• TABLE OF CONTENTS
HIDE
 Front Cover
 Title Page
 Copyright
 Table of Contents
 List of Tables
 List of Figures
 Acronyms and abbreviations
 Acknowledgement
 Executive summary
 Introduction
 The household survey
 Socioeconomic characteristics
 Gender differentials in farm management...
 Gender differentials in access...
 Gender patterns of labor utili...
 Gender differentials in agricultural...
 Gender differentials in agricultural...
 Gender differentials in technology...
 Gender differentials in access...
 Conclusion and policy implicat...
 Reference






Title: Gender differentials in agricultural production and decision-makeing among smallholders in Ada, Lume, and Gimbichu Woredas of the Central Highlands of Ethiopia
CITATION PDF VIEWER PAGE IMAGE ZOOMABLE PAGE TEXT
Full Citation
STANDARD VIEW MARC VIEW
Permanent Link: http://ufdc.ufl.edu/UF00077464/00001
 Material Information
Title: Gender differentials in agricultural production and decision-makeing among smallholders in Ada, Lume, and Gimbichu Woredas of the Central Highlands of Ethiopia
Physical Description: Book
Language: English
Creator: Tiruneh, Addis
Publisher: International Maize and Wheat Improvement Center (CIMMYT)
Publication Date: 2001
 Subjects
Subject: Africa   ( lcsh )
Farming   ( lcsh )
Spatial Coverage: Africa
 Record Information
Bibliographic ID: UF00077464
Volume ID: VID00001
Source Institution: University of Florida
Holding Location: African Studies Collections in the Department of Special Collections and Area Studies, George A. Smathers Libraries, University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: isbn - 970-648-061-7

Downloads

This item has the following downloads:

CentralHEth ( PDF )


Table of Contents
    Front Cover
        Front cover
    Title Page
        Page i
    Copyright
        Page ii
    Table of Contents
        Page iii
    List of Tables
        Page iv
    List of Figures
        Page v
    Acronyms and abbreviations
        Page vi
    Acknowledgement
        Page vi
    Executive summary
        Page vii
        Page viii
        Page ix
    Introduction
        Page 1
    The household survey
        Page 2
        Page 3
        Page 4
        Page 5
    Socioeconomic characteristics
        Page 6
        Page 7
        Page 8
        Page 9
        Page 10
        Page 11
        Page 12
    Gender differentials in farm management practices
        Page 13
        Page 14
        Page 15
        Page 16
        Page 17
        Page 18
        Page 19
        Page 20
        Page 21
        Page 22
        Page 23
        Page 24
        Page 25
    Gender differentials in access to land
        Page 26
        Page 27
        Page 28
        Page 29
    Gender patterns of labor utilization
        Page 30
        Page 31
        Page 32
        Page 33
        Page 34
        Page 35
        Page 36
        Page 37
        Page 38
        Page 39
        Page 40
        Page 41
        Page 42
    Gender differentials in agricultural production, utilization, and food availability
        Page 43
        Page 44
        Page 45
    Gender differentials in agricultural productivity
        Page 46
        Page 47
        Page 48
        Page 49
    Gender differentials in technology adoption
        Page 50
        Page 51
    Gender differentials in access to rural institutions
        Page 52
        Page 53
        Page 54
        Page 55
        Page 56
        Page 57
    Conclusion and policy implications
        Page 58
        Page 59
        Page 60
    Reference
        Page 61
        Page 62
Full Text




Gender Differentials in Agricultural

Production and Decision-Making Among

Smallholders in Ada, Lume, and

Gimbichu Woredas of the

Central Highlands of Ethiopia





Addis Tiruneh,
Teklu Tesfaye,
Wilfred Mwangi, and
Hugo Verkuiji




February 2001


) CIMMYT
iral INTERNATIONAL MAIZE AND
ion
WHEAT IMPROVEMENT CENTER


Funded by the
European Union









Gender Differentials in Agricultural

Production and Decision-Making

Among Smallholders in Ada, Lume,

and Gimbichu Woredas of the

Central Highlands of Ethiopia






Addis Tiruneh,
Teklu Tesfaye,
Wilfred Mwangi, and
Hugo Verkuijl*





February 2001




* Addis Tiruneh is with the Center for Education, Research and Training on Women in Development, of the Institute of
Development Research (IDR)/Department of Economics at the Addis Ababa University in Ethiopia. Teklu Tesfaye is with the
Debre Zeit Agricultural Research Centre of the Ethiopian Agricultural Research Organization. Wilfred Mwangi is a principal
economist with the International Maize and Wheat Improvement Center (CIMMYT) and also Director of Agriculture, Ministry
of Agriculture, Kenya. At the time this paper was drafted, Hugo Verkuijl was an associate scientist with CIMMYT, based in
Addis Ababa, Ethiopia. The views presented in this paper are those of the authors and do not necessarily reflect policies of
their respective organizations.













CIMMYT (www.cimmvt.orQ) is an internationally funded, nonprofit, scientific research and training organization.
Headquartered in Mexico, CIMMYT works with agricultural research institutions worldwide to improve the productivity,
profitability, and sustainability of maize and wheat systems for poor farmers in developing countries. It is one of 16 food
and environmental organizations known as the Future Harvest Centers. Located around the world, the Future Harvest
Centers conduct research in partnership with farmers, scientists, and policymakers to help alleviate poverty and increase
food security while protecting natural resources. The centers are supported by the Consultative Group on International
Agricultural Research (CGIAR) (www.cQiar.orQ), whose members include nearly 60 countries, private foundations, and
regional and international organizations. Financial support for CIMMYT's research agenda also comes from many other
sources, including foundations, development banks, and public and private agencies.

F U T U R E" Future Harvest builds awareness and support for food and environmental research for a world with
HARV/ES T less poverty, a healthier human family, well-nourished children, and a better environment. It supports
research, promotes partnerships, and sponsors projects that bring the results of research to rural communities, farmers,
and families in Africa, Asia, and Latin America (www.futureharvest.orQ).

International Maize and Wheat Improvement Center (CIMMYT) 2001. All rights reserved. The opinions expressed in
this publication are the sole responsibility of the authors. The designations employed in the presentation of materials in
this publication do not imply the expression of any opinion whatsoever on the part of CIMMYT or its contributory
organizations concerning the legal status of any country, territory, city, or area, or of its authorities, or concerning the
delimitation of its frontiers or boundaries. CIMMYT encourages fair use of this material. Proper citation is requested.

Printed in Mexico.

Correct citation: A. Tiruneh, T. Tesfaye, W. Mwangi, and H. Verkuijl. 2001. Gender Differentials in Agricultural
Production and Decision-Making Among Smallholders in Ada, Lume, and Gimbichu Woredas of the Central
Highlands of Ethiopia. Mexico, D.F.: International Maize and Wheat Improvement Center (CIMMYT) and Ethiopian
Agricultural Research Organization (EARO).

Abstract: This study provides concrete information about the role of gender in resource ownership and decision-
making in the mixed farming systems of Ada, Lume, and Gimbichu woredas in the central highlands of Ethiopia. A
multistage purposive sampling method was used to select male-and female-headed households based on population,
crops grown, altitude, and distance from the research center. Of a sample of 180 households, 81 (45%) were headed by
women. On average, male-headed households (MHHs) were larger than female-headed households (FHHs). Male heads
of households were more educated than female heads of household, and they owned more ox-plows and livestock. The
average area cultivated by MHHs was larger than that cultivated by FHHs for almost all crops. On the other hand, the
average per capital land holding was almost equal between MHHs and FHHs. Both types of households acquired land
through government allocation and used credit to purchase seed and fertilizer. Significant factors affecting gross value of
output for MHHs were the farmer's age, family labor, farm size, livestock units, and inorganic fertilizer. The significant
factors affecting gross value of output for FHHs were family labor, farm size, livestock units, inorganic fertilizer, hired
labor, and extension contact. The marginal value product (MVP) of family labor is higher in MHHs compared to its price
(wage rate), but it is lower in FHHs, indicating that MHHs were able to increase their productivity by using more family
labor. The MVP of farm size was lower than its factor price for MHHs and higher for FHHs, indicating that FHHs could
increase their productivity by cultivating more land. The MVP for inorganic fertilizer was higher than its factor cost for
both MHHs and FHHs, so both types of households could increase productivity by increasing their use of inorganic
fertilizer. The gender difference in gross output was considerable, partly because FHHs used fewer inputs. If MHHs and
FHHs had equal access to inputs, it is likely their levels of productivity would be similar. In 1997, about 59% of MHHs
and 42% of FHHs grew wheat, and most of these grew local varieties (70% of MHHs and 86% of FHHs). A significantly
higher proportion (t=5.7; p<0.05) of MHHs (30%) grew improved wheat varieties than FHHs (14%). In MHHs,
extension services and farm size had a positive effect on the adoption of improved wheats, whereas radio ownership and
farm size increased the odds in favor of adopting improved wheats for FHHs.

ISBN: 970-648-061-7
AGROVOC descriptors: Ethiopia; Production economics; Home economics; Plant production; Animal production;
Farming systems; Farm management; Land management; Socioeconomic environment; Labour allocation; Work
organization; Technology transfer; Innovation adoption; Decision making; Role of women; Small farms; Highlands
AGRIS category codes: E16 Production Economics;
E14 Development Economics and Policies
Dewey decimal classification: 338.16










CONTENTS


Tables ........................................................................................................................... iv
Figures ....................................................................................................................... v
Acronyms and abbreviations ............................................................................................vi
Acknowledgements .....................................................................................................vi
Executive Summary ..................................................................................................... vii

1.0 Introduction ................................................................................................................... 1
2.0 The Household Survey ..................................................................................................... 2
2.1 The Study Area .................................... ...................................................... ........................ 2
2 .2 M methodology ................ ............................................................................................. ......... 3

3.0 Socioeconomic Characteristics ......................................................................................... 6
3.1 Household Characteristics ......................................................... ...................................... 6
3.2 Fertility Characteristics .......................................................... .................................................. 7
3 .3 H household A m entities .............................................................................................................. 1 1
3.4 Farm Implements ................................................. 12

4.0 Gender Differentials in Farm Management Practices ...................................................... 13
4.1 Crop Production ..................................... ............. 13
4.2 Livestock Production ......... ................................... ...... ........ 22
4.3 Tree Crop Production .......... ................................................ ........ ..................... ................. 26

5.0 Gender Differentials in Access to Land ............................................................................ 26
5 .1 L an d R ig h ts ........................................................................................................................... 2 7
5.2 Land Acquisition ..................................... ............. 27
5.3 Land Quality .................................................................... ............... 28
5.4 Decision-Making for Land Improvement, Acquisition, and Rental .............................................. 29

6.0 Gender Patterns of Labor Utilization .......................................................................... 30
6.1 Use of Family Labor in Crop Production ............... ........................................................... 30
6.2 Use of Family Labor in Livestock Production ...................................... ......................... 36
6.3 Use of Hired Labor in Crop Production ................................ ................................................. 38
6.4 Use of Family Labor in Non-Agricultural Tasks .............................. .... .......................... 40

7.0 Gender Differentials in Agricultural Production, Utilization, and Food Availability ........... 43
7.1 Methodology ......................................................................................................................... 44
7.2 Results and Discussion ......... ............................... ...... ........ 44

8.0 Gender Differentials in Agricultural Productivity .......................................... ........... .. 46
8 .1 T he Production Function ............................................................ ........................................ 47
8.2 Model Results ............. .. .......... ........................ ...... .......... 48

9.0 Gender Differentials in Technology Adoption ............................................................ 50
9.1 Factors Affecting the Adoption of Wheat Production Technologies ......................................... 50
9.2 Model Results ...................................................................................... ................ 51

10.0 Gender Differentials in Access to Rural Institutions ...................................... ......... .. 52
10.1 Credit Services ........... ..... ..................................... 52
10.2 Extension Services .................................... ............................................................................ 54

11.0 Conclusion and Policy Implications ............................................................................ 58
11.1 Access to Resources ................................................. 58
1 1 .2 D ivisio n o f L ab o r ........................................................................................... ....................... 5 8
11.3 Decision-Making .......................................... .. ......... ............ 58
11.4 Adoption of Improved Wheat Varieties ........................ ................................ ......................... 58
11.5 Gender Differentials in Agricultural Productivity............... ..................................... ................. 59

References ........................................................................................................................ 61

iii










TABLES


Table 1. Distribution of household types by production system, land area, and crop
production in Ethiopia, 1983 .................... .......................................... .............. ... ............ 1
Table 2. Population size, density and area in Ada, Lume, and Gimbichu woredas, Ethiopia ......................... 3
Table 3. Socioeconomic characteristics of households in Ada, Lume, and Gimbichu woredas, Ethiopia ......... 6
Table 4. Statistics on childbearing and reproductive goals in Ada, Lume, and Gimbichu woredas, Ethiopia .... 7
Table 5. Reasons for having or not having more children in Ada, Lume, and Gimbichu woredas, Ethiopia .... 8
Table 6. Reasons that household members in MHHs gave for not sending girls to school in Ada, Lume,
and Gimbichu woredas, Ethiopia ..................................................... ............ ............... 8
Table 7. Method and source of information for family planning in Ada, Lume, and
G im bichu w oredas, Ethiopia .................... ............................................................................... 9
Table 8. Childbearing and adoption data in Ada, Lume, and Gimbichu woredas, Ethiopia ....................... 10
Table 9. Decision-making on the number of children and sending girls to school in Ada,
Lume, and Gimbichu woredas, Ethiopia....................................................... ..... ................ 10
Table 10. Types of dwelling in Ada, Lume, and Gimbichu woredas, Ethiopia ........................................... 11
Table 11. Household amenities of MHHs and FHHs in Ada, Lume, and Gimbichu woredas, Ethiopia ........... 12
Table 12. Ownership of farm implements in Ada, Lume, and Gimbichu woredas, Ethiopia ........................ 12
Table 13. Mean crop area cultivated (in kert) in Ada, Lume, and Gimbichu woredas, Ethiopia .................... 13
Table 14. Use of local and improved varieties in Ada, Lume, and Gimbichu woredas, Ethiopia ..................... 15
Table 15. Farmers' sources of information about improved varieties in Ada, Lume,
and G im bichu w oredas, Ethiopia ................................................. .. ........... ..... .......... 15
Table 16. Farmers' reasons for growing crops in Ada, Lume, and Gimbichu woredas, Ethiopia .................. 16
Table 17. Average quantity of improved and local seed varieties used in Ada, Lume,
and G im bichu w oredas, Ethiopia ........................................................................ .. ....... ...... 17
Table 18. Mode of transport of farm products in Ada, Lume, and Gimbichu woredas, Ethiopia .................. 17
Table 19. Fertilizer and insecticide use and farmers' reasons for not using insecticides in Ada,
Lume, and Gimbichu woredas, Ethiopia...................................................... ....... ........... 18
Table 20. Crop production constraints in Ada, Lume, and Gimbichu woredas, Ethiopia ............................... 19
Table 21. Decision-making of households on sale and consumption of agricultural products in Ada,
Lum e, and Gimbichu woredas, Ethiopia.................. ................. .... .. .............. ............. .. 20
Table 22. Responsibility for input use decisions in Ada, Lume, and Gimbichu woredas, Ethiopia ................. 21
Table 23. Payment for farm inputs by gender of household heads in Ada, Lume,
and G im bichu woredas, Ethiopia .................................................................. ..... ............ 22
Table 24. Mean number of livestock owned and bought and cost of purchase (Birr) in Ada, Lume,
and Gim bichu woredas, Ethiopia ........................................................................................ 23
Table 25. Livestock consumed and sold and their value (Birr) in Ada, Lume,
and Gimbichu woredas, Ethiopia .................................... .............. ........ .......... ...... ......... 23
Table 26. Costs of keeping livestock (Birr) in Ada, Lume, and Gimbichu woredas, Ethiopia .......................... 24
Table 27. Responsibility for animal husbandry in Ada, Lume, and Gimbichu woredas, Ethiopia .................. 24
Table 28. Reasons for keeping different types of livestock in Ada, Lume, and Gimbichu woredas, Ethiopia ... 25
Table 29. Most important tree crop in Ada, Lume, and Gimbichu woredas, Ethiopia .................................. 26
Table 30. Landholdings by gender of household head in Ada, Lume, and Gimbichu woredas, Ethiopia ......... 27
Table 31. Methods of land acquisition by gender of household head in Ada, Lume,
and Gimbichu woredas, Ethiopia ...................................... ....................... ....................... 28
Table 32. Soil fertility status and irrigation of farmland in Ada, Lume, and Gimbichu woredas, Ethiopia ........ 29
Table 33. Responsibility for decisions on land improvement, acquisition, and renting by gender
of household head in Ada, Lume, and Gimbichu woredas, Ethiopia .................. ........... ......... 30











Table 34. Division of household labor by gender and crop operation in Ada, Lume,
and G im bichu w oredas, Ethiopia ................................ ............................ ....... .. ....... ...... 31
Table 35. Division of family labor by gender and livestock production task in Ada, Lume,
and G im bichu w oredas, Ethiopia ............................... ................ .... ...................... 37
Table 36. Hired labor and reasons for not hiring labor in Ada, Lume, and Gimbichu woredas, Ethiopia ........ 38
Table 37. Division of hired labor by gender and agricultural production tasks in Ada, Lume,
and Gimbichu woredas, Ethiopia ................................. ..................................... ........... .... 39
Table 38. Division of household labor by gender and non-agricultural tasks in Ada, Lume,
and Gim bichu woredas, Ethiopia .................................. ..... ............................. ......... ........ 40
Table 39. Mean quantities of agricultural production and consumption (quintal) by type of crop in Ada,
Lume, and Gimbichu woredas, Ethiopia................ ............................................ .. ............. 45
Table 40. Estimates of Cobb-Douglas production function by gender in Ada, Lume,
and Gim bichu woredas, Ethiopia ............................... ........ ....... ......... ................. ...... 48
Table 41. MVP and factor prices (in Birr) for significant variables by gender of household head in Ada,
Lume, and Gimbichu woredas, Ethiopia........................................................... ....................... 49
Table 42. Parameter estimates of a logistic model for factors affecting adoption of improved wheat
varieties in Ada, Lume, and Gimbichu woredas, Ethiopia ................................... ................ .. 51
Table 43. Some selected variables on credit facilities by gender of household head in Ada, Lume,
and Gimbichu woredas, Ethiopia ................................... .................. .................. 53
Table 44. Decision-making on the use and payment of credit by gender of household head in Ada,
Lume, and Gimbichu woredas, Ethiopia................... .............................. .................... 54
Table 45. Type of extension services provided by gender of household head in Ada, Lume,
and Gimbichu woredas, Ethiopia .................................................... ........ ........... ....... 55
Table 46. Selected variables on extension activities by gender of household head in Ada, Lume,
and Gim bichu woredas, Ethiopia ................................ ................... .................................. 56
Table 47. Gender of household heads with whom extension agents make contact in Ada, Lume,
and Gimbichu woredas, Ethiopia .................................. .... ............................... 56
Table 48. Frequency of adopting recommended practices in Ada, Lume, and
G im bichu woredas, Ethiopia ..................................... ........... ..... .............................. 57
Table 49. Is the household head a contact farmer, and reasons for wanting to be a contact farmer in Ada,
Lum e, and G im bichu w oredas, Ethiopia..................................................................... ............... 57



FIGURES


Figure 1. Ada, Lume, and Gimbichu woredas, Ethiopia ...................... .................................................... 2
Figure 2. Utilization of wheat by gender of household head in Ada, Lume, and
G im bichu w oredas, Ethiopia .................... .. ....... ... ........................................ .......... 46
Figure 3. Utilization of tef by gender of household head in Ada, Lume, and
G im bichu w oredas, Ethiopia ........................................................................... ........... .... 46









ACRONYMS AND ABBREVIATIONS

CSA Central Statistical Authority
CIMMYT International Maize and Wheat Improvement Center
DZARC Debre Zeit Agricultural Research Center
EARO Ethiopian Agricultural Research Organisation
FAO Food and Agricultural Organization
FHHs Female-Headed Households
FDRE Federal Democratic Republic of Ethiopia
GDP Gross Domestic Product
IDR Institute of Development Research
MHHs Male-Headed Households
MVP Marginal Value Product
OLS Ordinary Least Squares
PA Peasant Association
TBA Traditional Birth Attendant
UNDP United Nations Development Program
WHO World Health Organization



ACKNOWLEDGEMENTS

We thank Efrem Bechere, former Vice-President for Research and Development of Alemaya
University of Agriculture; Tekalign Mamo and Negussie Taddesse, former Directors of the Debre Zeit
Agricultural Research Centre (DZARC), for supporting this study. We would like to thank other
colleagues at DZARC who assisted in carrying out this study. Special thanks go to Gezahegn Ayele,
who was one of the initiators of the project before he left for further studies, and Hailemariam
Teklewold for assisting in the fieldwork and data compilation.

The immense support given to the team by the offices of the Eastern Oromia Zone, Agricultural
Development Offices of Ada, Lume, and Gimbichu woredas, and the co-operation of the
development agents and officials of the surveyed Peasant Associations is highly appreciated.

We are grateful to the CIMMYT-European Union Project on Strengthening Economics and Policy
Research in National Agricultural Research Systems in Eastern Africa for funding this project. We
appreciate the editorial assistance of Satwant Kaur in producing this paper, and Marcelo Ortiz for
layout and design.

Our deepest and heartfelt thanks go to the men and women farmers who willingly spared their time
for interviews and shared their experiences generously, and the enumerators who walked during the
rainy and dry seasons through the villages.









ACRONYMS AND ABBREVIATIONS

CSA Central Statistical Authority
CIMMYT International Maize and Wheat Improvement Center
DZARC Debre Zeit Agricultural Research Center
EARO Ethiopian Agricultural Research Organisation
FAO Food and Agricultural Organization
FHHs Female-Headed Households
FDRE Federal Democratic Republic of Ethiopia
GDP Gross Domestic Product
IDR Institute of Development Research
MHHs Male-Headed Households
MVP Marginal Value Product
OLS Ordinary Least Squares
PA Peasant Association
TBA Traditional Birth Attendant
UNDP United Nations Development Program
WHO World Health Organization



ACKNOWLEDGEMENTS

We thank Efrem Bechere, former Vice-President for Research and Development of Alemaya
University of Agriculture; Tekalign Mamo and Negussie Taddesse, former Directors of the Debre Zeit
Agricultural Research Centre (DZARC), for supporting this study. We would like to thank other
colleagues at DZARC who assisted in carrying out this study. Special thanks go to Gezahegn Ayele,
who was one of the initiators of the project before he left for further studies, and Hailemariam
Teklewold for assisting in the fieldwork and data compilation.

The immense support given to the team by the offices of the Eastern Oromia Zone, Agricultural
Development Offices of Ada, Lume, and Gimbichu woredas, and the co-operation of the
development agents and officials of the surveyed Peasant Associations is highly appreciated.

We are grateful to the CIMMYT-European Union Project on Strengthening Economics and Policy
Research in National Agricultural Research Systems in Eastern Africa for funding this project. We
appreciate the editorial assistance of Satwant Kaur in producing this paper, and Marcelo Ortiz for
layout and design.

Our deepest and heartfelt thanks go to the men and women farmers who willingly spared their time
for interviews and shared their experiences generously, and the enumerators who walked during the
rainy and dry seasons through the villages.









EXECUTIVE SUMMARY


Gender analysis has been at the heart of many studies which have sought to outline the complexity,
flexibility, and political aspects of access to and control of resources. It is recognized and empirical
evidence indicates that women do play an important role in decision-making in agriculture and in the
adoption of agricultural technologies. The transfer and adoption of agricultural technologies is
therefore affected by who owns productive resources and who decides what to produce, when to
produce, and how much to produce. Information provided through research is, therefore, vital for
policymakers to be informed of the basis on which decisions are made at the micro-level. This type of
information is lacking in Ethiopia, and this study attempts to fill that gap and provide concrete and
statistical information about gender's role in agricultural production and decision-making in the
household economy. The main focus of the research was to assess the role of gender in terms of
resource ownership and decision-making power in the mixed farming systems of Ada, Lume, and
Gimbichu woredas in the central highlands of Ethiopia.

A multi-stage purposive sampling method was used to select male- and female-headed households.
The households were selected based on the total population, types of crops grown, altitude, and
distance from the research center. Out of a sample of 180 households, 81 (45%) were headed by
females. Female-headed households (FHHs) were those that were managed by a widowed, divorced,
or single woman without the mediation of a husband, father, or male relative in the routine day-to-
day activities of that household. Male-headed households (MHHs) were those where a husband was
present and was the final decision-maker in important issues pertaining to the household (Starkey et
al. 1994). The survey was carried out during the cropping season of 1997.

Survey results indicated a range of similarities and differences among FHHs and MHHs. The average
size of MHHs was larger than FHHs, and male heads of household were more educated than female
heads of household. In MHHs, the decision to have more children was made jointly by husband and
wife. With respect to wealth and resource ownership, both MHHs and FHHs lived in thatched roof
houses, but the number of ox plows owned by MHHs was significantly higher than FHHs and they
were therefore better able to prepare their land. The mean number of livestock owned by MHHs was
higher than FHHs.

The average area cultivated by MHHs was larger than FHHs for almost all crops. The main crop
cultivated in Ada and Lume was tef, while in Gimbichu it was wheat. In Ada, both types of
households grew only local tef varieties, while 53% and 58% of the MHHs and FHHs in Lume,
respectively, grew both local and improved varieties. In Lume, more MHHs (38.7%) grew improved
wheat varieties than FHHs (22.2%). In Ada, both types of households learned about improved
varieties through extension services provided by the Ministry of Agriculture; in Lume 52.6% of FHHs
learned about improved varieties from the market and 47% of MHHs through the extension service.
In almost all FHHs it was the head who decided what to plant. In MHHs, it was mostly a joint
decision by the head and wife in Ada (48.4%) and Gimbichu (54.4%), while in Lume (83.3%) it was
the head who decided. The mean daily energy production per person per day was about 2,278 and
2,291 calories for MHHs and FHHs, respectively.










All MHHs and FHHs used fertilizer. For MHHs, this was a joint decision in Ada (50%), and a
household head decision in Lume (97.1%) and Gimbichu (59.4%). In FHHs, the head decided on
the use of fertilizer. About 38% and 44% of MHHs and FHHs in Ada, 66% and 50% in Lume, and
21% and 4% in Gimbichu, respectively, used herbicides. Both male and female heads of household
decided on the use of herbicides.

The average per capital landholding was almost equal between MHHs and FHHs, and they acquired
their land from the government. MHHs indicated that their land was more fertile than FHHs. About
68%, 100%, and 61% of male heads of household were responsible for deciding whether to
improve land in Ada, Lume, and Gimbichu, respectively. In FHHs, the majority of heads of
household in Ada (96%), Lume (96%) and Gimbichu (100%) decided whether to improve land. The
decision to lend land in Lume was the full responsibility of the wife in MHHs, while in Gimbichu it
was a joint decision between the husband and wife (75%). In FHHs in Gimbichu, the head and the
son decided whether to lend land. The average amount of labor (h/ha) used for crop production
was higher for MHHs than FHHs. Also, MHHs hired more labor for crop production than FHHs.
Wives and daughters in MHHs and heads and daughters in FHHs were primarily responsible for
non-agricultural tasks.

Most farmers obtained credit to purchase seed and fertilizer from the Ministry of Agriculture. Many
farmers were also members of a service cooperative that helped them obtain inputs. The decision
to use credit was made primarily by the male and female heads of household in Lume and
Gimbichu. In Ada, the decision to use credit was made jointly in MHHs, while most female heads of
household decided whether to obtain credit. More MHHs had access to extension services than
FHHs, and extension contact actually had a negative effect on FHHs' use of recommended
technology.

Analysis based on the Cobb-Douglas production function indicated that variation in gross value of
output per hectare associated with the factors of production was 72% and 82% in MHHs and
FHHs, respectively. The significant factors affecting gross value of output for MHHs were the
farmer's age, family labor, farm size, livestock units, and inorganic fertilizer. The significant factors
affecting gross value of output for FHHs were family labor, farm size, livestock units, inorganic
fertilizer, hired labor, and extension contact.

The marginal value product (MVP) of family labor compared to its price (wage rate) is higher in
MHHs and lower in FHHs, which indicates that MHHs could increase their productivity by using
more family labor. On the other hand, the MVP for farm size was lower than its factor price in
MHHs and higher in FHHs. Thus, FHHs could increase their productivity by cultivating more land.
The MVP for inorganic fertilizer was higher than its factor cost for both MHHs and FHHs, which
indicates that both types of household could increase their productivity by increasing the use of
inorganic fertilizer.

The gender difference in gross output was considerable. MHHs had a gross output of Birr 6,456/
ha, while FHHs had a gross output of Birr 4,776/ha. These differences can be explained partly by










the lower quantities of inputs used by FHHs. The use of average values of these inputs from
MHHs resulted in a gross output of Birr 6,541/ha for FHHs-1.3% higher than MHHs. This
suggests that no productivity differences would have existed between both households if they had
equal access to inputs.

In 1997, about 59% and 42% of MHHs and FHHs, respectively, grew wheat. Local wheat
varieties were grown by 70% and 86% of MHHs and FHHs, respectively. A significantly higher
proportion (t=5.7; p<0.05) of MHHs (30%) grew improved wheat varieties than FHHs (14%).
The logit model explains 84% and 89% of the total variation specified in the model for MHHs and
FHHs, respectively. The chi-square indicates that the parameters are significantly different from
zero at the 1% level for both households. In MHHs, extension services and farm size had a
positive effect on the adoption of improved wheat varieties, while radio ownership and farm size
increased the odds in favor of adoption of improved wheat varieties in FHHs.

On the whole, the study found that the MHHs and FHHs in the three woredas had differences in
endowments (land rights, education) and differential access to technologies, factors of production,
and support services. These differences had implications for the productivity levels and adoption
capacities of both types of households.

To address some of these differences, efforts should be made to provide credit and improve the
supply of preferred improved seed in time for planting. It is also recommended that technologies
should take into account the resource base of female farmers; extension services should be
targeted specifically to them; and the decision-making power of female household heads should be
harnessed by exposing them to different opportunities.








Gender Differentials in Agricultural Production and
Decision-Making Among Smallholders in Ada, Lume, and
Gimbichu Woredas of the Central Highlands of Ethiopia

Addis Tiruneh, Teklu Tesfaye, Wilfred Mwangi, and Hugo VerkuijI



1.0 INTRODUCTION

This study is an attempt to address the lack of concrete and statistical information on gender roles in
agricultural production and decision-making in the household economy of Ethiopia, focusing
primarily on the role of gender in resource ownership and decision-making power in the mixed
farming systems of Ada, Lume, and Gimbichu woredas in the central highlands.

Agriculture is a dominant sector in Ethiopia. It contributes 51% to the GDP, employs nearly 80% of
the total labor force and generates the bulk of foreign exchange. Smallholder farms are predominant
and account for more than 90% of agricultural production and over 95% of the total area under
cultivation (Table 1). However, given the poor performance of the agricultural sector vis-g-vis the
growing population estimated at over 55 million and growing at a rate of 3% per annum-the
intensification of agriculture is very critical. As a result, there has been an overall effort to increase
agricultural productivity to meet the growing food demand.

Much valuable research already exists on the different roles of women and men in various farm
activities and non-farm activities like food preparation, household maintenance, and childcare, and
there is now growing recognition that they often have very different rights and responsibilities with
respect to resource use (Adepoju and Oppong 1994; Bryceson 1995; Dey 1981; McSweeny 1979;
Whitehead 1985). The transfer and adoption of agricultural technologies is affected by who owns
productive resources and who decides what to produce, when to produce, and how much to
produce. Empirical evidence show that women do play a greater role than previously thought in
decision-making in agriculture and in the adoption of agricultural technologies. Information provided
through research is vital for policymakers to be informed of the basis on which decisions are made at
the micro-level.


Table 1. Distribution of household types by production system, land area, and crop production in Ethiopia, 1983
Production of major Production of
food grains all crops
Households Cultivated area (% of cultivated (% of cultivated
(No.) (%) (ha) (%) area) area)
Smallholders 8,206,000 98.7 5,987,000 94.7 95.1 94.4
Producer cooperatives 94,000 1.1 114,000 1.8 1.9 2.0
State farms 18,000 0.2 222,000 3.8 3.0 3.6
Source: Cohen and Isaksson (1988); cited in Franzel (1993).
a Excluding coffee.










Gender also constitutes an important factor in the growing trend of widening disparities in the
distribution of income and assets in many low-income countries, which reflects both the erosion of
traditional rights of access to resources and increasing population pressures. Female-headed
households (FHHs) typically have a much smaller asset base than male-headed households (MHHs),
and it is not coincidental that relative poverty in the sense of relative deprivation bears most heavily
on women (Dasgupta 1993; UNDP 1990).

While gender is culture neutral, interest in gender relations also derives from gender's explanatory
power as a primary organizing principle of society, including agricultural society. For this study, we
look at gender not as a means of categorizing household headship, but as a basic key to
understanding structures and actions, including production relationships within and across
households, goal setting and priorities, mobilization of resources, willingness to take risks, and the
decision-making process vis-g-vis the rights to benefits derived from increased farm production.

The specific objectives of the research were to:
* Look into the structure and patterns of production and composition of crops grown, farm sizes
and tenure arrangements, types of technology used, and distribution of output across producers
and end-users.
* Take stock of the different access of male and female farmers to resources-land, credit, and
technology.
* Estimate production functions using both conventional inputs like land, labor, capital, and
purchased inputs, and non-conventional inputs like education, extension, and infrastructure.
* Test the heterogeneity of labor-adult male labor, adult female labor, child labor, hired male labor,
hired female labor, and hired child labor.
* Compare differences in managerial efficiency and decision-making power between male and
female household heads.
* Determine how socioeconomic factors and infrastructure contribute to the adoption of
technology.


2.0 THE HOUSEHOLD SURVEY

2.1 The Study Area
Ada, Lume, and Gimbichu
woredas were selected as study
areas because their accessibility
facilitated the organization and
monitoring of the field study
(Figure 1). Ada woreda, about 40
km southeast of Addis Ababa,
covers 1,750 km2; two-thirds of
this area lies above 1,800 m
(Gryseels and Anderson 1983).










Much of the land in Ada is eroded and poorly drained. July and August are the wettest months of the
year and April and May are the hottest. The major soil type is vertisol and major crops are tef, wheat,
barley, faba beans, chickpeas, and lentils (Workneh Negatu 1989).

Lume woreda lies northeast of Debre Zeit at an altitude ranging from 1,700 to 2,100 m. July and
August are the wettest months and April, May, and June are the hottest. The major soil type is vertisol.
The major crops grown are tef, wheat, haricot beans, maize, chickpeas, barley, and faba beans.

Gimbichu woreda, at an average altitude of 2,450 m, borders Ada on the northern side of Debre Zeit.
July and August are, on average, the wettest months. The major soil type is vertisol and the major
crops grown are wheat, tef, chickpeas, and faba beans.

The male-female ratio in all three areas is almost 1:1 (Table 2).

2.2 Methodology
"Household," "gender," or other social constructs influence the way in which field research is
structured. Experience has shown that both "household" and "gender" have practical value as
conceptual frameworks around which to structure field research. An appropriate balance must,
however, be struck between adopting practical research methodologies and accurately describing the
complexity of African rural society (Warner, Hassan, and Kydd 1997).

Since the choice of research methodology depends on the goals of each particular research project, the
resources available, time constraints, and a host of other factors, it is not appropriate to advocate one
research approach over others. It is vital to chose methods that make it possible to elicit as much
information as possible about individual members of rural societies, rather than to choose methods
based on preconceptions of the significance of any one social construct (Warner, Hassan, and Kydd
1997).

Boserup was one of the first scholars to provide a comparative analysis of women's work based on data
from a wide range of societies. In her 1970 book, Women's Role in Economic Development, she
emphasized that in spite of sex-role stereotyping and cross-cultural regularities in the sexual division of
labor, women's work differs from society to society.

In much of the anthropological literature, "household" is the term used to refer to the basic unit of
society involved in production, reproduction, consumption, and socialization. The exact nature and


Table 2. Population size, density and area in Ada, Lume, and Gimbichu woredas, Ethiopia
Population
Woreda Male Female Total Density Area (km2)
Lume 56,849 53,518 110,367 155.5 709.85
Gimbichu 36,754 35,236 71,990 101.8 707.49
Ada 141,265 137,837 279,102 170.7 1,635.16
Source: FDRE, CSA. StatisticalAbstract, Ethiopia (1998).










function of the household clearly varies from culture to culture and in different periods, but the
anthropological definition usually rests on what the people themselves regard as the significant unit
of their society. It is important to recognize that although recruitment to households is often
through kinship and marriage, household units are not necessarily the same as family units. Leaving
aside the definitional difficulties, households are important in feminist analysis because they
organize a large part of women's domestic/reproductive labor. As a result, both the composition
and the organization of households have a direct impact on women's lives, and in particular on
their ability to gain access to resources, labor, and income (Moore 1988).

Female-headed and female-centered domestic groups have been identified in a wide range of
communities all over the world (Smith 1973; Tanner 1974), and FHHs have emerged in increasing
numbers in recent times. It is important to consider under what conditions-social, economic,
political, and ideological-FHHs become a significant proportion of the total number of
households. The evidence is complex, but it seems that FHHs are more common in situations of
urban poverty, in societies with a high level of male labor migration, and in situations where general
insecurity and vulnerability prevail (Youssef and Hefler 1983; Merrick and Schmink 1983).

In Africa, preliminary studies from rural areas and national sample survey data suggest that the
incidence of FHHs varies inversely with the economic potential of the area. Incidence tends to be
high in areas where agricultural productivity is low, either due to population pressure or unfavorable
ecological factors. For example, a study conducted by Chipande (1987) in the Lilongwe Land
Development Program, Malawi, an area of very high agricultural potential, showed that women
headed only 16% of the 160 households sampled. These households were largely composed of
older women (45 years and over) who were mostly widowed, divorced, single, or living alone or
with their unmarried children. The prevailing view in the literature is that this trend results from
male labor migration. It is clear that in some rural economies the strain placed on conjugal relations
by the exploitation of rural areas as labor reserves is producing an enormously high proportion of
FHHs (Murray 1981; Bush, Cliffe, and Jansen 1986).

In addition to male labor migration, there is evidence that increasing socioeconomic differentiation
in rural communities is producing FHHs (Cliffe 1978). Changes in kinship systems and in the
organization of agricultural production have meant that many poorer women have lost the security
provided by former kinship networks and relationships.

It is true that FHHs are very poor, but as Peters (1994) points out, this is not the case for all of
them, and we have to be very careful to avoid any analytical illusion: lack of males = FHHs =
marginal = poor. The situation is more complex and requires more research. For example, there is
evidence from Africa and other parts of the world that some women are choosing not to marry
(Allison 1986; Nelson 1978; Obbo 1980) and that significant numbers of married women are
choosing to live separately from their husbands (Bukh 1979). This trend is perhaps more a feature
of urban than rural life, but it highlights the dangers of easy generalization and reinforces the
importance of historically and socially grounded research.










When households headed by women are taken into account, total female participation in agriculture
is greater in all developing regions. It is important to consider the role of headship because available
data show that female headship is relatively high and increasing in many places. In sub-Saharan
Africa, for example, it is estimated that women head one-fourth of rural households; in some areas
they head almost half (Due and Gladwin 1991). In Central America, nearly 20% of rural households
are headed by women (Yudelman 1994), and in Bangladesh the proportion of FHHs rose from 5-7%
to 16% over 20 years (Mekra 1995).

Women become household heads as a result of death, divorce, separation, and, increasingly, male
migration, with the frequent result that more women participate in the labor force. For instance,
women's participation in agriculture in Tunisia more than doubled between 1970 and 1985 because
male migration from rural areas left women as household heads (United Nations 1989) who were
fully responsible for farm production and management. The type and variety of farm tasks performed
by women may also increase, as well as their role in decision-making. Staudt (1979) found this to be
the case in 40% of farms headed by women in two areas of western Kenya.

A multi-stage sampling method was used for this study to select male- and female-headed households.
Formal contacts were made with officials in Peasant Associations (PAs). Informal contacts were made
with farmers in different PAs in the three woredas to supplement secondary information gathered
prior to the actual survey. After collection of the secondary data and the preliminary survey, the
sample PAs were selected. Selection was based on total population, types of crops grown, altitude,
and distance from the center of the Ministry of Agriculture.

In each woreda, 60 households were selected from six centers of the Ministry of Agriculture. Out of
the three PAs, ten households were selected, among which three were headed by women. The list of
dejure female household heads was made purposively, but the samples were selected randomly
using the prepared list to make sure that the women selected were full-time farmers. In general, the
households were more or less homogeneous in types of crops grown, farming operations, and socio-
economic characteristics. Enumerators were recruited and trained to administer a formal
questionnaire. The survey was conducted after the cropping calendar in the three woredas, with close
supervision.

Of a sample of 180 households, 81 (45%) were headed by females. FHHs were defined as
households managed by a widowed, divorced, or a single woman without the mediation of a
husband, father, or male relative in the routine day-to-day activities of that household. MHHs were
those in which a husband was present and was the final decision-maker in the important issues
pertaining to the household (Starkey, Mweamyae amd Stares 1994).

Researchers developed the questionnaire in November 1995 and it was pre-tested in February 1996.
The questionnaire had four different and related parts designed to fit in the different agricultural
activities phase by phase. The information gathered during the pre-testing was used to modify the
questionnaire. Part one was pre-tested with assistance from development agents in the three PAs per
woreda. After the pre-testing, the questionnaire was modified to its final version.













3.0 SOCIOECONOMIC CHARACTERISTICS


3.1 Household Characteristics

Of thel80 households interviewed across the three woredas, 99 (55%) were headed by males and 81
(45%) were headed by females. Table 3 shows the socioeconomic characteristics of households in Ada,
Lume, and Gimbichu. On average, MHHs had larger households and more family members (sons,
daughters, relatives, and non-relatives') than FHHs in all three woredas. The average number of
daughters was significantly higher in MHHs than FHHs.This situation is similar in most developing
countries where FHHs have been found to be smaller than MHHs (Buvinic and Gupta 1997). The
average age of the head in MHHs was about 43, 47, and 51 years in Ada, Lume, and Gimbichu,
respectively; in FHHs it was about 46, 51, and 47 years.


Table 3. Socioeconomic characteristics of households in Ada, Lume, and Gimbichu woredas, Ethiopia

Ada Lume Gimbichu
MHH FHH MHH FHH MHH FHH

Characteristic (Respondents) (Respondents) (Respondents) (Respondents) (Respondents) (Respondents)


Household size (No.) 8.1
Sons 3.0
Daughters 2.7
Relatives 1.2
Non-relatives 1.4
Age of head (yr) 43.4
Education of head (%)
Illiterate 28.1 9
Literacy class 40.6 13
Primary school 28.1 9
Secondary school 3.1 1
Education of wife (%)
Illiterate 37.5 12
Literacy class 50.0 16
Primary school 9.4 3
Secondary school 3.1 1
Education of son (%)
Illiterate 61.8 21
Literacy class 59.1 13
Primary school 54.5 6
Secondary school 45.5 5
Education of daughter (%)
Illiterate 65.5 19
Literacy class 57.9 11
Primary school 40.0 2
Secondary class 54.4 3


50.0 14 17.1 6
46.4 13 48.6 17
28.6 10
3.6 1 5.7 2

47.1 16
47.1 16
5.9 2


38.2 13
40.9 9
45.5 5
54.5 6

34.5 10
42.1 8
60.0 3
55.6 5


58.3 14
57.1 12
81.8 9
45.5 5

69.6 16
36.4 4
80.0 4
20.0 1


70.8 17 48.5 16 85.2 23
29.2 7 30.3 10 11.1 3
12.1 4 3.7 1
9.1 3

48.4 15
48.4 15
3.2 1


41.7 10 100.0 31 100.0 29
42.9 9
18.2 2 -
54.5 6 -

30.4 7 100.0 31 100.0 29
63.6 7
20.0 1
80.0 4 -


100.0 32 100.0 28 100.0 35 100.0 24 100.0 33 96.3 26
- 3.7 1


Ethnic group (%)
Oromo 71.1 22
Amhara 29.0 9
Marital status of head (%)
Married 96.9 31
Single 3.1 1
Divorced -
Widowed


64.0 18 74.3 26 87.5 21 90.6 29 96.3 26
36.0 10 25.7 9 12.5 3 9.4 3 3.7 1


3.6 1
35.7 10
7.1 2
53.6 15


97.1 34
2.9 1


4.2 1 78.8 26
18.2 6
20.8 5 -
75.0 18 -


3.8 1
15.4 4
80.8 21


1 Non-relatives refer to those who live in the household but are not related to the husband or wife.


Religion (%)
Christian
Muslim










MHHs and FHHs in all three woredas showed significant difference in access to education and
literacy. This scenario is also found in some African countries. In Uganda, for instance, more than
half of the female household heads received no schooling compared to less than a quarter of their
male counterparts (Appleton 1996, Bisanda and Mwangi 1996). All farmers in MHHs and FHHs in
Ada and Lume were Christians; in Gimbichu 3.7% of FHHs were Muslims. The dominant ethnic
groups were Oromo and Amhara. The proportion of MHHs whose ethnic group is Oromo is
greater than that of Amhara in all the woredas. The marital status of household heads differed
widely between MHHs and FHHs.

3.2 Fertility Characteristics

The survey also gathered information on child bearing. About 66% of MHHs and 77% of FHHs in
Ada, 54.2% of MHHs and all FHHs in Lume, and 37.1% of MHHs and 81.8% of FHHs in
Gimbichu did not want more children. The average number of the children they wanted varied from
4.7 in MHHs and FHHs in Ada to 3.4 for MHHs in Lume, and 6.6 and 6.0 for MHHs and FHHs
in Gimbichu, respectively. Women in both types of household delivered at home with no service
from a health center because of poor access. In most MHHs and FHHs in all woredas, traditional
birth attendants, neighbors, or the women's mothers assisted them during delivery (Table 4).

The main reasons cited for having more children were the lack of male or female children, that
women could still reproduce, because husbands wanted more children, or that more children were
needed for work (Table 5). In Ada, 33.3% of MHHs said that the number of males in the household
was not enough. In Lume, the main reasons for wanting to have more children were shortage of
males (20.68%) and the wife could still reproduce (20.68%). In MHHs in Gimbichu, about 36.4%
reported that there were not enough males and 22.7% reported there were not enough females. In
FHHs in Ada, all women wanted more children because they could still reproduce. In Gimbichu,
60% wanted more children because of a shortage of females, while 40% wanted more children
because they could still reproduce. The main reason cited for not having more children in MHHs
was that they had enough children (about 76% in Ada, 58% in Lume, and 54% in Gimbichu).
About 65%, 46%, and 36% of FHHs in Ada, Lume, and Gimbichu did not have more children
because they could not reproduce anymore.

Table 4. Statistics on childbearing and reproductive goals in Ada, Lume, Gimbichu woredas, Ethiopia
Ada Lume Gimbichu
MHH FHH MHH FHH MHH FHH
Characteristic (Respondents) (Respondents) (Respondents) (Respondents) (Respondents) (Respondents)
Place of delivery (%)
At home 100.0 32 100.0 26 100.0 34 100.0 24 100.0 30 100.0 26
At health centers
No. of children wanted 4.66 4.66 3.37 6.57 6.0
Who assists? (%)
Mother 25.0 8 42.3 11 17.6 6 2.5 3 25.8 8 25.0 7
Traditional birth 30.8 8 38.2 13 37.5 9 51.6 16 50.0 14
attendant 34.4 11 7.7 2 19.4 6 17.9 5
Mother-in-law 3.1 1 19.2 5 44.1 15 50.0 12 3.2 1 7.1 2
Neighbor 37.5 12 3.3 1
Home worker











Table 5. Reasons for having or not having more children in Ada, Lume, Gimbichu woredas, Ethiopia

Ada Lume Gimbichu
MHH FHH MHH FHH MHH FHH
(%) (Respondents) (%) (Respondents) (%) (Respondents) (%) (Respondents) (%) (Respondents) (%) (Respondents)
Reason to have:
Not enough males 33.3 3 20.68 6 36.36 8 -
Not enough females 51.72 15 22.72 5 60.0 3
Can still reproduce 22.2 2 100.0 6 20.68 6 -- 13.63 3 40.0 2
Husband wants 22.2 2 9.09 2 -
more children
Needed for work 22.2 2 6.89 2 18.18 4

Reason not to have:
Have enough 76.2 16 15.0 3 57.9 11 16.7 4 53.8 7 40.9 9
children
For health reasons 5.0 1 15.8 3 -- 15.4 2 13.6 3
Can't reproduce
anymore 4.8 1 65.0 13 15.8 3 45.8 11 15.4 2 36.4 8
Expensive 19.0 4 5.0 10.5 2 25.0 6 15.4 2
Don't like to bear from
the other person 10.0 2 12.5 3 9.1 2



Across all woredas, girls born in MHHs had better access to education. In Ada and Lume, over 70%
of MHHs and over 60% of FHHs approved of sending girls to school. In Gimbichu, almost 60% and
26% of MHHs and FHHs respectively, approved of sending girls to school. The main reasons for not
sending girls to school were that girls were needed for housework; they were not interested in school;
the fear of forced marriage; the distance from school; and not having enough children to spare for
schooling (Table 6). About 71% of MHHs in Ada did not send girls to school because they were
needed for household work, while 80% of MHHs in Lume reported that the school was too far away.
About 46% of MHHs in Gimbichu did not have enough girls to spare for schooling.



Table 6. Reasons that household members in MHHs gave for not sending girls to school in Ada, Lume, Gimbichu
woredas, Ethiopia

Ada MHH Lume MHH Gimbichu MHH

Household member (%) (Respondents) (%) (Respondents) (%) (Respondents)

Head
Needed for housework 71.4 5 10.0 1 23.1 3
Not interested in school 14.3 1 10.0 1 23.1 3
Fear of forced marriage 14.3 1
School is too far away -80.0 8 23.1 3
Don't have enough 46.2 6

Wife
Needed for housework 71.4 5 10.0 1 23.1 3
Not interested in school 14.3 1 10.0 1 23.1 3
Fear of forced marriage 14.3 1 23.1 3
School is too far away -80.0 8 23.1 3
Don't have enough -- 46.2 6










About 3% of MHHs in Ada and Gimbichu and 34% of MHHs in Lume practiced family planning, while
none of the FHHs in Ada, 8.3% in Lume, and 3.7% in Gimbichu practiced family planning. Of those
farmers practicing family planning in Lume, all FHHs and 75% of MHHs used birth control pills. The
most important source of information for contraception was the family planning office (Table 7).

Table 8 shows numbers of boys and girls who were born, adopted, and died; age at first marriage and
pregnancy; age at last pregnancy; and the numbers of adopted girls and boys. On average, more girls
were born in MHHs than FHHs. About 1.5 boys and girls died in both types of households. The
average age at first marriage was higher for MHHs than FHHs. In all woredas, girls married when they
were teenagers, perhaps because of the prevailing culture of girls marrying while they are young. The
average age of heads at first marriage was 21.9, 22.5, and 22.1 in MHHs and 15.1, 15.3, and 16.8
in FHHs in Ada, Lume, and Gimbichu, respectively. The average age of first marriage of the wife was
15.9, 16.8, and 16.9 in Ada, Lume and Gimbichu, respectively. The average age at first pregnancy
was 23.5, 24.2, and 24.1 in MHHs and 17.9, 19.3, and 19.3 in FHHs in Ada, Lume, and Gimbichu,
respectively. The average age of the wife at first pregnancy was 17.7, 18.5, and 19 in Ada, Lume,
and Gimbichu, respectively. In contrast to the age at first marriage, female household heads were older
than wives at first pregnancy. The age at last pregnancy was higher in MHHs and lower in FHHs. The
average age of the wife at last pregnancy was 31.1 years in Ada, 34 years in Lume, and 38.6 years in
Gimbichu. Wives were younger than female heads of household at last pregnancy. In Ada, Lume, and
Gimbichu, 46.2%, 62.5%, and 57.7% of FHHs, respectively, did not have more children due to the
death of a husband.

The mean number of adopted boys was 2.0 for Ada, 1.0 for Lume, and 0.25 for Gimbichu in MHHs.
In FHHs, it was 1.2, 2.0, and 1.5 for Ada, Lume, and Gimbichu, respectively. The mean number of
girls adopted in MHHs was 1.5 for Ada and 1.7 for Gimbichu, and 1.25, 1.2, and 1.5 for FHHs in
Ada, Lume, and Gimbichu, respectively.

Table 9 presents information for MHHs and FHHs on who determined the number of children and
decided whether to send girls to school.



Table 7. Method and source of information for family planning in Ada, Lume, Gimbichu woredas, Ethiopia
Ada Lume Gimbichu
MHH FHH MHH FHH MHH FHH
(%) (Respondents) (%) (Respondents) (%) (Respondents) (%) (Respondents) (%) (Respondents) (%) (Respondents)
Family planning 3.1 1 34.3 12 8.3 2 3.0 1 3.7 1
Method:
Pills 100.0 1 75.0 9 100.0 2
Traditional method 25.0 3 100.0 1 100.0 1
Source of information:
Family planning 100.0 1 91.7 11 100.0 2 -
Mass media 8.3 1
Friends and relatives 100.0 1 100.0 1












Table 8. Childbearing and adoption data in Ada, Lume, Gimbichu woredas, Ethiopia

Ada Lume Gimbichu

MHH FHH MHH FHH MHH FHH


No. of girls born
Head
Wife
Daughter
Relative
No. of boys born
Head
Wife
Daughter
Relative
No. of boys died
Head
Wife
Daughter
Relative
No. of girls died
Head
Wife
Daughter
Relative
Age at first marriage
Head
Wife
Daughter
Relative
Age at first pregnancy
Head
Wife
Daughter
Relative
Age at last pregnancy
Head
Wife
Daughter
Relative
No. of foster/adopted children


1.00


Table 9. Decision-making on the number of children and sending girls to school in Ada, Lume, Gimbichu
woredas, Ethiopia


Lume


Gimbichu


MHH FHH MHH FHH MHH FHH

Who decides on: (%) (Respondents) (%) (Respondents) (%) (Respondents) (%) (Respondents) (%) (Respondents) (%) (Respondents)

No. of children
Husband 21.9 7 3.8 1 5.9 2 10.3 3 3.8 1
Wife 15.4 4 2.9 1 29.2 7 3.4 1
Both 65.6 21 34.6 9 64.7 22 8.3 2 51.7 15 38.5 10
Nature 12.5 4 46.2 12 26.5 9 62.5 15 34.5 10 57.7 15

Sending girls to school
Head 65.4 17 62.5 15 59.4 19 26.9 7
Joint 78.1 25 71.4 25 55.2 16










3.3 Household Amenities


The characteristics of household dwellings in Ada, Lume, and Gimbichu are shown in Table 10. The
"simple family home" is a single hut where all family members live together, while "several family
homes" are a number of huts in which household members live. Household heads own all dwellings. In
Ada, about 66% of MHHs and 44% of FHHs lived in a simple family home. The proportion of MHHs
that lived as a single family in several huts and the proportion living in several family homes was roughly
similar and lower than the proportion of FHHs that had the same living arrangements. In Lume, 60% of
MHHs and 37.5% of FHHs lived in a single-family home; about 34% of MHHs and 50% of FHHs lived
as a single family in several huts, and 6% of MHHs and 13% of FHHs lived in several family homes. In
Gimbichu, about 70% of MHHs and 81.5% of FHHs lived in a simple family home.

Aside from gathering information on housing arrangements, the sample survey also gathered
information on household wealth indicators. The roof of a dwelling is an indicator of wealth: an
aluminum sheet roof indicates more wealth than a grass roof. MHHs and FHHs differed little with
respect to this wealth indicator. In Ada and Lume, 65% of both types of households had grass roofs. In
Gimbichu, about 71% of MHHs and 77% of FHHs had houses with thatch roofs, while 29.3% of
MHHs and about 23% of FHHs had houses roofed with aluminum.

Mud walls (made from mud, straw, and wood) and cement walls are also wealth indicators. In Ada, all
MHHs and 87.1% of FHHs had houses with mud walls whereas about 13% of FHHs had houses with
cement walls. In Lume, both types of households had mud walls. In Gimbichu, about 94% of MHHs
and all FHHs had houses with mud walls and only 6.1% of MHHs had houses with cement walls.

Table 11 presents additional data on household amenities in the three woredas. In Ada, 21.7% of
MHHs and 10% of FHHs owned a radio, while 1.7% of both households owned a tape recorder.
None of the households owned a TV, gas stove, or refrigerator. In Lume, about 29% of MHHs and
14% of FHHs owned a radio, while only 10.2% and 1.7% of MHHs owned a tape recorder and
bicycle, respectively. About 5% of MHHs and 3% of FHHs owned a gas stove. In Gimbichu, 20% of
MHHs and 8.3% of FHHs owned a radio, while 10% of MHHs and no FHHs owned a tape recorder.
Only one MHH owned a gas stove.


Table 10. Types of dwelling in Ada, Lume, and Gimbichu woredas, Ethiopia
Ada Lume Gimbichu
MHH FHH MHH FHH MHH FHH
(%) (Respondents) (%) (Respondents) (%) (Respondents) (%) (Respondents) (%) (Respondents) (%) (Respondents)
Type of dwelling
Simple family home 65.5 21 44.4 12 60.0 21 37.5 9 69.7 23 81.5 22
Single family
in several huts 18.8 6 33.3 9 34.3 12 50.0 12 18.2 6 11.1 3
Several family home 15.6 5 22.2 6 5.7 2 12.5 3 12.1 4 7.4 2
Type of roof
Grass thatch 65 26 64.5 22 64.5 20 64.5 20 70.7 29 77.4 24
Aluminum sheet 35 20 35.5 19 35.5 11 35.5 11 29.3 12 22.6 7
Type of wall
Mud 100.0 32 87.1 27 100.0 32 100.0 28 93.9 31 100.0 27
Cement 12.9 4 6.1 2











Table 11. Household amenities of farmers in Ada, Lume, and Gimbichu woredas, Ethiopia
Ada Lume Gimbichu
MHH FHH MHH FHH MHH FHH
(%) (Respondents) (%) (Respondents) (%) (Respondents) (%) (Respondents) (%) (Respondents) (%) (Respondents)
Radio 28.8 17 13.6 8 21.7 13 10.0 6 20.0 12 8.3 5
Radio tape recorder 10.2 6 1.7 1 1.7 1 10.0 6 -
Bicycle 1.7 1
Gas stove 5.1 3 3.4 2 1.7 1
Type of toilet
Pit latrine 5.7 2 3.1 1 3.7 1 -
Open community 25.7 9 25.0 6
None 68.6 24 75.0 25 96.9 31 96.3 26 100.0 33 100.0 27
Source of lighting
Electricity 29.0 1 6.3 2
Tile lamp 37.1 1.3 20.8 5 18.5 5 5.9 2
Oil lamp 60.0 21 79.2 19 93.7 30 81.5 22 94.1 32 100.0 27
Type of fuel
Firewood 62.8 22 60.0 15 44.6 25 46.9 23 62.9 22 53.5 15
Cow dung 37.2 13 36.0 9 55.4 31 51.0 25 37.1 13 42.9 12
Kerosene stove 4.0 1 2.1 1 3.6 1
Source of firewood
Collection 72.0 18 87.0 20 50.0 11 71.4 10 27.3 6 13.3 2
Purchase 4.0 1 8.7 2 9.1 2 7.1 10 27.3 6 33.3 5
Own 24.0 6 4.3 1 40.9 9 21.4 3 45.5 10 53.3 8
Source of water
Collection 90.6 29 96.3 26 39.5 15 55.2 16 94.0 31 96.3 26
Water vendor 9.4 3 3.7 1 28.9 11 34.5 10 3.0 1 3.7 1
Own 31.6 12 10.3 3 3.0 1



Most households did not use any type of toilet. In Ada, 3.1% of MHHs and 3.7% of FHHs had a
pit latrine, while in Lume about 6% of MHHs had a pit latrine. About 25% of both types of
households in Lume used open community areas. Most households in all woredas used oil lamps
for lighting. In Ada and Lume, a very small percentage of MHHs and none of the FHHs had
electricity. About 37% and 21% of the MHHs and FHHs in Lume used a tile lamp, respectively,
while 18.5% of FHHs in Ada and 6% of MHHs in
Gimbichu used a tile lamp. Table 12. Number of farm implements owned by MHHs
and FHHs in Ada, Lume, and Gimbichu woredas, Ethiopia


Most farmers used wood or cow dung for fuel. Very
few households in all woredas used a kerosene
stove. Most farmers collected water from rivers,
ponds, and other natural sources of water, although
some purchased it from the water vendor.

3.4 Farm Implements

The number of farm implements owned by both
types of households across all the woredas was low
(Table 12). All households used ox plows for
cultivation and sickles for harvesting. For


Ada Lume Gimbichu
Implement MHH FHH MHH FHH MHH FHH

Hoe 1.3 1.0 1.0 1.5 1.4 1.1
Shovel 1.1 1.0 1.0 1.0 2.6 1.8
Spade 1.0 1.0 1.0 1.2 1.0
Digging implement 1.1 1.1 1.1 1.0 1.3 1.0
Sickle 2.2 1.9 1.4 1.0 2.3 1.6
Sprayer 1.0 1.0 1.0
Plow 2.6 1.7 1.2 1.1 2.2 1.6
Mensha 2.3 1.6 1.3 1.1 1.5 1.1
Lydaa 1.3 1.2 1.1 1.0 1.4 1.3
Axe 1.7 1.4 1.1 1.0 1.2 1.0

a Amharic words for implements that farmers use for winnowing
their produce while threshing.










winnowing, they used the mensh and lyda. In Ada, the average number of ox plows owned by
MHHs (2.6) was significantly higher than FHHs (1.7) (t=2.9; p<0.01); in Lume and Gimbichu, FHHs
had fewer ox plows but the difference was not significant. Because MHHs generally had more ox
plows than FHHs, they were better able to prepare their land on time. In Lume, the average number
of sickles owned by MHHs was significantly higher than FHHs (t=2.5; p<0.05). Similarly, the
average number of hoes, spades and digging implements owned by MHHs was higher than FHHs
(t=2.2, p<0.05; t=2.6; p<0.05; and t=2.2, p<=0.05, respectively).,



4.0 GENDER DIFFERENTIALS IN FARM MANAGEMENT PRACTICES


4.1 Crop Production

Agriculture in the study area is based on the cultivation of small grain cereals (tef and wheat), with the
exception of lowlands in Lume, where farmers grew maize and haricot beans (Table 13). Depending
on the type of crops grown, most farmers used seedbed preparation techniques such as flatbeds,
ridge and furrow, and broad bed and furrow during land preparation. The majority of households in
all woredas used ox plows to form ridges and furrows.

In Ada and Lume, tef is the most important crop, followed by wheat and highland pulses; in
Gimbichu, wheat is the most important crop, followed by tef and highland pulses. Wheat and tef
seed and highland pulses were broadcasted whereas maize was usually planted in rows. All MHHs
and FHHs reported broadcasting small grain cereals. Very few MHHs (2.9%) and FHHs (8.3%) that
grew maize in Lume broadcasted.


Households weed the main crops by hand or apply herbicide. In Ada, 9.1% of households used
herbicides, while 27.3% of MHHs and 36.4 % of FHHs weeded by hand. The other households did
not weed. About 63.3% of the MHHs and 54.6 % of FHHs used both herbicides and hand weeding;


Table 13. Mean crop area cultivated (in kerta) in Ada, Lume, and Gimbichu woredas, Ethiopia

Ada Lume Gimbichu

MHH FHH MHH FHH MHH FHH

Crop (in kert) (Respondents) (Respondents) (Respondents) (Respondents) (Respondents) (Respondents)

Tef 5.43 32 4.74 27 7.64 34 6.47 24 2.79 29 2.35 24
Wheat 1.72 25 1.33 21 2.07 31 1.20 18 5.92 33 4.92 26
Chickpeas 0.88 18 0.94 13 1.32 14 1.50 10 2.00 11 1.85 7
Lentils 1.00 2 0.60 5 0.60 5 3.19 31 2.86 25
Barley 1.62 12 1.08 12 1.00 31 1.00 21 1.00 1
Faba beans 1.33 6 1.00 2 0.64 7 0.43 4 0.56 2 1.00 1
Haricot beans 0.50 3 0.50 1 1.14 21 1.00 10
Horticultural crops 0.30 1 0.81 4
Other crops 0.75 4 0.37 2

a A kert is one-fourth of a hectare.










hand weeding, supplemented by herbicides, was used particularly to control broadleaf weeds. In
Lume and Gimbichu, most MHHs (98.2%) and all FHHs weeded by hand. Only about 2% of the
MHHs used herbicides. MHHs and FHHs in all woredas harvested crops manually. Harvesting was
usually done by hired labor because labor was scarce at that time.

The major crops grown in the study area were tef, wheat, chickpeas, lentils, barley, faba beans, field
peas, haricot beans, maize, rough peas, and horticultural crops, mainly tomatoes (Table 13). MHHs
in Ada had higher average area of tef, wheat, barley, and faba beans than FHHs. In MHHs in Lume,
the average area under tef (7.6 kert) and wheat (2.1 kert) was the highest, followed by chickpeas
(1.3 kert), haricot beans (1.1 kert), and barley (1.0 kert). The FHHs in Lume grew an average of 6.5
kert of tef, chickpeas (1.5 kert), wheat (1.2 kert), barley (1.0 kert), and haricot beans (1.0 kert).
Lume is different from the other woredas because farmers grew horticultural crops probably because
they have irrigation. In Gimbichu, wheat covers the largest area (5.9 kert and 4.9 kert for MHHs and
FHHs, respectively), followed by lentils and tef.

Households were questioned about their use of local and improved varieties (Table 14). In Lume,
about 53% of MHHs and 58% of FHHs plant both local and improved tef varieties. In Ada, all
farmers grew local tef varieties, while in Gimbichu all FHHs and 93.1% of MHHs planted local
varieties. The use of wheat varieties followed a similar pattern, in which more farmers grew local
varieties. FHHs grew fewer improved wheat varieties. All farmers in Ada and Gimbichu grew local
chickpea varieties, while 21% of MHHs and 20% of FHHs in Lume grew improved chickpea
varieties. Very few farmers planted improved varieties of lentils, barley, faba beans, and rough peas.
In Lume, households grew improved varieties of field peas, maize, and haricot beans. Less than one-
third of MHHs and FHHs grew improved field pea varieties and only 13% of MHHs and 11% of
FHHs grew improved maize varieties. Only 10% of the MHHs grew improved haricot bean varieties.

Farmers' sources of information about improved varieties are shown in Table 15. In Lume, where
farmers grew more improved varieties than farmers in the other woredas, extension agents were the
main source of information (47% of MHHs and 21% of FHHs). About 53% of FHHs and 24% of
MHHs learned about new varieties from the market. Other farmers were another source of
information on improved varieties (29% of MHHs and 26% of FHHs). The different responses
reflected the bias of extension services in favor of MHHs.

Farmers reported that their choice of crops depends on their profitability and their use as food.
Some farmers cited good soil as a reason to grow certain crops (Table 16). Tef, wheat, barley,
maize, haricot beans, and horticultural crops were grown for food and also because they were
profitable. Chickpeas, lentil, faba beans, rough peas, and field peas were grown for similar reasons,
although their compatibility with soils was a third important reason.













Table 14. Use of local and improved varieties in Ada, Lume, and Gimbichu woredas,Ethiopia

Ada Lume Gimbichu

MHH FHH MHH FHH MHH FHH

Crop/type of variety (%) (Respondents) (%) (Respondents) (%) (Respondents) (%) (Respondents) (%) (Respondents) (%) (Respondents)

Tef
Local 100.0 32 100.0 27 44.1 15 41.7 10 93.1 27 100.0 24
Improved 2.9 1 6.9 2
Both 52.9 18 58.3 14

Wheat
Local 84.0 21 95.2 20 38.7 12 72.2 13 87.9 29 88.5 23
Improved 12.0 3 4.8 1 38.7 12 22.2 4 3.0 1
Both 4.0 1 22.6 7 5.6 1 9.1 3 11.5 3

Chickpeas
Local 100.0 18 100.0 13 78.6 11 80.0 8 100.0 10 100.0 7
Improved 21.4 3 20.0 2 -

Lentils
Local 100.0 2 100.0 5 80.0 4 100.0 32 100.0 25
Improved 20.0 1

Barley
Local 100.0 12 100.0 13 100.0 31 100.0 21 100.0 1
Improved

Faba beans
Local 100.0 14 100.0 1 100.0 6 100.0 4 100.0 2 100.0 2
Improved

Field peas
Local 100.0 14 100.0 18 62.5 10 68.8 11 100.0 5 100.0 7
Improved 25.0 4 31.3 5 -
Both 12.5 2 -

Rough peas
Local 100.0 12 100.0 7 100.0 3 100.0 1 100.0 7 100.0 5
Improved

Maize
Local 100.0 1 87.1 27 89.5 17
Improved 12.9 4 10.5 2

Haricot beans
Local 100.0 3 100.0 1 85.7 18 100.0 2 -
Improved 9.5 2 -
Both 4.8 1


Table 15. Farmers' sources of information about improved varieties in Ada, Lume, and Gimbichu woredas, Ethiopia

Ada Lume Gimbichu

MHH FHH MHH FHH MHH FHH
Source
of information (%) (Respondents) (%) (Respondents) (%) (Respondents) (%) (Respondents) (%) (Respondents) (%) (Respondents)

Extension 100.0 4 100.0 1 47.3 18 21.0 4
Other farmers 28.9 11 26.3 5
Market 23.6 9 52.6 10













Table 16. Farmers' reasons for growing crops in Ada, Lume, and Gimbichu woredas, Ethiopia

Ada Lume Gimbichu

MHH FHH MHH FHH MHH FHH

Reason (%) (Respondents) (%) (Respondents) (%) (Respondents) (%) (Respondents) (%) (Respondents) (%) (Respondents)


42.8 18 43.3 26 42.5 17 28.9 11 30.3 10
57.1 24 56.6 34 57.5 23 71.0 27 69.6 23


38.4 15 36.3 12
61.5 24 63.6 21
3.0 1


38.0 19 22.7 5 31.1 14 34.2 12
62.0 31 77.2 17 68.8 31 65.7 23


13.3 2
26.6 4
60.0 9


42.8 3
42.8 3
14.2 1


16.6 2 33.3 4 33.3 3
50.0 6 41.6 5 55.5 5
33.3 4 25.0 3 11.1 1


30.2 13 25.0 4
60.0 3 60.4 26 56.3 9
40.0 2 9.3 4 18.8 3


Profitable
Food crop

Wheat
Profitable
Food crop
Good soil

Chickpeas
Profitable
Food crop
Good soil

Lentils
Profitable
Food crop
Good soil

Barley
Profitable
Food crop

Faba beans
Profitable
Food crop
Good soil

Rough peas
Profitable
Food crop
Good soil

Maize
Profitable
Food crop

Haricot beans
Profitable
Food crop
Good soil


100.0 5 100.0 1



20.0 3 27.2 3


18.1 2
54.5 6
27.2 3


25.0 1
75.0 3


50.0 1

100.0 2 50.0


25.0 2 12.5


54.5 6 100.0 1 100.0 1 50.0 4 50.0 1
18.1 2 100.0 1 100.0 1 25.0 2 37.5 3


44.0 22 37.0 10
56.0 28 62.9 17


100.0 3 100.0 1


91.3 21 60.0 6
8.69 2 40.0 4


Horticultural crops
Profitable
Food crop

Field peas
Profitable
Food crop
Good soil


32.0 8
52.0 13
16.0 4


28.0 7
44.0 11
28.0 7


50.0 1
50.0 1


25.0 6
58.3 14
16.6 4


100.0 4



15.7 3
57.8 11 50.0 2 25.0 1
26.3 5 50.0 2 75.0 3


36.9 17
63.0 29


18.8 3
56.2 9
25.0 4


9.5 2
71.4 15
19.0 4



50.0 1
50.0 1


45.0 9
55.0 11


45.4 5 36.1 17 20.8 5
68.8 11 63.8 30 79.1 19 100.0 1


60.0 9
20.0 3












Table 17 shows the average quantities of local and improved seed varieties used in the study area. In
Ada, the average quantity of improved wheat seed was 56 kg for MHHs and 50 kg for FHHs, while
the average quantity of local wheat seed was 80 kg for MHHs and 60 kg for FHHs. In Gimbichu,
MHHs (181 kg) and FHHs (88 kg) used less improved wheat seed than local wheat seed (282 kg and
230 kg for MHHs and FHHs, respectively). In Lume, however, MHHs (81 kg) and FHHs (56 kg) used
more improved wheat seed on average than local wheat seed (61 kg for MHHs and 42 kg for FHHs).
Generally, FHHs used less local and improved seed of any crop (wheat, tef, barley, maize, faba beans,
rough peas, field peas, and haricot beans) than MHHs. FHHs used slightly more local chickpea
varieties than MHHs in all woredas.


Farmers commonly transport their produce from the field to the home and market using animals they
own, hire, or borrow; hired lorries; owned or hired carts; or owned vehicles (Table 18). In Ada, most
MHHs (89.3%) and FHHs (69.2%) used their own animals (donkeys and mules) to transport farm


Table 17. Average quantity of improved and local seed varieties used in Ada, Lume, and Gimbichu woredas, Ethiopia
Ada Lume Gimbichu
MHH FHH MHH FHH MHH FHH
(Respondents) (Respondents) (Respondents) (Respondents) Respondents) (Respondents)
Improved variety (kg/ha)
Barley
Chickpeas 37.0 3 34.0 2
Faba beans
Field peas 46.0 5 31.0 5
Haricot beans 20.5 4 30.0 1
Lentils 8.0 1 -
Maize 35.0 5 33.0 2
Rough peas
Tef 84.1 18 75.9 14 30.0 12 100.0 1
Wheat 56.3 4 50.0 80.4 19 56.0 5 181.3 4 87.5 2
Local variety (kg/ha)
Barley 80.8 12 51.7 12 48.7 31 45.0 21 30.0 1
Chickpeas 27.6 17 29.6 13 35.3 11 36.6 8 70.5 10 81.4 7
Faba beans 35.0 5 37.5 2 25.8 6 13.8 4 33.0 2 22.5 2
Field peas 94.6 14 39.7 18 46.7 13 36.3 11 66.3 4 56.7 6
Haricot beans 19.0 3 15.0 1 26.9 18 20.5 8 -
Lentils 12.5 2 6.4 5 5.5 4 104.7 32 100.8 25
Maize 20.0 1 24.6 25 21.2 17 -
Rough peas 32.7 11 22.4 7 21.7 3 10.0 1 40.0 7 27.5 6
Tef 84.2 31 74.8 27 68.1 33 56.9 24 47.7 27 36.9 24
Wheat 80.2 21 60.3 20 60.8 19 42.6 14 281.7 32 230.0 26



Table 18. Mode of transport of farm products in Ada, Lume, and Gimbichu woredas, Ethiopia
Ada Lume Gimbichu
MHH FHH MHH FHH MHH FHH
Mode of transport (%) (Respondents) (%) (Respondents) (%) (Respondents) (%) (Respondents) (%) (Respondents) (%) (Respondents)
Owned animal 89.3 25 69.2 18 28.6 10 37.5 9 20.0 1 27.3 6
Hired animal 3.8 1 2.9 1 40.0 2 36.4 8
Hired pick-up (lorry) 7.1 2 15.4 4 45.7 16 37.5 9 -
Own cart 3.8 1
Hired cart 3.6 1 3.8 1 4.2 1
Own lorry 3.8 120.0 7.0 16.7 4.0
Borrowed animal 2.9 1 1.7 1 40.0 2 36.4 8










produce. In Lume, most farmers transported produce with a hired lorry (46% of MHHs and 38% of
FHHs) or with their own lorry (20% of MHHs and 17% of FHHs). In Gimbichu, most farmers hired
or borrowed animals to transport farm produce.

4.1.1 Use of fertilizer and insecticides
Farmers in all the woredas used fertilizer (Table 19). About 38%, 66%, and 21% of MHHs in Ada,
Lume, and Gimbichu, respectively, used insecticides compared to about 44%, 50%, and 4% of FHHs
in Ada, Lume, and Gimbichu. In Ada and Lume, insecticide was not used because there was no pest
problem (50% of MHHs and 47% of FHHs in Ada and 42% of MHHs and 25% of FHHs in Lume).
In Gimbichu, the main reason farmers gave for not applying insecticide was lack of knowledge of its
application (65% of MHHs and 73% of FHHs). Other reasons included lack of cash, infrequent
supply of insecticides, and that insecticides were sometimes not effective to control pests.

4.1.2 Major crop production problems
The main crop production problems mentioned by sample households were the lack of land, shortage of
family household labor, high price of inputs, lack of loans from formal and informal sources, poor access
to markets, shortage of appropriate storage facilities, and lack of extension services (Table 20).

Shortage of land was the main constraint for 64% of MHHs and 63% of FHHs in Ada, 91% of
MHHs and 63% of FHHs in Lume, and 61% of MHHs and 50% of FHHs in Gimbichu. Other
constraints included poor quality of soils and erosion. Most farmers in Ada (82% and 90% of MHHs
and FHHs, respectively) and Lume (67% of MHHs and FHHs) reported a shortage of family
household labor as their main labor constraint. In Gimbichu, the main constraint was the high cost of
hired labor (59% of MHHs and 50% of FHHs). Only a few farmers had problems obtaining fertilizer.

The main constraint with agrochemicals in Ada and Lume was their high price; in Gimbichu, it was
the shortage of chemicals. In Ada, about 44% of both types of households reported the lack of
informal credit as the main credit constraint, while farmers in Lume reported the lack of formal credit
as their main credit constraint (72% of MHHs and 44% of FHHs). Few farmers in Gimbichu had


Table 19. Fertilizer and insecticide use and farmers' reasons for not using insecticides in Ada, Lume, and Gimbichu
woredas, Ethiopia

Ada Lume Gimbichu
MHH FHH MHH FHH MHH FHH
Input (%) (Respondents) (%) (Respondents) (%) (Respondents) (%) (Respondents) (%) (Respondents) (%) (Respondents)
Fertilizer 100.0 32 100.0 27 100.0 35 100.0 24 100.0 33 100.0 27
Insecticide 37.5 12 44.4 12 65.7 23 50.0 12 21.2 7 3.7 1
Reasons for not using insecticide
Cash shortage 15.0 3 6.7 1 16.7 2 25.2 3 7.7 2 7.7 2
No pest problem 50.0 10 46.7 7 41.7 5 25.0 3 26.9 7 19.2 5
No knowledge of use 20.0 4 40.0 6 8.3 1 25.0 3 65.4 17 73.1 19
Infrequent supply 15.0 3 6.7 1 25.0 3 25.0 3 -
Do not avoid pest 8.3 1
<>













Table 20. Crop production constraints in Ada, Lume, and Gimbichu woredas, Ethiopia

Ada Lume Gimbichu

MHH FHH MHH FHH MHH FHH

Constraint (%) (Respondents) (%) (Respondents) (%) (Respondents) (%) (Respondents) (%) (Respondents) (%) (Respondents)

Land
Unavailability 63.6 7 62.5 5 90.9 20 63.2 12 61.1 11 50.0 9
Poor quality of land 9.1 2 15.8 3 33.3 6 33.3 6
Serious erosion 36.4 4 37.5 3 21.0 4 5.6 1 16.7 3

Labor
Not enough family
labor 81.8 9 90.0 9 66.7 6 66.7 8 41.2 7 35.7 5
Hired labor not
available 18.2 2 11.1 1 14.3 2
Hired labor
expensive 10.0 1 22.2 2 33.3 4 58.8 10 50.0 7

Fertilizer
Untimely delivery 33.3 1 100.0 1 100.0 1 -
Infrequent
availability 66.7 2

Chemicals
Chemicals not
available 7.1 2 8.3 2 30.0 3 10.0 1 50.0 6 60.0 12
Price too high 89.6 25 91.7 22 40.0 4 60.0 6 8.3 1 35.0 7
Infrequent supply 3.6 1 30.0 3 30.0 3 14.7 5 15.0 1

Credit
Don't have required
collateral 6.3 1 100.0 3 75.0 3
Bank loans not
available 72.2 13 44.4 4 -
Informal loans not
available 43.8 7 44.4 4 5.6 1
Unfavorable
repayment terms 6.3 1 11.1 1 16.7 3 44.4 4 -
High interest rates 31.3 5 22.2 2 5.6 1 11.1 1
Not aware about
credit 12.5 2 22.2 2 25.0

Market
Market centers are
too far 88.9 16 78.3 18 50.0 7 66.7 6 -
High transportation
cost 11.1 2 17.4 4 22.2 2
Poor farm roads 4.3 1 7.1 1
Price instability 42.9 6 11.1 1 100.0 3 100.0 1

Storage
No appropriate
storage
facilities 100.0 1 100.0 5 100.0 6 20.0 1
Available storage
space not suitable 80.0 4

Extension
None 9.4 3 11.1 3 37.1 13 12.5 3 36.4 12 14.8 4
Extension service
unavailable 71.9 23 81.5 22 25.7 9 37.5 9 45.5 15 66.7 18
Extension not
appropriate 3.1 1 3.7 1 14.3 5 16.7 4 3.0 1 3.7 1
Extension services
nottimely 3.1 1 5.7 2 4.2 1 12.1 4 11.1 3
Infrequent visits 18.75 6 7.4 3 17.1 6 29.2 7 3.0 1 3.7 1












credit problems. The main marketing problem was the distance between the farm and the market.
About 89% of MHHs and 78% of FHHs in Ada, and 50% of MHHs and 67% of FHHs in Lume,
reported this as their main problem, while all farmers in Gimbichu reported price instability as a
marketing constraint. Inappropriate storage structures were primarily a problem for farmers in Lume.
More FHHs than MHHs said they suffered from a lack of extension services.


4.1.3 Decision-Making in Crop Production
Table 21 shows the decision-making pattern of households for the sale and consumption of agricultural
production by woreda. Both husbands and wives made decisions on consumption in 68%, 80%, and
63% of MHHs in Ada, Lume, and Gimbichu, respectively. Otherwise, the head made the decision. In
FHHs the head of the household usually made the decision (88% in Ada, 100% in Lume, and 96% in
Gimbichu). Sons participated in decision-making in 11% of FHHs in Ada and 3.7% in Gimbichu.


About 60% of heads in MHHs in Lume made the decision to sell the produce. The head and wife in
69% of MHHs in Ada and 58% of MHHs in Gimbichu made this decision jointly. About 89% of


Table 21. Decision-making of households on sale and consumption of agricultural produce in Ada, Lume, and
Gimbichu woredas, Ethiopia

Ada Lume Gimbichu

MHH FHH MHH FHH MHH FHH

Constraint (%) (Respondents) (%) (Respondents) (%) (Respondents) (%) (Respondents) (%) (Respondents) (%)(Respondents)

Who decides how much to consume?
Head 31.3 10 88.9 24 20.0 7 100.0 24 36.4 12 96.3 26
Head and son 11.1 3 3.7 1
Head and wife 68.8 22 80.0 28 63.6 21 -
Who decides how much to sell?
Head 31.3 10 88.9 24 60.0 21 100.0 24 42.4 14 92.3 24
Head and son 11.1 3 7.7 2
Head and wife 68.8 22 40.0 14 57.6 19
Who decides how to use stored produce?
Head 31.3 10 92.6 25 80.0 28 66.7 16 -
Wife 80.0 4 -
Son 7.4 2 20.0 1
Head and son 2.9 1 33.3 8 100.0 1
Head and wife 68.8 22
Non-relative
Who actually sells?
Head 93.8 30 55.6 15 97.1 34 75.0 18 72.7 24 42.3 11
Son 29.6 8 12.5 3 3.0 1 23.1 6
Head and son 7.4 2 12.5 3 6.1 2 11.5 3
Head and wife 6.3 2 2.9 15.2 5 -
Relative 3.7 1 15.4 4
Non relative 3.0 1 7.7 2
Place of sale
Local market 100.0 32 100.0 27 100.0 34 100.0 24 100.0 33 100.0 26
Farm gate & service 2.9 1
Who keeps revenues?
Head 34.4 11 92.6 25 100.0 32 39.4 13 100.0 26
Son 100.0 1 -
Wife 24.2 8 -
Husband and wife 65.6 21 36.4 12
Head and son 7.4 2










female heads of households in Ada, 100% in Lume, and 92% in Gimbichu decided how much
produce to sell. In MHHs in Ada, the decision to utilize stored produce was made jointly (69%),
while in FHHs, heads almost always made the decision (93%). In Lume, it was mostly the head of
the household who made this decision (about 80% and 67% of the MHHs and FHHs, respectively).

In MHHs, the head was primarily responsible for selling the produce, while in FHHs, the head and
her son played an important role. About 94%, 97%, and 73% of males heads of household in Ada,
Lume, and Gimbichu, respectively, were responsible for selling the produce. The head was
responsible in around 56%, 75%, and 42% of FHHs in Ada, Lume, and Gimbichu, respectively; in
37%, 25%, and 35% of FHHs, the son alone or the head and the son were responsible. Most
produce was sold at the local market. In 65.6% of MHHs in Ada and 36.4% of MHHs in Gimbichu,
husbands and wives kept the revenue from the sale of produce. In MHHs in Lume the head kept the
revenues. Household heads usually kept the revenue in FHHs (92.6% and 100% of the FHHs in
Ada and Gimbichu, respectively).

The decision to use agricultural inputs by gender is shown in Table 22. In MHHs in Ada and
Gimbichu, the decision on the type of crops grown was either made jointly by the head and wife or
by the head alone, while in Lume it was mainly the head (83.3%) who decided what to plant. In
FHHs, the head often decided what to plant (77%, 100%, and 96% in Ada, Lume, and Gimbichu,
respectively).

Male heads of household in Lume (97.1%) were often the sole decision-makers in the use of
fertilizers, while in Gimbichu they made the decision with their wives (59.4%). In 50% of MHHs in
Ada, the decision was made either jointly or by the household head. More than 80% of female
heads of households in Ada, Lume, and Gimbichu are responsible for decisions about fertilizer use.
The decision to use improved seed was primarily a joint decision in MHHs in Ada (75%) and
Gimbichu (75%), while in 50% of the MHHs in Lume it was either a joint decision or a decision by
the head. In FHHs in all woredas, the head decided on the use of improved seed. In MHHs in Lume
and Gimbichu, heads of households decided on the use of chemicals, while in 80% of MHHs in
Ada, the head and wife decided together. Household heads made this decision in all FHHs. In
MHHs in Lume, the decision to use manure was made by the head alone (67%) or jointly (23%).


Table 22. Responsibility for input use decisions in Ada, Lume, and Gimbichu woredas, Ethiopia (%)

MHH FHH
Head Head and wife Head Head and son
Who decides: Ada Lume Gimbichu Ada Lume Gimbichu Ada Lume Gimbichu Ada Lume Gimbichu
What to plant 51.6 83.3 45.5 48.4 16.7 54.5 76.9 100.0 96.2 19.2 -3.8
To use fertilizer 50.0 97.1 31.3 50.0 59.4 82.1 95.7 88.9 -7.4
To use manure 66.7 22.2 100.0 100.0 100.0 -
To use improved seed 25.0 50.0 25.0 75.0 50.0 75.0 100.0 100.0 100.0
To use chemicals 20.0 100.0 100.0 80.0 100.0 100.0










In Lume, other members of the household (sons, daughters, relatives, and non-relatives) also
participated in making decisions on the use of chemicals (2.9%) and manure (11.1%). Female heads
of household in all woredas made the decision on the use of manure.

The decision to use farm inputs and the decision on who pays for these inputs are two different
issues within households. Table 23 shows who pays for the farm inputs by gender of household head.
In MHHs in Lume, heads (94.3%) usually paid for fertilizer, while about 56% and 54% of the heads
in Ada and Gimbichu respectively, did so. In about 44% of MHHs in Ada and in Gimbichu, the head
and wife paid for fertilizer. In Lume and Gimbichu the male household head always paid for improved
seeds and chemicals. In Ada, the male household head (66.7%) or the head and wife (33.3%) paid
for improved seed. Both head and wife (80%) usually purchased the chemicals. Female household
heads in all woredas usually purchased inputs.

4.2 Livestock Production

The average number of livestock owned and bought is shown in Table 24. The average number of
cattle owned by MHHs was higher than FHHs in all woredas. In Lume and Gimbichu, MHHs had
slightly more sheep and goats than FHHs, while in Ada, FHHs had more. On average, more MHHs
owned poultry, donkeys, and mules than FHHs.

Farmers purchased cattle, donkeys, goats, mules, poultry, and sheep. In Ada, farmers purchased
cattle (9 in MHHs and and 4 in FHHs) at an average cost of Birr 990 for MHHs and Birr 973 for
FHHs. In Lume, MHHs purchased more livestock (17 animals) than FHHs (8 animals) at a lower cost
(Birr 824 for MHHs and Birr 1,061 Birr for FHHs). In Gimbichu, MHHs purchased 9 animals
compared to 6 for FHHs at an average cost of Birr 736 and Birr 449, respectively.

Livestock sales and consumption are shown in Table 25. In all woredas, MHHs consumed more
livestock than FHHs. In Ada, however, the value of the livestock consumed in FHHs (Birr 120) was
higher than the value in MHHs (Birr 66); the opposite was true in Lume and Gimbichu.

The total number of livestock sold in MHHs was 28, 13, and 8 animals for Ada, Lume, and
Gimbichu, respectively, while FHHs sold 18, 14, and 8 animals, respectively. In Ada, the mean value
of the livestock sold was higher in FHHs (Birr 818) than MHHs (Birr 529). The average value of the
livestock sold was higher for MHHs (Birr 1,329 and Birr 1,036 in Lume and Gimbichu, respectively)
than for FHHs (Birr 950 in Lume and Birr 448) in Gimbichu.

Table 23. Payment for farm inputs by gender of household heads in Ada, Lume, and Gimbichu woredas, Ethiopia (%)
MHH FHH
Ada Lume Gimbichu Ada Lume Gimbichu
Head and Head and Head and Head and
Who pays for: Head Wife Wife Head Wife Head Wife wife Head Head son Head son
Fertilizer 56.3 43.8 94.3 2.9 54.3 3.3 43.8 96.2 95.7 4.3 96.3 3.7
Improved seed 66.7 33.3 100.0 100.0 100.0 100.0 100.0
Chemical 20.0 80.0 100.0 100.0 100.0 100.0












Table 24. Mean number of livestock owned and bought and cost of purchase (Birr) in Ada, Lume, and Gimbichu
woredas, Ethiopia

Ada Lume Gimbichu

MHH FHH MHH FHH MHH FHH

Mean no. Mean no. Mean no. Mean no. Mean no. Mean no.

Livestock
owned
Cattle 5.6 4.9 7.9 5.9 5.7 4.8
Sheep 2.4 3.3 7.3 6.0 4.3 2.7
Goats 3.1 4.0 6.1 5.3 4.7 3.3
Poultry 4.2 3.9 6.5 5.3 4.5 4.3
Donkeys 2.0 1.2 1.9 1.3 1.2 1.4
Mules 1.1 1.0

Livestock Mean Respon- Mean Respon- Mean Respon- Mean Respon- Mean Respon- Mean Respon-
bought no. dents no. dents no. dents no. dents no. dents no. dents
Cattle 1.1 8 1.0 4 2.0 8 1.2 6 1.0 2 2.0 1
Donkeys 1.0 3 1.0 1 1.0 3 1.0 1
Goats 1.0 1 1.0 1 -
Mules
Poultry 1.3 4 1.0 1 2.0 2
Sheep 2.0 1 4.0 3 3.0 2
Mean cost of purchased stock

Cost Respon- Cost Respon- Cost Respon- Cost Respon- Cost Respon- Cost Respon-
(Birr) dents (Birr) dents (Birr) dents (Birr) dents (Birr) dents (Birr) dents

990 8 973 4 824 17 1,051 8 736 9 449 6





Table 25. Livestock consumed and sold and their value (Birr) in Ada, Lume, and Gimbichu woredas, Ethiopia

Ada Lume Gimbichu

MHH FHH MHH FHH MHH FHH

Mean Mean Mean Mean Mean Mean
no. Respondents no. Respondents no. Respondents no. Respondents no. Respondents no. Respondents


Livestock consumed
Cattle 1.0
Goats 1.0
Poultry 1.8
Sheep 2.5

Livestock sold


1
2 1.5
8 3.3
4 1.5


1.0 1
1.0 1
1


1.6 12 1.3 10 1.1
1.0 1 1.0
1.5 4 1.0 1 1.0


11.1
2.4


3 2.0
3 1.0


9 1.5


1.1 6 1.2 5


1 3.5 2 1.0 2 1.0 1


1 1.0 10


3.0 2


Mean value of stock sold during the year
Cost Respondents Cost Respondents Cost Respondents Cost Respondents Cost Respondents Cost Respondents
(Birr) (Birr) (Birr) (Birr) (Birr) (Birr)
529 28 818 18 1,329 13 950 14 1,036 8 448 8
Mean value of stock consumed at home
Cost Respondents Cost Respondents Cost Respondents Cost Respondents Cost Respondents Cost Respondents
(Birr) (Birr) (Birr) (Birr) (Birr) (Birr)
66 20 120 15 115 27 89 22 193 7 148 2


Cattle
Donkeys
Goat
Mule
Poultry
Sheep













The main costs of keeping livestock were hiring labor and feeding (Table 26). The cost of livestock

feed is almost the same for MHHs and FHHs in Ada, but in Lume it was much higher for MHHs

than FHHs, and in Gimbichu FHHs paid more than MHHs. Other costs included veterinary services,

drugs and vaccines, and building costs.


Table 27 shows the division of responsibilities of household members for animal husbandry activities.

In MHHs in Ada and Lume, the head or son were mainly responsible for animal feeding, while in


Table 26. Costs of keeping livestock (Birr) in Ada, Lume, and Gimbichu woredas, Ethiopia


Lume


Gimbichu


MHH FHH MHH FHH MHH FHH
Mean Mean Mean Mean Mean Mean
no. Respondents no. Respondents no. Respondents no. Respondents no. Respondents no. Respondents
Feeding 110.1 17 119.5 11 301.7 17 95.7 9 188.7 26 217.3 10
Veterinary
attention 1.0 3 14.5 10 9.8 6 95.0 3 17.5 2
Drugs and
vaccine 13.3 6 9.0 6 20.2 5 17.0 5
Hired labor 390.0 10 353.3 15 256.2 17 127.1 6 315.8 6 280.0 5
Building 60.0 1 33.8 5 35.0 3 15.0 1



Table 27. Responsibility for animal husbandry in Ada, Lume, and Gimbichu woredas, Ethiopia
Ada Lume Gimbichu
MHH FHH MHH FHH MHH FHH
Activity/
household member (%) (Respondents) (%) (Respondents) (%) (Respondents) (%) (Respondents) (%) (Respondents) (%) (Respondents)


Animal feeding
Wife 34.4 11
Son 71.9 23
Daughter 15.6 5
Head 78.1 25
Relative/non-relative 37.5 12
Animal treatment


Daughter
Head
Relative/non-relative
Animal transport
Wife
Son


6.3 2

34.4 11



3.1 1
9.4 3


Daughter
Head 56.3 18
Relative/non-relative 9.4 3
Animal slaughter


Son
Daughter
Head
Relative/non-relative
Animal production
Wife
Son
Daughter
Head
Relative/non-relative


3.1 1

43.8 14



62.5 20
6.3 2
6.3 2
3.1 1


64.3 18 57.1 20
17.1 6
60.7 17 80.0 28
64.3 18 28.6 10


17.9 5

17.9 5
3.6 1



25.0 7

28.6 8
10.7 3



28.6 8
7.1 2
3.0


7.1 2
7.1 2
50.0 14
7.1 2


73.9 17 45.5 15
13.0 3 21.2 7
30.4 7 9.1 3
47.8 11 39.4 13


44.0 11
16.0 4
20.0 5
48.0 12


19.4 6 -
26.1 6 6.1 2 12.0 3


48.6 17


26.1 6 51.5 17
4.3 1 6.1 2


3.2 1
14.3 5 26.1 6 6.1 2

51.4 18 34.8 8 39.4 13
13.0 3 15.2 5


2.9 1

51.4 18



79.4 27


4.0 1
40.0 10
32.0 8



20.0 5

8.0 2
32.0 8


43.5 10 3.2 1 20.0 5


60.6 20
6.1 2


45.2 14


25.7 9 39.1 9
87.0 20
2.9 1 -


9.1 3 48.0 12
12.1 4 20.0 5












Gimbichu it was mainly the son, non-relatives, and relatives. In FHHs in Ada, the female head and
her son were mainly responsible for animal feeding, while in Lume and Gimbichu it was mainly the
son or relatives. The head of the household was mainly responsible for the treatment of sick animals
and the transport of animals. In MHHs in all woredas, the household head slaughtered the animals,
and in FHHs, the son or head were primarily responsible. In MHHs, the wife was mainly
responsible for animal production, while in FHHs it was the household head.


Table 28 gives respondents' reasons for keeping different types of livestock. Households mainly
kept livestock for draft power and dairy products. MHHs also kept cattle for dowries, but none of
the FHHs did so. Donkeys were mainly used for transport although a few households sold them.
Mules and horses were reported only in Gimbichu and were used for transport. Goats were both
consumed and sold.



Table 28. Reasons for keeping different types of livestock in Ada, Lume, and Gimbichu woredas, Ethiopia

Ada Lume Gimbichu

MHH FHH MHH FHH MHH FHH

(%) (Respondents) (%) (Respondents) (%) (Respondents) (%) (Respondents) (%) (Respondents) (%) (Respondents)


Cattle
Gift
Consumption
main products
Draft power
Dowry
Consumption
of by-products
Transportation

Donkeys
Gift
Sale
Transportation


12.5 4 14.3 4


31.3 10
100.0 32
100.0 32


3.1 1


42.9 12 61.1 22
100.0 28 97.2 35
100.0 32


84.4 27 100.0 28
100.0 32 71.4 20


11.1 3
18.5 5
100.0 27


43.5 10 18.8 6
91.3 21 100.0 32
100.0 32


50.0 18 100.0 23
43.5 10


9.5 2 35.5 11 15.8 3
100.0 21 100.0 31 100.0 19


12.0 3
100.0 25


81.3 26 100.0 25
- 80 20


100.0 27 100.0 19


Mules and horses
Transportation -100 15 100 2


Goat
Consume by-products
Gift
Consume main
products
Sale

Poultry
Consume main
products
Sale

Sheep
Consume by-products
Dowry
Gift
Consume main
products
Sale


28.6
14.3

100.0
57.1


2 20.0 1
1 20.0 1


36.4 4 42.9 3


7 100.0 5 100.0 8 100.0 4 90.9 10 85.7 6
4 60.0 3 100.0 8 100.0 4 72.7 8 85.7 6


100.0 19 100.0 16 96.0 24 100.0 13 57.1 12 72.2 13
63.2 12 75.0 10 96.0 24 92.3 12 57.1 12 38.9 7


11.1 1
12.5 1
33.3 3 25.0 2

100.0 9 100.0 8 90.9 10 84.6 11 94.1 16 69.2 9
77.8 7 87.5 7 100 11 100.0 13 41.2 7 69.2 9











Almost all of the MHHs and FHHs in Ada and Lume kept poultry for consumption; fewer
households in Gimbichu kept poultry for this purpose. More FHHs than MHHs kept poultry for sale.
Sheep were kept for consumption and sale.

4.3 Tree Crop Production

Table 29 shows tree crops on farmers' fields in the study area. More households in Ada and Lume
had trees in their fields than Gimbichu. Acacia was the most common species, although eucalyptus
was also important.


5.0 GENDER DIFFERENTIALS IN ACCESS TO LAND


Studies of African concepts of land tenure during the colonial period report that land was conceived
of primarily as an inalienable community property which was indispensable for the cohesion of
social relations. Land, the most important means of production, was the basis of the family, the
source of life, and formed the link between social organization and ideology. For these reasons,
access to land was subjected to strictly patriarchal control, and land was inalienable to anyone (Berg
van den 1997). The total amount of land available per household and its use are crucial, given the
population growth and the need for increased agricultural productivity.

In all three woredas, MHHs had relatively larger farm sizes than FHHs, but there was no significant
difference between the two groups of households. The difference in farm size could have resulted
from the nature of land ownership in the study area. Traditionally, in Oromo culture, women had
access to land only through marriage, and a widow's land was still the property of the husband, but a
PA policy allowed them to allocate land to FHHs. In general, the law in Ethiopia does not discrimate
overtly against women in terms of land inheritance, ownership, and management, but women's
rights have not been fully asserted in major legislation (Daniel Haile 1980). In Africa in general,
women's access to land is not a problem where social institutions allocate land to both men and
women or where women can borrow or claim unused land (Bryson 1981).


Table 29. Most important tree crop in Ada, Lume, and Gimbichu woredas, Ethiopia
Ada Lume Gimbichu
MHH FHH MHH FHH MHH FHH
Tree crop (%) (Respondents) (%) (Respondents) (%) (Respondents) (%) (Respondents) (%) (Respondents) (%) (Respondents)
Most important tree crop
Eucalyptus 10.0 20 5.3 1 3.6 1 6.7 10
Acacia 85.0 17 94.7 18 53.6 15 46.7 7 50.0 1
Wacho 5.0 1 6.7 1 100.0 1 50.0 1
True man tree 3.6 1 13.3 2
Other 39.3 11 26.7 4
Second most important tree crop
Acacia 50.0 1 50.0 1 20.0 2 28.6 2
Wacho 50.0 1 14.3 1
Others 50.0 1 60.0 6 57.1 40
Eucalyptus 10.0 1
True man tree 10.0 1










Per capital landholdings of both MHHs and FHHs did not vary significantly. In Lume, MHHs held
more land per capital than FHHs, while in Gimbichu FHHs had larger land holdings per capital than
MHHs (Table 30). This difference can be explained by the fact that, in Ethiopia, there is no individual
ownership of land/plots within the farm household. Instead, all productive resources owned by the
family are used collectively, with the exception of small plots in the backyard that women use to grow
vegetables.

5.1 Land Rights

An increasing number of African farmers are finding themselves with insufficient land to feed their
families, or worse still, no land at all. An abundant and available resource prior to colonial
occupation, land was rarely owned in the western sense of exclusive, inalienable rights vested in one
individual in exchange for cash. There is confusion over terms related to land rights, such as tenure,
usufruct, freehold, and ownership when they are applied in the African context. "Tenure" refers to
landholding rights, including land that is passed on through inheritance, through a loan or rent for an
established exchange value, or through an outright sale. "Usufruct" refers to a collective group or
individual right to use land. "Freehold" refers to an individual's or a corporate body's exclusive right
to hold a piece of land that can be transferred. "Ownership" refers to land that has cash or
commodity value and is registered through a process of entitlement to an individual or corporate
group (Davis 1988).

A critical issue for smallholder agriculture throughout Africa is the shortage of good quality farming
land. Increasing population pressures and fragmentation of holdings have sharply reduced cultivated
area per person. Within this context of rapid population growth and the need for increased
productivity of land, there is a growing debate about whether women farmers' access to land and
their decision-making power are constraints to agricultural transformation (Saito, Mekonnen, and
Spurling 1994). Historically, men gained access to land as lineage members, but in the majority of
cases, women gained access as wives; in a few cases, women inherited land as lineage daughters
(Davis 1988). Both men and women farmers are generally better off when they have the right
to land.

5.2 Land Acquisition

Unlike other African countries, in Ethiopia MHHs and FHHs acquired land primarily through
allocation by the government or the village head. In Kenya, for instance, a higher percentage of
MHHs and FHHs obtained land through inheritance and purchase, and in Nigeria, most MHHs and
FHHs acquired land through inheritance. In Burkina Faso, government interventions in land tenure
have ignored the rights of women (Saito, Mekonnen, and Spurling 1994).

Table 30. Landholdings by gender of household head in Ada, Lume, and Gimbichu woredas, Ethiopia
Ada Lume Gimbichu
MHH FHH MHH FHH MHH FHH
Farm size (ha) 2.35 1.90 3.15 2.82 3.16 2.64
Household size (no. persons) 8.10 6.50 7.60 5.50 7.80 5.60
Per capital holding (ha/person) 0.29 0.29 0.46 0.51 0.41 0.47










The process of land acquisition in the study area was analyzed to determine whether MHHs and
FHHs had differential access to land. Land was allocated to MHHs in Ada (42%) and Lume (74%) by
the village head. In Gimbichu, 64% of MHHs acquired land from the government. Land was also
acquired through rent or lease. A very small proportion of MHHs in Ada. (3.2%) purchased land
(Table 31).

About 68% of FHHs in Lume and 89% in Gimbichu acquired land through government allocation,
while about 13% of FHHs in Lume acquired land through inheritance. In Ada, 48% of FHHs were
allocated land by the village head and around 36% from the government. Other forms of land
acquision for FHHs in Ada (16%) included renting or leasing.

5.3 Land Quality

Concomitant with decreasing per capital landholdings, the deterioration in the quality of land is a
concern among farmers in many developing countries. Continuous monocropping, lack of soil-
ameliorating practices, and increasing populations of humans and livestock are among the prominent
factors exacerbating the situation. The shift in farming practices from the traditional land-extensive,
low-input cultivation systems that maintained ecological balance to modern, labor-intensive systems is
also considered as one of the factors leading to the deteriorating quality of land (Saito, Mekonnen,
and Spurling 1994).

The quality of land is good in Ada and Lume because soils in these woredas are mainly fertile
vertisols. Gimbichu has a degraded topograph; hence a large proportion of its total area is not
suitable for agricultural production.

Table 32 shows farmers' opinions of the fertility of their soils and the proportion of irrigated land in
the study area. Most farmers in Ada and Lume said that their soils ranged from fertile to very fertile,
while farmers in Gimbichu reported that their soils were poor to very poor. In general, the study
shows that more MHHs than FHHs thought they had fertile soils.

Irrigation is one of the factors that improves the quality of land. Even though most farmland is
rainfed, efforts have been made to introduce irrigation, particularly during the off-season for the
production of horticultural crops, which play an important role in providing cash. Few farmers in Ada

Table 31. Methods of land acquisition by gender of household head in Ada, Lume, and Gimbichu woredas, Ethiopia

MHH FHH
Ada Lume Gimbichu Ada Lume Gimbichu
Methods of land
acquisition (%) (Respondents) (%) (Respondents) (%) (Respondents) (%) (Respondents) (%) (Respondents) (%) (Respondents)
Purchase 3.2 1
Inheritance 3.2 1 3.0 1 20.8 6 -
Allocated by
government 94.6 29 100.0 25 97.0 30 100.0 25 79.2 24 100.0 25
Others 32.2 10 25.7 9 33.3 11 16.0 4 8.3 2 7.4 2

a Rented/leased,joint allocation by village head and government.










Table 32. Soil fertility status and irrigation of farmland in Ada, Lume, and Gimbichu woredas, Ethiopia

Ada Lume Gimbichu
MHH FHH MHH FHH MHH FHH

(%) (Respondents) (%) (Respondents) (%) (Respondents) (%) (Respondents) (%) (Respondents) (%) (Respondents)
Very fertile 32.3 20 18.8 9 22.4 13 26.0 13 3.3 3 1.3 1
Fertile 38.7 24 41.7 20 56.9 33 42.0 21 31.9 29 29.9 23
Poor 24.2 15 31.3 15 19.0 11 26.0 13 28.6 26 33.8 26
Very poor 4.8 3 8.3 4 1.7 1 6.0 3 36.3 33 35.1 27
Irrigated 18.4 7 4.0 1 16.7 5 21.4 6 9.7 3 11.1 3
Not irrigated 81.6 31 96.0 24 83.3 25 78.6 22 90.3 28 88.9 24



and Gimbichu practiced irrigation. In Lume more FHHs (21%) practiced irrigation than MHHs (17%),
possibly because in Lume, women were responsible for horticultural crop production on small plots of
land using irrigation.

5.4 Decision-Making for Land Improvement, Acquisition, and Rental

Some researchers regard the indigenous tenure system as a static constraint to agricultural
development, providing insufficient tenure security to induce farmers to make necessary land-
improving investments (Dorner 1972; World Bank 1974; Harrison 1987). Others counter that
indigenous tenure arrangements are dynamic and evolve in response to changes in factor prices
(Cohen 1980; Boserup 1981; Nornha 1985; Bruce 1988). In Kenya, for instance, women
traditionally did not inherit land, but their rights to use land belonging to a male relative were assured.
Today, unless the land is registered in her own name, a women's right to use land is threatened by
land commercialization. In fact, it could be argued that there is a spontaneous individualization of land
rights over time, whereby farm households acquire a broader and more powerful set of transfer and
exclusionary rights over land as population pressure and agricultural commercialization proceed. This
trend in turn raises the question of whether or not there is a need for expensive land registration and
titling programs at this stage of economic development in sub-Saharan Africa. On the other hand, if
indigenous tenure systems are dynamic, then it is relevant to ask if governments can take more useful
measures to facilitate the process of technology adoption.

Table 33 shows the decision-making responsibilities to improve, lend, or rent land in the three
woredas. About 68%, 100%, and 61% of male heads of household are responsible for decisions to
improve land in Ada, Lume, and Gimbichu, respectively. Almost all female heads of household are
responsible for this decision in all woredas. The decision to lend land in Lume was the full
responsibility of the wife in MHHs, while in Gimbichu it was mostly a joint decision between the
husband and wife (75%). In Gimbichu, female household heads and their sons were the ones who
decided to lend land.










Table 33. Responsibility for decisions on land improvement, acquisition, and renting by gender of household head in
Ada, Lume, and Gimbichu woredas, Ethiopia

MHH FHH
Ada Lume Gimbichu Ada Lume Gimbichu

Who Head Head Head Head
decides to: Head Wife Head Wife and wife Head Wife and wife Head Son Head and son Head and son
Improve land 67.7 32.3 100 60.6 39.4 96.3 3.7 95.7 4.3 100
Lend land 100 25.0 75.0 100
Rent land 33.3 66.7 100 100 100



In Ada, there is no lending or renting of land. The decision to rent land in Lume was made by
husband and wife (67%), while in FHHs the decision was the joint responsibility of the female head
and son. In MHHs in Gimbichu, the decision to rent land was made by the husband and wife (100%)
and in FHHs, by the female head and son (100%).



6.0 GENDER PATTERNS OF LABOR UTILIZATION


People often refer to the "triple roles" of gender: production, reproduction, and community
management. The productive role involves the production of goods and services for consumption
and trade. Work related to crop and livestock production (agricultural production) and other income-
generating activities is regarded as productive. Reproductive work involves the care and
maintenance of the household and its members. Collecting fuelwood and water, caring for children,
cooking, and cleaning are all considered reproductive work. These activities do not directly bring
cash into the household and are often not even recognized as work. If it is clear that a woman or
groups of women are collecting fuelwood for sale, however, then this activity is considered
productive. Community management work involves the collective organization of social events and
services. Thus group work such as cleaning irrigation channels and terracing is viewed as community
management.

6.1 Use of Family Labor in Crop Production

6.1.1 Land preparation
Unlike some African countries where the gender division of family labor in agricultural production is
becoming less distinct (Saito, Mekonnen, and Spurling 1994), in Ethiopia in the study area, no female
farmers engaged in land preparation. Across all three woredas, land preparation was the responsibility
of the male household head, son, male relatives, and male non-relatives in MHHs; in FHHs, the son,
male relatives, and male non-relatives were responsible (Table 34).

In Ada and Gimbichu, male household heads prepared most of the land. The average number of days
male heads spent preparing land was 34, 21.5, and 30 in Ada, Lume, and Gimbichu, respectively.
The difference between Ada and Lume and between Ada and Gimbichu was significant at p<0.01.
The average number of days allotted for land preparation in Ada is probably higher because tef is











Table 34. Division of household labor by gender and crop operation in Ada, Lume, and Gimbichu woredas, Ethiopia

Ada Lume Gimbichu
Crop operation MHH FHH MHH FHH MHH FHH

Land preparation
Average number of days
Head 34.0 21.5 30.0
Son 29.3 29.8 26.6 28.2 20.3 19.6
Relatives 28.7 40.0 32.0 14.3 19.0
Non-relatives 27.8 31.4 16.4 28.2 29.0 20.8
Average h/day
Head 7.4 6.6 6.9
Son 7.1 7.5 6.6 6.5 7.8 7.6
Relatives 6.7 6.2 6.8 6.4 6.2
Non-relatives 6.5 8.0 7.4 6.4 6.9 7.1
Planting
Average number of days
Head 17.0 11.4 15.5
Son 14.1 13.1 9.2 13.6 14.7 13.0
Relatives 15.7 8.0 18.7 15.2
Non-relatives 16.5 15.0 9.2 12.0 14.0 14.1
Average h/day
Head 8.6 7.9 8.6
Son 8.5 8.8 7.6 7.7 7.4 8.5
Relatives 11.3 4.0 7.3 7.5
Non-relatives 9.0 9.2 8.0 7.1 8.4 9.0
First weeding
Average number of days
Head
W ife 14.3 13.7 21.3 18.6 14.5 15.9
Son 13.7 18.0 16.9
Daughter 22.6 20.9 26.3 26.0 19.3 20.2
Relatives 18.3 20.7 20.2 19.0 16.0 21.3
Non-relatives 3.3 12.7 93.0 8.0 18.3 21.3
Average h/day
Head 8.7 8.8 7.8 7.2 7.6 7.3
Wife 8.9 7.7 7.0
Son 9.1 8.5 7.7 7.8 7.3 7.8
Daughter 8.9 8.3 7.8 7.6 7.8 7.3
Relatives 9.5 8.5 8.0 9.0 6.3 8.0
Non-relatives 8.4 9.2 7.9 7.3 7.7 7.2
Second weeding
Average number of days
Head 6.7 8.6 7.8 9.6 4.7 3.0
Wife 7.1 7.9 4.5
Son 10.6 9.3 13.9 13.0 10.5
Daughter 12.4 9.0 9.2 8.3 8.0 10.0
Relatives 11.0 7.7 4.0 -
Non-relatives 13.0 10.3 10.3 10.7 2.0
Average h/day
Head 7.7 9.0 6.8 7.8 9.0 7.0
Wife 8.0 7.2 9.0
Son 8.4 8.5 6.6 8.8 8.9
Daughter 7.2 8.3 6.1 7.8 8.5 9.0
Relatives 7.5 5.5 9.5
Non-relatives 7.7 8.3 7.5 9.3 8.0

(Continued) -











Table 34. (Cont'd.)

Ada Lume Gimbichu
Crop operation MHH FHH MHH FHH MHH FHH

Harvesting
Average number of days
Head 12.2 3.2 11.2 6.4 18.9 7.2
Wife 4.1 6.1 5.3
Son 16.5 16.0 12.1 11.0 21.5 23.4
Daughter 6.3 6.0 5.2 8.2 9.9 10.7
Relatives 20.3 16.5 9.1 9.8 17.1 20.7
Non-relatives 10.2 15.4 12.0 8.3 25.0 27.0
Average h/day
Head 7.7 7.4 7.2 6.5 8.8 7.1
Wife 7.2 6.8 7.8
Son 7.5 8.2 7.1 7.0 8.7 8.6
Daughter 7.3 5.3 7.5 8.1 8.1 8.1
Relatives 8.0 8.2 7.8 7.8 8.5 8.1
Non-relatives 7.9 7.9 7.3 7.4 8.4 7.5
Transporting
Average number of days
Head 9.6 9.4 5.7 10.6 6.9
Wife 8.5 8.9 8.3
Son 16.9 11.0 12.1 9.2 13.8 12.5
Daughter 10.1 7.7 9.3 7.3 7.8
Relatives 17.1 9.2 21.7 19.2 10.4 20.1
Non-relatives 9.7 11.8 11.3 5.2 15.3 17.6
Average h/day
Head 6.9 7.0 6.6 6.2 5.8
Wife 6.4 7.2 6.5
Son 6.7 7.0 7.1 6.8 6.2 5.7
Daughter 6.5 7.3 7.2 6.8 5.8
Relatives 7.4 6.8 7.8 8.5 6.3 6.1
Non-relatives 7.2 6.7 6.5 7.5 5.9 5.9
Threshing/storage
Average number of days
Head 13.0 4.7 16.4 9.5 11.9 7.2
Wife 7.1 10.8 8.3
Son 17.8 14.1 19.5 19.1 14.5 12.6
Daughter 12.9 3.3 8.5 16.4 9.3 7.9
Relatives 22.5 10.4 9.6 14.0 6.2 11.1
Non-relatives 18.3 13.3 19.1 12.0 18.3 9.9
Average h/day
Head 8.6 6.5 8.5 7.9 8.7 6.5
Wife 6.9 7.7 7.9
Son 8.8 8.8 8.6 8.5 8.8 8.6
Daughter 7.0 9.0 8.1 7.3 7.6 7.6
Relatives 7.7 7.8 8.4 8.4 7.3 8.2
Non-relatives 10.0 8.4 8.2 8.9 8.6 8.2










the dominant crop and the land requires several plowings before it is planted. Even though farmers
in Lume and Gimbichu also grow tef, tef performs so well in Ada that farmers give it top priority in
the allocation of labor. The average number of days spent on land preparation was relatively higher
in Gimbichu than in Lume because of waterlogging, which is the dominant crop production
problem there.

In Ada and Lume, FHHs spent 30 days on land preparation while in Gimbichu they spent 20 days.
MHHs in Lume and Gimbichu spent more time on land preparation than FHHs, while in Ada both
types of household spent roughly the same amount of time on land preparation.

There was little difference in the number of hours spent on land preparation between MHHs and FHHs
across the three woredas. In Ada, male household heads spent more hours per day preparing land than
other members of the household, while in Lume and Gimbichu the male head spent fewer hours per
day than other members of the household, although the difference was not significant. In FHHs, sons,
relatives, and non-relatives spent more than six hours per day on land preparation.

6.1.2 Planting
There was little difference between MHHs and FHHs in the average number of hours spent on planting
per day. The male household head, sons, and non-relatives in MHHs in Ada and Lume did the
planting, while relatives also participated in Gimbichu. Female heads of household did not engage in
planting. In all three woredas, the son, relatives, and non-relatives were engaged in planting. In Ada
and Lume, male heads spent more days on planting than other members of the household. In
Gimbichu, relatives spent more days planting than male heads. In FHHs in all woredas, sons spent on
average 13 days on planting. With the exception of Lume, relatives in FHHs spent more days planting
than relatives in MHHs.

6.1.3 Weeding
All members of farm households were involved in weeding (Table 34). In MHHs, the son worked the
highest number of days (more than 20 days) on the first weeding, followed by non-relatives (close to 20
days) and daughters (more than 16 days). The number of days spent on weeding ranged from about 3 in
Ada to 18 in Gimbichu. In all woredas, there was little difference in the number of days husbands and
wives spent on weeding. In FHHs, the average number of days spent on weeding was greatest for sons
(more than 20 days). In Ada and Lume, daughters and relatives spent on average 21 days on weeding.
Female heads spent the fewest hours on weeding. The average number of days for the first weeding in
FHHs ranged from 8 days by relatives to 26 days by sons in Lume. Female heads on average spent about
14, 19, and 16 days on weeding in Ada, Lume, and Gimbichu, respectively. There was no marked
difference in the number of days spent by male and female heads on first weeding.

In MHHs, the average hours spent by relatives per day on the first weeding ranged from around 6 hours
in Gimbichu to around 10 hours in Ada. There was no significant difference in the average number of
hours per day among members of the household. However, household members spent on average more
hours per day on weeding in Ada, followed by Lume and Gimbichu. There was no significant difference
between the average number of hours that male heads and wives spent per day spent on weeding in the










three woredas. In FHHs, the average hours spent on weeding ranged from about 7 by the female head
in Lume and non-relatives in Gimbichu to 9 by non-relatives in Ada per day. Female heads spent fewer
hours per day on weeding than male heads in all the three woredas, although the difference was not
significant.

The average number of days spent on the second weeding was less than the average number of days
spent on first weeding in all woredas. In MHHs, it ranged from 2 days by non-relatives in Gimbichu to 14
days by sons in Lume. Heads spent on average 6.7, 7.8, and 4.7 days on the second weeding in Ada,
Lume, and Gimbichu, respectively. In Ada and Lume, wives spent more days on the second weeding
than the heads, but again, the difference was not significant. The average number of days spent on the
second weeding was the lowest in Gimbichu. Relatives did not participate in the second weeding in Lume
and Gimbichu.

In FHHs, the average number of days spent on the second weeding ranged from 3 days by the head in
Gimbichu to about 11 days by non-relatives in Lume. Female heads spent an average of 8.6 days in Ada,
9.6 days in Lume, and 3 days in Gimbichu on the second weeding. In FHHs in Gimbichu, the second
weeding was limited to the head (3 days) and daughters (10 days). MHHs spent more days on the second
weeding than FHHs, although the difference was not significant.

The average hours per day spent on the second weeding in MHHs ranged from about 6 hours by
daughters in Lume to 9 hours by the head and wife in Gimbichu. Heads spent on average 7.7, 6.8, and
9 hours per day on the second weeding in Ada, Lume, and Gimbichu, respectively. The average number
of hours per day that wives spent on the second weeding was higher than the time spent by household
heads in Ada and Lume, while wives and heads spent the same number of hours per day (9) in
Gimbichu. The average number of hours that household members spent each day on the second
weeding was higher in Gimbichu, followed by Ada and Lume.

In FHHs, the average number of hours per day spent on second weeding ranged from about 6
hours by relatives to 9 hours by the head in Ada and daughters in Gimbichu. Heads spent on
average 9, 7.8, and 7 hours per day in Ada, Lume, and Gimbichu, respectively, on the second
weeding. FHHs spent more time on second weeding than MHHs, although the difference was not
significant.

6.1.4 Harvesting
All household members in the study area took part in harvesting (Table 34). Non-relatives were
more involved in harvesting than other members of the household. On average, MHHs spent much
more time on harvesting than FHHs in all woredas (p<0.01). There was little difference between
MHHs and FHHs in the average number of days that household members (sons, daughters,
relatives, and non-relatives) spent on harvesting. In MHHs, the average number of days spent on
harvesting ranged from 4 days by the wife in Ada to 25 days by non-relatives in Gimbichu, while in
FHHs it ranged from 3 days by the head in Ada to 27 days by non-relatives in Gimbichu.










The average hours spent per day on harvesting in MHHs ranged from about 7 hours by the wife in
Lume to about 9 hours by the head in Gimbichu. On average, all household members in MHHs in
Ada and Lume spent more time on harvesting than FHHs. In Ada, this was because daughters in
MHHs spent more hours (7) harvesting. This difference was significant (p<0.01). In Lume, there was
little difference in the average number of hours household members spent each day on harvesting
between MHHs and FHHs. In Gimbichu, there was a significant difference in the average number of
hours that male and female heads of households spent on harvesting (p<0.01).

6.1.5 Transporting produce
This activity entails the transportation of agricultural produce from the field to the threshing area and
to the nearest market. There was no significant difference between MHHs and FHHs in the number of
days spent transporting produce (Table 34). The average number of days household members spent
transporting produce ranged from 5.2 days by non-relatives in FHHs to 21.7 days by relatives in
MHHs in Lume. In MHHs in Ada, household heads spent an average of 9.6 days transporting
produce, while wives spent 8.5 days. Relatives, followed by sons in MHHs, undertook most of the
transporting. In FHHs, non-relatives and sons did most of the transporting. Heads and daughters in
FHHs in Ada were not involved in transporting at all.

In Lume, heads and non-relatives in MHHs spent significantly more time transporting produce than
heads and non-relatives in FHHs (p<0.01). In FHHs in Lume and Gimbichu, all household members
were involved in transporting produce. More household members in MHHs transported produce than
in FHHs. In Gimbichu, there was a significant difference in the average number of days that male and
female heads of households (p< 0.05) and relatives (p<=0.05) spent transporting produce. On
average, household members in FHHs in Gimbichu spent more days transporting produce than
household members in MHHs.

The average hours spent per day transporting produce ranged from about 6 hours by sons in FHHs in
Gimbichu to about 9 hours by relatives in FHHs in Lume. Except for relatives and non-relatives in
MHHs in Ada, Lume, and Gimbichu, the average number of hours that farm household members
spent transporting produce per day was higher in MHHs than in FHHs.

6.1.6 Threshing/storage
In the study area,threshing of small grain cereals involves both animal power and human labor.
Animals are rotated over the thinly spread crop on a specially prepared threshing ground, and
household members winnow the grain from the straw. In Ada, the average number of days spent
threshing and storing grain ranged from about 5 days by the head in FHHs to about 23 days by
relatives in MHHs (Table 34). With the exception of daughters and relatives in Lume and relatives in
Gimbichu, the average number of days that household members spent on threshing and storing grain
was higher in MHHs than FHHs in all woredas.

In Ada, relatives followed by non-relatives in MHHs recorded the highest average number of days
spent on threshing/storage, while in FHHs the son followed by non-relatives and relatives recorded
the highest average number of days. Time spent on this operation differed significantly only between










the heads (p<0.01) and daughters (p<0.07) of MHHs and FHHs. Female heads (4.7 days) and
daughters (3.3 days) spent the least number of days on threshing/storage.

In MHHs in Lume, the head, son, and non-relatives spent more than 15 days on threshing/storage,
while in FHHs, only sons and daughters spent more than 15 days on threshing/storage. There was
a significant difference between MHHs and FHHs in the case of the heads (p<0.01), daughters
(p<0.00), and relatives (p<0.00). Female heads spent the least number of days on threshing/
storage.

The average hours per day of threshing/storage ranged from about 7 hours by the female head in
Ada and Gimbichu to 10 hours by non-relatives in MHHs in Ada. Sons spent more hours per day
on threshing/storage than other members of the household with the exception of non-relatives in
MHHs in Ada and Lume. Household members in MHHs spent on average more hours per day on
threshing/storage than household members in FHHs. On average, daughters in MHHs and FHHs
in Ada spent significantly fewer hours per day on threshing/storage than daughters in Lume and
Gimbichu.

6.2 Use of Family Labor in Livestock Production
All household members in MHHs and FHHs were involved in livestock activities though different
household members were responsible for different activities (Table 35).

In most MHHs in Ada, the head (78%) and the son (72%) were responsible for animal feeding,
while the wife (34%) and relatives/non-relatives (36%) were less active. In about 80% of the MHHs
in Lume, the head was responsible for livestock feeding, while in Gimbichu, only 9% of heads in
MHHs were responsible. In Gimbichu, the son (46%) was mainly responbile for livestock feeding,
followed by relatives/non-relatives (39%), the daughter (21%), and the wife (7%). In FHHs, the
responsibility for animal feeding rested mainly with the son and relatives/non-relatives across all
woredas.

Heads of households were primarily responsible for treating sick animals in all woredas. In MHHs,
heads were responsible for slaughtering and selling animals. In Ada, Lume, and Gimbichu, the head
was responsible for slaughtering and selling animals in about 44%, 51%, and 61% of the MHHs,
respectively; in FHHs, sons and relatives/non-relatives were responsible. Heads of household were
responsible for transporting animals to the market in about 56%, 51%, and 39% of MHHs in Ada,
Lume, and Gimbichu, respectively, while sons were responsible in 9% of MHHs in Ada, 14% in
Lume, and 6% in Gimbichu. In FHHs, 29% of heads in Ada, 35% in Lume, and 8% in Gimbichu
transported animals to the market, while sons did this in 25% of FHHs in Ada, 26% in Lume, and
20% in Gimbichu. Relatives/non-relatives transported animals to the market in 11%, 13%, and
32% of FHHs in Ada, Lume, and Gimbichu, respectively.











Wives in MHHs were mainly responsible for processing animal products. Heads of households were
responsible in 50% of FHHs in Ada, 87% in Lume, and 48% in Gimbichu. In Lume, daughters also
participated considerably in processing animal products.


In general, the head was mainly responsible for feeding, treating, slaughtering, and transporting
animals in MHHs, while the wife processed animal products. Other family members assisted only in
treating, slaughtering, and transporting animals. On the other hand, other family members were
responsible for all livestock activities in FHHs. It is important to note, however, that in Ethiopia the
sexual division of labor and gender roles varies from one cultural setting to another depending on
whether the plow or the hoe are used for cultivation. For instance, in some parts of Ethiopia, women
do not sell or buy bulls, oxen, heifers, or cows (Rahamato 1991).




Table 35. Division of family labor by gender and livestock production task in Ada, Lume, and Gimbichu woredas, Ethiopia (%)

Ada Lume Gimbichu

Task MHH FHH MHH FHH MHH FHH

Feeding
Head 78.1 60.7 80.0 30.4 9.1 20.0
Wife 34.4 22.9 6.5
Son 71.9 64.3 57.1 73.9 45.5 44.0
Daughter 15.6 17.1 13.0 21.2 16.0
Relative/non-relative 37.5 64.3 28.6 47.8 39.4 48.0

Treating sick animals
Head 34.4 17.9 48.6 26.1 51.5 40.0
Wife 19.4
Son 6.3 17.9 -26.1 6.1 12.0
Daughter 4.0
Relative/non-relative 3.6 4.3 6.1 32.0

Slaughter and sale
Head 43.8 7.1 51.4 60.6
Wife
Son 3.1 26.6 2.9 43.5 3.2 40.0
Daughter
Relative/non-relative 10.7 8.7 6.1 36.0

Transport to market
Head 56.3 28.6 51.4 34.8 39.4 8.0
Wife 3.1 3.2
Son 9.4 25.0 14.3 26.1 6.1 20.0
Daughter
Relative/non-relative 9.4 10.7 -13.0 15.2 32.0

Product processing
Head 3.1 50.0 87.0 9.1 48.0
Wife 62.5 79.4 45.2
Son 6.3 7.1
Daughter 6.3 7.1 25.7 39.1
Relative/non-relative 7.1 2.9 12.1 20.0










6.3 Use of Hired Labor in Crop Production

The use of hired labor in crop production is shown in Table 36. Hired labor in agricultural operations
is limited to weeding, harvesting, threshing and transporting. Both male and female workers were
hired for these operations. Children were also hired for weeding. In Ada, Lume, and Gimbichu, about
88%, 97%, and 82% of MHHs, respectively, hired labor for crop production compared to about
78%, 96%, and 74% of FHHs. The main reason given by both types of households for not hiring
labor was that the labor supply was sufficient. Other reasons were that cash was short or that hiring
labor was not profitable.

The division of hired labor by gender and agricultural production tasks in Ada, Lume, and Gimbichu
is shown in Table 37. In MHHs in Ada, male and female labor was hired for weeding for 19 and 15
days, respectively, and for 11 and 10 days in FHHs. In Lume, only male labor was used (17 days in
MHHs and 25 days in FHHs). In Gimbichu, households hired male labor (33 days in MHHs and 19
days, in FHHs) and child labor (2 days in MHH and 8 days in FHHs). Laborers were paid Birr 4-5
per day for weeding.

In Ada, both types of households hired mainly male labor for harvesting. Only MHHs in Ada (6 days)
and Lume (8 days) hired female labor. Male laborers were paid Birr 5-9 for harvesting per day, while
female laborers were paid Birr 5- 6 per day.

In MHHs in Ada, Lume, and Gimbichu, male labor was hired for about 10, 19, and 7 days for
transportation, while FHHs in Gimbichu hired male labor for an average of 28 days. Hired laborers
were paid Birr 5-7 per day.

Most households hired only male labor for threshing, with the exception of FHHs in Lume, which
hired men as well as women. In Lume, the cash payment and cash equivalent per day for male
laborers were Birr 5.7 and 3.8, respectively, in both MHHs and FHHs, while that of female labor
was Birr 7. In Gimbichu, MHHs and FHHs hired male labor for an average of 8 and 9 days,
respectively. The cash payment per day was Birr 8.8 in MHHs and Birr 4.3 in FHHs. The cash
equivalent was Birr 6.8 in MHHs and Birr 5 in FHHs.




Table 36. Hired labor and reasons for not hiring labor in Ada, Lume, and Gimbichu woredas, Ethiopia

Ada Lume Gimbichu
MHH FHH MHH FHH MHH FHH

(%) (Respondents) (%) (Respondents) (%) (Respondents) (%) (Respondents) (%) (Respondents) (%) (Respondents)

Households using
hired labor 87.5 28 77.8 21 97.1 34 95.8 23 81.8 27 74.1 20

Reasons for not hiring labor
Have enough labor 100.0 4 83.3 5 100.0 1 50.0 3 71.4 5
Cash shortage 16.7 1 100.0 1 33.3 2 26.6 2
Not profitable 16.7 1











Table 37. Division of hired labor by gender and agricultural production tasks in Ada, Lume, and Gimbichu
woredas, Ethiopia

Ada Lume Gimbichu

Production task MHH FHH MHH FHH MHH FHH

Weeding
Number of days
Male labor 19.0 11.0 16.9 25.0 33.3 19.0
Female labor 14.7 10.4
Child labor 2.0 2.0 8.0
Cash payment (Birr)
Male labor 4.5 4.0 4.2 5.0 4.6 4.5
Female labor 4.7 3.9
Child labor 1.0 3.0 4.0

Harvesting
Number of days
Male labor 28.1 23.2 38.4 22.8 30.6 28.2
Female labor 5.5 8.0 -
Cash payment (Birr)
Male labor 8.9 8.6 8.9 8.4 5.3 6.0
Female labor 6.0 5.0 -
Cash equivalent (Birr)
Male labor 5.2 5.2 4.2 4.4 4.4 4.8
Female labor 6.0 5.3 -

Transporting
Number of days
Male labor 9.7 -19.0 1.0 7.0 28.0
Female labor
Cash payment (Birr)
Male labor 5.5 -5 7.0 5.0 5.0
Female labor
Cash equivalent (Birr)
Male labor 6.5 -4 6.8 3.0 3.0
Female labor -

Threshing
Number of days
Male labor 9.7 -28.4 22.3 8.0 9.0
Female labor 8.0 -
Cash payment (Birr)
Male labor 6.8 -5.7 5.7 8.8 4.3
Female labor 7.0 -
Cash equivalent (Birr)
Male labor 5.4 -3.8 3.8 6.8 5.0
Female labor -











6.4 Use of Family Labor in Non-Agricultural Production Tasks


6.4.1 Food preparation
Household members' involvement in various non-agricultural tasks is shown in Table 38. Food
preparation is generally the responsibility of women in Ethiopia and no male heads of household
were involved. In all woredas, wives, daughters, and relatives participated in food preparation. Non-
relatives also participated in MHHs in Gimbichu and in FHHs in Ada and Lume. Relatives were more
involved in food preparation than wives across all three woredas, perhaps because of a cultural


Table 38. Division of household labor by gender and non-agricultural tasks in Ada, Lume, and Gimbichu woredas, Ethiopia

Ada Lume Gimbichu
Task MHH FHH MHH FHH MHH FHH


Food preparation
Average number of days/week
Head
Wife
Son
Daughter
Relatives
Non-relatives
Average hours per day
Head
Wife
Son
Daughter
Relatives
Non-relatives
Fetching water
Average number of days/week
Head
Wife
Son
Daughter
Relatives
Non-relatives
Average hours per day
Head
Wife
Son
Daughter
Relatives
Non-relatives
Childcare
Average number of days/week
Head
Wife
Son
Daughter
Relatives
Non-relatives
Average hours per day
Head
Wife
Son
Daughter
Relatives
Non-relatives


3.02


(Continued) -+











Table 38. (Cont'd.)
Ada Lume Gimbichu
Task MHH FHH MHH FHH MHH FHH
Social activities
Average number of days/week
Head 2.3 2.3 2.0 4.4 2.7 2.0
Wife 2.1 4.7 2.0 0.7
Son 1.5 1.8 2.0 2.1 4.5
Daughter 2.0 1.0 4.8 5.2 3.6 4.0
Relatives 1.0 3.5 4.0
Non-relatives 2.8 1.3 3.3 2.0
Average hours per day 2.7
Head 4.5 4.8 4.2 3.6 5.6 4.5
Wife 4.7 3.6 3.9
Son 2.5 3.0 5.0 4.0 6.5 5.1
Daughter 6.5 3.0 3.4 2.7 6.0 3.4
Relatives 5.0 6.5 5.0
Non-relatives 2.1 3.3 5.3 3.8
Making handicrafts
Average number of days/week
Head 3.3 3.0
Wife 2.9 2.3
Son
Daughter 4.3 2.6 3.8 4.0 4.0
Relatives 4.0
Non-relatives 2.0 1.0
Average hours per day
Head 3.0 5.3
Wife 3.0 3.5
Son
Daughter 3.3 3.0 2.5 5.0 6.0
Relatives 3.0
Non-relatives -- 2.0 -1.0



tradition that considers it disrespectful for wives to cook in the presence of relatives. The average
number of days spent per week on food preparation in MHHs varied from 3.6 days by daughters in
Gimbichu to 7 days by relatives in both Ada and Lume.


In FHHs, the average number of days spent on food preparation per week ranged from 4.1 days by
daughters in Gimbichu to 7 days by heads, relatives, and non-relatives in Ada and non-relatives in
Lume. With the exception of Gimbichu, female heads of households, like wives in MHHs, spent
fewer days per week on food preparation than other household members. Again, relatives and non-
relatives spent more days per week on food preparation than other family members. In general,
household members spent on average relatively fewer days on food preparation in FHHs than in
MHHs, although the difference was not significant.


The average hours spent per day on food preparation in MHHs ranged from 2.9 hours by daughters
in Gimbichu to 6.5 hours by daughters in Lume. The average number of hours spent by all household
members on food preparation per day was highest in Lume followed by Gimbichu and Ada. Among
members of the farm household, wives spent fewer hours per day on food preparation than
daughters, relatives, and non-relatives. With the exception of Lume, daughters in Ada and Gimbichu
spent more hours per day on food preparation than any other family member.










In FHHs, the average hours spent per day on food preparation ranged from 2.6 hours by heads and
daughters in Gimbichu to 6.5 hours by non-relatives in Lume. Female heads of household spent on
average 3.5, 4.2, and 2.6 hours on food preparation in Ada, Lume, and Gimbichu, respectively.
However, daughters in Ada, non-relatives in Lume, and relatives in Gimbichu spent on average more
hours per day on food preparation than female heads of household. With the exception of Ada,
household members in MHHs spent on average more hours per day on food preparation than
household members in FHHs, although the difference was not significant.

6.4.2 Fetching water
In both MHHs and FHHs in all woredas, the average time spent per day on fetching water was
around 2 hours, an indication that water points were relatively close to homesteads. All household
members in MHHs spent more than 4 days each week fetching water, with the exception of relatives
in Lume and sons in Ada. The average number of days spent fetching water per week varied from 2
by relatives in Lume to 7 by the head, wife, relatives, and non-relatives in Ada. Heads of households
spent an average of 7 days per week in Ada and 4.3 days in Lume and Gimbichu fetching water,
while wives spent 7 days in Ada, 6.2 days in Lume, and 5.6 days in Gimbichu. With the exception of
Ada, wives spent more days per week fetching water than male heads. Daughters spent an average
of 6.8 days, 6 days, and 5.6 days per week fetching water in MHHs in Ada, Lume, and Gimbichu,
respectively.

In general, household members in FHHs spent more days per week fetching water than household
members in MHHs. In FHHs, the average number of days spent fetching water ranged from 3.5 by
non-relatives in Lume to 7 by relatives in Ada. Female heads spent on average 6.2, 6, and 4.7 days
fetching water in Ada, Lume, and Gimbichu, respectively, while relatives spent 7, 5, and 5.8 days
per week fetching water in Ada, Lume, and Gimbichu. Daughters and relatives spent 5.6 and 7 days
in Ada, 6 and 5 days in Lume, and 5.5 and 5.8 days in Gimbichu, respectively.

6.4.3 Childcare
Childcare is one of the reproductive roles of gender. In both MHHs and FHHs in Ada and Lume,
daughters were usually responsible for childcare, while in Gimbichu it was the wife in MHHs and the
head in FHHs.

In MHHs, household heads spent on average 6.7 days per week in Ada and 3 days per week in
Gimbichu taking care of children. Male heads in Lume did not spend time caring for children. Wives
spent 6.4, 6.8 and 5.5 days per week in Ada, Lume, and Gimbichu, respectively, looking after
children. Daughters spent 6.1, 6.8 and 3.5 days per week in Ada, Lume and Gimbichu, respectively,
while non-relatives spent 7 and 1 days per week in Lume and Gimbichu, respectively.

In FHHs, the responsibility for children rested with the head, daughters, and relatives. Heads of
household and daughters spent on average 7 days per week caring for children in both Ada and
Lume while the head spent 5.8 days per week in Gimbichu. In Gimbichu only the household head
cared for the children. Relatives spent 1 day in Ada and 7 days per week in Lume caring for
children. In Lume, non-relatives spent on average 7 days per week caring for children.










In MHHs, household heads spent on average 3 and 3.5 hours per day on childcare in Ada and
Gimbichu, respectively, while in Lume, the head did not spend any time on childcare. Wives, on the
other hand, spent 6, 5.1, and 5.8 hours per day in Ada, Lume, and Gimbichu, respectively.
Daughters spent 7.5, 5.5, and 1.1 hours per day in Ada, Lume, and Gimbichu, respectively. Non-
relatives spent 4 hours per day in Lume and Gimbichu while they did not spend any time on
childcare in Ada.

Female household heads spent on average 6.1 hours per day in Ada, 5.3 hours per day in Lume and
7.1 hours per day in Gimbichu on childcare while daughters spent 7.5 and 5.7 hours per day in Ada
and Lume, respectively. Relatives spent 6 and 8 hours per day in Ada and Lume, respectively, while
non-relatives spent 6 hours per day in Lume.

6.4.4 Firewood collection
Almost all household members collected firewood. In MHHs, the head spent 5 and 1 days per week
collecting firewood in Ada and Lume, respectively, while in Gimbichu the head did not collect
firewood. Wives spent 3.6, 3.8, and 2.6 days per week collecting firewood in Ada, Lume, and
Gimbichu, respectively, while daughters spent 4.1, 2, and 2.5 days per week. Relatives and non-
relatives also gathered firewood.

In FHHs, the head and daughters were mainly responsible for collecting firewood. Household heads
spent 3.6, 3.3, and 3.8 days per week collecting firewood in Ada, Lume, and Gimbichu,
respectively, while daughters spend 5.2 and 3.3 days per week in Ada and Lume.

6.4.5 Social activities and handicraft making
Household members also spent time attending local festivals, funerals, wedding ceremonies, and
other social gatherings. In general, MHHs and FHHs spent the same amount of time on social
activities.

Household members in all three woredas spent little time making handicrafts. Across woredas, more
daughters participated in this activity than any other household member.



7.0 GENDER DIFFERENTIALS IN AGRICULTURAL PRODUCTION,
UTILIZATION, AND FOOD AVAILABILITY

In sub-Saharan Africa, agriculture is the livelihood of 69% of the economically active population. The
productive capacities of natural resources on the continent depend on the productive capacities of its
people. In rural communities, producers and consumers live in the same household and are often the
same people. The way rural households function and make decisions and their visions of the future
have long been recognized as essential information for planners and policymakers in the agricultural
sector. What is less frequently recognized is the significance of the consequences of different levels
and patterns of consumption and the effects of agricultural decisions on household food security and
nutritional status of both the producers and consumers in rural and urban areas (FAO 1997).










Food security may have different meanings for different people. The International Conference on
Nutrition, held in Rome in 1992, defined food security as "access by all people at all times to the
food needed for a healthy life" (FAO/WHO 1992). Essentially, to achieve food security a country
must achieve three basic aims: ensure adequate quantity, quality, and variety of food; optimize
stability in the flow of food supplies; and secure sustainable access to available food supplies by all
who need them. Adequate food availability at the national, regional, and household levels, obtained
through markets and other channels, is the cornerstone of nutritional well being. At the household
level, food security implies physical and economic access to foods that are adequate in terms of
quantity, nutritional quality, safety, and cultural acceptability to meet each person's needs. Household
food security depends on adequate income and assets, including land and other productive resources.
It is ultimately associated with access to nutritionally adequate food at the household level, i.e., the
ability of households or individuals to acquire a nutritionally adequate diet at all times.

Agricultural decisions may result in a direct change in diet or in the quantity, quality, variety, and
safety of food available in a particular community. Frequently, the effect is a change in access to food
for a particular sector of society or community because of fluctuations in food prices or in household
income. All of these changes affect consumers' consumption of food, health, and productivity. The
poor, who do not have the resources or adequate stocks to withstand a crisis and to maintain
household food security on a sustainable basis, are especially affected by these changes (FAO 1997).

7.1 Methodology
Food balance data simply provide information on the quantity of food available to the consumer.
Actual food consumption data, on the other hand, help to identify the types of food consumed and
nutritional problems experienced by a certain population group. Food consumption data may come
from two types of surveys: by actually weighing the food consumed in the household and/or by
collecting data on household food expenditure (the quantity and monetary value of food items
acquired and/or consumed by households). Surveys of actual consumption rather than expenditure
yield more representative and complete estimates of "habitual" food intake, as information such as
food wastage and food produced at home is accounted for (FAO 1997).

Table 39 shows the production and consumption patterns of each household in Ada, Lume, and
Gimbichu for main food crops. From these quantities of grains and pulses the calorie equivalent was
calculated (using the conversion factors of FAO's Country Nutrition Profile of Ethiopia 1997). It
should be noted from the outset that these data by no means represent the food security status of the
household. Rather, they give a rough estimate of whether there were gender differentials in daily
energy production and consumption from the major food crops produced in the study area.

7.2 Results and Discussion
The Ethiopian diet is based mainly on cereals, which provide about 70% of calorie intake. Ada,
Lume, and Gimbichu are known for their cereal and pulse production, and thus the calorie intake is
based on cereal and pulse production (Mulat et al. 1995).











The mean daily energy (in calories) of production per person per day was about 2,278 for MHHs
and 2,291 for FHHs. These amounts are higher than the average national energy supply from
cereals (1,713 calories per person per day) (FAO 1996). MHHs and FHHs did not differ
significantly in total calorie production per person per day. This finding is different from other
findings reported in the literature, in which FHHs are usually categorized as being at the chronic
stage in terms of access to food (Debebe and Maxwell 1992).


The total cereal calorie consumption of both MHHs and FHHs was lower than the daily
requirement recommended by FAO/WHO/UNU (1,540 calories per person per day for cereals
only). The amount of calories consumed per person per day was about 894 and 877 for MHHs
and FHHs, respectively. The main explanation for the low calorie consumption per person per day



Table 39. Mean quantities of agricultural production and consumption (quintal) by type of crop in Ada, Lume, and
Gimbichu woredas, Ethiopia

Ada Lume Gimbichu
Crop production and
utilization MHH FHH MHH FHH MHH FHH

Barley
Grain produced 7.66 4.92 4.67 3.81 3.00
Grain consumed 3.08 2.13 2.73 2.01 1.00
Chickpea
Grain produced 3.66 3.73 5.78 5.30 4.22 4.16
Grain consumed 2.03 2.06 1.51 1.43 1.01 0.95
Faba beans
Grain produced 3.83 4.00 2.16 1.33 1.25 2.31
Grain consumed 1.96 2.00 0.99 0.70 0.87 0.50
Field peas
Grain produced 3.13 1.56 2.48 1.49 2.16 1.70
Grain consumed 1.35 0.82 0.95 0.74 0.41 0.65
Haricot beans
Grain produced 1.36 1.50 3.85 3.95
Grain consumed 0.98 1.12 1.38
Lentils
Grain produced 2.75 0.65 1.08 8.00 6.40
Grain consumed 1.60 0.50 0.70 0.68 0.61
Maize
Grain produced 4.00 7.62 5.73
Grain consumed 2.40 3.29 2.28
Rough peas
Grain produced 2.52 3.28 2.86 0.50 2.25 1.41
Grain consumed 1.93 1.50 0.90 0.35 0.90 0.79
Tef
Grain produced 14.82 12.19 16.70 13.65 6.44 5.05
Grain consumed 7.02 5.55 6.13 4.69 2.62 1.81
Wheat
Grain produced 7.54 5.65 8.56 4.79 20.16 15.05
Grain consumed 3.47 2.57 3.68 2.65 6.80 5.65










is that the households consumed small portions of the cereals and pulses they produce. Figures 2 and
3 show the wheat and teff utilization by gender of the household head. About 38% of MHHs and
42% of FHHs consumed the wheat they produced, while 40% of MHHs and 39% of FHHs
consumed their tef produce. The amount of food that farmers purchased after they sold their
produce was not included in the calculation of calories consumed, and therefore the daily calorie
intake is underestimated.

The three woredas differed significantly with respect to the calorie production per person per day.
The mean calorie production per person per day was 1,707, 2,668, and 2,478 in Ada, Lume, and
Gimbichu, respectively. The daily calorie production per person was below the joint recommendation
of FAO/WHO/UNU only in Ada. The daily cereal calorie consumption per person was about 810 in
Ada, 775 in Gimbichu, and 1,083 in Lume. The calorie consumption per person per day was
significantly higher in Lume than in Ada and Gimbichu (p<0.05). In Ada, environmental degradation
has significantly reduced land availability, and hence contributed to food insecurity (Debebe and
Maxwell 1992).



8.0 GENDER DIFFERENTIALS IN AGRICULTURAL PRODUCTIVITY


Overcoming stagnating agricultural productivity and food insecurity hinges on increasing agricultural
productivity. In many parts of sub-Saharan Africa, where subsistence agriculture predominates, it is
paramount to place strong emphasis on increasing the productivity of labor, land, capital, and other
resources. Agricultural productivity may vary by gender if men and women use different technologies
or different quantities of production factors, or if there are differences in the quality of these factors
(Saito, Mekonnen, and Spurling 1994). In this section, we use a Cobb-Douglas production function
to assess whether agricultural productivity varies in MHHs and FHHs.


45 50
40.-------------- 450-------------------------------------
40 ______ ____________ 45 --




2525
3 M25
20--
20
15-- 15

10 -- 10

5 -- 5 ---

0 0
Consumed Kept as seed Sold Given away Consumed Kept as seed Sold Given away

Figure 2. Utilization of wheat by gender of household Figure 3. Utilization of tef by gender of household head
head in Ada, Lume, and Gimbichu woredas, Ethiopia. in Ada, Lume, and Gimbichu woredas, Ethiopia.










8.1 The Production Function


A Cobb-Douglas production function, in which coefficients of input variables were estimated from
survey data, was used to used to assess the extent to which productivity was affected by various
inputs. The variables were hypothesized to influence the adoption of improved wheat varieties either
positively (+), negatively (-), or positively and/or negatively (+/-). The production function can be
expressed as:

[1] Y= Poi *X X 1* X02,... .Pn,i
~ ,i 1,i 2,i "'" n,i

i =1,2 (1= MHH; 2= FHH),
where:
Y = gross value of farm output in Birr2;
X, = age of the household head (yr) (+/-);
X2 = total family labor used for agricultural production (weighted by 1 h/yr for
men and women and 0.5 h/yr for children) (+);
X3 = farm size (ha) (+);
X4 = number of tropical livestock units (weighted by 1 for cattle, 0.14 for goats
and sheep, 0.43 for donkeys, and 0.02 for poultry) (+);
X, = amount of inorganic fertilizer used (kg N/ha) (+);
X, = quantity of herbicide (ha) (+);
X, = quantity of insecticide (ha) (+);
X, = hired labor for agricultural production (h/yr) (+);
X, = extension contact (dummy variable = 0 if no contact, 1 if contact) (+);
X10 = education of household head (dummy variable = 0 if illiterate, 1 if literate) (+);
b0 = constant;
b,'s = estimated parameters.

The Cobb-Douglas production function is used widely because of a number of desirable properties. One
of these desirable properties is that 3P's are the elasticities of output with respect to the relevant input. A
critical assumption is that 3P's are positive and each is less than one. The sum of the 3P's also provides
the returns to scale parameter. Another attractive property of the production function is that,
econometrically, it is easy to estimate, because in its log form, the parameters are linear and can be
estimated easily using the Ordinary Least Squares method. However, since it treats input choices as
exogenous, it is susceptible to management bias. Ideally, input choices should be modeled
simultaneously with the production function, but this usually requires price variation. In addition, the
Cobb-Douglas production functional form is not flexible for modeling complements and substitutes,
such as the relationship of land and labor or the role of labor availability in choosing variable inputs
(Saito, Mekonnen, and Spurling 1994).





2 Output is calculated as the farmgate price multiplied by yields (including straw yields). The crops included in the total output
were barley, chickpeas, faba beans, field peas, lentils, maize, rough peas, teff, and wheat.











8.2 Model Results


The estimates of the Cobb-Douglas production function are shown in Table 40. The coefficients of
multiple determination adjusted for degrees of freedom indicated that the variation in gross value of
output per hectare associated with the factors of production specified in the model was 72% and
82% in MHHs and FHHs, respectively.


The significant factors affecting gross value of output for MHHs were farmer's age, family labor,
farm size, livestock units, and inorganic fertilizer; in FHHs, they were family labor, farm size,
livestock units, inorganic fertilizer, hired labor, and extension contact.


In MHHs, a farmer's age had a significant and negative impact on gross output. A 10% increase in
the farmer's age resulted in a 2.1% decrease in gross output. Family labor had a significant and
positive impact on the gross value of output for both MHHs and FHHs. A 10% increase in the
amount of labor resulted in a 2.3% and 1.8% increase in gross value of output for MHHs and FHHs,
respectively. Farm size had a positive and significant impact on the gross value of output for MHHs
and FHHs. A 10% increase in farm size resulted in a 5.6% and a 7.2% increase in gross value of
output for MHHs and FHHs, respectively.


The number of livestock had a positive and significant impact on the gross value of output for both
households. A 10% increase in the number of livestock for MHHs and FHHs resulted in a 1.3% and
0.7% increase in the gross value of output, respectively. The amount of inorganic fertilizer used had a
positive and significant impact on the gross value of output in both types of households as well. A 10%
increase in the amount of fertilizer resulted in a 1% increase in gross value of output for both MHHs and
FHHs. Saito, Mekonnen, and Spurling (1994) found that the use of inorganic fertilizer on plots managed
by women in Kenya increased the gross output to 1.6% compared to 1.3% for plots managed by men.


Table 40. Estimates of Cobb-Douglas production function by gender in Ada, Lume, and Gimbichu woredas, Ethiopia

MHH FHH

Regression Regression
Variable coefficient (b) T-statistic coefficient (b) T-statistic


Intercept 6.6271 9.41*
Age of household head (yr) -0.2112 1.80***
Family labor (h/yr) 0.2342 2.54**
Farm size (ha) 0.5592 7.46*
Tropical livestock units (number) 0.1332 2.67*
Inorganic fertilizer (kg N/ha) 0.1012 1.69***
Herbicide (I/ha) 0.0679 1.36
Insecticide (I/ha) 0.0367 0.43
Hired labor (h/yr) -0.0007 0.05
Extension contact (dummy) -0.0461 0.77
Education of head (dummy) -0.0288 0.44
Adjusted R2 0.72
F-test 26.2*
Sample size (N) 100
Note: = significant at p<0.01; = significant at p<0.05; and ** = significant at p<0.1.


6.3330
-0.0817
0.1809
0.7551
0.0719
0.1027
0.0386
0.1520
0.0278
-0.2844
-0.0922


8.38*
0.74
2.05**
10.49*
1.63***
1.92***
0.54
0.89
2.56**
4.04*
1.48


0.82
35.8*
77










The use of hired labor for agricultural production positively and significantly affected the gross output
for FHHs, although the impact was only very marginal, because a 10% increase in the amount of
hired labor resulted in a 0.03% increase in gross output. Extension contact had a negative and
significant impact on the gross output for FHHs. The gross output was lower for FHHs that had
contact with extension services.

Allocative efficiency can be determined by comparing the marginal value product (MVP) of a factor with its
opportunity cost (factor price). The MVP of a factor is the additional return from adding one more unit of
that factor, holding all other inputs constant. In this study we have calculated the MVP using a 10%
increase in the use of that factor. An MVP that exceeds its opportunity cost suggests that there is scope
for raising productivity by increasing the use of that factor. Conversely, increasing the use of a factor for
which the MVP is less than the associated opportunity cost will decrease productivity.

MVPs for family labor, farm size, and fertilizer were determined for MHHs and FHHs. Table 41 shows the
MVPs and factor prices for the significant variables for MHHs and FHHs using a 10% increase in the
actual use of inputs. Thus, for FHHs 126.3 hours (10% of 1,263 h/yr) were added. The factor price was
then calculated using a daily wage rate of Birr 5.5/day. The duration of a working day was 8 hours.
Factor prices for farm size and fertilizer were calculated by using the average rent (Birr 31 1/kert) and
actual fertilizer prices (Birr 5.2 per kilogram of N/ha). These prices were multiplied by a 10% increase in
farm size (0.3 ha for MHHs and 0.24 ha for FHHs) and fertilizer use (6.4 kg N/ha for MHHs and 4.4 kg
N/ha for FHHs). The MVP of family labor compared to its price (wage rate) is higher in MHHs and lower
in FHHs, indicating that MHHs could increase their productivity by using more family labor. A study by
Quisumbing (1993) also found that the marginal product of women's labor was lower than that of men's.
Also, Quisumbing (1996) cites a study in India in which the marginal product of male labor is greater than
that of female labor.

The MVP of farm size was lower than its factor price for MHHs and higher for FHHs; thus FHHs could
increase their productivity by cultivating more land. The MVP for inorganic fertilizer was higher than its
factor cost for both MHHs and FHHs, which indicates that both types of household could increase their
productivity by increasing their use of inorganic fertilizer.

In general, MHHs had more land, labor, and capital (particularly cattle) than FHHs. It was also shown that
MHHs had more access to formal education than FHHs.




Table 41. MVP and factor prices (in Birr) for significant variables by gender of household head in Ada, Lume, and
Gimbichu woredas, Ethiopia

MHH FHH
Factor MVP Factor price MVP Factor price
Family labor (hlyr) 145.70 114.74 82.05 86.72
Farm size (ha) 353.36 372.96 356.33 299.32
Inorganic fertilizer (kg N/ha) 62.55 33.07 47.02 23.11









The production function analysis showed that elasticities for the significant factors affecting the gross value
of output for MHHs were farmer's age (-0.21), fertilizer (0.10), farm size (0.56), labor (0.23), and livestock
(0.13). For FHHs the elasticities were fertilizer (0.10), farm size (0.76), labor (0.18), livestock (0.07), hired
labor (0.03), and extension (-0.28). The negative elasticity for extension implies that policymakers and the
Ministry of Agriculture should specifically target FHHs to mitigate extension's negative effect on gross value
of output for FHHs.

The comparison of MVP with the factor cost showed that MHHs could increase productivity by using
more labor and fertilizer, while FHHs could do so by using more land and fertilizer. MHHs (Birr 6,456/
ha) had a higher gross output compared to FHHs (Birr 4,776/ha). However, the gross value of the
output was 1.3% higher for FHHs if the average value of inputs from MHHs were used. This suggests
that no significant productivity differences between MHHs and FHHs would exist if FHHs had equal
access. Moock (1976) in Kenya also found that women obtained 6.6% more output at the mean levels
of input compared to men. Saito, Mekonnen, and Spurling (1994) found that women in Kenya
obtained about 22% more output than men if they had equal access to resources as men. Quisumbing
(1996), however, stated that these kinds of simulations should be interpreted with caution, since we do
not know how the levels of inputs could be raised for FHHs.


9.0 GENDER DIFFERENTIALS IN TECHNOLOGY ADOPTION

9.1 Factors Affecting the Adoption of Wheat Production Technologies
In 1997, about 59% and 42% of MHHs and FHHs, respectively, grew wheat. Local wheat varieties
were grown by 70% of MHHs and 86% of FHHs. A significantly higher proportion (t=5.7, p<0.05) of
MHHs (30%) grew improved wheat varieties than FHHs (14%). In FHHs, the decision to grow
improved wheat varieties was always made by the head; in MHHs it was either a joint decision between
the head and wife (55.6%) or a decision by the head alone (44.4%). About five years ago, the adoption
of improved wheat varieties was over 90% in the same study area (Workneh et al. 1994). The decline
can be partly attributed to farmers who considered their recycled improved wheat seed as local.
Furthermore, most farmers reported that the seed of the preferred wheat variety was not avalailable.

The two models frequently used in adoption studies are logit and probit (Feder, Just and Zilberman
1985). For this study, the logit model is used to analyze factors affecting decisions in MHHs and FHHs
to adopt improved wheat varieties. It is hypothesized that the use of improved wheat varieties is
influenced by a combined (simultaneous) effect of a number of factors related to farmers' objectives and
constraints (CIMMYT 1993). The variables were hypothesized to influence the adoption of improved
wheat varieties positively (+), negatively (-), or positively and/or negatively (+/-).

Following Gujarati (1988) the model is specified as:

[2] In (P/(1-P)) = Po + P1 X1 + P2 X2 + 3X,+4 X4+5 Xs6 X6+47 X7+ X8 + 1P9 X+ e

where:

X, = RADIO (owned) (+);
X2 = LUNITS (number of livestock owned) (+);










X, = MEMBCOOP (membership in a service cooperative) (+);
X4 = HHSIZE (household size) (+);
Xs = EXTCON (extension contact during the last 3 months) (+);
X, = EDUHEAD (education level of the head) (+);
X, = DISTMKT (average distance from farm to market in km) (-);
X8 = AREALAND (area of farm in hectares) (+);
X, = AGEHH (age of the household head in years) (+/-); and
e = error term.

9.2 Model Results

The logistic model explains 84% and 89% of the total variation specified in the model for MHHs and
FHHs, respectively (Table 42). The chi-square indicates that the parameters are significantly different
from zero at the 1% level for MHHs and FHHs.

The odds in favor of adopting improved wheat varieties increased by a factor of 22.1 for MHHs that
had access to extension services. MHHs benefited more from extension than FHHs. Other studies
(Moock 1976; Saito, Mekonnen and Spurling 1994) have found similar results.

Radio ownership increased the odds in favor of adopting improved wheat varieties by a factor of 5.9
for FHHs. Chilot, Shapiro and Demeke (1996) also found that radio ownership significantly and
positively affected the adoption of improved wheat varieties.


Table 42.Parameter estimates of a logistic model for factors affecting adoption of improved wheat varieties in Ada,
Lume, and Gimbichu woredas, Ethiopia

MHH FHH

Parameter Wald Parameter
Household characteristic estimate (0) Statistic estimate (1) Wald statistic

Intercept -5.2808** 5.01 -12.5913*** 0.08
Radio 0.1031 0.0136 -1.7676 0.8156
Livestock (TLU) 0.0517 0.30 0.0609 0.11
Membership in cooperative -1.8079 1.48 8.9540 0.04
Household size (number of people) 0.1958 1.71 -0.0012
Extension contact 3.0973*** 18.27 1.5066 2.27
Education 1.1862 2.08 1.8381 2.48
Distance to market (km) -0.0155 0.04 -0.1714 1.82
Farm size (kert) 0.1285* 3.21 0.1650* 3.74
Age (yr) 0.0012 -0.0160 0.09
Chi-square 47.74*** 20.42***
Overall predicted 84.3% 89.2%
N 89.0 65.0

Note: = significant at p<0.1; = significant at p<0.05; and ** = significant at p<0.01.










If farm sizes are increased, the odds in favor of adopting improved wheat varieties increase by a factor
of 1.1 for MHHs and 1.2 for FHHs. The average farm size, area under wheat, and livestock units were
significantly higher for MHHs than FHHs. If the farm size of both MHHs and FHHs increased, the
probability of adopting improved wheat varieties would increase almost equally.

The results showed that MHHs were more likely to adopt improved wheat varieties than FHHs.
However, the availability of seed of improved wheat varieties should be addressed, because lack of seed
has discouraged adoption. In FHHs, the decision to grow improved wheat varieties was always made by
the head, while in MHHs it was either a joint decision between the head and the wife (55.6%) or a
decision by the head alone (44.4%). Some studies have also shown that women and men are faced by
differential access to new technology, education, health care, and other resources (Ahmed 1985,
Stamp 1989, Abu and Oppong 1987).



10.0 GENDER DIFFERENTIALS IN ACCESS TO RURAL INSTITUTIONS

10.1 Credit Services
Credit availability, by increasing risk-taking capacity, increasing the ability to invest, and improving access
to other productive inputs and assets, is very important for improving farm productivity and returns.

The high price of fertilizer and agro-chemical supplies (implying the need for more cash) and the non-
availability of loans (as much as required) from the bank and informal credit sources are among the major
problems of crop production. This problem is very much reflected in the list of items (fertilizer, seed,
chemicals, and equipment, in order of importance) purchased by farmers with the credit they obtained
(Table 43).

All the sampled MHHs, and 96.3% and 95.7% FHHs in Ada and Lume, respectively, said they used
some kind of credit facility. Although credit sources included the bank, co-operative, local lenders, and
relatives, farmers' use of credit varied among the woredas.

More than 85% of MHHs and FHHs across the woredas required collateral to obtain credit. The type of
collateral used was generally co-signature in Ada (71.9% for MHHs and 92% for FHHs) and Lume (97%
for MHHs and 95.5% for FHHs). This form of collateral was less common in Gimbichu where group
guarantee was used more frequently (76.7% for MHHs and 82.6% for FHHs). Land and future crop were
used less frequently as collateral. The head of the household was mainly responsible for putting up
collateral in all households.

All farmers in Ada reported that they obtained credit from the bank and not from the service co-
operative, whereas farmers in the other two woredas (except 4.2% of FHH in Lume) said they obtained
credit from their respective service co-operatives. Over 85% of both MHHs and FHHs in the three
woredas did not secure any credit from local moneylenders. Sample households in Lume woreda are
different because they use irrigation throughout the year and plant a lot of cash crops, especially
horticultural crops like tomatoes.













The credit that farmers received from the co-operative was tied to the extension program. Access to credit

in this case was possible if one was willing to buy inputs included in the extension program at a given

price. As Chipande (1987) has pointed out, extension programs use credit as a means of persuading

farmers to adopt a certain package of technology. When the initial credit strategy is analyzed, two




Table 43. Some selected variables on credit facilities by gender of household head in Ada, Lume, and Gimbichu
woredas, Ethiopia

Ada Lume Gimbichu

MHH FHH MHH FHH MHH FHH

(%) (Respondents) (%) (Respondents) (%) (Respondents) (%) (Respondents) (%) (Respondents) (%) (Respondents)


Use credit 100.0 32
Collateral is required 100.0 32
Type of collateral
Co-signature 71.9 23
Group guarantee 15.6 5


Land
Future crop
Who put up collateral?
Head
Wife
Head and wife


9.4 3
3.1 1

100.0 32


96.3 26
96.2 25

92.0 23

8.0 2


91.7 22


Son -8.3 2
Where was credit obtained?
Bank 100.0 32 100.0 27
Co-operatives
Local money lenders 15.6 5 7.4 2
Relatives 3.7 1
Items purchased by credit
Fertilizer 100.0 32 100.0 26
Seed 12.5 4 11.5 3
Chemicals 3.1 1
Equipment


100.0 35 95.7 23 100.0 33 100.0 27
94.3 33 95.7 22 90.9 30 85.2 23


97.0 32
3.0 1




97.0 32
3.0 1


95.5 21 23.3 7 17.4 4
4.5 1 76.7 23 82.6 19




100.0 22 96.7 29 87.0 20


13.0 3


100.0 35
14.3 5
8.6 3

100.0 35
42.9 15
42.9 15
34.3 12


95.8 23 100.0 33 100.0 27
4.2 1 -3.7 1


100.0 23 100.0 33 100.0 27
21.7 5 3 1
30.4 7
17.4 4


Reason for not using credit
No need
Lack of awareness
Membership in a
service co-operative 75.0 24
Co-operative Services
Marketing services
Technical services 8.3 2
Credit services
Farm input 62.5 15


Membership in a
women's group
Main activities of


46.9 15


100.0 1


100.0 1


74.1 20 100.0 35 100.0 24 97.0 32 100.0 27


85.0 17

51.9 14


women's group
Business activities
Social activities
Women's group affiliation
with another agency or institution
Research 100.0 15 78.6 11
Aid agencies 100.0 15 100.0 14
Average amount (Birr) and
duration (months) of credit
from co-operatives and bank
Amount 762.71 32 635.00 26
(t -value) NS


68.6 24
34.3 12
97.1 34
77.1 27


58.3 14 6.3 2 14.8 4
16.7 4 50.0 16 51.9 14
95.8 23 100.0 32 96.3 26
54.2 13 75.0 24 74.1 20


94.3 33 95.8 23 93.1 27 92.6 25


84.8 28 91.3 21
100.0 33 100.0 23 100.0 27 100.0 25


18.2 6
100.0 6



1,046.57 35
2.09*


9.1 2
100.0 2



736.69 23 1,296.21 33 1,056.77 27
NS










conclusions can be drawn. First, the credit package did not take into account that land resources,
managerial ability, and labor resources varied across households, so a number of farmers received credit
packages that were inappropriate for their needs and resources. Naturally such households defaulted
(Kinsey 1974). Second, there was a close relationship between defaulting and the mean labor units
available to the household. This finding suggests that labor-deficient households were offered labor-
intensive packages that they were unable to manage adequately. This resulted in poor yields that in turn led
to defaults. The observation that labor-deficient households were likely to default led to a reorientation of
the criteria for granting credit. During the second stage, the emphasis was on the "ability to repay"
principle. Credit was granted only to those deemed credit-worthy and FHHs lost out.


The average amount of credit obtained from the bank by households in Ada amounted to Birr 762.71
(MHHs) and Birr 635 (FHHs). The amount obtained from cooperatives in Lume was relatively higher.
However, FHHs received smaller amounts than their male counterparts from either the bank or
cooperative. No credit programs were designed especially for women.


More women were members of women's groups in Lume and Gimbichu than Ada. All group activities
were social in Gimbichu but included business actitivies as well in Lume. The women's group in Ada
was affiliated to a research institute, while in Lume it was affiliated to aid agencies.

In MHHs, the decision to use credit was made either by the head alone or in conjunction with the
wife. In FHHs, the head made the decision with some participation from the son (Table 44).

10.2 Extension Services

Adams (1983) defined extension as assistance to farmers to help them to identify and analyze their
production problems and to become aware of the opportunities for improvement. Rolling (1988) on the
other hand defined agricultural extension as a communication process geared towards bringing voluntary
behavioral change. Ban van den and Hawkins (1988) have defined extension as the conscious
communication of information to help people form sound opinions and make good decisions.


Table 44. Decision-making on the use and payment of credit by gender of household head in Ada, Lume, and
Gimbichu woredas, Ethiopia

Ada Lume Gimbichu
MHH FHH MHH FHH MHH FHH

(%) (Respondents) (%) (Respondents) (%) (Respondents) (%) (Respondents) (%) (Respondents) (%) (Respondents)
Who decides whether to use credit?
Head 37.5 12 84.6 22 80.0 28 100.0 23 78.8 26 92.6 25
Head and wife 62.5 20 20.0 7 -3.0 1
Son 15.4 4 3.7 1
Who repays the credit?
Head 97.1 34 100.0 23 40.6 13 100.0 26 87.9 29 88.9 24
Wife 2.9 1 3.0 1
Head and wife -59.4 19
Son 59.4 19










Extension provided agricultural and vocational training on the use of fertilizer, insecticides, improved
seed, land use practices, animal husbandry, and home economics. In Lume, 48.6% of MHHs
reported that they received agricultural and vocational training as compared to 9.4% in Ada. In
FHHs, it was 3.7% in Ada and 25% in Lume (Table 45).

All households in Lume and 37.5% of MHHs and FHHs in Ada indicated that they were taught how to
use fertilizer. In Lume 62.9% of MHHs and 45.8% of FHHs received information on the use of
insecticides and 51.4% of MHHs and 16.7% of FHHs received information on the use of improved
seed. About 91% of MHHs and 62.5% of FHHs in Lume were taught land use practices, and all
households were taught animal husbandry. In Ada farmers were taught neither land use practices nor
animal husbandry. In Lume 54.2% of FHHs were taught home economics but none were taught
in Ada.

In Gimbichu, the extension messages were related to credit services (all MHHs and 96.3% of FHHs), farm
inputs (75% of MHHs and 74.1% of FHHs), and marketing services (6.3% of MHHs and 14.8% of FHHs).

The extension contact of households differed from woreda to woreda in the last three months. About
81% (MHHs) and 88.9% (FHHs) in Ada and 45.7% (MHHs) and 58.3% (FHHs) in Lume reported
that they had no contact with extension agents (Table 46). In both cases the figure for FHHs was
slightly lower, indicating again their lack of access to extension facilities. Farmers in Lume had better
contact with extension agents than in Ada, but here as well, it was relatively less for FHHs.

In Kakamega District of Kenya, 40% of the women interviewed knew nothing about the extension
services' credit program and no woman manager had ever obtained a loan (Due, Mollel, and Malone
1987). A survey in Nigeria's Ogun State Agricultural Development Project revealed that extension
agents visited only 10% of women farmers every week, whereas 70% of male farmers were visited
weekly (Elabour-Idemudia 1991). Apart from outright discrimination, a number of factors account for
women's low participation in extension programs. The methods used to disseminate technical
information, such as the contact farmer approach and the use of training centers, tend to channel
information to farmers who have more resources and who are generally men (Berger, Delancey, and
Mellencamp 1984).


Table 45. Type of extension services provided by gender of household head in Ada, Lume, and Gimbichu
woredas Ethiopia
Ada Lume Gimbichu
MHH FHH MHH FHH MHH FHH
(%) (Respondents) (%) (Respondents) (%) (Respondents) (%) (Respondents) (%) (Respondents) (%) (Respondents)
Agricultural and
vocational training 9.4 3 3.7 1 48.6 17 25.0 6 33.3 11
Fertilizer use 37.5 12 37.0 10 100.0 35 100.0 24 87.9 29 51.9 14
Insecticide use 3.1 1 62.9 22 45.8 11 36.4 12 11.1 3
Improved seed use 12.5 4 11.1 3 51.4 18 16.7 4 30.3 10 14.8 4
Land use practices 91.4 32 62.5 15 18.2 6 11.1 3
Animal husbandry 100.0 35 100.0 24 9.1 3
Home economics 3.1 1 54.2 13 15.2 5 3.7 1











Table 46. Selected variables on extension activities by gender of household head in Ada, Lume, and Gimbichu
woredas, Ethiopia

Ada Lume Gimbichu

MHH FHH MHH FHH MHH FHH

(%) (Respondents) (%) (Respondents) (%) (Respondents) (%) (Respondents) (%) (Respondents) (%) (Respondents)

Extension contact in
the last three months 18.8 6 11.1 3 54.3 19 41.7 10 36.4 12 14.8 4
Once in three months 33.3 2 33.3 1 15.8 3 30.0 3 30.8 4
Once a month -47.4 9 50.0 5 15.4 2 100.0 4
Twice a month 16.7 1 15.8 3 38.5 5
Thrice a month -15.8 3 10.0 1 -
Four or more times 50.0 3 66.7 2 5.3 1 10.0 1 15.4 2

Gender of extension agent
Male -57.9 11 60.0 6 76.9 10
Female 100.0 6 100.0 3 42.1 8 40.0 4 23.1 4 100.0 4
Individual
extension visit 100.0 6 100.0 3 100.0 20 100.0 10 100.0 13 100.0 4
Preferred gender of agent
Male 3.2 1 7.4 2 37.1 13 30.4 7 9.5 2
No preference 96.8 30 92.6 25 62.9 22 69.6 16 90.5 19 100.0 17


The frequency of contact for households varied from one to four or more times a month (Table 46).
Respondents from Lume had more extension contact (where the highest frequency reported was once a
month), although some said once in three months and others three times a month. What can be discerned
from this pattern is that there is no definite visit schedule as the replies were so divergent.


All households in Ada indicated that they had a female extension agent. In Lume about 40% of MHHs
and FHHs had contact with a female extension agent. Visits were usually on an individual basis. A majority
of the sampled households were indifferent to the gender of the extension agent, with the exception of
Lume, where 37.1% of MHHs and 30.4% of FHHs indicated they preferred male extension agents.


In Ada, Lume, and Gimbichu, more than half of MHHs indicated that extension agents talked to male
members only, while most FHHs in Lume and Gimbichu said that extension agents talked to female
members only. On the other hand, 50%, 47.4%, and 50% of MHHs in Ada, Lume, and Gimbichu,
respectively, indicated that extension agents talked to both male and female members as compared to
100%, 20%, and 50% of FHHs in Ada, Lume, and Gimbichu (Table 47).


Table 47. Gender of household heads with whom extension agents make contact in Ada, Lume, and Gimbichu
woredas, Ethiopia
Ada Lume Gimbichu
MHH FHH MHH FHH MHH FHH

(%) (Respondents) (%) (Respondents) (%) (Respondents) (%) (Respondents) (%) (Respondents) (%) (Respondents)
Male members only 50.0 3 52.6 10 10.0 1 50 6
Female members only 70.0 7 100.0 4
Male and female
members 50.0 3 100.0 3 47.4 9 20.0 2 50 6











Eighty percent of MHHs and 87.5% of FHHs in Ada indicated that they had not adopted
recommended practices at all (Table 48). On the other hand, more than half of MHHs and FHHs in
Lume said they adopted recommended practices sometimes. Recommended practices were very
often adopted by 25.7% of MHHs and 4.2% of FHHs in Lume, whereas 11.4% of MHHs and
12.5% of FHHs said they seldom adopted them.


The reasons given for not adopting recommended practices in Ada were the lack of awareness, that
the practice was not used by anybody, and that there was no need. A few observed that the
recommendation was too technical. In Lume, some respondents said it was too risky and not
applicable.


Very few farmers in Ada and Lume (except 22.9% of MHHs in Lume and 6.1% in MHHs in
Gimbichu) were contact farmers. When asked whether they would be interested in being contact
farmers, 15.6% of MHHs in Ada, and 22.2% of MHHs and 12.5% of FHHs in Lume, were
interested. The main incentive to be a contact farmer was to get new ideas (Table 49).



Table 48. Frequency of adopting recommended practices in Ada, Lume, and Gimbichu woredas, Ethiopia
Ada Lume Gimbichu
MHH FHH MHH FHH MHH FHH

(%) (Respondents) (%) (Respondents) (%) (Respondents) (%) (Respondents) (%) (Respondents) (%) (Respondents)
Frequency
Very often 10.0 3 25.7 9 4.2 1 18.2 6 3.7 1
Sometimes 6.7 2 8.3 2 54.3 10 50.0 12 9.1 3 1.1 3
Seldom 3.3 1 4.2 1 11.4 4 12.5 3 18.2 6 25.9 7
Not at all 80.0 24 87.5 21 8.6 3 33.3 8 54.5 18 59.3 16
Reason for not adopting recommended practices
Too expensive 2.5 1 6.3 1
Too technical 4.2 1 44.4 8 37.5 6
Don't apply 8.3 2 33.3 1 37.5 3 -
Too risky 33.3 1 12.5 1 11.1 2 37.5 60
No need for it 33.3 8 28.6 6 33.3 1 12.5 1 11.1 2 6.3 1
No one use them 33.3 8 42.9 9 25.0 2 27.8 5 12.5 2
Lackofawareness 20.8 5 28.6 6 -- 5.6 1




Table 49. Is the household head a contact farmer, and reasons for wanting to be a contact fanner in Ada, Lume, and
Gimbichu woredas, Ethiopia
Ada Lume Gimbichu
MHH FHH MHH FHH MHH FHH

(%) (Respondents) (%) (Respondents) (%) (Respondents) (%) (Respondents) (%) (Respondents) (%) (Respondents)
Is contact farmer 22.9 8 6.1 2
Is interested in being
a contact farmer 15.6 5 22.2 6 12.5 3 22.6 7 7.7 2
Reason to be a contact farmer
To get new ideas 100.0 5 100.0 6 66.7 2 66.7 2 100.0 1
To get improved
seed 33.3 1 33.3 1










11.0 CONCLUSION AND POLICY IMPLICATIONS


On the whole the differences in endowments (land rights, education), access to technologies, factors
of production, and support services had implications for agricultural productivity in both types of
households.

The major problems faced by all farmers, male or female, were: insufficient land; shortage of family
labor; high price of fertilizer and agro-chemical supplies; lack of loans from formal and informal
sources; distance from market centers; the shortage of appropriate storage facilities; and inadequate
extension services. Of these, the biggest constraints were the shortage of land and labor, the high
cost of labor, and poor soils. While these problems affected all respondents equally, the degree and
magnitude varied between households. Differences and similarities were observed between the two
groups with respect to access to resources, division of labor, decision-making, and adoption of
improved wheat varieties.

11.1 Access to Resources
Male heads of household were more educated than female heads of household. MHHs had better
land, bigger households, and owned more cattle than FHHs. MHHs also had better access to formal
and informal financial institutions and extension services. They owned more ox plows and were thus
better able to prepare their land on time. Both types of households owned the same number of farm
implements.

11.2 Division of Labor
There was a very high and varied involvement of all household members, including male and female
heads, in various farm activities. Across all three woredas, land preparation was the responsibility of
male members of both types of households. Female household members (wives and daughters in
MHHs and heads and daughters in FHHs) were primarily responsible for non-agricultural tasks such
as food preparation, household maintenance, and childcare. Sons, relatives and non-relatives were
mainly responsible for planting. All household members participated in weeding, harvesting, and
animal husbandry.

11.3 Decision-Making
The inter- and intra-household decision-making pattern used as a proxy to portray the control that
one has over a certain resource in a household also gave some indication of the degree of power the
head of the household exercised over other household members. As result of the differences in
endowment and access to resources, however, the capacity of FHHs and MHHs to adopt improved
agricultural technology is different (see next section).

11.4 Adoption of Improved Wheat Varieties
About five years ago, the adoption of improved wheat varieties was over 90% in the study area
(Workneh et al. 1994). Sample farmers in this study, however, reported growing a smaller
percentage of improved wheat. The apparent decline in adoption of improved wheat varieties can be










attributed to the fact that many farmers considered their recycled improved wheat seed to be local,
and to the unavailability of the preferred variety of seed in the market.

The findings indicate that MHHs were more likely to adopt improved wheat varieties than FHHs.
However, the availability of seed of improved wheat varieties should be addressed, because lack of
seed has discouraged adoption

In FHHs, the decision to grow improved wheat varieties was always made by the head; slightly more
than half of MHHs made the decision jointly.

The odds in favor of adopting improved wheat varieties increased by a factor of 22.1 for MHHs who
had access to extension services, while in FHHs, radio ownership increased the adoption of improved
wheat varieties by a factor of 5.9. These data reflect the bias of extension services in favor of MHHs.
Other farmers (29% of MHHs and 26% of FHHs) constituted another source of information about
improved varieties.

With increasing farm size, adoption of improved wheat varieties increased by a factor of 1.1 for
MHHs and 1.2 for FHHs. While the average farm size, area under wheat, and livestock units were
significantly higher for MHHs than FHHs, the probability of adopting improved wheat varieties would
increase almost equally for MHHs and FHHs if farm sizes were increased.

11.5 Gender Differentials in Agricultural Productivity
Gender differences in gross output were considerable. MHHs had a gross output of Birr 6,456/ha,
while FHHs had a gross output of Birr 4,776/ha. These differences can be explained partly by the
lower quantities of inputs used by FHHs. The use of average values of these inputs from MHHs
resulted in a gross output of Birr 6,541/ha in FHHs (1.3% higher than MHHs). This suggests that if
both types of households had equal access to inputs, no productivity differences would exist.

The significant factors affecting gross value of output for MHHs were the farmer's age, family labor,
farm size, livestock units, and the use of inorganic fertilizer; in FHHs, they were family labor, farm
size, livestock units, the use of inorganic fertilizer, hired labor, and extension contact.

Labor and farm size had a significant and positive impact on the gross value of output for both
MHHs and FHHs. A 10% increase in labor resulted in a 2.3% and 1.8% increase in gross value of
output for both types of households; a 10% increase in farm size resulted in a 5.6% and 7.2%
increase in gross value of output for MHHs and FHHs, respectively. In MHHs, a farmer's age had a
significant and negative impact on the gross output. A 10% increase in farmer's age resulted in a
2.1% decrease in gross output.

The number of livestock owned and amount of fertilizer used had a positive and significant impact on
the gross value of output for both types of households. A 10% increase in the number of livestock
resulted in a 1.3% increase in the gross value of output for MHHs and a 0.7% increase for FHHs; a
10% increase in fertilizer use resulted in a 1% increase in gross value of output in both types of










household. Extension contact also had a negative and significant impact on gross output in FHHs-
gross output was lower for FHHs that had contact with extension services.

The production function analysis showed that the elasticities for the significant factors affecting the
gross value of output for MHHs were farmer's age (-0.21), fertilizer (0.10), farm size (0.56), labor
(0.23), and livestock (0.13). For FHHs the elasticities were fertilizer (0.10), farm size (0.76), labor
(0.18), livestock (0.07), hired labor (0.03), and extension (-0.28). The negative elasticity for extension
implies that policymakers should specifically target extension services for FFHs to mitigate
extension's negative effect on gross value of output in these households.

The production function analysis therefore shows that the significant factors affecting gross value of
output per hectare in FHHs were household size, farm size, livestock units, inorganic fertilizer, hired
labor, and extension contact.

The logit analysis showed that in MHHs, farm size and extension contact significantly and positively
affected the adoption of improved wheat variety, while in FHHs it was farm size and ownership of a
radio.

Based on these findings, it is recommended that to increase agricultural productivity in both MHHs
and FHHs, it is crucial to improve the provision of credit and the supply of preferred variety of
improved seed in time for planting. To increase agricultural productivity in FHHs specifically, it is
recommended that technologies take into account their resource base and improve FHHs access to
resources. Extension services should also target FHHs to ease their constraints and facilitate adoption
of improved varieties. Furthermore, female farmers' decision-making power and control of resources
should be positively exploited by exposing them to different opportunities.












REFERENCES


Abu, K., and C. Oppong. 1987. Seven Roles of Women: Impact
of Education, Migration and Employment on Ghanaian
Mothers. Women Switzerland, Work and Development Series
No. 13. Geneva: International Labor Organization (ILO).
Adams, M.E. 1983. Agricultural Extension in Developing
Countries. Intermediate Tropical Agriculture Series. Essex,
UK: Longman Scientific and Technical.
Adepoju, A., and C. Oppong. 1994. Gender, Work and Population
in sub-Saharan Africa. London, UK: James Currey.
Ahmed, I. 1985. Technology and rural women: Conceptual and
empirical issues. In L. N. Kimenye. Analysis of Technology
Dissemination and Utilization by Women and Men Farmers:
Case study ofEmbu and Mbeere Districts, Kenya. London,
Boston and Sydney: George Allen and Unwin.
Allison, C. 1986. It's Like Holding the Key to Your Own Jail:
Women in Namibia. Geneva, Switzerland: World Council of
Churches
Appleton, S. 1996. Women-headed households and household
welfare: An empirical deconstruction for Uganda. World
Development. 24 (12):1811-1827.
Ban, A.W. van den, and H.S. Hawkins. 1988. Agricultural
Extension. Essex, UK: Longman Scientific and Technical.
Berg, A. van den. 1997. Land Right Marriage left: Women's
Management of Insecurity in North Cameroon. Leiden,
Netherlands: CNWS Publications.
Berger, M., V. Delancey, and A. Mellencamp. 1984. Bridging the
Gender Gap in Agricultural Extension. Report prepared for
the Office of Women in Development/USAID, International
Conference on Research for Women.
Bisanda, S., and W. Mwangi. 1996. Adoption of Recommended
Maize Technologies in Myeba Region of the Southern
Highlands of Tanzania. Addis Ababa, Ethiopia: CIMMYT
(International Maize and Wheat Improvement Center) and
Ministry of Agriculture, United Republic of Tanzania
Boserup, E. 1970. Women's Role in Economic Development.
New York, NY: St Martin's Press.
Boserup, E. 1981. Population and Technological Change.
Chicago, Illinois: University of Chicago Press.
Bruce, J., 1988. Homes Divided: Women and Income in the
Third World. Sanford, North Carolina: Sanford University
Press.
Bryceson, D.F (Ed.). 1995. Women Wielding the Hoe. Lessons
from Rural Africa for Feminist Theory and Development
Practice. Oxford, UK: Berg Publishers.
Bryson, J.C. 1981. Women and agriculture in sub-Saharan Africa:
Implications for development (An exploratory study). In N.
Nelson (Ed.), African Women in the Development Process.
Totowa, New Jersey: Cass F Publishers.
Bukh, J. 1979. The Village Woman in Ghana. Uppsala:
Scandinavia Institute of African Studies
Bush, R., L. Cliffe, and V. Jansen. 1986. The crisis in the
reproduction of migrant labor in Southern Africa. In P.
Lawrence (Ed.). World Recession and the Food Crisis in
Africa. London, UK: James Currey
Buvinic, M., and G.R. Gupta. 1997. Female-headed households and
female-maintained families: Are they worth targeting to reduce
poverty in developing countries? Economic Development and
Cultural Change. 45(2): 259-280.


Central Statistical Authority (CSA). 1998. Statistical Abstract,
Addis Ababa, Ethiopia.
Chilot, Y., B.I. Shapiro, and M. Demeke. 1996. Factors
influencing adoption of new wheat technologies in Wolmera
and Addis Alem areas of Ethiopia. Ethiopian Journal of
Agricultural Economics 1 (1): 63-84
Chipande, G.H. 1987. Innovation adoption among female-headed
households: The case of Malawi. Development and Change
18: 315-27.
Cliffe, L. 1978. Labor migration and peasant Differentiation:
Zambian experiences. Journal of Peasant Studies 5(3).
CIMMYT Economics Program. 1993. The Adoption of
Agricultural Technology: A Guide for Survey Design.
Mexico, D.F.: International Maize and Wheat Improvement
Center (CIMMYT).
Cohen, J. 1980. Land Tenure and Rural Development in Africa.
In R. Bates and M. Lofchie (Eds.). Agricultural Development
In Africa. New York, NY: Praeger.
Dasgupta, P. 1993. An Inquiry into Well-being and Destitution.
New York, NY: Oxford University Press.
Davis, J. 1988. Agriculture, Women, and Land: The African
Experience. Boulder, Colorada: Westview USA.
Debebe, H. and S. Maxwell. 1992. Vulnerability Profiles and
Risk Mapping in Ethiopia. Addis Ababa: Food and Nutrition
Unit, Ministry of Planning and Economic Development, Addis
Ababa, Ethiopia. Mimeo.
Demeke M., A. Wolday, S. Ehui, and T. Zegeye. 1995. Food
Security, Nutrition, and Poverty Alleviation in Ethiopia.
Proceedings of the Inaugural and First Annual Conference of
the Agricultural Economics Society of Ethiopia, 8-9 June
1995, Addis Ababa, Ethiopia. Addis Ababa, Ethiopia:
Agricultural Economics Society of Ethiopia.
Dey, J. 1981. Gambian Women: Unequal partners in rice
development projects. Journal of Development Studies
17(3): 109-122
Dorner, P. 1972. Land Reform and Economic Development.
Kingsport, Tennessee: Kingsport Press.
Due, J. and C. H. Gladwin. 1991. Impacts of Structural
Adjustment Programs on African Women Farmers and
Female-Headed Households. American Journal of
Agricultural Economics. 73(5):1431-1439.
Due J.M., N. Mollel and V. Malone. 1987. Does the T & V system
reach female-headed families? Some evidence from
Tanzanian Agriculture. Administration and Extension 26:
209-217
Elabour-Idemudia, P. 1991. The impact of structural adjustment
program on women and their households in Bender and
Ogun States, Nigeria. In C.H. Gladwin (Ed.). Structural
Adjustment and African Women Farmers. Gainesville,
Florida: University of Florida Press.
FAO (Food and Agricultural Organization). 1997. Technical Notes
for Country Nutrition Profile (ESNA). Rome, Italy: FAO.
FAO. 1996. Mapping Under Nutrition. Poster for the World
Food Summit, Nov 13-17, 1996. Rome, Italy: FAO
FAO/WHO (World Health Organization). 1992. Assessing,
analyzing and monitoring nutrition situation. In Major Issues
for Nutrition Strategies. Rome, Italy: FAO/WHO.
Feder,G., R.E. Just, and D. Zilberman. 1985. Adoption of
agricultural innovations in developing countries: A survey.
Economic Development and Cultural Change 33(2):255-
297












Franzel, S. and H. van Houten. 1992. Research with Farmers:
Lessons from Ethiopia. Addis Ababa: Ethiopia. CAB
International for the Institute of Agricultural Research,
Ethiopia.
Gryseels, G., and F.M. Anderson. 1983. Research on Farm and
Livestock Productivity in the Central Ethiopian Highlands:
Initial Results, 1977-1980. ILCA Research Report No. 4.
Addis Ababa, Ethiopia: International Livestock Center for
Africa (ILCA).
Gujarati, D.N. 1988. Basic Econometrics. 2nd Edition. New
York, NY: McGraw-Hill.
Haile, D. 1980. Law and the Status of Women in Ethiopia.
Addis Ababa: Economic Commission for Africa, United
Nations.
Harrison, B. 1987. The Greening of Africa. London, UK:
Paladin Grafton.
Kinsey, B.H. 1974. Rural Development in Malawi: A Review of
the Lilongwe Land Development Program. Washington,
D.C.: World Bank.
Mekra, R. 1995. Raising agricultural productivity: The role of
women farmers. In G.H. Peters and D.D. Hedley (Eds.).
Agricultural Competitiveness: Market Forces and Policy
Choice. Proceedings of the 22nd International Conference of
Agricultural Economists, International Association of
Agricultural Economics (IAAE), 22-29 August 1994, Harare,
Zimbabwe. Oxford, UK: University of Oxford.
Merrick, T., and M. Schmink. 1983. Households headed by
women and urban poverty in Brazil. In M. Buvinic et al
(Eds.). Women and Poverty in the Third World. Baltimore:
Johns Hopkins University Press.
McSweeney, B., and M. Freedman.1980. Lack of time as an
obstacle to women's education: The case of Upper Volta.
Comparative Education Review. P.p.
134-139.
Moock, P.R. 1976. The efficiency of women as farm managers in
Kenya. American Journal of Agricultural Economics.
Dec.:831-835
Moore, H. 1988. Feminism and Anthropology. Minneapolis,
Minnesota: University of Minnesota. Pp. 42-72.
Mulat, D., A. Wolday, S. Ehui and T. Zegeye (Eds.). 1995. Food
Security, Nutrition, and Poverty Alleviation in Ethiopia:
Problems and Prospects. Proceedings of the Inaugural and
First Annual Conferencce of the Agricultural Economics
Society of Ethiopia, Addis Ababa, Ethiopia, 8-9 June, 1995.
Addis Ababa, Ethiopia: Agricultural Economics Society of
Ethiopia
Murray, C. 1981. Class, gender and the household: The
development cycle in Southern Africa. Development and
Change 18:235-50.
Negatu, W. 1989. Diagnostic Survey of the Farming Systems in
Ada, Lume, and Akaki (Unpublished). Debre Zeit, Ethiopia:
Alemaya University of Agriculture and Debre Zeit Agricultural
Research Center.
Negatu, W, W. Mwangi, and T. Tessema. 1994. Cultural
Practices and Varietal Preferences for Durum Wheat by
Farmers of Ada, Lume and Gimbichu Woredas of
Ethiopia. Research Report Series No. 1. Debre Zeit,
Ethiopia: Alemaya University of Agriculture and Debre Zeit
Agricultural Research Center.


Nornha, R. 1985. A Review of the Literature on Land Tenure
Systems in sub-Saharan Africa. Report No. ARU 43.
World Bank Agriculture and Rural Development
Department. Washington, D.C.: World Bank.
Obbo, C. 1980. African Women: Their Struggle for Economic
Independence. London, UK: Zed Press
Peters, G.H., and D. Hedley (Eds.) 1994. Proceedings of the
22nd International Conference of Agricultural
Economists, 22-29 August 1994, Harare, Zimbabwe.
Quisumbing, A.R. 1993. Improving women's agricultural
productivity as farmers and workers. Paper prepared for a
special study on Women in Development, World Bank,
Washington, D.C.
Quisumbing, A.R. 1996. Male-female differences in agricultural
productivity: Methodological issues and empirical evidence.
World Development 24(10):1579-1595.
Rahamato D.1991. Rural women in Ethiopia: Problems and
prospects. In T. Berhane-Selassie (Ed.). Gender Issues in
Ethiopia. Addis Ababa: Institute of Ethiopian Studies, Addis
Ababa University. Pp. 31-45.
Rolling, N. 1988. Extension Science: Information Systems in
Agricultural Development. WYE Studies in Agricultural
and Rural Development. Cambridge, UK: Cambridge
University Press.
Saito, K.A., H. Mekonnen, and D. Spurling. 1994. Raising the
productivity of women farmers in sub-Saharan Africa.
World Bank Discussion Paper 230. Washington D.C.: World
Bank.
Stamp, P. 1989. Gender, Power and Technology. Ottawa:
IDRC (International Development Research Centre).
Starkey, P., E. Mwenyae, and J. Stares. 1994. Improving
Animal Traction Technology. Proceedings of the First
Workshop of Animal Traction Network for Eastern and
Southern Africa, 18-23 January 1992, Lusaka, Zambia.
Staudt, K.A. 1979. Class and sex in the politics of women
farmers. J. Politics 41 (2):492-512
United Nations. 1989. Compendium of Statistics and
Indicators on the Situation of Women. New York, NY:
UN.
Warner, M.W., R.M.Hassan, and J.G. Kydd. 1997. Beyond
gender roles? Conceptualizing the social and economic lives
of rural peoples in sub-Saharan Africa. Development and
Change 28:143-168.
Whitehead, A. 1985. Effects of technological change on rural
women: A review of analysis and concepts. In I. Ahmed
(Ed.). Technology and Rural Women: Conceptual and
Empirical Issues London, UK: George Alien and Unwin.
Pp. 27-64.
Yirga, C., B.I. Shapiro and M. Demeke. 1996. Factors
influencing adoption of new wheat technologies in Wolmera
and Addis Alem areas of Ethiopia. Ethiopian Journal of
Agricultural Economics 1(1): 63-84.
Workneh, N., W. Mwangi, and T. Tesema. 1994. Cultural
Practices and Varietal Preferences for Durum Wheat by
Farmers of Ada, Lume, and Gimbichu Woredas of Ethiopia.
Research Report Series No. 1. Debre Zeit, Ethiopia: AUA,
DZARC
World Bank. 1996. Ethiopia: An Economic Overview.
Washington, D.C.: World Bank.
Yudelman, S. 1994 Women farmers in Central America.
Myths, roles, reality. Grassroots Development 17(2).




University of Florida Home Page
© 2004 - 2010 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Last updated October 10, 2010 - - mvs