December 2007
Agenda Item 4
AFRICAN COMMISSION ON AGRICULTURAL STATISTICSTwentieth Session
Algiers, Algeria, 10 - 13 December 2007
AGRI-GENDER DATABASETOOL KIT FOR COLLECTION AND USE OF SEX DISAGGREGATED DATA
- WORLD PROGRAMME FOR THE CENSUS OF AGRICULTURAL (WCA) 1930–’90
The first six rounds of the World programme for the Census of Agricultural
(WCA 1930 1980) focussed mainly on data collection concerning the quantities produced of selected principal agricultural products (crops and livestock). Data collection related to the human factor in agricultural production, if at all collected, pertained to the head of the household or agricultural holder, initially without distinction of his/her sex. Women’s involvement in agricultural production was usually perceived as domestic or reproductive work rather than economic or productive workand was therefore seldom recorded (FAO, 2005b).
Male and female farmers are affected differently by agricultural policies and programmes because of their diverse yet often complementary roles and responsibilities in agricultural production, disparities in their access to and control over productive resources and the existence of social norms and legal legislations that often favour men over women. Sex-disaggregated agricultural data can be used to illustrate economic, social and political differences that may exist between male and female farmers, to assess the possible impacts of these differences on their production and productivity, and to better understand and recognize men and women’s (changing) roles and responsibilities related to the agricultural sector, rural development and food security.
With the 7th round of the programme (1990), more attention was being paid to the collection of socio-economic and sex-disaggregated agricultural data. New data requirements had come up because of the expanding role of the private sector and civil society, increased decentralization in decision-making, a rising demand for greater transparency in decision-making, the emergence of poverty reduction and food security programmes and by the end of that decade, the establishment of the Millennium Development Goals. Data producers responded by expanding the number of topics covered in line with the change in focus from agricultural to rural development, presenting data at sub-national levels and accepting the need to reflect gender concerns through the collection of sex-disaggregated data at household and sub-household levels. Agricultural censuses carried out during this round of the WCA provided more data on human resources required for agricultural production, including data on women’s contributions to agricultural production and their access to productive resources (FAO, 1995).
- GENDER RELATED DATA IN THE 2000 WCA
The 2000 round of the programmeexpanded data collection to also cover small-scale and peri-urban agricultural activitiesandstarted to address gender biases in statistical tools used.
During this round of WCA, several countries in Africa[1] experimented with analysing and presenting data atsub-holding[2] level,introducing the sub-holder[3]conceptto try and reduce the under-representation of women farmers’ efforts in agricultural production in statistical data collection. These concepts allow for a better assessment of the role of all household members, particularly women, in the management of anagricultural holding, as their productive activities are statistically no longer attributed to the male head of households, as was previously the case.Data presentation at sub-holding level and per sex of sub-holder allows for extensive cross-tabulation of production factors such as plot/field sizes, cultures and inputs used with socio-demographic factors. The data thus obtained provides the necessary basis forin-depth analysis of intra-household sex and age-based differences in agricultural production.
Furthermore, many countries started collecting data in such a way that presentation of the results became possible at sub-national level, be it provincial, regional and at times at district level. This further contributed to greater visibility of female farmers work as most gender-based differences, like any other variation,show more clearly at sub-national levels.
The resulting data provides useful insights in and confirms trends earlier identified by incidental case study material in relation to, among others:
The feminisation of the agricultural sector;
The male-dominated rural out-migration;
Differences between male and female farmers’ access to productive resources like land, water and animals, but also differences in access to extension services, credit and marketing facilities;
Differences in availability of family –unpaid- labour and its impact on female farmers’ production costs;
Differences in preferences in cropscultivated and animals reared.
Obviously such differences vary between countries and between different regions within the countries. Most important however is that data are now becoming available, allowing for targeted planning of agricultural and rural development programmes and gender-specific evaluation and monitoring of the impact of such programmes.
Noteworthy here is the publication of thematic census reports providing insight into gender relations in the agricultural sector. In September 2007, Tanzania launched Volume IX of its 2002/2003 agricultural sample census, entitled “Gender profile of smallholder rural agricultural population in Tanzania Mainland” and Niger and Cape Verde are expected to follow shortly with similar thematic census reports.
The PowerPoint presentation with give further details of data collected by selected countries
- GENDER RELATED DATA IN THE 2010 WCA
For the current round of the World Programme for the Census of Agriculture (WCA 2010), FAO recommends the use of a modular approach, consisting of: -i- a core census module covering a limited number of variables on which data is to be collected in an exhaustive manner and –ii-twelve supplementary modules[4], collecting data in greater detail through sample surveys,coveringsub-sectorsselected on the basis of the country-specific circumstances. This way the organisation hopes to support countries in reducing costs and the time required fordata processing and analysis thereby facilitating early dissemination of the results.
The approach proposed by FAO for the 2010 round of the programme is expected to further enhance the production and use of sex-disaggregated agricultural data for a number of reasons:
- First of all, this roundtakes into account the need to use agricultural census data for monitoringcountries’ achievements towards attainingthe Millennium Development Goals, includingMDG 3 Promote gender equality and empower women.
- Secondly, the agricultural holder[5]concept has been modified to better reflect the realities of farm management practices such as differences in men’s and women’s managerial and financial control over the production, storage, processing and marketing of agricultural products. It is now recognized that more than one person, for example a husband and a wife, could manage a holding as joint holders.[6]
- Thirdly, the programme encourages countries to include items in the supplementary modules of their census that provide greater insight into the roles and responsibilities of men and women in agricultural production. Theme 12 of the supplementary modules covers data collection about the Management of the holding, including reference to any sub-holders mentioned before, that may be operating on a farm. Information obtained under Theme 12 becomes particular relevant if the data collection is linked to other modules, especially those pertaining to access to productive resources (themes 01, 02, 04, 10 and 11), agricultural practices and inputs (themes 05 and 06) and farm labour (theme 08).
- Finally, employment concepts have been amended in line with standards of the International Labour Organization to better reflect the structure of employment in rural areas (FAO, 2005a).
- AGRI-GENDER DATABASE – STATISTICAL TOOL FOR COLLECTION OF SEX-DISAGGREGATED DATA
This database is being prepared insupport of increased production and analysis of sex-disaggregated agricultural data. It presentsgender-sensitive questions/questionnaire components and tables obtainedfrom agricultural censusesthat have been implemented in Africaduring the 2000 round of the WCA programme.
The database covers the following nine data items:
Data itemsessential for gender specific analysis of the agricultural sector1
2
3
4
5
6
7
8
9 / Agricultural population and households
Access to productive resources
Production and productivity
Destination of agricultural produce
Labour and time-use
Income and expenditures
Membership of agricultural/farmer organisations
Food security
Poverty indicators
The list of items is not exhaustive but highlights subjects which are considered essential for gender specific analysis of the agricultural sector. Data items 1, 2, 3 and 5 represent minimum data requirementstoadequately reflect the roles and responsibilities of men and women farmers in the agricultural sector. Collection of sex-disaggregated data on data items 4, 6, 7 and 8 are considered vital for planning food security and poverty reductionprogrammes and advancing gender equality and the empowerment of women. They are crucial for measuring achievements made towards the Millennium Development Goals and other internationally set development targets.
The Agri-gender database comprises two sections. Section 1presents examples of gender-sensitivequestions and questionnaire components obtained from recent agricultural censuses, whileSection 2presents examples of tables that can facilitate analysis and presentation of the data collected.Each table provides sex-specific information and as such builds on the more classical presentation of agricultural census data.
This presentation will look into data items 1, 2, 3, and 5 for reasons of time.
4.1Data item 1 - Agricultural population and households
The demographic data collected on agricultural population and households provide important information on the gender-based structure of anagricultural population, the composition of agricultural households andsocio-economic characteristics of the household members.
Analysis of the agricultural population by sex and age groups at national and sub-national levels offers insight into the effects of rural out-migration, civil conflicts and the HIV/AIDS pandemic on the agricultural labour force and availability of farm labour.
Agricultural census data from Guinea, clearly reflect a phenomenon commonly called “feminisation of the agricultural sector”
Figure 1: Agricultural population at the nationallevelper sex and age group in Guinea
/ Figure 2: Agricultural population per sex and age group inthe Labé Region of Guinea
Source: National Agricultural Census (RNA), Guinea, 2000
Agricultural census data from Tanzania illustratesthe impact male rural out-migration can have on the availability of male active members in female-headed households.
Table 1 Active agriculture Population by Sex in Male and Female Headed Agricultural HouseholdsSelected Regions / Male- headed households / Female- headed households
male: female ratio / Total per hh / males per hh / females per hh / Male: female ratio / Total per hh / males per hh / females per hh
Arusha / 110:100 / 2.8 / 1.5 / 1.3 / 65:100 / 2.3 / 0.9 / 1.4
Mtwara / 97:100 / 2.4 / 1.2 / 1.2 / 38:100 / 1.7 / 0.5 / 1.2
Ruvuma / 102:100 / 2.6 / 1.3 / 1.3 / 38:100 / 1.8 / 0.5 / 1.3
Iringa / 107:100 / 2.6 / 1.3 / 1.2 / 38:100 / 1.8 / 0.5 / 1.3
Mbeya / 105:100 / 2.5 / 1.3 / 1.2 / 43:100 / 1.7 / 0.5 / 1.2
Mwanza / 107:100 / 3.3 / 1.7 / 1.6 / 56:100 / 2.5 / 0.9 / 1.6
Mara / 105:100 / 3.1 / 1.6 / 1.5 / 60:100 / 2.3 / 0.9 / 1.4
Manyara / 115:100 / 3.0 / 1.6 / 1.4 / 64:100 / 2.4 / 0.9 / 1.5
Tanzania Mainland / 106:100 / 2.8 / 1.4 / 1.4 / 49:100 / 2.0 / 0.7 / 1.3
Source: National sample census of agriculture, Tanzania, 2002/03
The following chart then illustrates how the lack of male adult family labour effects the use of credit in female headed households.
/ Source: National sample census of agriculture, Tanzania, 2002/03These examples clearly show how cross-tabulating demographic data with other agricultural related data contributes to greater insights in gender-related issues prevailing in the agricultural sector.
4.2Data item 2 - Access to productive resources
Sex-disaggregated data on men and women’s access to and control over productive resources[7] provide vital information for planners and policy makers. Theyillustrate gender differences in agricultural production and productivity and provide insight into support measures needed.
Land is one if not the most important productive resource for farmers. The Agri-gender database presents an example from Nigeron how tocollect data about male and female sub-holders’ access to land within the family agricultural holding. Column 1 and 2 record the number of the plots and fields, column 3 and 4 record the name and sex of the sub-holder or plot manager with column 5 collecting data on whether the land is cultivated jointly, as “family land” or individually, (col. 5) and how it was obtained (col. 8).
Example: Question regarding landownership by sub-holder
INVENTAIRE DES PARCELLES DU MËNAGE AGRICOLEIdentification
champs, parcelles / Nom et prénoms duresponsable de la parcelle / Sexe du responsable de la parcelle / Type de gestion de la parcelle
1 / 2 / 3 / 4 / 5
Inscrire le numéro d'ordre
du champ / Inscrire le numéro d'ordre de la parcelle / Inscrire d'abord le nom, puis les prénoms du responsable
de la parcelle en commençant parle Chef de Ménage / 1 = Masculin
2 = Féminin / 1 = Individuel
2 = Collectif
|__|__| / |__|__| / |__| / |__|
|__|__| / |__|__| / |__| / |__|
|__|__| / |__|__| / |__| / |__|
Etc.
Continue
INVENTAIRE DES PARCELLES DU MËNAGE AGRICOLEPassé cultural
de la parcelle / Système de
culture / Mode
d'acquisition / Type de relief
6 / 7 / 8 / 9
1 = Cultivé
2 = Jachère / 1 = Culture pure
2 = Cult. associé / 1 = Héritage
2 = Achat
3 = Fermage ou métayage
4 = Prêt
5 = Don
6 = Autre / 1 = Plaine ou plateau
2 = Bas-fonds
3 = Versant colline montagne
|__| / |__| / |__| / |__|
|__| / |__| / |__| / |__|
|__| / |__| / |__| / |__|
Etc.
Source: Recensement général de l’agriculture et du cheptel, Niger, 2005/06
Access to land is particular important give that landownership may have an impact on access to: credit,membership in cooperatives and the services these may provide. Moreover, insecure land-rights reduces farmers’incentives to invest inhigher yielding agricultural practices or to preserve and regenerate the land.
The Nigercensus also provided interesting results about the ownership of livestock and poultry, which was recorded by sex of the owner (individual household member). Based on the country- specific circumstances, the results given in Table 2 indicate a relatively high percentage of sedentary cows belonging to a woman, whereas the percentage of female owned chicken was much lower than expected.
Table 2 – Ownership of selected animals by sex of owner– Niger
Region / Cows / Sheep / Goats / Birdsmen / women / men / women / men / women / men / women
% / % / % / % / % / % / % / %
Agadez / 74.2 / 25.8 / 73.1 / 26.9 / 44.8 / 55.2 / 46.0 / 54.0
Diffa / 83.6 / 16.4 / 76.0 / 24.0 / 72.1 / 27.9 / 68.3 / 31.7
Dosso / 84.1 / 15.9 / 60.9 / 39.1 / 35.2 / 64.8 / 66.1 / 33.9
Maradi / 76.7 / 23.3 / 48.3 / 51.7 / 25.2 / 74.8 / 72.4 / 27.6
Tahoua / 65.1 / 34.9 / 62.4 / 37.6 / 51.1 / 48.9 / 74.7 / 25.3
Tillabery / 80.5 / 19.5 / 61.7 / 38.3 / 58.0 / 42.0 / 61.3 / 38.7
Zinder / 76.9 / 23.1 / 57.5 / 42.5 / 42.9 / 57.1 / 74.4 / 25.6
Niamey / 69.9 / 30.1 / 64.1 / 35.9 / 58.4 / 41.6 / 74.3 / 25.7
Ensemble Niger / 77.6 / 22.4 / 59.9 / 40.1 / 45.2 / 54.8 / 70.3 / 29.7
Source: Recensement général de l’agriculture et du cheptel, Niger, 2005/06
Senegal’slast agricultural census recorded data on men and women’s access to agricultural extension, which showed that at national level, male sub-holders receive three times more extension services that female sub-holders.
Table 3: Access to agricultural extension services by plot manager –Senegal
Region / Male Plot Managers / Female Plot Managers / Total / %N / % / N / %
Region of Diourbel / 16197 / 4.62 / 4722 / 1.35 / 20919 / 3.9
Region of Thiès / 7826 / 1.37 / 5895 / 1.03 / 13721 / 1.2
Region of Koalack / 39889 / 6.55 / 9075 / 1.47 / 48964 / 5.6
Region of Kolda / 18364 / 4.05 / 4612 / 1.02 / 22976 / 3.4
National / 155072 / 4.08 / 53253 / 1.40 / 208325 / 3.4
Source: RecensementNational de l'Agriculture, Sénégal,1998-99
4.3Data item 3 - Production and productivity
Sex-disaggregated data on agricultural production and productivity, especially at sub-holding level, can give insight into who produces what, the amount produced and -when linked with other production factors- the constraints encountered by men and women farmers in this regard.
The following section from the 1999/2000 Mali agricultural census gives a good example of how data on planted area and kind of culture can be recapitulated and then be used for calculations about the production level of sub-holders. The Agri-gender database recommends that column 156 is added to ensure that the information on the sex of the sub-holder remains linked to the sub-holding as this link could easily get lost when demographic data is collected on a separate questionnaire sheet.
RECAPITULATIF DES BLOCS ET PARCELLES DE L’EXPLOITATION TRADITIONELLEN° Bloc / N° parcelle / Saison:
1= Hivern.
2= Hors hivern. / Culture principale / Superficie en ares / N° d’ordre du responsable de mise en val. de la parcelle / Nom duresponsable de la parcelle / Sexe duresponsable de la parcelle
1 = M
2 = F
Nom / Code
148 / 149 / 150 / 151 / 152 / 153 / 154 / 155 / 156
I__I__I / I__I__I / I___I / _____ / I__|__|__I / I__I__I__I__I,I__I__I / I___I___I / ______/ |__|
I__I__I / I__I__I / I___I / _____ / I__|__|__I / I__I__I__I__I,I__I__I / I___I___I / ______/ |__|
I__I__I / I__I__I / I___I / _____ / I__|__|__I / I__I__I__I__I,I__I__I / I___I___I / ______/ |__|
Etc.
Total (1) / I___I___I___I___I___I, I___I___I
Source: République du Mali – Recensement Général de l’Agriculture 1999/ 2000
Data on productivity levels per sex of (sub-) holder is more difficult to establish, as most census programmes apply onecommon multiplier factor determined as an average for a particular agro-ecological zone. In this regard it will be interesting to see the results of the Niger census, which tried to collect information about quantities produced at plot level. When compared with plot sizes and the sex of the plot manager this could give valuable information regarding differences, if any, in productivity levels of male and female sub-holders.
Code
/Système de culture
(code) / Sexe du responsable de parcelle(1 = M; 2 = F)
/Nom de la culture
/Code de la culture
/Poids brut en kilogramme
(kg) /Poids net en kilogramme
(kg)Champ
/Parcelle
1
/ 2 /3
/3’
/4
/5
/6
/7
|__|__|
/ |__|__| / |__| / |__| /………….
/ |__|__| /|__|__|,|__|__|__|
/|__|__|,|__|__|__|
|__|__| / |__|__| / |__| / |__| / …………. / |__|__| /|__|__|,|__|__|__|
/|__|__|,|__|__|__|
|__|__| / |__|__| / |__| / |__| / …………. / |__|__| /|__|__|,|__|__|__|
/|__|__|,|__|__|__|
|__|__| / |__|__| / |__| / |__| / …………. / |__|__| /|__|__|,|__|__|__|
/|__|__|,|__|__|__|
Etc.
Source République du Niger – Recensement Général de l’Agriculture et du Cheptel 2004-2006
The Agri-gender database also comprises an example from the 2001 agricultural census of the Gambiaonhow to collect data about ownership of animals by male and female members of an agricultural household.
11.Enter in the tables below the number of livestock and/or poultry that the holder has on this day of filing the questionnaire
A. CATTLE AND CALVES / Number of Cattle / Total Cattle / Number of Cattle managed or owned by sex of membersMales / Females / Male / Female
(a) / (b) / (c) / (d) / (e) / (f)
Total number of cattle of all ages
(If “None” enter X and move to B)
Total number under 2 years of age
Total number 2 years of age and over
B. GOATS / Number of Goats / Total Goats / Number of Goats managed or owned by sex of members
Males / Females / Male / Female
(a) / (b) / (c) / (d) / (e) / (f)
Total number of goats of all ages
(If “None” enter X and move to C)
Total number under 1 years of age
Total number1 year of age and over
C. SHEEP AND LAMBS / Number of Sheep / Total Sheep / Number of Sheep managed or owned by sex of members
Males / Females / Male / Female
(a) / (b) / (c) / (d) / (e) / (f)
Total number of sheep of all ages
(If “None” enter X and move to D)
Total number under 1 years of age
Total number 1 year of age and over
D. PIGS / Number of Pigs / Total Pigs / Number of Pigs managed or owned by sex of members
Males / Females / Male / Female
(a) / (b) / (c) / (d) / (e) / (f)
Total number of pigs of all ages
(If “None” enter X and move to E)
Total number under 6 months of age
Total number 6 months of age and over
E. POULTRY AND RABBITS / Hens, Cocks, Pullets and Chicks / Ducks and Ducklings / Other Poultry such as Turkeys, Guinea Fowls, etc. / Other Farm Animals such as beehives, etc. (Specify) / Rabbits
(a) / (b) / (c) / (d) / (e)
Total number
Total under 6 months / XXXX / XXXXXXXXXXX / XXXXXXXX / XXXXXXXX
6 months and over / XXXX / XXXXXXXXXXX / XXXXXXXX / XXXXXXXX
Source: The Republic of the Gambia – Agricultural Census 2001