A meta-analysis of the impact of agricultural extension services

Abstract

PurposeThis study aims to address two research questions. First, do the agricultural extension services have an impact on the potential outcomes considered in the primary studies, and to what extent? Second, how sensitive is the reported impact to the study-specific characteristics in the primary studies?Design/methodology/approachThe paper synthesizes 45 studies that assessed the causal impact of agricultural extension services published in 2004–2021, using meta-regression analysis. It considers three measures of effect sizes – Cohen’s, Hedges and principal correlation coefficient (PCC) – to standardize the reported impact of agricultural extension services in the primary studies.FindingsThe empirical results show that, on average, agricultural extension services have statistically significant and positive impacts on the potential outcomes identified in the primary studies. However, the magnitude of the impact is considered medium-sized. Other results show that the effect size estimates of agricultural extension services' impact significantly vary with the data type (cross-sectional data vs. panel data), research design (non-experimental vs. experimental design) and econometric methods employed in the primary studies.Practical implicationsOne can argue that the medium-sized impact we estimated indicates evidence of a moderate, weak relationship between agricultural extension services and the potential outcomes considered in the primary studies. This means that agricultural extension services need to be restructured in the current form to stimulate change in the agricultural sector globally. In addition, the sensitivity of effect sizes to study attributes (i.e. data types, research design and econometric methods) shows that researchers and academicians need to pay attention to these attributes to provide more reliable estimates for policy purposes.Originality/valueThis is the first study that attempts to shed light on the overall performance of agricultural extension services using a meta-regression analysis approach.

Yield effects of agricultural cooperative membership in developing countries: A meta‐analysis

“…The standard approach to dealing with this problem is to convert the estimated yield effects to a partial correlation coefficient, or PCC (Ogundari 2022;Ogundari & Bolarinwa 2018;Xue et al 2020). The PCC is a scale-free metric, providing a unified interpretation across different studies (Anwar & Mang 2022;Ogundari & Bolarinwa 2018;Ridhwan et al 2022).…”

“…This provides further confirmation of the presence of publication bias. According to Doucouliagos and Stanley (2013) and Ogundari (2022), estimated coefficients larger than one in absolute value for the standard error variable indicate substantial selectivity bias. The respective estimate in Table 4 for the WLS-RE model is 1.963 (2.867 for the WLS-FE model), indicating strong publication bias.…”

“…Meta-analysis is a statistical tool that allows researchers to appropriately weight contrasting estimates while accounting for the influence of publication bias (Anwar & Mang 2022;Ogundari 2022;Ogundari & Bolarinwa 2018;Ridhwan et al 2022). The goal is to produce an estimate of the overall, average effect of a given treatment or intervention.…”

“…previous studies(Anwar & Mang 2022;Duan et al 2020;Ogundari 2022;Ridhwan et al 2022;Xue et al 2021), we employ two weighted least squares (WLS) estimators to address this problem: WLS-Fixed Effects (FE) and WLS-Random Effects (RE).2 The WLS-FE model assumes that there is a single underlying true effect and the reason for different estimates across studies is solely due to sampling error. The corresponding weight is given by the inverse of the variance of the estimated effect, that is, 1∕[𝑆𝐸(𝑃𝐶𝐶 𝑖 ) 2 ] (Xue et al 2021).…”

Effect of Agricultural Extension Services and Development on Productivity among Small Scale Farmers in India

“…Since the history of mankind, agriculture evolved round the basic food needs of man and developed principles and practices in crop production including field management (Ogundari, 2022). With changing situation, technological development have shifted the traditional agriculture from more crop production to mixed farming involving allied activities like animal husbandry and sericulture, pisciculture therefore need for technological development in agriculture (Netter, Luedeling & Whitney, 2022).…”