Annual Agricultural Survey 2019-2020

The sample from the previous campaign was renewed for the 2019/2020 campaign. As a reminder, the EAA was built on a two-stage survey, with census districts (DR) as primary units (PU) and agricultural households as secondary units (US), as defined in the 2013 RGPHAE. results of the last RGPHAE count 755,532 agricultural households practicing agriculture in the broad sense. On this basis, 455,916 farm households practice rain-fed agriculture. The latter constitute the frame file for the agricultural survey. The farm household file was used as the sampling frame for the first stage CD draws. The secondary sampling frame is made up, at the level of each primary unit (UP) or DR drawn, of the exhaustive list of agricultural households who live there and represent the secondary units (US).

In line with the broadening of the scope of the survey recommended by the AGRIS approach, the sampling plan integrated from this 2019/2020 campaign, a first degree stratification, induced by that of the second degree, for better reflect different agricultural activities and improve the efficiency of estimates. The choice of a first degree stratification induced by quadratic stratification, although less efficient than an independent first degree stratification, was guided by the constraint of non-existence of relevant variables of interest in the sampling frame of the RGPHAE to discriminate against DRs.

The stratification took into account the relative importance of the main agricultural activities (in terms of household size) identified during the RGPHAE, namely rain-fed agriculture, animal husbandry and horticulture. Four strata were thus constituted as follows:

• the “rain-fed stratum only” which groups together all the households practicing only crops under rain;
• the “Breeding only” stratum for households that practice breeding only;
• the “Horticulture and other crops” stratum which groups together households which mainly practice horticulture and incidentally other crops (forestry, fruit growing, etc.);
• the “Pluvial-Livestock” stratum made up of households which practice both rain-fed agriculture and livestock.

The size of the sample of agricultural households to be surveyed was calculated by department (field of study) by setting a relative error of 10% on the variable of interest (see technical note in documentation).

At the national level, this resulted in a theoretical sample size of 7,300 agricultural households, distributed among 1,460 physical DRs, at the rate of 5 households per DR. At the end of the enumeration operation carried out in the physical sample DRs, adjustments were made to take into account the actual updated size of the DRs, which led to a final size of 7755, or 1527 DR.

The advantage of this method is that it is not necessary to have an exhaustive list of secondary units for the whole of the national territory, but only those residing in the primary sample units, which makes it possible to reduce the travel and consequently the costs of the investigation.

The sample draws are organized independently from one field of study (department) to another. Primary units (or DRs) are drawn with unequal probabilities and with replacement (PIAR draw). The probability of leaving a CD at each drawing is chosen proportional to its size expressed in terms of the number of agricultural households.

At the second stage, the secondary units (or agricultural households) are drawn with equal probabilities and without replacement (PESR draw). A constant number of secondary units is selected from each DR in the first stage sample. This constant number of secondary units is chosen equal to 5.