86  0 .5 6 0. 41  0 .3 6 0. 71  0 .0 7 0. 59 � 5 -0 .9 5* 0. 51  0 .0 6 0. 20 -0 .0 8 0. 72 1. 00 9 0. 84  0 .5 3 1. 10  0 .0 6 0. 74 � 6 -0 .2 8 0. 55 -0 .0 2 0. 42 -0 .4 4 03 1 -0 .6 ** * 0. 19 -0 .4 7 0. 45  0 .0 5 0. 42 N o ob se rv at io ns 73 10 5 25 41 37 35 57 Page 22 of 26Zenbaba et al. Journal of Innovation and Entrepreneurship (2025) 14:48 Ta bl e  10  S ec on d st ag e es tim at es o f M ES R fo r h ou se ho ld w he at p ro du ct io n in co m e (ln In co m e) W he re * ** , * *, & * re pr es en t t he s ig ni fic an ce a t 1 % , 5 % , & 1 0% p ro ba bi lit y le ve ls , r es pe ct iv el y Va ri ab le s I 0R 0F 0 I 1R 1F 1 I 1R 1F 0 I 1R 0F 1 I 0R 1F 1 I 1R 0F 0 I 0R 0F 1 Co ef St d. D ev Co ef St d. D ev Co ef St d. D ev Co ef St d. D ev Co ef St d. D ev Co ef St d. D ev Co ef St d. D ev A ge  0 .0 03 0. 02  0 .0 2 0. 02  0 .0 5 0. 03  0 .0 5 0. 05 -0 .0 8 0. 05  0 .0 2 0. 08 -0 .0 07 0. 01 Se x  0 .6 3 0. 87  0 .9 8 0. 56  0 .7 8 2. 67  2 .0 6 3. 76 -2 .0 9 1. 27  1 .3 4 1. 13 -0 .4 6 0. 82 Ed uc at io n  0 .1 7 0. 11 -0 .0 1 0. 03  0 .2 1* 0. 12 -0 .0 8 0. 41  0 .0 7 0. 33 -0 .1 1 0. 36  0 .1 7 0. 18 M ar ke t d is ta nc e  0 .0 07 0. 02  0 .0 04 0. 01  0 .0 4 0. 04  0 .0 7 0. 10 -0 .0 1 0. 08  0 .0 1 0. 04  0 .0 2 0. 05 Tr ai ni ng d is ta nc e  0 .0 1 0. 03  0 .0 09 0. 01  0 .0 2 0. 4  0 .0 6 0. 7 -0 .0 2 0. 14  0 .0 2 0. 03  0 .0 5 0. 09 Fa rm d is ta nc e  0 .0 02 0. 02  0 .0 4 0. 02  0 .0 6* 0. 03  0 .1 1* ** 0. 04 -0 .0 3 0. 09  0 .0 06 0. 07 -0 .0 02 0. 01 M ob ile o w ne rs hi p -0 .4 7 0. 34 -0 .5 1 0. 24 -0 .9 2* 0. 52 -1 .1 6 1. 45  0 .0 6 1. 67  0 .0 9 1. 46 -1 .0 4 0. 95 Ra di o ow ne rs hi p  0 .9 5* ** 0. 29 -0 .0 4 0. 26  0 .3 8 0. 45  0 .1 5 2. 47  0 .1 7 1. 45  0 .0 5 1. 57  1 .1 4* * 0. 54 C re di t s er vi ce s -0 .1 0 0. 28 -0 .3 6 0. 22 -1 .5 2 0. 95 -1 .1 6 0. 91  0 .0 8 1. 13  0 .3 6 1. 19  0 .4 2 0. 41 Ex te ns io n se rv ic es -0 .4 6 0. 37 -0 .3 1 0. 17  0 .0 1 1. 15  0 .0 2 0. 45  0 .7 6* 0. 44 -0 .1 1 0. 70  0 .3 7 0. 90 Co op er at iv e m em be rs hi p -0 .0 4 0. 12  0 .2 6 0. 20 -0 .4 6* 0. 27 -0 .0 5 0. 51  0 .4 4 0. 64 -0 .1 1 0. 66 -0 .4 8 0. 98 La nd s iz e -0 .0 7 0. 10 -0 .2 1 0. 15 -0 .5 6 0. 48 -0 .8 2 0. 84  0 .3 3 0. 49  0 .0 8 0. 58  0 .7 8 1. 17 Li ve st oc k -0 .0 7* 0. 04 -0 .1 0 0. 03 -0 .4 8 0. 31 -0 .3 5 0. 34  0 .0 2 0. 18 -0 .0 6 0. 38 -0 .2 0 0. 25 O ff- fa rm in co m e -0 .4 5 0. 70  0 .1 9 0. 26 -1 .0 8* * 0. 45 -1 .0 8 1. 28  0 .0 8 0. 61  0 .2 8 0. 30 -0 .8 1 0. 98 An ci lla ry σ 2 22 .8 9 23 .9 0 14 .9 0 4. 72 14 0. 5* ** 4. 13 16 8. 9* ** 54 .9 4 37 .2 2 95 .0 1 25 .7 6 88 .9 0 72 .1 4 22 5. 80 � 0 -0 .7 3 0. 61  0 .2 1 0. 84  0 .2 8 0. 95  0 .1 4 0. 85  0 .0 5 0. 67  0 .6 1 0. 73 � 1  0 .0 05 0. 99 -0 .3 5 0. 59 -0 .9 5 0. 60  0 .7 4 0. 63 -0 .6 4 0. 47  0 .2 9 0. 35 � 2  0 .4 3 0. 84  0 .7 8 0. 32  0 .6 2 0. 42  0 .3 3 0. 35 -0 .2 2 0. 65  0 .3 2 0. 76 � 3  0 .0 8 0. 50 -0 .8 9 0. 77 -0 .8 1 0. 23 -1 .2 6 0. 83  1 .2 2 1. 03 3 -1 .2 9* 0. 68 � 4 -1 .2 3 0. 90  0 .0 3 0. 96 -0 .3 1 1. 18 -0 .6 5 0. 72 -0 .6 5 0. 45 -0 .3 6 1. 08 � 5  0 .9 4* * 0. 45  0 .8 4 0. 92  1 .2 9* ** 0. 24  0 .7 8 0. 50 -0 .4 0 1. 04  0 .3 3 0. 89 � 6 -0 .2 8 0. 25  0 .3 8 0. 21 -0 .1 6 0. 59 -0 .0 5 0. 28  0 .1 9 0. 49 0. 29 0. 43 N o ob se rv at io ns 73 10 5 25 41 37 35 57 Page 23 of 26Zenbaba et al. Journal of Innovation and Entrepreneurship (2025) 14:48 Abbreviations ATT​ � Average Treatment on Treated ATU​ � Average Treatment on Untreated ATA​ � Agricultural Transformation Agency CSA � Central Statistical Authority ETB � Ethiopian Birr FAO � Food and Agricultural Organization FCS � Food Consumption Score FSIN � Food Security Information Network HDDS � Households’ Dietary Diversity Score HGWZAO � Horo Guduru Wollega Zone Agricultural Offices IAP � Improved Agronomic Practices MESR � Multinomial Endogenous Switching Regression MNL � Multinomial Logit NGO � Non-Governmental Organizations SNP � Safety Net Program WFP � World Food Program Acknowledgements The authors of this study greatly acknowledge enumerators and respondents of all districts. Moreover, the authors are also thankful to Wollega University and Haramaya University for the facilitation of the research work. Finally, the authors would like to thank both editor and reviewers who devoted their time to provide constructive comments and sugges- tions for the further improvement of the quality of the research. Author contributions OSZ conceptualized the study, organized research methodology, supervised data collection, analyzed the data, inter- preted the model outcomes, and wrote and prepared the manuscript. MK, MJ, & KJ provided suggestions on the initial stage of the study and commented and provided detail corrections at all stages of the manuscript preparation starting from proposal writing to the final work of the manuscript and finally approved the manuscript. Funding The Ethiopian Ministry of Education financially supported this study. Availability of data and materials Upon request, the corresponding author of this study will provide data used for this study. Declarations Ethics approval and consent to participate The Haramaya University Post-Graduate Research Office Committee approved the right to conduct this research. Competing interests The authors declare that we have no conflict of interest. Received: 24 March 2024 Accepted: 23 February 2025 References Abay, K. A., Abay, M. H., Amare, M., Berhane, G., & Aynekulu, E. (2021). Mismatch between soil nutrient deficiencies and fertilizer applications: Implications for yield responses in Ethiopia. Agricultural Economics, 53, 1–16. https://​doi.​org/​ 10.​1111/​agec.​12689 Abro, Z. A., Alemu, B. A., & Hanjra, M. A. (2014). Policies for agricultural productivity growth and poverty reduction in rural Ethiopia. World Development, 59, 461–474. https://​doi.​org/​10.​1016/j.​world​dev.​2014.​01.​033 Adhikari, S. P., Ghimire, Y. N., Timsina, K. P., Subedi, S., & Khare, M. (2021). Technical efficiency of wheat growing farmers of Nepal. Journal of Agriculture and Natural Resources, 4(2), 246–254. https://​doi.​org/​10.​3126/​janr.​v4i2.​33857 Ahmed, M. H. (2022). Impact of improved seed and inorganic fertilizer on maize yield and welfare: Evidence from Eastern Ethiopia. Journal of Agriculture and Food Research, 7, 100266. https://​doi.​org/​10.​1016/j.​jafr.​2021.​100266 Ahmed, M. H., Mesfin, H. M., Abady, S., Mesfin, W., & Kebede, A. (2016). Adoption of improved groundnut seed and its impact on rural households’ welfare in Eastern Ethiopia. Cogent Economics & Finance, 4, 1268747. https://​doi.​org/​10.​ 1080/​23322​039.​2016.​12687​47 Alemu, T., & Mengistu, A. (2019). Impacts of climate change on food security in Ethiopia: adaptation and mitigation options: A review. Climate Change-Resilient Agriculture and Agroforestry. Climate Change Management. https://​doi.​ org/​10.​1007/​978-3-​319-​75004-0_​23 Ali, H., Menza, M., Hagos, F., & Haileslassie, A. (2022). Impact of climate-smart agriculture adoption on food security and multidimensional poverty of rural farm households in the Central Rift Valley of Ethiopia. Agriculture & Food Security. https://​doi.​org/​10.​1186/​s40066-​022-​00401-5 https://doi.org/10.1111/agec.12689 https://doi.org/10.1111/agec.12689 https://doi.org/10.1016/j.worlddev.2014.01.033 https://doi.org/10.3126/janr.v4i2.33857 https://doi.org/10.1016/j.jafr.2021.100266 https://doi.org/10.1080/23322039.2016.1268747 https://doi.org/10.1080/23322039.2016.1268747 https://doi.org/10.1007/978-3-319-75004-0_23 https://doi.org/10.1007/978-3-319-75004-0_23 https://doi.org/10.1186/s40066-022-00401-5 Page 24 of 26Zenbaba et al. Journal of Innovation and Entrepreneurship (2025) 14:48 Anteneh, A., & Asrat, D. (2020). Wheat production and marketing in Ethiopia: Review study. Cogent Food & Agriculture. https://​doi.​org/​10.​1080/​23311​932.​2020.​17788​93 Arouna, A., Lokossou, J., Wopereis, M., Bruce-Oliver, S., & Roy-Macauley, H. (2017). Contribution of improved rice varieties to poverty reduction and food security in sub-Saharan Africa. Global Food Security, 14, 54–60. Assaye, A., Habte, E., Sakurai, S., & Alemu, D. (2022). Impact assessment of adopting improved rice variety on farm house- hold welfare in Ethiopia. Journal of Agriculture and Food Research, 10, 100428. https://​doi.​org/​10.​1016/j.​jafr.​2022.​ 100428 ATA (Agricultural Transformation Agency). (2018). Annual Report. http://​www.​ata.​gov.​et/​wp-​conte​nt/​uploa​ds/​2019/​01/​ ATA_Annual Report_2010. Atinafu, A., Lejebo, M., & Alemu, A. (2022). Adoption of improved wheat production technology in Gorche district, Ethio- pia. Agriculture & Food Security. https://​doi.​org/​10.​1186/​s40066-​021-​00343-4 Aweke, C. S., Hassen, J. Y., Wordofa, M. G., Moges, D. K., Endris, G. S., & Rorisa, D. T. (2021). Impact assessment of agricultural technologies on household food consumption and dietary diversity in eastern Ethiopia. Journal of Agriculture and Food Research. https://​doi.​org/​10.​1016/j.​jafr.​2021.​100141 Ayenew, W., Lakew, T., & Kristos, E. H. (2020). Agricultural technology adoption and its impact on smallholder farmer’s welfare in Ethiopia. African Journal of Agricultural Research, 15(3), 431–445. https://​doi.​org/​10.​5897/​AJAR2​019.​14302 Biru, W., D., Zeller, M.,& Loos, T. K. (2020). The Impact of Agricultural Technologies on Poverty and Vulnerability of Small- holders in Ethiopia: A Panel Data Analysis. Social Indicators Research, 147:517–544. https://​doi.​org/​10.​1007/​ s11205-​019-​02166-0 Bourguignon, F., Fournier, M., & Gurgand, M. (2007). Selection bias corrections based on the multinomial logit model. Journal of Economic Surveys, 21, 174–205. Brasesco, F., Asgedom, G., Sommacal, V., & Casari G. (2019). Strategic analysis and intervention plan for wheat and wheat products in the Agro-Commodities Procurement zone of the pilot Integrated Agro-Industrial Park in Central-Eastern Oromia, Ethiopia. Addis Ababa. FAO. 104. Cafiero, C., Melgar-Quiñonez, H. R., Ballard, T. J., & Kepple, A. W. (2014). Validity and reliability of food security measures. Annals of the New York Academy of Sciences, 1331, 230–248. Cordero-Ahiman, O. V., Vanegas, J. L., Beltrán-Romero, P., & Quinde-Lituma, M. E. (2020). Determinants of food insecurity in rural households: the case of the Paute River Basin of Azuay Province. Ecuador. Sustainability, 12(946), 1–18. CSA (Central Statistical Authority). (2007). Population and Housing Census of Ethiopia Administrative Report. Addis Ababa. Danso-Abbeam, G., & Baiyegunhi, L. J. S. (2018). Welfare impact of pesticides management practices among smallholder cocoa farmers in Ghana. Technology in Society, 54, 10–19. https://​doi.​org/​10.​1016/j.​techs​oc.​2018.​01.​011 Di Falco, S., Veronesi, M., & Yesuf, M. (2011). Does adaptation to climate change provide food security? A micro-perspec- tive from Ethiopia. American Journal of Agricultural Economics, 93(3), 829–846. Ehiakpor, D. S., Danso-Abbeam, G., & Mubashiru, Y. (2021). Adoption of interrelated sustainable agricultural practices among smallholder farmers in Ghana. Land Use Policy, 10, 105142. FAO, ECA and AUC. (2021). Africa regional overview of food security and nutrition 2020: Transforming food systems for affordable healthy diets. Accra, FAO. https://​doi.​org/​10.​4060/​cb483​1en FAO (Food and Agricultural Organization). (2021). The impact of disasters and crises on agriculture and food security, Rome. https://​doi.​org/​10.​4060/​cb367​3en FSIN (Food Security Information Network). (2017). Global Report on Food Crises. Gebremariam, G., & Wünscher, T. (2016). Combining sustainable agricultural practices pays off: evidence on welfare effects from Northern Ghana. African Association of Agricultural Economists (AAAE) Gedefe, K. & Bekele, F. (2019). Pathway to sustainable Land-Use and Food Systems in Ethiopia by 2050. In: FABLE, Pathways to Sustainable Land-Use and Food Systems. Report of the FABLE Consortium, Laxenburg and Paris: International Institute for Applied Systems Analysis (IIASA) and Sustainable Development Solutions Network (SDSN). 166–179. Giller, K. E., Andersen, J. A., & Sumberg, J. (2017). A golden age for agronomy? In J. Sumberg (Ed.), Agronomy for develop- ment: The politics of knowledge in agricultural research (pp. 150–160). Earth scan. Glover, D., Sumberg, J., Ton, G., Andersson, J., & Badstue, L. (2019). Rethinking technological change in smallholder agricul- ture. Outlook on Agriculture, 48(3), 169–180. Hailu., M., Tolossa, D., Girma, A., & Kassa, B. (2021). The impact of improved agricultural technologies on household food security of smallholders in Central Ethiopia: An endogenous switching estimation. World Food Policy, 7(3). https://​ doi.​org/​10.​1002/​wfp2.​12029 Hailu, M., Tolossa, D., Girma, A., & Kassa, B. (2021). The impacts of improved agricultural technologies on household food security of smallholders in central Ethiopia An endogenous switching estimation. World Food Policy, 7(2), 111–127. https://​doi.​org/​10.​1002/​wfp2.​12029 HGWZAO (Horo Guduru Wollega Zone Agricultural Offices). (2023). Hoddinott, J. & Yohannes, Y. (2002). Dietary diversity as a food security indicator Food Consumption and Nutrition. Discus- sion Paper No. 136. International Food Policy Research Institute, Washington, DC. https://​doi.​org/​10.​22004/​ag.​econ.​ 16474. INDDEX Project. (2018). Data4Diets: Building Blocks for Diet-related Food Security Analysis. Tufts University, Boston, MA. https://​inddex.​nutri​tion.​tufts.​edu/​data4​diets. Izraelov, M., & Silber, J. (2019). An assessment of the global food security index. Food Secur, 11, 1135–1152. Jaleta, M., Kassie, M., Marenya, P., Yirga, C., & Erenstein, O. (2018). Impact of improved maize adoption on household food security of maize producing smallholder farmers in Ethiopia. Food Security, 10, 81–93. https://​doi.​org/​10.​1007/​ s12571-​017-​0759-y Jambo, Y., Alemu, A., & Tasew, W. (2021). Agriculture & Food Security. https://​doi.​org/​10.​1186/​s40066-​021-​00294-w John, A. O., Afolake, A. C., & Lawrence, B. O. (2020). Effect of livelihood diversification and technology adoption on food security status of rice farming households in Ogun state, Nigeria. Agricultural Socio-Economics Journal, XX, 3, 233–244. https://doi.org/10.1080/23311932.2020.1778893 https://doi.org/10.1016/j.jafr.2022.100428 https://doi.org/10.1016/j.jafr.2022.100428 http://www.ata.gov.et/wp-content/uploads/2019/01/ATA_ http://www.ata.gov.et/wp-content/uploads/2019/01/ATA_ https://doi.org/10.1186/s40066-021-00343-4 https://doi.org/10.1016/j.jafr.2021.100141 https://doi.org/10.5897/AJAR2019.14302 https://doi.org/10.1007/s11205-019-02166-0 https://doi.org/10.1007/s11205-019-02166-0 https://doi.org/10.1016/j.techsoc.2018.01.011 https://doi.org/10.4060/cb4831en https://doi.org/10.4060/cb3673en https://doi.org/10.1002/wfp2.12029 https://doi.org/10.1002/wfp2.12029 https://doi.org/10.1002/wfp2.12029 https://doi.org/10.22004/ag.econ.16474 https://doi.org/10.22004/ag.econ.16474 https://inddex.nutrition.tufts.edu/data4diets https://doi.org/10.1007/s12571-017-0759-y https://doi.org/10.1007/s12571-017-0759-y https://doi.org/10.1186/s40066-021-00294-w Page 25 of 26Zenbaba et al. Journal of Innovation and Entrepreneurship (2025) 14:48 Jones, A.D., Ngure, F.M., Pelto, G. & Young, S. L. (2013). What are we assessing when we measure food security? A compen- dium and review of current metrics. Adv Nutr 4(5), 481–506 Kaliba, A. R., Mazvimavi, K., Gregory, T. L., Mgonja, F. M., & Mgonja, M. (2018). Factors affecting adoption of improved sor- ghum varieties in Tanzania under information and capital constraints. Agricultural and Food Economics, 6, 18. https://​ doi.​org/​10.​1186/​s40100-​018-​0114-4 Kassa, B., Kassa, B., & Aregawi, K. (2014). Adoption and impact of agricultural technologies on farm income: Evidence from Southern Tigray, Northern Ethiopia. International Journal of Food and Agricultural Economics, 2(4), 91–106. Kassie, M., Teklewolde, H., Erenstein, O., Jaleta, M., Marenya, P., & Mekura, M. (2015). Technology diversification: Assess- ing impacts on crop income and agrochemical uses in Malawi. ICAE (International Conference of Agricultural Economic), 29th Milan Italy. Kassie, M., Marenya, P., Tessema, Y., Jaleta, M., Zeng, D., Erenstein, O., & Rahut, D. (2018). Measuring farm and market level economic impacts of improved maize production technologies in Ethiopia: Evidence from panel data. Journal of Agricultural Economics, 69(1), 76–95. https://​doi.​org/​10.​1111/​1477-​9552.​12221 Kennedy, G., Ballard, T., & Dop, M. (2011). Guidelines for measuring household and individual dietary diversity. Food and Agriculture Organization of the United Nations. Khonje, M., Manda, J., Arega, A., & Kassie, M. (2015). Analysis of adoption and impacts of improved maize varieties in Eastern Zambia. World Development, 66, 695–706. Khonje, M. G., Manda, J., & Mkandawire, P. (2018). Adoption and welfare impacts of multiple agricultural technologies: Evidence from eastern Zambia. Agricultural Economics, 49, 599–609. https://​doi.​org/​10.​1111/​agec.​12445 Kothari, C. R. (2004). Research methodology: Methods and techniques. New Delhi: New Age International (P) Limited, Publishers Manda, J., Arega, D. A., Gardebroek, C., Kassie, M., & Tembo, G. (2016). Adoption and impacts of sustainable agricultural practices on maize yields and incomes: evidence from rural Zambia. Journal of Agricultural Economics, 67(10), 130–153. https://​doi.​org/​10.​1111/​1477-​9552.​12127 Marenya, P. P., Gebremariam, G., Jaleta, M., & Rahut, D. B. (2020). Sustainable intensification among smallholder maize farmers in Ethiopia: Adoption and impacts under rainfall and unobserved heterogeneity. Food Policy, 95, 101941. https://​doi.​org/​10.​1016/j.​foodp​ol.​2020.​101941 Mark, W. R., Jawoo, K., Nicola, C., Claudia, R., Richard, R., Fisher, M., Cindy, C., Karen, G., Nicostrato, D.P., Pascale, S. (2014). Food Security in a World of Natural Resource Scarcity. The Role of Agricultural Technologies. International Food Policy Research Institute. Maxwell, D., Vaitla, B., & Coates, J. (2014). How do indicators of household food insecurity measure up? An empirical comparison from Ethiopia. Food Policy, 47, 107–116. https://​doi.​org/​10.​1016/j.​foodp​ol.​2014.​04.​003 Maxwell, S. (1996). Food security: A post-modern perspective. Food Policy, 21, 155–170. Mc-Fadden, D. (1973). Conditional logit analysis of qualitative choice behavior. In P. Zarembka (Ed.), Frontiers in economet- rics (pp. 105–142). Academic Press. Mekonnen, T. (2017). Productivity and Household Welfare Impact of Technology Adoption: Micro-Level Evidence from Rural Ethiopia. UNU-MERIT Working Papers. Merga, G., Sileshi, M., & Zeleke, F. (2023). Welfare impact of improved maize varieties adoption among smallholder farmers in Amuru district of Horo Guduru Wollega. Ethiopia. Cogent Economics & Finance, 11, 2207923. https://​doi.​org/​10.​ 1080/​23322​039.​2023.​22079​23 Mulugeta T., & Hundie, B. (2012). Impacts of Adoption of Improved Wheat Technologies on Households’ Food Consump- tion in Southeastern Ethiopia. Selected Poster prepared for presentation at the International Association of Agricul- tural Economists (IAAE) Triennial Conference, Foz do Iguaçu, Brazil. Ngango, J., Nkurunziza, F., Mbaraka, S. R., & Cyamweshi, A. R. (2022). Determinants of sustainable agricultural intensifica- tion adoption and impacts on household productivity and consumption in Rwanda. Journal of Agriculture and Rural Development in the Tropics and Subtropics, 123(1), 39–50. https://​doi.​org/​10.​17170/​kobra-​20220​11955​71 Ngomi, C.S.T., Fadikpe, A. A. A., Ngaba, M. J. Y., Chen, Q. P., Nfonbeu, M. F. M., & Yang J. Z. (2020). Impact of adoption of agricultural extension services on farm households food security in Cameroon. IOP Conference Series: Earth and Environmental Science 601, 012001. https://​doi.​org/​10.​1088/​1755-​1315/​601/1/​012001. Nirgude, R. R., & Sonawane, K. G. (2017). An estimation of impact of wheat production technology. Trends in Biosciences, 10(27), 5759–5766. Ogundari, K. (2017). Categorizing households into different food security states in Nigeria: The socio-economic and demographic determinants. Agricultural and Food Economic, 5(1), 1–20. https://​doi.​org/​10.​1186/​s40100-​017-​0076-y Oyetunde-Usman, Z., Olagunju, K. O., & Ogunpaimo, O. R. (2021). Determinants of adoption of multiple sustainable agricultural practices among smallholder farmers in Nigeria. International Soil and Water Conservation Research, 9, 241–248. Richard, K., Gupta, A., Helena, O., & Abawiera, C. (2020). Adoption and impact of modern rice varieties on poverty in Eastern India. Science Direct, 27(1), 56–66. https://​doi.​org/​10.​1016/j.​rsci.​2019.​12.​006 Setsoafa, E. D., Ma, W., & Renwick, A. (2022). Effects of sustainable agricultural practices on farm income and food security in northern Ghana. Agricultural and Food Economics, 10, 9. https://​doi.​org/​10.​1186/​s40100-​022-​00216-9 Shiferaw, B., Kassie, M., Jaleta, M., & Yirga, C. (2014). Adoption of improved wheat varieties and impacts on household food security in Ethiopia. Food Policy, 44, 272–284. https://​doi.​org/​10.​1016/j.​foodp​ol.​2013.​09.​012 Shikur, Z. H. (2020). Agricultural policies, agricultural production and rural households’ welfare in Ethiopia. Economic Structures, 9, 50. https://​doi.​org/​10.​1186/​s40008-​020-​00228-y Shimeles, A., Verdier-Chouchane, A., & Boly, A. (2018). Introduction: Understanding the Challenges of the Agricultural Sec- tor in Sub-Saharan Africa. In: Shimeles, A., Verdier-Chouchane, A., Boly, A. (eds) Building a Resilient and Sustainable Agriculture in Sub-Saharan Africa. Palgrave Macmillan, Cham. https://​doi.​org/​10.​1007/​978-3-​319-​76222-7_1 Sisay, K. (2024). Impacts of multiple livelihood diversification strategies on diet quality and welfare of smallholder farmers: Insight from Kaffa zone of Ethiopia. Cleaner and Responsible Consumption, 12, 100161. https://​doi.​org/​10.​1016/j.​clrc.​ 2023.​100161 https://doi.org/10.1186/s40100-018-0114-4 https://doi.org/10.1186/s40100-018-0114-4 https://doi.org/10.1111/1477-9552.12221 https://doi.org/10.1111/agec.12445 https://doi.org/10.1111/1477-9552.12127 https://doi.org/10.1016/j.foodpol.2020.101941 https://doi.org/10.1016/j.foodpol.2014.04.003 https://doi.org/10.1080/23322039.2023.2207923 https://doi.org/10.1080/23322039.2023.2207923 https://doi.org/10.17170/kobra-202201195571 https://doi.org/10.1088/1755-1315/601/1/012001 https://doi.org/10.1186/s40100-017-0076-y https://doi.org/10.1016/j.rsci.2019.12.006 https://doi.org/10.1186/s40100-022-00216-9 https://doi.org/10.1016/j.foodpol.2013.09.012 https://doi.org/10.1186/s40008-020-00228-y https://doi.org/10.1007/978-3-319-76222-7_1 https://doi.org/10.1016/j.clrc.2023.100161 https://doi.org/10.1016/j.clrc.2023.100161 Page 26 of 26Zenbaba et al. Journal of Innovation and Entrepreneurship (2025) 14:48 Solomon, T., Bedada, B., & Gurmu, M. Y. (2016). Impact of improved wheat technology adoption on productivity and income in Ethiopia. African Crop Science Journal, 24(s1), 127–135. Ssozi J., Asongu, S., & Amavilah, V. (2017). Is Aid for Agriculture Effective in Sub-Saharan Africa?," Working Papers of the African Governance and Development Institute. 17/035, African Governance and Development Institute (AGDI). Ssozi, J., Asongu, S., & Amavilah, V. H. (2019). The effectiveness of development aid for agriculture in Sub-Saharan Africa. Journal of Economic Studies, 46(2), 284–305. https://​doi.​org/​10.​1108/​JES-​11-​2017-​0324 Swindale, A., & Bilinsky, P. (2006). Household dietary diversity scores (HDDS) for measurement of household food access: indicator guide Washington, DC: Food and Nutrition Technical Assistance Project, Academy for Educational Development. Tadesse, G., Bernard, T., de Alan, B., & Minot, N. (2018a). The impact of the use of new technologies on farmers’ wheat yield in Ethiopia: Evidence from a randomized control trial. Agricultural Economics, 49, 409–421. Tadesse, W., Bishaw, Z., & Assefa, S. (2018b). Wheat production and breeding in sub-saharan africa challenges and oppor- tunities in the face of climate change. International Journal of Climate Change Strategies and Management, 11(5), 696–715. Teka, A., & Lee, S. K. (2020). Do agricultural package programs improve the welfare of rural people? Evidence from Small- holder Welfare of Rural People? Evidence from Smallholder Farmers in Ethiopia. Agriculture, 10, 190. https://​doi.​org/​ 10.​3390/​agric​ultur​e1005​0190 Wake, R., & Goshu, D. (2019). Impact of high yielding wheat varieties adoption on farm income of smallholder farmers in Ethiopia. International Journal of Agricultural Extension, 07(01), 45–59. https://​doi.​org/​10.​33687/​ijae.​007.​01.​2490 Wang, H., Pandey, S., & Feng, L. (2020). Econometric analyses of adoption and household-level impacts of improved rice varieties in the uplands of Yunnan, China. Sustainability, 12, 6873. https://​doi.​org/​10.​3390/​su121​76873 Ward, P. S., Bell, A. R., Droppelmann, K., & Benton, T. G. (2018). Early adoption of conservation agriculture practices: Under- standing partial compliance in programs with multiple adoption decisions. Land Use Policy, 70, 27–37. Webb, P., Coates, J., Frongillo, E. A., Rogers, B. L., Swindale, A., & Bilinsky, P. (2006). Measuring household food insecurity: Why it’s so important and yet so difficult to do. Journal of Nutrition, 136(5), 1404s–1408s. Welteji, D. (2018). A critical review of rural development policy of Ethiopia: Access, utilization and coverage. Agriculture and Food Security, 7, 55. https://​doi.​org/​10.​1186/​s40066-​018-​0208-y WFP (World Food Programme). (2008). Calculation and use of the food consumption score in food security analysis. Prepared by VAM unit HQ Rome, Italy WFP (World Food Program). (2023). Saving lives changing lives. Ethiopia Country Brief October. Wordofa, M. G., Hassen, J. Y., Endris, G. S., Aweke, C. S., Moges, D. K., & Rorisa, D. T. (2021). Adoption of improved agricultural technology and its impact on household income: A propensity score matching estimation in eastern Ethiopia. Agriculture and Food Security, 10, 5. https://​doi.​org/​10.​1186/​s40066-​020-​00278-2 Yirga, C., & Alemu, D. (2016). Adoption of Crop Technologies among Smallholder Farmers in Ethiopia: Implications for Research and Development. EIAR 50th Year Jubilee Anniversary Special Issue: 1–16. Zegeye, B. M., Fikire, H. A., & Asefa, A. B. (2022). Impact of agricultural technology adoption on food consumption expenditure: Evidence from rural Amhara Region, Ethiopia. Cogent Economics & Finance, 10(1), 2012988. https://​doi.​ org/​10.​1080/​23322​039.​2021.​20129​88 Zegeye, M. B. (2021). Adoption and ex-post impact of agricultural technologies on rural poverty: evidence from Amhara Region, Ethiopia. Cogent Economics and Finance. https://​doi.​org/​10.​1080/​23322​039.​2021.​19697​59 Zegeye, M. B., Fikire, A. K., & Meshesha, G. B. (2022). Determinants of multiple agricultural technology adoption: Evidence from rural Amhara region, Ethiopia. Cogent Economics & Finance, 10(1), 2058189. https://​doi.​org/​10.​1080/​23322​039.​ 2022.​20581​89 Zerssa, G., Feyssa, D., Kim, D. G., & Eichler-Lobermann, B. (2021). Challenges of smallholder farming in Ethiopia and oppor- tunities by adopting climate-smart agriculture. Agriculture, 11, 192. https://​doi.​org/​10.​3390/​agric​ultur​e1103​0192 Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. https://doi.org/10.1108/JES-11-2017-0324 https://doi.org/10.3390/agriculture10050190 https://doi.org/10.3390/agriculture10050190 https://doi.org/10.33687/ijae.007.01.2490 https://doi.org/10.3390/su12176873 https://doi.org/10.1186/s40066-018-0208-y https://doi.org/10.1186/s40066-020-00278-2 https://doi.org/10.1080/23322039.2021.2012988 https://doi.org/10.1080/23322039.2021.2012988 https://doi.org/10.1080/23322039.2021.1969759 https://doi.org/10.1080/23322039.2022.2058189 https://doi.org/10.1080/23322039.2022.2058189 https://doi.org/10.3390/agriculture11030192 Impact of wheat production technology packages adoption on smallholder farmers’ food security and income in Horo Guduru Wollega Zone, Ethiopia Abstract Introduction Empirical review Research methodology Description of the study area Methods of data collection Sampling techniques and sample size Estimation strategy Multinomial logit selection model Multinomial endogenous switching regression Estimation of average treatment effects Measurement of outcome variables Results and discussion Descriptive statistics result Empirical results Factors affecting adoption decision of technology packages Average treatment effects of wheat technology packages adoption Conclusions and policy suggestions Appendix Acknowledgements References