Types of Research
- (-) Remove Data Analysis filter Data Analysis
- (-) Remove Research Brief filter Research Brief
- (-) Remove South Asia Region and Selected Countries filter South Asia Region and Selected Countries
- (-) Remove Market & Value Chain Analysis filter Market & Value Chain Analysis
- (-) Remove Risk, Preferences, & Decision-Making filter Risk, Preferences, & Decision-Making
- (-) Remove Monitoring & Evaluation filter Monitoring & Evaluation
- (-) Remove Education & Training filter Education & Training
Studies of improved seed adoption in developing countries almost always draw from household surveys and are premised on the assumption that farmers are able to self-report their use of improved seed varieties. However, recent studies suggest that farmers’ reports of the seed varieties planted, or even whether seed is local or improved, are sometimes inconsistent with the results of DNA fingerprinting of farmers' crops. We use household survey data from Tanzania to test the alignment between farmer-reported and DNA-identified maize seed types planted in fields. In the sample, 70% of maize seed observations are correctly reported as local or improved, while 16% are type I errors (falsely reported as improved) and 14% are type II errors (falsely reported as local). Type I errors are more likely to have been sourced from other farmers, rather than formal channels. An analysis of input use, including seed, fertilizer, and labor allocations, reveals that farmers tend to treat improved maize differently, depending on whether they correctly perceive it as improved. This suggests that errors in farmers' seed type awareness may translate into suboptimal management practices. In econometric analysis, the measured yield benefit of improved seed use is smaller in magnitude with a DNA-derived categorization, as compared with farmer reports. The greatest yield benefit is with correctly identified improved seed. This indicates that investments in farmers' access to information, seed labeling, and seed system oversight are needed to complement investments in seed variety development.
Precise agricultural statistics are necessary to track productivity and design sound agricultural policies. Yet, in settings where intercropping is prevalent, even crop yield can be challenging to measure. In a systematic survey of the literature on crop yield in low-income settings, we find that scholars specify how they estimate the yield denominator in under 10% of cases. Using household survey data from Tanzania, we consider four alternative methods of allocating land area on plots that contain multiple crops, and explore the implications of this measurement decision for analyses of maize and rice yield. We find that 64% of cultivated plots contain more than one crop, and average yield estimates vary with different methods of calculating area planted. This pattern is more pronounced for maize, which is more likely than rice to share a plot with other crops. The choice among area methods influences which of these two staple crops is found to be more calorie-productive per ha, as well as the extent to which fertilizer is expected to be profitable for maize production. Given that construction decisions can influence the results of analysis, we conclude that the literature would benefit from greater clarity regarding how yield is measured across studies.
Crop yield is one of the most commonly used partial factor productivity measures. It is used to estimate the ratio of quantity of crop output, generally measured in kilograms or tons, to a sole input, land area. Ongoing EPAR research explores the policy implications of measuring yield by area planted versus area harvested. In this brief, we consider implications for crop yield estimates of other decisions in how to construct yield measures from household survey microdata. Using data from three waves of the Tanzania National Panel Survey (TNPS) and two waves of the Ethiopia Socioeconomic Survey (ESS), both part of the World Bank’s Living Standards Measurement Study-Integrated Surveys on Agriculture (LSMS-ISA), we calculate separate crop yield estimates across survey waves following different decisions on disaggregating yield by gender(s) of the plot decision-maker(s) and for pure-stand and mixed stand (intercropped) plots, on including crop production from multiple growing seasons, and on how to treat outlier observations.
We use OLS and logistic regression to investigate variation in husband and wife perspectives on the division of authority over agriculture-related decisions within households in rural Tanzania. Using original data from husbands and wives (interviewed separately) in 1,851 Tanzanian households, the analysis examines differences in the wife’s authority over 13 household and farming decisions. The study finds that the level of decision-making authority allocated to wives by their husbands, and the authority allocated by wives to themselves, both vary significantly across households. In addition to commonly considered assets such as women’s age and education, in rural agricultural households women’s health and labour activities also appear to matter for perceptions of authority. We also find husbands and wives interviewed separately frequently disagree with each other over who holds authority over key farming, family, and livelihood decisions. Further, the results of OLS and logistic regression suggest that even after controlling for various individual, household, and regional characteristics, husband and wife claims to decision-making authority continue to vary systematically by decision – suggesting decision characteristics themselves also matter. The absence of spousal agreement over the allocation of authority (i.e., a lack of “intrahousehold accord”) over different farm and household decisions is problematic for interventions seeking to use survey data to develop and inform strategies for reducing gender inequalities or empowering women in rural agricultural households. Findings provide policy and program insights into when studies interviewing only a single spouse or considering only a single decision may inaccurately characterize intra-household decision-making dynamics.
Common estimates of agricultural productivity rely upon crude measures of crop yield, typically defined as the weight harvested of a crop divided by the area harvested. But this common yield measure poorly reflects performance among farm systems combining multiple crops in one area (e.g., intercropping), and also ignores the possibility that farmers might lose crop area between planting and harvest (e.g., partial crop failure). Drawing on detailed plot-level data from Tanzania’s National Panel Survey, our research contrasts measures of smallholder productivity using production per hectare harvested and production per hectare planted.
An initial analysis (Research Brief - Rice Productivity Measurement) looking at rice production finds that yield by area planted differs significantly from yield by area harvested, particularly for smaller farms and female-headed households. OLS regression further reveals different demographic and management-related drivers of variability in yield gains – and thus different implications for policy and development interventions – depending on the yield measurement used. Findings suggest a need to better specify “yield” to more effectively guide agricultural development efforts.
A farmer’s decision of how much land to dedicate to each crop reflects their farming options at the extensive and intensive margins. The extensive margin represents the total amount of agricultural land area that a farmer has available in a given year (referred to interchangeably as ‘farm size’ or ‘agricultural land’). A farmer increases land use on the extensive margin by planting on new agricultural land. The intensive margin represents area planted of crops as a proportion of total farm size. A farmer increases the intensive margin by increasing output within a fixed area. This analysis examines cropping patterns for households in Tanzania between 2008 and 2010 using data from the Tanzania National Panel Survey (TZNPS). This brief describes changes in farm size, total area planted, and area planted of select annual crops to highlight the dynamic nature of farmer’s cropping choices for a sample population of 2,246 agricultural households that reported having any agricultural land in 2008 or 2010. Throughout the brief, we present summary statistics at the national level and compare them with household-level data to show how results vary depending on how the sub-population is defined and how average measures can mask household level changes. We analyze these questions in the context of smallholders (defined as households with total agricultural land area as less than two hectares) and farming systems.