The aim of these guidelines is to provide National Statistical Offices with a practical guide of possible modifications to the main aspects of their regular national agricultural survey programmes leading to the collection and dissemination of more adequate gender-disaggregated data.
National agricultural censuses and sample surveys constitute the most important source of basic data in the agricultural sector. They are, in particular, a source of information on the agriculture-related population. For these reasons, and in view of FAO's objective to cooperate with member countries to obtain better data on women in agriculture, these guidelines suggest possible improvements to the national agricultural survey programmes in order to obtain and disseminate more adequate gender-disaggregated statistics.
The terms agricultural holding, holding parcel and agricultural holder, which are basic for an agricultural survey programme, are defined in the next chapter which also includes a number of comments in relation to their use and possible gender biases.
The national agricultural census and sample survey programmes considered in these guidelines are assumed to be the regular periodic agricultural production surveys which are conducted in order to obtain the basic data on the agricultural sector. The agricultural surveys are conducted in order to obtain estimates of a given group of agricultural items (which are numbers) from the totality or a sample of reporting units. A reporting unit is a unit for which a numerical value of each agricultural characteristic considered is defined. The survey population (or sampled population) is the set of survey units that are actually sampled (or it may involve 100 percent of the units) and are called sampling units or enumeration units. A sampling unit must either be associated with one, and only one, reporting unit, or there must be known rules to associate sampling units with reporting units of the survey.
The main objective of the agricultural production survey is to obtain estimates for all or some of the following items:
· planted and harvested areas, areas intended for harvest, potential and actual crop yields; crop production and number of trees;
· livestock and poultry inventories (e.g. type, age, sex, breed and use); production of milk, eggs, honey and seeds;
· characteristics of the holdings (e.g. number, location, legal status, land tenure) and characteristics of the holder (e.g. age, sex, marital status, level of education, whether worked on the holding, amount of time worked on the holding, whether received any payment in cash or kind, main activity and secondary activity);
· population involved in agriculture (e.g. basic demographic characteristics of the holder, holder's household members working on the holding, hired workers on the holding, days of work, etc.);
· farming practices and agricultural inputs: labour involved, amount and type of seeds, fertilizers and pesticides, source of irrigation, drainage, shifting cultivation, grain stocks; machinery, equipment and agricultural buildings; costs of production and value of sales.
The agricultural survey items are defined in each reporting unit. The most common type of reporting unit is the agricultural holding, since agricultural items are naturally defined for the holdings. But the holdings are not the only possible reporting units.
The agricultural population studied in the agricultural surveys consists of.
· holders and members of the holder's household engaged in agricultural activities;
· agricultural workers other than members of the holder's household.
The population variables generally studied are age, sex, days of work, family and hired workers, and access to credit.
The survey field data collection may cover all agricultural holdings (in which case the survey is called agricultural census) or only a sample (called agricultural sample survey). A sample survey could be based on a selection of holdings (list sample survey), or could be based on the selection of a sample of segments that are land areas (area sample survey), or a sample survey could be based on a combination of an area sample and a list sample survey, in which case it is called a multiple frame survey.
For agricultural censuses and list sample surveys:
· the enumeration unit and the reporting unit of the survey will be the agricultural holding;
· the survey data collection will be accomplished by enumerators through direct interviews with the holders or with the hired managers using a questionnaire for each agricultural holding.
For agricultural area sample surveys:
· the reporting unit of the survey is the agricultural holding or the tract, which is the part(s) of holdings (or non-agricultural areas) included in the segments;
· the survey data collection will be accomplished by enumerators through direct interviews with the holders or with the hired managers using a questionnaire for each tract.
A general-purpose agricultural sample survey may cover all the above-mentioned variables. An agricultural census might not include, for example, crop yields and/or livestock estimates. In any case, a general-purpose agricultural survey usually covers only the basic characteristics of the population concerned with agricultural activities on the holdings.
The suggestions contained in these guidelines for such national agricultural census and sample survey programmes refer specifically to the collection and dissemination of data on the agriculture-related population.
The improvements suggested consider the different aspects of such national agricultural survey programmes and relate to the production of more adequate gender-disaggregated data, such as concepts, definitions and classifications used, questionnaire items, questionnaire construction, instruction manuals, selection and training of field staff, pilot surveys, field work, tabulation and dissemination of results and postenumeration survey and publicity campaigns.