. 453:: r

v. r.

:4.
o.
a... ..i z, u.
E...
I :01, .

l «n.

5

gram»... ‘9:
Af

. {In .1!)
7. vii

 

 

 

. 1‘ u ‘ 1&9. £51 13...! F:fi:’w¥i. . ism. . . inn.) 4 114,15...‘ .3
V n: . , . u.. _ . , , 3r
Ira—3132.33}. Bu“ Lu”? 5.1!..E Artiunaf. Ex p. . . an. . . i

. «is MICHIGAIVLSg‘fié'RIES
E UNIVERSITY
A EAST LANSING, MICH 48824-1048

This is to certify that the
thesis entitled

ANALYSIS OF THE ADOPTION AND PRODUCTION OF
ORANGE-FLESHED SWEETPOTATOES: THE CASE STUDY
OF GAZA PROVINCE IN MOZAMBIQUE

presented by

FELICIANO M. MAZUZE

has been accepted towards fulfillment
of the requirements for the

MS. degree in _Agricultural Economics

 

 

“law l/l. 3WD

Major Professor’s Signature

“42.1404

Date

 

MSU is an Affirmative Action/Equal Opportunity Institution

 

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6/01 c1/ClRC/DateDuep65-p. 1 5

ANALYSIS OF THE ADOPTION AND PRODUCTION OF ORANGE-FLESHED
SWEETPOTATOES: THE CASE STUDY OF GAZA PROVINCE IN
MOZAMBIQUE

By

FELICIANO M. MAZUZE

A THESIS

Submitted to
Michigan State University
In partial fiilfillment of the requirements
For the degree of

MASTER OF SCIENCE

Department of Agricultural Economics

2004

ABSTRACT
ANALYSIS OF ADOPTION AND PRODUCTION OF ORANGE-FLESHED
SWEETPOTATOES: THE CASE STUDY OF GAZA PROVINCE IN MOZAMBIQUE
By

FELICIANO M. MAZUZE

Vitamin A Deficiency is a major threat to women and children’s health in
Mozambique. In 2000, the Ministry of Agriculture and Rural Development launched a
program to multiply and widely distribute vines of orange-fleshed sweetpotato varieties
(OFSVS) to smallholders throughout the country. Since sweetpotatoes are grown and
consumed by almost all small farmers, it was anticipated that widespread farmer adoption

of OFSVS would have a major impact on reducing vitamin A deficiency.

Farmers (150) who had received vines in 2000 of recently-released OFSVs were
surveyed to assess the adoption rate and investigate the key factors determining farmer

adoption Of the nutritionally-improved sweetpotato varieties.

Maj or factors associated with adoption include sweetpotato cultivated area,
participation in field days/demonstration activities, number of OFSVs grown by farmers,
and the number Of times the respondent received vines. To spur adoption of OF SVs, it is
recommended that more resources should be allocated to post-harvest research to identify
apprOpriate storage and processing techniques, market Opportunities for processed
sweetpotato products, and effective ways to link OF SVs growers to potential processors

and output markets, and funding to rehabilitate soil drainage systems in lowlands.

ACKNOWLEDGEMENTS

I sincerely express my gratitude to Dr. Richard Bemsten, my major professor,
who tirelessly provided useful support, encouragement, and timely constructive
comments and critically suggestions during all phases of the elaboration of the thesis. I
also extent my gratitude to other members of my Committee, Dr. Eric Crawford and Dr.

John Kerr, for their timely and useful comments and suggestions.

I also thank the USAID/Mozambique (Grant 656-00-00-00050-00: Capacity
Building in Public Sector Research and Extension) for financial support during my M.S.
program degree at the Department of Agricultural Economics, Michigan State University,
and for logistical support for my fieldwork in Mozambique. I also thank the International
Sorghum and Millet Collaborative Research Support Program (INTSORMIL CRSP)
Management Entity for the successful implementation of the INIA/CRSP graduate degree
program in the US. I also thank my colleagues at National Institute for Agriculture
Research for logistical support during my fieldwork, especially to Dr. Isabel Andrade and

Dr. Tom Walker for their research support during the thesis process.

I am deeply grateful for the support, sacrifice, love, and inspiration provided by

my family.

iii

TABLE OF CONTENTS

LIST OF TABLES ......................................................................... v
LIST OF FIGURES ........................................................................ viii
KEY TO ABBREVIATIONS ............................................................ x
CHAPTER I: INTRODUCTION
1.1 Background ............................................................... 1
1.2 Problem Statement ........................................................ 6
1.3 Study Objective ............................................................ 7
1.4 Summary of Data and Research Methods .............................. 9
1.5 Thesis Outline .............................................................. 10

CHAPTER [1: DESCRIPTION OF THE STUDY AREA

2.1 Agroclimatic Features ..................................................... 11
2.1.1 Soils ................................................................... 11
2.1.2 Climate ................................................................ 15
2.1.3 Major Farming Systems ............................................ 17
2.2 Sweetpotato Production ................................................... 19
2.3 Historical Agriculture Development and
Farmers’ Characteristics ................................................... 21
2.4 Sweetpotato Multiplication and Distribution Program ................ 24
2.5 Chapter Summary ................................. 1 ........................ 26

CHAPTER III: LITERATURE REVIEW

3.1 Beta-carotene as Precursor of Vitamin A ............................... 28
3.2 Impact of F ood-based Interventions to Mitigate

Vitamin A Deficiency ...................................................... 32
3.3 Adoption of New Technologies .......................................... 36
3.4 Chapter Summary ........................................................... 42

iv

 

CHAPTER IV: RESEARCH METHODS AND DESIGN

4.1

4.2

4.3

4.4

Technology Adoption Decision (Objective 2) ........................... 44
4.1.1 Theoretical Model ................................................. 44
4.1.2 Empirical Model ................................................... 49
Prices Traders Offer for Roots of Orange—Fleshed

Sweetpotato Varieties (Objective 3) ...................................... 53
Description of Data and Sampling Approach .......................... 54
4.3.1 Key Informants Interviews ....................................... 54
4.3.2 Sweetpotato Growers Survey .................................... 54
4.3.3 Sweetpotato Traders .............................................. 55
Chapter Summary ........................................................... 56

CHAPTER V: DESCRIPTIVE ANALYSIS OF SWEETPOTATO

5.1

5.2

5.3

5.4

5.5

PRODUCERS, PRODUCTION AND FARMERS’

EVALUATION OF OFSVs
Socioeconomic Characteristics of Sweetpotato Growers ............ 57
5.1.1 Education Level, Age and Gender .............................. 57
5.1.2 Household Size and Number of Children ..................... 58
5.1.3 Livestock Ownership and Total Farmland ..................... 60
5.1.4 Awareness and Extension Services Provided to Farmers 62
5.1.5 Land Tenure and Cropping Systems ........................... 65
Descriptive Statistics of Sweetpotato Production ...................... 66
5.2.1 Identified Varieties Grown by Households ..................... 66
5.2.2 Sweetpotato Cultivated Area .................................... 67
5.2.3 Planting Methods of Sweetpotato .............................. 71
5.2.4 Orange-fleshed Varieties Grew and Vines

Received by Households ......................................... 72
Producers’ Evaluation of Orange-fleshed Varieties .................. 76
5.3.1 Producers’ Perceptions about Performance of

Orange-fleshed Sweetpotato Varieties ......................... 76
5.3.2 Major Constraints to Sweetpotato Production ................ 79
5.3.3 Proposed Actions to Improve Orange-fleshed

Sweetpotato Production .......................................... 81
Consumption and Commercialization of Sweet Potato ............... 82
5.4.1 Sweetpotato Consumption Patterns ............................ 82
5.4.2 Commercialization of Sweetpotato ............................ 87
Chapter Summary ......................................................... 91

CHAPTER VI: FACTORS INF UENCING F ARMER ADOPTION

6.1

6.2
6.3
6.4

OF OF SVs
The Adoption of Orange-fleshed Varieties .............................. 95
6.1.1 Definition of Explanatory Variables ........................... 95
6.1.2 Results of the Probit Model ..................................... 98
Discussion of the Results ................................................ 101
Goodness-of Fit ............................................................ 102
Chapter Summary .......................................................... 103

CHAPTER VII: SUMMARY AND POLICY IMPLICATIONS

7.1 Summary .................................................................. 104

7.2 Policy Implications ......................................................... 109

7.2.1 Agricultural Research & Development .......................... 109

7.2.2 Rural Extension & Development ................................. 111

7.2.3 DDARDS and NGOs ............................................... 113

7.2.4 Health Institutions ................................................. 1 14

7.3 Limitations of the Study ................................................... 114

7.4 Further Research Needs .................................................... 1 15
APPENDICE A .............................................................................. 118
APPENDICE B .............................................................................. 132
APPENDICE C .............................................................................. 140
APPENDICE D .............................................................................. 157
APPENDICE E .............................................................................. 161
REFERENCES ............................................................................... 164

vi

 

 

LIST OF TABLES

Table 1: Average Yields of Principal Crops in Mozambique and

Sub-Saharan Africa ............................................................ 1

Table 2: Characteristics of Eight Released OFSVs ................................. 4

Table 3: Descriptive Statistics of Total Annual Rainfall and Total
Rainfall during Wet Season in Chokwe and Xai-Xai Districts ........ 15

Table 4: Farming Systems and Main Characteristics of R2 and R3 ................ 18

Table 5: Number of Farms, Total Area of Food Crops, Average Area and
Sweetpotato Cultivated Area by Small, Medium, and Commercial
Scale Farmers in Gaza Province (2000-2001 Cropping Season ......... 22

Table 6: Multiplication Plots, Districts Covered and Beneficiaries of
Sweetpotato Vines .............................................................. 25

Table 7: Frequency, Sources and Type of Technical Assistance
Received by Farmers .......................................................... 63

Table 8: Sweetpotato Plots Cultivation across Different Soil Types ............. 66

Table 9: Four Most Grown White/Cream and Orange-fleshed Varieties in
Chokwe and Xai-Xai Districts .............................................. 67

Table 10: Sweetpotato Cultivated Area by Smallholders in Chokwe and
Xai-Xai Districts ............................................................. 68

vii

Table l 1:

Table 13:

Table 13:

Table 14:

Table l5:

Table l6:

Table 1'};

Table 1 8:

Table 19.

lable :0.

T

a", ‘\
Die ‘1:

abi ,
re ‘2

 

 

 

 

"r1 OW

n

f/)

Table 1 1:

Table 12:

Table 13:

Table 14:

Table 15:

Table 16:

Table 17:

Table 18:

Table 19:

Table 20:

Table 21:

Table 22:

 

Proportion of WFSVs Relative to OFSVS Planted by
Smallholders in Chokwe and Xai-Xai Districts ........................... 69

Farmers’ Reasons for Planting Different Proportions of
Sweetpotato Varieties ......................................................... 70

Adoption (%) of OFSVS in Each Zone Covered by NGO, and by
Season in Chokwe and Xai-Xai Districts ..................................... 71

Major Causes of Loss of Planting Materials of OF SVs in the
Chokwe and Xai-Xai Districts ................................................. 73

Frequency of OFSPV Vines Distribution, and Number of Varieties
Each Household Received in Chokwe and Xai-Xai Districts ............. 74

Respondents’ First and Second Comments Regarding Performance
Of OF SPVS in Chokwe and Xai-Xai Districts ............................. 76

Farmers’ Evaluation of Performance of OF SVs versus “Local”
Cultivated Varieties ............................................................. 78

Constraints of Sweetpotato Production in Chokwe and
Xai-Xai Districts ............................................................... 80

Main Actions Proposed by Respondents for Improving
Sweetpotato Production in Chokwe and Xai-Xai Districts ................ 82

Consumption of Roots, Awareness and Household Experience in
Processing Products out of Sweetpotato .................................... 83

Level of Awareness and Practice of Processing Techniques of
OFSVS in Chokwe and Xai-Xai Districts ................................. 87

Indicators of Sweetpotato Commercialization in Chokwe and
Xai-Xai Districts .............................................................. 88

viii

—_‘,..‘. I.

 

Table 23:

Table 24:

Table 25:

Table 36:

Table D]: A

Table D3. I

.\'

\
\\

 

 

Table 23:

Table 24:

Table 25:

Table 26:

Table A:

Table D-l:

Table D-2:

Table D3:

Table E:

Summary Statistics of Selected Explanatory Variables .................. 98

Maximum Likelihood Estimates of the Unrestricted (full) Model for
Factors Influencing Adoption of OFSVS in Chokwe and
Xai-Xai Districts ............................................................... 99

Maximum Likelihood Estimates of the Restricted Model for
Factors Influencing Adoption of OFSVS in Chokwe and
Xai-Xai Districts ............................................................... 100

Comparison of Observed and Predicted Outcomes of
Adoption of OFSVS ............................................................ 103

Household, Cultivated Area and Average Cultivated Area in
2000/01 Cropping Season ...................................................... 131

Adoption (%) of OFSPVS in Villages Covered by Different
NGOs Operating in Chokwe and Xai-Xai Districts, 2002-2004 ......... 158

Frequency of OF SPV Vines Distribution, and Number of

Varieties Each Household Received from 2000/01 to 2003/04
Wet Season by Village, Chokwe and Xai-Xai Districts .................... 159

Levels of Awareness and Practice of Processing Techniques of
OF SPVs in 2003/04 Wet Season, Chokwe and Xai-Xai Districts ....... 160

Correlation Matrix of Variables in the Probit Model ...................... 161

ix

‘LL-

Figure 1:

Tizure 2:

iv

1 I
(I J

Tim 4:

Flgure 5:

Pi's’ure 6:

Figme 7:

Figure 8:

figure 9;

Figure 10

 

LIST OF FIGURES

Figure 1: Monthly Average Precipitation in Chokwe and Xai-Xai Districts ....... 16
Figure 2: Area cultivated of Sweetpotato in Mozambique in
2000/01 Cropping Season ...................................................... 19

Figure 3: Area Cultivated (ha) by Cash Crops in Gaza Province in
2000/01 Cropping Season ..................................................... 23

Figure 4: Area Cultivated (ha) by Cash Crop in Gaza Province in
2000/01 Cropping Season ....................................................... 24

Figure 5: Distribution of Years of School Completed, Xai-Xai and Chokwe

Districts ........................................................................... 57
Figure 6: Respondents’ Age Distribution, Xai-Xai and Chokwe
Districts ............................................................................ 58

Figure 7: Respondents’ Family Size Distribution, Xai-Xai and Chokwe
Districts ............................................................................. 59

Figure 8: Distribution of Young Children in Households, Xai-Xai and Chokwe
Districts ........................................................................... 59

Figure 9: Distribution of Households’ Total Cattle, Xai-Xai and Chokwe
Districts ........................................................................... 61

Figure 10: Distribution of Households’ Total Farmland, Xai-Xai and Chokwe
Districts ........................................................................... 62

 

figure A:

Figure B:

 

 

 

Figure A:

Figure B:

 

Major Economic Sectors and Their Contribution to
Mozambican GDP in 2002 .................................................... 122

Relationship among Different Types of Nutrition Interventions ........ 137

xi

_. gull—h” ‘

 

 

MRDC
CAP
CEF
(11’
(RC
DDARD
DNER
HO
GDP
COM
Hlfl
ICRw
1er
HA)!
UTA

[\IF

[Vii
0'1 TE
IPA

ch

 

AVRDC

CAP

CEF

CIP

CRC

DDARD

DN ER

FAO

GDP

GOM

HKI

ICRW

IF PRI

IIAM

IITA

IMF

INE

INIA

INIVE

IPA

IPC

KEY TO ABBREVIATIONS

Asian Vegetable Research and Development Center
Agriculture and Livestock Census

Center for Forestry Experimentation

Centro Intemacional de la Papa (International Potato Center)
Caritas Regional Chokwe

District Directorate of Agriculture and Rural Development
National Directorate of Rural Extension

Food and Agriculture Organization (of the United Nations)
Gross Domestic Product

Government of Mozambique

Helen Keller International

International Center for Research on Women

International Food Policy Research Institute

Mozambique Institute for Agricultural Research
International Institute of Tropical Agriculture

International Monetary Fund

National Institute of Statistics

National Institute for Agronomic Research

National Institute for Livestock Research

National Institute for Animal Production

International Potato Center

xii

 

L\\I-'
IIIDER
MAI
IIISAL’
MSU
.VGO

OFSPVS

 

PENI
PROAG R1
SARRVET
SNA

SIN SE

SAP

SSA

STC
[NICE]:
[SAID
I‘m
lTIM
“TSPVS
“THO

“RS

“1"

 

 

LWF

MADER

MAF

MISAU

MSU

NGO

OFSPVS

PEM

PROAGRI

SARRN ET

SIMA

SINSE

SAP

SSA

STC

UNICEF

USAID

VAD

VITAA

WFSPVS

WHO

Lutherans World Federation

Ministry of Agriculture and Rural Development
Ministry of Agriculture and Fisheries

Ministry of Health in Mozambique

Michigan State University

Non-Governmental Organization
Orange-Fleshed Sweetpotato Varieties

Protein Energy Malnutrition

National Agricultural Development Programme
Southern African Roots Crops Research Network
Agricultural Market Information System
Integrated National Extension System
Structural Adjustment Program

Sub-Saharan Africa

Save the Children

United Nations Children’s Fund

United States Agency for International Development
Vitamin A Deficiency

Vitamin A for Africa

White-fleshed Sweetpotato Varieties

World Health Organization

World Relief Services

World Vision

xiii

..._..... _#., -—lM—.4-‘ ' -

 

llozam'i‘bl '

1992.). 1}ch

    

policies. 110
aierage an!)
annual rate

production 1
rather than d

{tier-age 1e\ e

Table l: A'
'------

Commodin

\

Maize

Beans
COUQn

n2:

 

CHAPTER I

INTRODUCTION

1.1 Background

During the past three decades, the performance of the agricultural sector in
Mozambique has been drastically affected by more than 15 years Of civil war (1975-
1992), frequent occurrence of drought, and inappropriate macroeconomic and agricultural
policies. However, between 1992 and 1997, agricultural production grew at an estimated
average annual rate of 6.3 %, while marketed surplus through formal markets grew at an
annual rate of 34 % (PROAGRI 1997). This substantial recovery in agricultural
production has mainly resulted from area expansion and favorable climatic conditions,
rather than an increase in productivity. Yield levels continue to be low and below the

average levels for Sub-Saharan Africa (SSA), as shown in Table 1 (Mucavele F, 1998).

Table 1: Average Yields of Principal Crops in Mozambique and Sub-Saharan Africa.

 

 

 

Average yield (tons / ha)
Commodity
Mozambique Sub-Saharan Africa

Maize 0.3 — 1.3 1.2
Sorghum 0.3 — 0.6 0.8
Cassava 5.0 — 7.0 7.5
Rice 0.5 — 1.0 1.6
Beans 0.2 -— 0.4 0.7
Cotton 0.3 - 0.6 0.8

 

Source: Ministry of Agriculture and Fisheries, PROAGRI, 1997.

 

 

On.
1135 been 11‘.
Qpieall} l;
At‘pcntlix .‘
01-310mm?
Pm

Vitamin A ,.

azure main;
”$1110 (8--
The Sing}. a,
361 had We

. ~.~.

I
1

 

One of the major challenges to Mozambique’s economy growth and development
has been the absolute poverty1 Of the vast majority of the population (80%), who
typically lack access to markets and are dependent on subsistence rainfed agriculture (see
Appendix A for a brief description of the main agroclimatic and socioeconomic features
of Mozambique).

Poverty is a primary cause of widespread malnutrition in Mozambique, including
Vitamin A Deficiency (VAD). For example, a multi-sectoral assessment, conducted in
2002 by several national and provincial Mozambican governmental institutions in
selected areas of several provinces (e. g. Maputo, Gaza, Inhambane, Sofala, Manica and
Tete) estimated that an average of 6.4 % of children 6 to 59 months old suffered from
acute malnutrition. Gaza province exhibited the highest levels (11.2%), followed by
Maputo (8%); and Sofala revealed the lowest level (4%), followed by Inhambane (4.6%).
The study also found that 41 % of the targeted population (children 6 to 50 months of
age) had received at least one dose of Vitamin A supplement during the second half of
2002. Sofala province exhibited the lowest percentage (27%) and Maputo the highest
(56%)(GOM, 2003). For a detailed discussion of Nutrition Policy in Mozambique and
recent studies that document the incidence of micronutrient deficiency, see Appendix B.

It is important to point out that in order to be effective, Vitamin A
supplementation programs must be implemented twice a year. However, in countries like
Mozambique, where the majority of rural population lives in a very scattered pattern and

characterized by poor network of roads and undeveloped health system, Vitamin A

 

In Mozambique, poverty is defined as the “inability of individuals to ensure for themselves and their
dependants a set of basic minimum conditions necessary for their subsistence and well-being in
accordance with the norms of society”. The absolute poverty line is defined as the sum of a food
poverty line (approximately 2,150 calories per person per day) plus an amount for non-food expenses
per person per day that varies from province to province (GOM, 2001).

 

supplemer
Alternatix:
production
human bod
rural popul

III II
oforange-f
chemical c3
satisf} cons
consumpm.

eliminate or

 

 

 

supplementation program is difficult to implement, expensive, and thus unsustainable.
Alternatively, a food-based approach - which promotes a more diversified crop
production including plants rich in B-carotene that is convertible to Vitamin A by the
human body - has the potential to improve (qualitatively and quantitatively) the diet of
rural population.

In the 19905, scientists at the International Potato Center (CIP) identified a group
of orange-fleshed sweetpotato varieties (OFSVS) with high content of B-carotene (a
chemical element used by the body to generate Vitamin A) and sufficient dry matter to
satisfy consumer preferences and taste. Subsequent studies demonstrated that the
consumption of just small amounts of foods derived from the new OF SVs could
eliminate or greatly reduce Vitamin A deficiencies in both young children and pregnant
and lactating women (HarvestPlus, 2003). In 2001/02, an international crop-based
initiative to combat VAD in Sub-Saharan Africa was launched —known as VITAA or
Vitamin A for Africa— involving agronomists, health experts, and nutritionists. This
initiative was aimed at expanding the impact of the OF SVs in seven countries: Kenya,
South Africa, Tanzania, Ethiopia, Uganda, Ghana and Mozambique.

During 1999/2000 and 2000/01 cropping seasons, the National Institute for
Agricultural Research (INIA), with financial and technical support of Southern Africa
Root Crops Research Network (SARRN ET), conducted two consecutive rounds of multi-
location trials to evaluate different orange-fleshed sweetpotato clones with respect to total
yield, commercial yield, weight of roots damaged by root rot, number of damaged roots,
harvest index, dry matter content, and their adaptability across various agro ecological

conditions of Mozambique. Based on the results Of these trials, eight clones were

selected, which performed well across a range of different environments (Andrade and

Ricardo, 1999; Andrade er al., 2002) (Table 2).

Table 2: Characteristics of Eight Released OF SVs.

 

 

Variety Name Total Total % of Dry Habit Virus
yield yield marketable matter growth resistance
(t/ha)a (t/ha)b roots content
(%)
Caromex l 5 .3 6.5 5 5.9 22.7 Prostrate Good
CN 1448-49 15.7 8.4 48.3 22.7 Erect Good
Japon Selecto 14.5 6.6 44.7 21.6 Erect Good
Jonathan 1 6. l 6.0 43 .4 21 .2 Erect Moderate
Kandee l 4.5 5 .5 3 6.6 2 5 .3 Prostrate Good
L0 323 13.6 6.0 43.4 21.0 Erect Good
Resisto 14.5 7.2 55 .9 27.2 Semi-erect Good
Taimung 64 13.9 6.0 48.4 20.9 Prostrate Good

 

Source: Adapted from IITA/SARRNET, 2003 and INIA/SARRNET (undated).

Average total root yield of trials carried out in 14 locations from January to August 1999.
Average total root yield of trials carried out in 7 locations from July 1999 to February 2000 in the dry
season, but irrigated until the plants were established.

b

In 2001, a two-year project—supported by USAID through the INIA/SARRNET
project with multiple sources of funding, including IITA and CIP—was launched to
alleviate the negative impacts of the adverse cycle of floods/drought on the rural
economy and on food security in the affected districts of Mozambique by increasing the
production of improved cassava and sweetpotato varieties. The project’s target
population was comprised Of 500,000 smallholders, distributed over 65 districts in nine
provinces. The project had five Objectives: (i) to provide improved sweetpotato and
cassava planting materials to farmers through the establishment Of a well-coordinated
network of project and partners fields in selected districts across the country; (ii) to

promote widespread cultivation and use of orange—fleshed sweetpotato, envisaged to

reduce the

 

processing
machines .;
variety tri..
Mozambic

Ac.

 

prox'inces .

 

household

i 7.]

k’fifl I.‘ T‘ O 9
heel-'90 at.

b," arrange:
8300 01‘ The“ I
3% sold m.
9013 sold 45
leaxe; Ah.
rfleaSed UT
Vaflfilleg‘ dL
be 19516 Oil
Elm OFSV
“Ci-“I'M 3 \‘ul

“ere 21A fin
‘ I

 

reduce the risk of VAD in Mozambique; (iii) to promote household utilization and agro-
processing of both crops through demonstrations of the use of improved tools and
machines at workshops and field days, and (iv) to establish cassava and sweetpotato
variety trials in order to select clones best adapted to the major agro ecological zones of
Mozambique (INIA-IITA/SARRN ET, 2003).

According to the baseline national survey, which was conducted in 2002 in five
provinces Of Mozambique and included 1,476 respondents, sweetpotato-growing
households averaged about nine persons and 53% of the growers were women. Of the
sweetpotato-growing farmers interviewed, 50% acquired land through inheritance, 29%
by arrangement with local authorities, 14% rented land, and only 5% bought land. About
83% of the sweetpotato producers consumed most of their harvested sweetpotato roots,
8% sold more of their harvested sweetpotato roots and leaves than they consumed, and
9% sold as much as they consumed. About 97% of producers consumed sweetpotato
leaves. About 34% Of the sweetpotato producers interviewed planted the recently-
released OF SVs. Approximately 68% reported that they did not grow orange-fleshed
varieties, due to unavailability of planting materials and 43% or because they did not like
the taste of OFSVS. Interestingly, about one-third of the respondents reported that they
grew OF SVs because they wanted to grow a variety that was sweet and high-yielding,
wanted a variety rich in pro-Vitamin A, wanted to mitigate hunger, or just because they

were given the vines by an NGO (INIA-IITA/SARRNET, 2003).

 

 

 

l
minimize '.

etiiciency
household
household

ueaith. Tn.

 

tolerant er...
communit;

ln.\

 

C0 ‘ i

‘.
0. 01'6ij in, r
‘1

1.2 Problem Statement

Although it is fully recognized that agrarian societies have developed strategies to
minimize the impact of adverse climatic conditions (e. g. floods and droughts), the
efficiency of these coping and survival mechanisms is strongly dependent on a variety of
household characteristics, including levels of knowledge and experience, gender Of
household head, family labor availability, off-farm job Opportunities, and accumulated
wealth. The main coping strategies applied by farmers include cultivating drought-
tolerant crops, raising livestock for sale, fishing, seeking casual labor within their
communities, and consuming wild fruits and tubers.

In Mozambique, as part of their strategy to insure household food security,
women grow mostly white-fleshed sweetpotato varieties (WFSVs)—primarily for
household consumption but also to generate cash income. Recognizing the potential of
OFSVS to reduce VAD, in 2000 the Ministry Of Agriculture and Rural Development
(MADER) launched a program to multiply and distribute planting material (vines) of
OFSVS to smallholders throughout the country. Since sweetpotatoes are grown and
consumed by almost all small farmers, it was anticipated that widespread farmer adoption
of these improved varieties would have a major impact on reducing VAD.

While the goals Of the OFSVS program are laudable, there exists an on-going
debate regarding whether or not African consumers would accept OF SVs. Some
researchers argue that consumers will not adopt these varieties because they are either too
watery and/or too sweet to meet local taste preferences (HarvestPlus, 2003). However, a
recent study conducted by the International Center for Research on Women (ICRW)

revealed that in Kenya, women would accept orange-fleshed varieties if the clones are

arid inten
socioeeor

em'ironm

adoption (
generati"
06V 1
eIopme
ad 1111) d;
01 at ailakt.

mod THNQQL

 

 

sufficiently high in starch, low in fiber, and are introduced through community-level
education programs that emphasized the health of young children.

Furthermore, some recent studies in developing countries revealed that the level
and intensity of farmer adoption of agricultural technologies are governed by farmer
socioeconomic characteristics, institutional arrangements, agricultural policy, and
environmental factors (Adesina and Moses, 1993, Sanders er al., 1996, Akhter and Rajah,
1992, Mbata, 2001 ). Thus, there is a need to investigate the key determinants of the
adoption of OF SVs from both the supply and demand sides. This study is aimed at
generating insights that will contribute to: (i) improving the reliability of technology
development and transfer; (ii) strengthening the linkages between research and extension,
and (iii) designing agricultural policy conducive to higher and more widespread adoption
of available improved varieties--thereby increasing sweetpotato production, mitigating

food insecurity, and reducing VAD.

1.3 Study Objective

The objective of this study is to assess the socio-economic and institutional
factors associated with the adoption Of OFSVS by smallholders in two districts Of the
Gaza province. Although it is recognized that the promotion of OFSVs rich in pro-
Vitamin A is likely to generate reasonable health impact, an analysis of these health

effects falls outside the realm of this studyz.

 

An on-going collaborative project involving the Nutrition Division ofthe Ministry of Health in
Mozambique, Michigan State University, World Vision lntemational, Helen Keller lntemational, the
National Institute for Agronomic Research and the Southern African Root Crops Research Network, is
addressing VAD and inadequate caloric intake through dietary diversification, improved feeding
practices and improved serum retinol status, particularly in young children under five years of age in
Zambezia province.

 

ll

4) ‘

 

The research questions to be addressed by this study are as follows:

1)

2)

3)

4)

5)

What are the socioeconomic characteristics of sweetpotato-growing farm

households?

What is the adoption rate of the available OFSVS and what are the major
determinants associated with farmer adoption/disadoption of the

recommended nutritionally improved varieties (e. g., characteristics of the
variety, contribution to cash income, perception of nutritional benefits of

OF SVS, educational level, household size and composition)?

Do traders offer sweetpotato sellers the same price, a lower price (price
discount), or a higher price (price premium) for the roots produced from the

improved OFSVS (compared to WFSVs) and why?

What institutional and agricultural policy interventions are needed to spur the
rate of technology adoption and thereby increase sweetpotato production and

household income?

Will increasing OFSV production sufficiently increase the bioavailability and

conversion Of B-carotene so as to alleviate VAD, or are additional initiatives

needed?

 

 

collected .;
the introd
infomiari‘»
sueetpotai
I
OFSVS. T2
collect dag
emphasis t.
nutritionaf I
collected t»
magnitude

A tr
c'F'ruzieteris:

traders' kn.

 

1.4 Summary of Data and Research Methods

In order to achieve the objectives above mentioned, existing secondary data were
collected and analyzed in order to gain insights needed to better understand the history of
the introduction and distribution Of OFSVS to farmers. Similarly, interviews with key
informants were conducted to gain insights about the history of the OFSVS program,
sweetpotato cropping patterns, and factors influencing farmers’ decision in adopting
OFSVS. This background information was used to design the farmer questionnaire to
collect data on farmers’ household characteristics, and production circumstances with
emphasis on understanding the farmers’ decision-making process in adopting the
nutritionally improved OF SVs. In addition, data on sweetpotato roots sales were
collected to determine the structure of the marketing channel, and gain insights about the
magnitude of marketed surplus and producer prices.

A trader questionnaire was designed to collect data on basic sweetpotato trading
characteristics, the retail prices of both orange and white-fleshed sweetpotato roots, and
traders’ knowledge and acceptability of the improved OF SVs.

A considerable amount of time was devoted to pretesting in the field both farmer
level and trader questionnaires in order to improve the ability to gather the needed data.

The data collected from the farmer questionnaire were used for descriptive
analysis of socio-economic characteristics of sweetpotato growers and sweetpotato
cropping patterns. In addition farm and farmers’ characteristics, and measures of the
institutional environment in which sweetpotato growers operate were used as explanatory
variables in the probit model to determine the key factors governing farmers’ adoption of

improved OFSV.

 

 

 

Fii
analysis 0

price pren‘

15 Th

Th:
of the agrit
multiplica:
bioavailabi
based inter
Chapter IV
results and
COHImercia

influencjn :

 

 

 

Finally, data collected through the trader questionnaire were used for a descriptive
analysis of sweetpotato commercialization, and to determine whether or not traders paid a

price premium for nutritionally improved OFSVs.

1.5 Thesis Outline

This thesis is divided into six chapters. Chapter 11 provides a general description
of the agriculture sector in Gaza province, and the orange-fleshed sweetpotato
multiplication and distribution program. Chapter III briefly reviews the literature on the
bioavailability and conversion of B-carotene to retinol (Vitamin A), the impact of food-
based interventions to mitigate VAD, and adoption of new agricultural technologies.
Chapter IV presents the research design and methodology. Chapter V reports survey
results and descriptive analysis of sweetpotato production and producers, consumption,
commercialization, and producers’ evaluation of OFSVS. Chapter VI presents the factors
influencing farmer adoption of nutritionally improved OF SVs, the probit model
specification for the adoption of OFSVs, and a discussion of the results of this binary-
choice model. Finally, Chapter VII provides the summary of the major findings and

draws policy implications.

10

 

Sc
low at 311,.
NC less
extremel}
0986 i. In
loulund .1:

WW} 11C 110,}:

 

CHAPTER II

DESCRIPTION OF THE STUDY AREA

2.1 Agro-climatic Features

2.1.1 Soils

Southern Mozambique is mostly characterized by light/coarse-textured soils with
low available water content (AWC). Approximately 50% of the area has soils with an
AWC less than 100 mm/m and 25% have less than 50 mm/m. Thus, the risk of drought is
extremely high and most of the area has over 50% chance of dryland crop failure (Reddy
(1986). In order to mitigate these adverse growing conditions, farmers exploit the
lowland areas in which the residual soil moisture can be effectively used for crop
production. While crop failure in the upland zone is due mainly to frequent soil water
deficits throughout the year, yield losses in the lowland areas are also common—due to
flooding and excessive soil moisture during the erratic and irregular rainy season.

In addition, hydromorphic soils are found in areas. of water seepage, at the lower
slopes of the escarpments, which form the transition between the beach ridges and coastal
margins, and low areas (e. g., flat marshy zones and swampy depressions). These soils are
saturated with water, either permanently or during most of the year. This results in lack of
oxygen, which slows the process by which organic matter is broken down by bacterial
activity. Although some mineralization is caused by anaerobic bacteria, this is much less
than under dryland conditions (Bleeker, 1983). The accumulation Of organic matter gives

rise to an organic peat horizon, which can vary from 0.4 to 1.0 m depth.

11

Two types of hydromorphic soils are found in southern Mozambique. First, the
organic (peat) soils, called machongos3, are generally very fertile and continuously wet,
as they receive fresh water all year round through seepage from the surrounding dune
areas due to high infiltration and high recharge rates. These soils—which have very good
soil structure for plant growth, a high water holding capacity, a high soil aeration, and
good workability—are intensively used for small-scale agriculture. However, excessive
drainage can contribute to mineralization of the peat, resulting in soil acidification.
Second, the hydromorphic sandy soils are less rich in organic matter than the machongos,
but their water table is closer to the soil surface. Due to the similarity of these soils, in
terms of occurrence, wetness/drainage, land use, and vegetation, farmers also commonly
call them machongos.

Due to a high accumulation of organic matter and water saturation of those soils
throughout the year, they commonly have an accentuated acid reaction (pH 3.8-5.5) due
to the release of hydrogen ions (HT) during the decomposition of organic matter (Gomes
et al., 1998).

Once drained, a variety of crops can be successfully grown on machongos--
mainly rice, maize, beans, and vegetables. However, rice4 is the main crop grown on
these soils, due to its rooting system which is well adapted to waterlogged conditions and

since rice is grown during the hot and rainy season, when the peat soils are likely to be

 

The importance of machongos for small-scale agriculture in southern Mozambique has been recognized
for a long time. Between 1951 and 1957, a total of 2,800 ha were identified, surveyed (scale 1: 10 000),
developed, and distributed among 4,200 farmers in Gaza province (Monteiro, 1957).

4 Some farmers recorded yields of 1.9-2.8 tons/ha for rice in the machongos of Gaza province (Monteiro,

1957).

12

 

”at: ('o r ‘UEU'I - l-ra

....4 __-

flooded.l
prodUCIlO

1i.
componei
agricultur

1n
properties
Xlahgueat
“ETC Colic

and others

basic print

01.111998)
’ Si“.
fai

Cur;
hm

0rdg

 

 

 

flooded. During the dry season, these soils are especially desired for vegetable
production, including sweetpotato cultivation.

However, since these soils are quite fragile, due to the absence of the mineral
component, mismanagement can lead to their degradation and permanent loss for
agriculture (Dykshoorn er al. , 1988, Gomes et al. , 1998).

In order to investigate the impact of excessive field drainage on peat soil
properties and to monitor the changes in pH values, a detailed soil survey in
Manguenhane (Gaza province) was conducted in which samples at different soil depths
were collected and exposed to a drying process (Mafalacusser er al. , 1997). This study,
and others carried out previously (Monteiro, 1957; Dykshoom er al., 1988), identified
basic principles for a sustainable management of these soils, as described by Gomes et
aL,(1998):

> Shallow drainage must be encouraged in order to prevent the water table from
falling lower than 0.2-0.4 m in depth; drain spacing and depth should be
determined for each case, according to the physical conditions of each site. Drains
should be dug by hand and artisanal gates can be used to control the water table at
the farm level;

> Fields should be irrigated as frequent as possible in order to maintain high water
content in the upper layers and decrease soil temperature, especially during the
hot season. The water should flow from the high to the low end of the field in

order to lixiviate the salts downwards and avoid adverse salinization;

l3

 

 

 

 

 

 

‘r Cr.
sh
IL:
ax
\\
‘r (
S
)
SK SI 3:
33's:

> Technology based on intensive use of manual tools should be promoted, since the
potential for mechanization is limited due to the low carrying capacity of these
soils;

> Land clearing should avoid burning vegetation and crop residues, as it also
destroys the peat topsoil; and burning should occur outside the machongo area or
in selected places with high water table levels to avoid its spread. Ashes can then

be spread on the soil surface to improve the soil nutrient status;

V

Crop selection and rotation should consider the conditions of these soils (e.g.

shallow water table and soil acidity). Rice is the most suitable crop during the

rainy season, and vegetables (with shallow rooting system and with tolerance to

acidity) are most suitable during the dry season. Depending on the depth of the

water table, crops such as maize and beans can also be considered.

‘9 Cultivation in mounds or hills permits the farming of some crops (for example
sweetpotato) while the soil is still poorly drained.

> Mulching with crop or vegetation residues will reduce the effect Of solar radiation

in the oxidation and decomposition of peat layers.

It is important to point out that since there is always a risk of floods during the
wet season, which if not taken into account, all investments will likely be lost. Thus,
intensive maintenance of the drainage system is crucial for the sustainable use and

development of these important soil resources.

14

I

-:- -“H' q. Nansen-'13

 

21:750-]

Steon A

Table 3; L

/,-

DlStrici

Choke

\

 

 

2.1.2 Climate

In the southern provinces of Mozambique, the mean annual rainfall pattern
decreases from 800-1,000 mm along the coastline to less than 400 mm in the interior of
Gaza province (Pafuri), with the rainy period concentrated between October and April
(Reddy, 1986). Due to this spatial and temporal rainfall pattern, the coastal zones are the
most heavily populated and characterized by high land use intensity (Snijders, 1986).
This high land use intensity associated with light/coarse-textured soils, low soil fertility
and low water holding capacity, leads to progressively lower yields over time.

Analysis Of more than a 30-years time-series data Of Chokwe (1961-2000) and
Xai-Xai (1951-2002), indicates that those two districts have different patterns of
precipitation (Figure 1). Xai-Xai had relatively more total annual precipitation and more
total precipitation during rainy season than Chokwe (Table 3). Although sweetpotato is
considered to be a drought-resistant crop, optimum annual total precipitation is estimated
at 750 -1,000 mm, with approximately 500 mm Of rainfall registered during the growing

season (Woolfe, 1992; Salunkhe et al., 1998).

Table 3: Descriptive Statistics of Total Annual Rainfall and Total Rainfall During Wet
Season in Chokwe and Xai-Xai Districts (mm).

 

 

 

Standard
District Description Minimum Maximum Mean Error
Chokwe Total Annual Rainfall 179.7 986.5 604.5 34.5
Total during Wet Season 145.0 1009.8 500.4 31.5
Xai-Xai Total Annual Rainfall 355.1 1744.3 1004.1 41.5
Total during Wet Season 293.8 1395.0 748.7 36.8

 

Source: Survey 2004

15

mi:

 

A
initial Sta
during th.
paribus. s
has both 1

than C hoi

1.40
120

1C

Precipitation (mm)
8

40

20

The

(The gr0“u»

dBVejopmer

EXCegSiVe \ .

er a1" 1998

/

AM ..
[éLOrqu
‘. Tespgk,

 

According to Bouwkamp (1985), sweetpotato requires 18 mm/week during the
initial stages of crop establishment, increasing gradually to approximately 44 mm/week
during the midstage of the crop’s vegetative cycle. Therefore, it is expected that ceteris
paribus, sweetpotato farming in Xai-Xai would be relatively less constrained--given it
has both higher total annual precipitation and more rainfall during the growing season

than Chokwes.

Figure 1 : Monthly Average Precipitation in Chokwe

and Xai-Xai Districts
160 -

140 .
120 J
100 -
80 ~

60+

Precipitation (mm)

40*

201

 

 

Months + Xai-Xai + Chokwe

Source of data: National Institute of Meteorology

The optimum temperature amplitude for obtaining good root yield and abundant
vine growth is 20-30° C. Temperatures lower than 10° C negatively affect root
development and cause chilling injury whilst temperatures above 300 C promote
excessive vegetative growth of aerial parts at the expenses of root development (Salunkhe

et al., 1998).

 

According to the agroecological zoning of Mozambique, Xai-Xai and Chokwe are classified as R2 and
R3, respectively (MAF, 1996).

 

[g -I.vr..xY-"'I ' 7‘

 

ecologiCJ

A,
belongs. tt
Sate Rix c
to norther:

central an.

 

 

2.1.3 Major Farming Systems
The southern region of Mozambique covers essentially three major agro-
ecological zones, namely:
. Inland Maputo Province (R1)
0 South Save Coastal Region (R2) and

o Central-North Gaza and Western Inhambane (R3)

Agroecological zones R2 and R3 cover Gaza province. The district of Xai-Xai
belongs to R3 and Chokwe mostly falls within R3. The Coastal Region South of the
Save River (R2), which comprises an extensive zone from southern Maputo province up
to northern Inhambane province, has the highest country’s population density. In contrast,
central and northern Gaza province and the western part of Inhambane province, which
are the most arid regions of the country, have a relatively low population density
(Snijders, 1986).

In 1995, the Gaza’s Provincial Land Use Planning team identified in Xai-Xai,
6,802 hectares of lowland alluvial fertile soils (Bilene system) and 3,363 hectares of peat
soils that were potentially suitable for crop production by smallholders, but lacked
drainage system rehabilitation (N yamuno et al., 1995). However, due to the severe
flooding that took place in recent years in Gaza province and because very little or no
investment has been made in the rehabilitation of drainage systems, the exploitation of
these land resources with high potential for agricultural production has been hampered.
Furthermore, frequent cases of flooding do occur in the region, due to the inadequacy of

the drainage systems and poor management Of those land resources.

17

aw

. e' «.2 Av

 

in uplanc

depletion

Table 4: F

‘
Fanning

‘

Mixed
cereal cams:
and cashext

Peat soils st

Bilene 5} 5k.

koconut and

mp5 5.15mi:

Chokwe jm'
scheme 5‘ stj

SOUTCe; Mm]

 

 

3L _ _77 _._fi _._.

Farmers in R2 and R3 follow several farming systems, depending on soil
characteristics and access to irrigation (Table 4). Although farmers cultivate sweetpotato
in upland sandy soils, the risk of crop failure is high due to drought and soil fertility

depletion as a result Of shorter or no fallow periods.

Table 4: Farming Systems and Main Characteristics of R2 and R3.

 

F arming Systems Main Characteristics

 

Mixed Upland sandy soils are cultivated with intercropped maize, cassava,
cereal/cassava/pulse cowpea, pigeon pea, groundnut, and pumpkin, along with cashew trees.
and cashew

Peat soils subsystem Main crops are maize, rice, vegetables, banana, and sugar cane.
Drainage is the most determinant factor for use and management of
these soils.

Bilene system Alluvial fertile soils used for maize grown in both cropping seasons,
including other crops such as beans, cowpea, pumpkin, sweetpotato,
sugar cane, and cotton.

Coconut and annual Sandy and loamy soils used to grow maize, cassava, sorghum,

crops system groundnut, cowpea, sweetpotato, coconut, and citrus.

Chokwe irrigation Traditionally rice is the most important annual crop with maize grown as
scheme system a second crop. Farmers grow maize, groundnut, cowpea, groundnut, and

sweetpotato in plots other than the irrigation scheme.

 

Source: Ministry of Agriculture and Fisheries, 1996

Alluvial fertile soils of bilene farming system are Often heavy clay in texture.
Consequently, a lack of access to animal traction or a tractor for land preparation and
drainage are major constraints for sweetpotato farming. Conversely, the cultivation of

peat soils is labor demanding for manual land preparation and is constrained by a lack of

18

 

investm

perform

prod ucti
S“ eetpol
“ illCl'l \t'.‘
CUlIl1aIc(

followed

 

50m.

 

investment in drainage system improvement and a lack of organization among farmers to

perform the regular maintenance of drainage systems.

2.2 Sweetpotato Production

Although there exists no current information regarding trends in sweetpotato
production by province, available data indicates that in the 2000/01 cropping season,
sweetpotato was cultivated on 7% of the total area devoted to roots and tubers crops,
which was equivalent to approximately 1% (48,000 hectares) of the country’s total area
cultivated. Gaza province was the country’s major sweetpotato cultivated area (29%),

followed by Zambezia (21%), Maputo (10%), and Manica (9%) (Figure 2).

Figure 2: Area Cultivated of Sweetpotato in Mozambique (2000/01
Cropping Season)

 

 

 

Inhambane 1—
Sofala i
Manica .

Tete
Zambezia . : ‘-
Nampula ,,

C. Delgado "

Niassa

 

 

  

 

   

im'ltmmilfiflm‘.’

 

 

r I r I I T

0 2.000 4.000 6,000 8.000 10,000 12,000 14,000 16,000

hectares

Source: INE, 2002

19

 

 

mama

indw

Acom'

Theme

"u ’n ‘
.‘LJicE OI j

her i
sif‘ “ll-:7"

iit

 

 

 

 

Moreover, data from Agricultural and Livestock Census 1999/2000 (CAP)
indicated that Mozambique produced a total of 196,646 metric tons (mt) of sweetpotato
in the 1999/2000 cropping season. Zambezia province accounted for 21%, followed by
Manica (16%), and Tete, Sofala and Gaza (12% each).

In recent years, there has been little change with respect to the sweetpotato area.
A comparison of data from Agriculture Survey 1996 (TIA 96) and the Agricultural and
Livestock Census 1999-2000 (CAP 2000) indicates that the total sweetpotato area —as a
percentage of total country’s cultivated area--was 1.3 and 1.2%, respectively. The
estimated percentages Of total holdings that cultivated sweetpotato apparently increased
slightly from 9.4% (TIA 96) to 12.3% (CAP 2000).

Sweetpotato plays an important role in smallholder farming systems. The 2002
baseline national survey on cassava and sweetpotato production, processing and
marketing-conducted in five of Mozambique’s ten provinces and covering a total of
1,467 respondents--identified sweetpotato as the third most important crop after maize
and cassava--based on their use for food (57% of respondents) and its value as either a
source of household income (19%) or for use in exchange (1 1%). A total of 59 varieties
were identified in the 43 districts of the five provinces surveyed. According to this
baseline survey, 50% of the surveyed sweetpotato-growing farmers acquired land through
inheritance, 29% by arrangement with local authorities, 14% rented land and only 5%

bought land (INIA-IITA/SARRNET, 2003).

20

 

irrigation
Ponugue:
reflected 1

Fl:
the large-:
Which u c]

combinati

(I?
(1

occurred

U

-.

01 Cljuun I
lie. 5011th
of labor \x'
PIOtlUCllor
USPS.

Fi :’
“LIT 1k Ullurg
Cl”Q'l’fi‘rati:

v
I l {3'
A1 i‘lc.

 

 

2.3 Historical Agricultural Development and Farmers’ Characteristics

Historically, smallholders in the region practiced rainfed agriculture because the
irrigation scheme were exclusively devoted to large-scale plantations or cropped by
Portuguese farmers. Over the past 50 years, the southern region of Mozambique
reflected the various phases of rural development that characterized Mozambique.

First, a differentiated peasantry was highly involved in providing forced labor for
the large—scale, monoculture export crop plantations (e. g., sugar, tea, copra, and cotton),
which were controlled by foreign capital, petty producers of cash crops, or often a
combination Of both.

Second, changes in the pattern of labor utilization in the peasant economy
occurred at the expense of subsistence production. As a consequence of the introduction
of cotton production, farmers switched from a more labor-demanding subsistence crop
(i.e. sorghum) to a less labor-consuming one (i.e. cassava). Also, changes in the division
of labor within the family occurred--women became responsible for subsistence
production and men became responsible for supplying wage labor or producing cash
crops.

Finally, successive changes have been implemented in post-independence
agricultural policy, from the organization and mobilization of people into state farms and
cooperatives to today’s emphasis on agriculture reform and increased market orientation
(Hermele, 1988; Wuyts, 1978).

According to a recent Agriculture and Livestock Census6 (CAP)-- conducted from

September 2000 to March 2001--the numbers of smallholder farms, medium-scale farms,

 

6 CAP did not collect data on land with permanent trees, pasture. land under fallow, forestry land, and

forestry areas used for social purposes.

21

 

 

and com:
estimatet
arproxin
190 0 01. 11
pro\ ince

crops. an

Table 5:

\

Fists "Lang
Heldmgy

Area
Cdlll\a[¢d

(Ullnalcd

FC‘Od CTOF‘
Area 1 ha 1

Cash Crop.
Area I: ha 1

SI." Bet p‘fll! at.
Area 1 h a .1

>0ur¢

e; [\l

and commercial farms (during the 2000/2001 cropping season) in Gaza province was

estimated at 216,917; 2,544; and 80, respectively. Gaza province accounted for

approximately 7% Of the country’s smallholders, 25% of its medium-scale farms, and

19% of its commercial holdings. During the same cropping season, smallholders in Gaza

province cultivated 99% of total farms, accounted for 11% of the total area in basic food

crops, and 29% of the total cultivated sweetpotato area, as shown in Table 5 (INE, 2002).

Table 5: Number of Farms, Total Food Crop Area, Average Area, and Sweetpotato
Cultivated Area by Small, Medium, and Commercial-scale Farmers in Gaza
Province, 2000-2001 Cropping Season.

 

 

Small-scale Sector Medium- Commercial Gaza’s
scale Sector Total Share of
Number Sohare Sector National
U0) (Number) (Number) (Number) Total (%)
Plots/Land 216. 917 98.8 2, 544 80 219, 541 7.1
Holdings
Area 435,865 94.8 20,792 3,282 459,939 11.9
Cultivated (ha)
Average Area 2.0 NA 8.2 41.0 2.1 NA
Cultivated (ha)
Food Crops 340,550 10.8 14,793 2,574 357,917 11.2
Area (ha)
Cash Crops 871 0.54 32 0 903 0.42
Area (ha)
Sweetpotato 13,407 29.0 412 16 13,835 29.3
Area (ha)

 

Source: INE, 2002

22

NA: Not applicable

 

 

1r.
cash crop
importan‘.
fanning r
C ompanii
cotton an.
market 11'

.\.1
province.
percentag-
security 1:”
ll.\'E .\1.-\;

and mu :

 

In terms of area cultivated in the 2000/01 rainy season, sugar cane was the major
cash crop grown by small-scale farmers, followed by sunflower, cotton, and tobacco. It is
important to point out that the absence of a commercial sector engaged in cash crop
farming reflects the recent shift from the large concession companies (Joint Ventures
Companies) towards contract farming in which small-scale holders cultivate primarily
cotton and tobacco. The main role of the companies is to provide inputs and a secure
market for cotton.

Maize was the most important food crop grown by small-scale farmers in Gaza
province, followed by cassava, peanut, cowpea, beans, and sweetpotato. The highest
percentage of land devoted to food crops reflects the paramount importance given to food
security through subsistence agriculture, due to failures in rural output markets
(INE/MADER, 2002). Thus, the small-scale farming sector is by far the most important

and most representative agriculture activity in Gaza province (Figures 3 and 4).

Figure 3: Area Cultivated (ha) of Cash Crops in Gaza Province
(2000/01 rainy season)

 

 

 

 

 

 

 

 

 

 

 

Tobacco
Soya 4
Sisal I
Sunflower A ., f . . UCommercial
I IMedium scale
Ginger ‘ DSmall scale
Tea
Sugar caneE 1..- g . - T
Cotton L j .
0 100 200 300 400 500
ha

Source: INE, 2002

23

.~ “Hi-i

 

 

SC';:

Ca:

1

E-

I

CC:

PeI

Sweet Lil
2.4 S“
In I
apmelma
Mature, 9'
CiOSe C0113 I

Wham-V t...
LIIOn

3'"?
«1' Our ,
U
3:111

 

Figure 4: Area Cultivated (ha) of Food Crops in Gaza Province
(2000/01 Rainy Season)

Rice
Corn
Millet
Sorghum
Cassava
Beans
Cowpea

Peanut

 

Sweet potato

0 20.000 40.000 60.000 80,000 100,00 120,00 140,00 160,00 180.00
M 0 0 0 0 0

E1 Small scale I Medium scale I Commercial

Source: INE, 2002

2.4 Sweetpotato Multiplication and Distribution Program

In February and March 2000, Mozambique experienced floods that devastated
approximately 92,000 ha of cultivated land in the southern and central provinces of
Maputo, Gaza, Inhambane, Manica, and Sofala. Soon after the floods, SARRNET (in
close collaboration with INIA and other relevant stakeholders) launched a program to
multiply and distribute planting materials of sweetpotato and cassava. This program was
primarily targeted towards flood-affected rural people in an effort to boost agricultural
production and mitigate the scourge of hunger, malnutrition, and food insecurity by

encouraging farmers to plant improved varieties of sweetpotato and cassava. It is

24

 

 

coordin:
sweetpol
ha of cas
source 01
(fig, uni
\aneues
institutioi
imT‘rox ed
managed
:nultiplic:
ngram C

which dat.

Table 6:
\

ReElfin

Reuben]
emral
' 0rthem

Smite: IN

.4' .
“Slab.
. ““011

 

To produce sufficient planting material, the program developed a well-
coordinated network of 124 partners who planted both conventional (222 ha of
sweetpotato, 218 ha of cassava) and rapid multiplication plots (3.2 ha of sweetpotato, 5.1
ha of cassava) (INIA-IITA/SARRNET, 2003). First, research stations identified the best
source of planting materials of improved varieties and multiplied them in primary plots,
(e. g., uniformity, purity, physical and pathological characteristics). Subsequently, these
varieties were multiplied at secondary plots, which were managed by district agriculture
institutions or NGOs. Finally, as the volume of planting materials increased, these
improved varieties were further multiplied at tertiary plots, which were operated and
managed by farmer groups or individual farmers. The estimated areas of these
multiplication plots, districts covered, and beneficiaries of the massive distribution
program of vines of improved sweetpotato varieties, varied by region (see Table 6, in

which data for Gaza province is in parentheses).

Table 6: Multiplication Plots, Districts Covered, and Beneficiaries of Sweetpotato

 

 

 

Vines.
Multiplication plots (ha)
Region Districts covered Beneficiaries
Rapid Conventional
Southern 8.2 (2. 4) 69.3 (45.8) 20 (11) 29,074 (10,150)
Central 3 .6 125 .6 3 7 74,677
Northern NA 30.55 25 46,765
Source: INIA-IITA/SARRNET, 2003 NA = Not Available

In Gaza province, ten institutions were actively involved in the multiplication and
distribution program: Chokwe Research Station, Rural Extension, the Foundation for

Community Development (F DC), Lutheran World Federation (LWF), Caritas Regional

25

(16 C110}:

 

i Chfldrer

Tecnico>

cohabor.

impnne.

 

edunqu.
hrpnnL.

“ hi¥h I‘kIt

 

 

iPVIA-lli

ZS (jj

IT:
Tecem pus:
mBJorin 0
P’C‘ducrio:

Ekha)€t.

 

de Chokwe (CRC), World Relief Services (WRS), World Vision (WV) and Save the
Children (STC), Christian Council of MOzambique (CCM), and “Associacao dos
Tecnicos Agro-Pecuarios” (ATAP).

In addition to providing leadership to the program, SARRNET in close
collaboration with its partners, conducted several training sessions to teach farmers about
improved sweetpotato agronomic practices, their nutritional value, agro-processing
techniques options for sweetpotato; developed teaching materials on nutritional concepts
for primary school curriculum; and sponsored field days/demonstration activities at
which food products made out of the roots of orange-fleshed sweetpotato were displayed

(INIA-IITA/SARRNET, 2003).

2.5 Chapter Summary

The Mozambican economy had shown relatively high GDP growth rate over the
recent post-war years (11.6% from 1998 to 2002). However, the livelihood of the
majority of the population relies on subsistence agriculture. Thus, the low levels of
production and productivity in conjunction with the relatively high population growth
rate have contributed substantially to increased unemployment and widespread poverty.

The small-scale farming sector is by far the most important and most
representative agriculture activity in Gaza province. The absence of a commercial sector
engaged in cash crop farming reflects the recent shift from the large concession
companies towards contract farming in which small-scale holders cultivate primarily
cotton and tobacco. The main role Of the companies is to provide inputs and a secure

market for cash crops.

26

 

 

 

 

multipli.
was laur: '
addition.
sessions '
nutrition.
materials
dais den.
fleshed 5
Gaza pro:
hléh risk 1
Varieties j.
conditiom

50:15 1mm.

 

Gaza province has experienced a cycle of droughts and flooding. A program of
multiplication and distribution program of OF SVs targeted to flood-affected rural people
was launched in 2000 aimed at mitigating the hunger, malnutrition and food insecurity. In
addition, SARRNET in close collaboration with its partners, conducted several training
sessions to teach farmers about improved sweetpotato agronomic practices, their
nutritional value, agro-processing techniques Options for sweetpotato; developed teaching
materials on nutritional concept for primary school curriculum; and sponsored field
days/demonstration activities at which food products made out of the roots of orange-
fleshed sweetpotato were displayed. One of the major constraints for crop production in
Gaza province is the low and irregular rainfall during the growing season, resulting in
high risk Of crop failure. Thus, the development of short-duration (early-maturing)
varieties is crucial towards an increased crop production under rainfed growing
conditions. Furthermore, the lack Of investments in drainage systems on lowland fertile

soils threatens the use Of these soils for agricultural development.

27

CHAPTER III

LITERATURE REVIEW

This chapter reviews the literature related to the potential contribution of 13-
carotene to alleviating VAD (the rationale for Mozambique’s OFSVS distribution
program), summarizes the findings of studies that have analyzed the impact of food-

based interventions on mitigating VAD in developing countries, and highlights factors

 

that have been found to affect farmer’s adoption of new agricultural technologies in order
to gain insights regarding constraints to the adoption of nutritionally improved OFSVS in

the study area.

3.1 B-Carotene 7 as Precursor of Vitamin A
VAD results in various diseases, including follicular keratosis in the skin and
pathological drying of the eye, leading to xerophthalmia8 and sometimes keratomalacia9

(Sommer, 1982, Davidson er al., 1979).

 

7 Beta-carotene is one of the sources of Vitamin A widely found in plant substances namely dark green

leaves such as amaranth, sweetpotato and cassava leaves, mangoes, papaws, tomatoes, carrots, yellow,
maize, and red palm oil (Lathan, 1979, West and Damton-Hill, 2001).

Xerophthalmia is an extreme dryness and thickening of the conjunctiva resulting from VAD,
characterized by lesions appearing as spots, either white or yellow, scattered along the sides of the
blood vessels (Davidson et al., 1979).

Keratomalacia is an advanced form of xerophthalmia frequently associated with protein-energy
malnutrition (PEM) in young children and is characterized by softening and dissolution of the comea
(Davidson er al., 1979).

28

43.x!
\

n" . 1 '( 1—'~L'

 

 

precu:
Rhich
contrc
the ab
1979 l.
\V
Vitam
1H0
alloua
of retin
ofage,

Eears 01

Inc} _
1111“]ch

 

 

 

Several studies have been conducted to elucidate the role of B-carotene as a
precursor of Vitamin A10 in human health. The conversion of B-carotene into Vitamin A,
which takes place in the walls of the intestines and liver, is strongly determined and
controlled by Vitamin A levels. Intestinal diseases (common in the tropics) tend to reduce
the ability of the body to convert carotene to Vitamin A (Davidson et al., 1979, Lathan,
1979).

While the daily requirement for an adult female ranges from 750 to 1,200 ug of
Vitamin A, children (0-59 months Of age) require 250 to 300 ug Of Vitamin A
(FAO/WHO, 1967). However, some countries have their own recommended daily
allowances (RDA). For instance, the RDA for adult women in the United States is 800 ug
of retinol, while for children the RDA is estimated at 375 ug from birth to about one year

of age, 400 ug from one to three years. and 500 pg of retinol per day from four to six
years of age (NRC, 1989).

Bioavailability of dietary carotenoids and their conversion to retinol are
influenced by several factors, including the type of carotenoid; amount of carotenoid in a
meal; matrix in which the carotenoid is incorporated; presence Of absorption modifiers
(e. g., fiber which inhibit carotenoid absorption, and dietary fat which enhances
absorption); nutritional status of population groups; and human-related factors (e. g. age,
gastro-intestinal infections and parasites (West, 2000; de Pee er al., 1998, van het Hof et

al., 2000).

 

10 . . . . . .
Vitamin A is an Important component of the Visual purple of the retina of the eye and appears to play

an important role in immune competence at the level of both the surface tissue (i.e., skin and mucous
membranes) and the immune system (Jamison, 2003).

29

De Pee and West (1996) suggested that studies are required to quantify the impact
of these factors, especially related to the matrix, human-related factors, and absorption
modifiers. According to Lathan, (1979), if no animal products are being consumed and
the human body entirely relies on carotene as a precursor Of Vitamin A, larger quantities
should be consumedl 1. Secondly, carotene is poorly utilized in a low fat diet and, in most
cases, diets that are deficient in Vitamin A are also deficient in fat. Thirdly, intestinal
diseases limit the conversion of carotene and the absorption of Vitamin A. Finally;
infants and young children do not convert carotene to Vitamin A as readily as adults. In
the human body, provitamin A carotenoids are converted to retinol by a cleavage enzyme
(located in the intestinal cells) that is directly exposed to various consumed food
components. However, the properties Of this cleavage enzyme are not yet well known and
the regulatory mechanism Of the dioxygenase still needs to be examined in terms of the
nutrition of pro-Vitamin A carotenoids (Olson and Hayaishi, 1965).

A study which evaluated the effects of dietary components on metabolism of B-
carotene in intestinal cells revealed that intestinal infection levels of young children
heavily hamper the absorption and conversion of sources of B-carotene to Vitamin A in
most developing countries (Nagao, er al., 2000).

A study conducted in Indonesia revealed that B-carotene fortified wafer
significantly improved the serum retinol concentrations in lactating women. However, no

improvement in Vitamin A status occurred when the same amount Of B-carotene was

 

” About one third of B-carotene in food is absorbed and available. Due to losses occurring during the

conversion of carotenes into Vitamin A, one pg of absorbed B-carotene has the biological activity of

half pg of retinol. Thus, one pg of B-carotene in a mixed diet is equivalent to one-sixth pg of retinol
(FAO/WHO Expert Group, 1967; Lathan, 1979, NRC, 1980). However, other scientists argue that 21
pg of B-carotene from fruits and vegetables are required to provide one pg of retinol (West, 2000).

30

provided in the form of dark-green leafy vegetables. The authors suggested that B-
carotene from dark-green leafy vegetables is not readily absorbed because it is trapped in
a complex matrix within plant cells (de Pee er al., 1995, de Pee et al., 1998).

Recent studies revealed that Vitamin A plays an important role in a number of
nutrient-nutrient interactions. Excessive doses of Vitamin K may reduce absorption of
Vitamin A. Vitamin C may impair the toxic effects of hypervitaminosis A. By protecting
against the oxidative destruction of Vitamin A, the therapeutic efficacy of this vitamin is
enhanced by Vitamin E. Zinc deficiency reduces the bioavailability of Vitamin A and
thus, limits its functionality (Jamison, 2003). According to Rahman et al., (2002),
combined zinc and Vitamin A supplementation improved Vitamin A nutriture in Vitamin
A-deficiency young children. Similarly, significant protein deficiency may impair the
impact of Vitamin A in mitigating related diseases, since protein plays an important role
in transporting Vitamin A from liver stores to target cells. Several studies revealed that
the presence of Protein Energy Malnutrition (PEM) increases the risk of xerophthalmia
by approximately the same order of magnitude, as does the occurrence of diarrhea and
respiratory diseases (Davidson er al., 1979). Thus, successful prevention and treatment of
VAD and its adverse consequences, requires a careful consideration of the targeted

population’s protein status (Sommer and West, 1996).

31

3.2 Impact of Food-based Interventions to Mitigate VAD

Several studies have documented the impact of food-based initiatives envisaged to
prevent VAD. In many developing countries, pro-Vitamin A carotenoids are the major
source of dietary Vitamin A, contributing more than 80 % of the total Vitamin A intake.
However, the lower bioavailability of Vitamin A in vegetables and fruits, and probably
also the seasonal variability of production of vegetables and fruits in household gardens,
are factors underlying the causes of VAD in these regions (Bloem et al., 1998).

Studies conducted in developing countries, such as Bangladesh and India,
documented low dietary Vitamin A intakes by preschool-age children and found that
most Of the nonbreast milk Vitamin A intake was derived from plant sources (Zeithin et
al. , 1992). Although mothers declared that their young children had frequently consumed
dark-green leafy vegetables, the amounts consumed were considered too small to satisfy
their Vitamin A requirements. In addition, it was Observed that young children were
likely to consume exorbitant quantities Of mangoes and papayas, which were available
for short period of time. Thus, given the poor and seasonal patterns of dietary Vitamin A
intakes, low fat intakes, parasitic infections, and the reduced bioavailability of Vitamin A
from dark-green leafy vegetables, these diets were inadequate to meet the Vitamin A
requirements of preschool—age children (one to three years Old) without the inclusion of
animal-based sources Of Vitamin A and prolonged breast-feeding (Ramakrishnan et al. ,

1999).

32

 

 

SW6:
Sout
by II:

who.

intcgm
4' .
clear}
Ol‘VOun
CODITOI)

lb? 11m,

T“-\
A. Boil,
fRAE
1RD‘

URL I
Gui-it,

’0 i_"-f

 

rile: f
lion ,
Fear,

 

A study, aimed at determining the efficacy of boiled and mashed orange-fleshed
sweetpotato'2 in improving the Vitamin A status of 5-10-year-old school children in
South Africa, concluded that the Vitamin A status of the targeted population (measured
by the modified relative dose response (MRDR13) improved relative to the control group,
who received boiled and mashed white-fleshed sweetpotato (van J aarsveld et al. , 2003).

A study conducted in South Africa showed that fortification of a biscuit14 with
iron, iodine, and B-carotene significantly improved the micronutrient status of primary
school children-- 40% of whom exhibited subclinical VAD at the baseline assessment,
due to low availability of B-carotene in the dark-green leafy vegetables which they
consumed on a regular basis (van Stuijvenberg et al., 1999).

A study carried out in South Africa, designed to determine the effect Of an
integrated home—gardening program plus primary health care program on improving
dietary intake of yellow and dark-green leafy vegetables and serum retinol concentrations
of young children, concluded that school children in both villages (experimental and
control) had comparable serum retinol concentrations. Consequently, it was assumed that

the two groups of 2-5-year-old children would have the same mean serum retinol

 

Boiled and mashed orange- fleshed sweetpotato provided 1,031 pg of Retinol Activity Equivalent
(RAE’Z) per 125 grams serving, which corresponds to 250% of Recommended Dietary Allowance
(RDA) (van Jaarsveld er al.. 2003).

MRDR value is the ratio of the serum concentrations of vitamin Az/vitamin A. as described by
Tanumihardjo er al., 1996. Although this MRDR is considered as the most practical method for field
collection is more sensitive to changes in vitamin A status (by distinguishing between normal (DR:R
<0.060) and abnormal (DR:R >0.060) vitamin A status) than in serum retinol concentrations alone.

The fortified biscuit was designed to provide 50% of the recommended daily dietary requirement of

iron (5 mg of ferrous fumarate), iodine (60 pg potassium iodate), and B-carotene (2.1 mg) to 7-10-
year-old school children.

33

9.
- .. run-nu.- i

 

 

COIlC

anthr

xerop
group
consu
hem e
garder

all frui

 

 

 

concentrations. Furthermore, a follow-up study found no significant difference in
anthropometric indexes of nutrition between children in both Villages (Faber et al., 2002).

A study by Shankar et al., (1998), designed to compare the relationship between
access to home gardening and animals among households with and without
xerophthalmic children in rural Nepal, found that although access to gardens for both
groups was high and not significantly different, there was no difference in the
consumption of Vitamin A-rich vegetables. This suggested that there was no relationship
between the frequency of consumption Of carotenoid-enriched vegetables'5 and home
garden size. Conversely, there was an increase of non-carotenoid-enriched vegetables and
all fruits.

In Bangladesh an integrated homestead gardening program, which also promoted
primary health care and nutritional education, demonstrated the potential impact of a
comprehensive program. Among 2-5-year-old children, the serum retinol concentrations
were strongly associated with increased intake of dark-green leafy vegetables (Bloem et
aL,1996)

Furthermore, West and Darnton-Hill (2001), argued that food—based interventions
that are implemented with the active participation of women with community
engagement through social marketing16 approaches are likely to introduce food habit

changes, improve Vitamin A intakes, and eventually enhance Vitamin A status. However,

 

In this study, carotenoid-enriched vegetables included dark-green leafy vegetables and ripe pumpkin
and carotenoid-rich fruits included papayas, mangoes, and jackfruit.

Social marketing is defined as “the design, implementation, and control of programs calculated to
influence the acceptability of social ideas, involving considerations of product, planning, pricing,
communications and market research” (Kotler and Zaltman, 1971). This approach requires the
community participation in the identification of problems and needs, implementation, monitoring and
evaluation of the programs. This approach increases the community responsibility in planning,
monitoring and assessment of collective actions that were agreed upon and proved necessary to be
undertaken (Neill, 1992).

34

. .
..~ g—u—-‘————
W . V."..E"fl‘.'ovn
_——-

 

Nei

had

done

comn

rdanx

0
E3
’3‘
(L

Hudies
cfiEcU\
unprOI;

2000 ,_

 

 

Neill, (1993) found no clear evidence that the Indonesian nutritional education project
had a long-term impact on the mortality of children under five years of age and continued
implementation of such a program is unlikely without long-term financial support from
donors.

De Pee and West (1996), in their review of the literature (a total Of sixteen studies
conducted in developing countries) regarding dietary carotenoids and their role in
combating VAD, pointed out that while many cross—sectional, case-control, and
community-based studies have shown that increased intake of fruits and vegetables is
related to improved Vitamin A status, this does not entirely prove causality. Furthermore,
many experimental studies that indicated a positive effect of fruits and vegetables intake
can be criticized for their poor experimental design. Furthermore, recent experimental
studies have found no effect of vegetable intake on Vitamin A status. Thus, the
effectiveness Of carotene-rich foods in improving Vitamin A status and ways of
improving carotene bioavailability need further investigation (de Pee et al., 1996, West,
2000)

Currently, researchers utilize the food frequency method, developed by the Helen
Keller International (HKI), to identify people at-risk of suffering from VAD. For
example, using this method, researchers (based on a positive change in HKI scores)
concluded that food-based interventions in Ethiopia, Kenya, and Tanzania enhanced
Vitamin A status (Johnson—Welch, 1999). However, findings Of research in other
countries showed that improvements in HKI scores, as a result of an increase in
consumption of dark-green leafy vegetables alone, did not result in improvement of

Vitamin A status.

35

countri

 

F;

,.J

n—— .

$4..
—

 

In fact, due to a variety of foods, HKI scores were improved (de Pee et al., 1995).
Thus, Johnson-Welch (1999) emphasized the need to develop methodologies to better
assess the nutritional impact of food-based intervention trials conducted in developing

countries.

3. 3 Adoption of New Technologies

Many studies have sought to explain farmer adoption of new technologies,
including specific traits of improved varieties. A Kenyan study, which evaluated the
effect of women farmers’ adoption of OF SVs in raising Vitamin A intake, found that
several of the new OFSVS grown in on-farm trials were adapted to the agro-ecological
conditions with respect to yield, pest and disease tolerance, as well as having reasonable
beta carotene content. This study found that women farmers were likely to adopt the
OFSVS if the clones were sufficiently high in starch, low in fiber, and if they were
introduced through community-level education programs that focused on the health of
young children (Hagenimana and Oyunga, 1999). According to this study, the new
OF SVs were widely accepted (with respect to their appearance, taste, and texture) by
both producers and consumers, and substantially contributed to the alleviation of VAD.

Furthermore, a recent ex ante impact case study by economists from Michigan
State University and the International Potato Center estimated that widespread
distribution Of the new orange-fleshed cultivars could benefit an estimated 50 million
children under age six, who currently suffer from VAD in Sub-Saharan Africa (Low et

aL,2001)

36

e . v
-. . L
“). ‘Ihfl'“'- '

  

\
F» r .
1 ., .2 '....-

 

enxirt
had be
in Eth

techno

conclus
Constrdg
conditit
a need 1

ent'imn

 

 

 

A study by Maredia and Minde (2002), which explored the relationship between
profitability of agricultural technologies and its adoption by farmers in Eastern Africa,
identified three groups Of technologies: First, some profitable technologies were widely
adopted, such as improved cassava varieties in Uganda and improved coffee varieties in
Kenya. Second, some technologies were profitable under researcher-controlled
environments, including technologies that were not as fully adopted as expected, and/or
had been restricted to on-farm demonstration plots (e. g., wheat variety and hybrid maize
in Ethiopia and the application of inorganic fertilizer on maize in Kenya). Third, some
technologies were unprofitable under specific circumstances, such as animal draught-
power technology for weed control when applied on farms less than 2 hectares. In these
studies, profitability was estimated using partial budget analysis, gross margin analysis,
net present values, and value cost ratios. Furthermore, the authors drew two major
conclusions: i) the productivity gap in Africa is heavily determined by non-technological
constraints (6. g. infrastructure, policies, input/output markets, and adverse climatic
conditions), which reduced profitability and adoption of new technologies, and ii) there is
a need for continuous efforts to supply technologies that are adapted to the prevailing
environmental conditions in order to make them profitable for farmers.

A study of the key factors associated with the adoption of hybrid maize in Latin
America and the Caribbean region by Kosarek et al., (2001), reported that farmers’
decision to adopt hybrid maize was determined by the expected returns (i.e. profitability)
of the technology, the availability of hybrid seed, and risks associated with uncertainty
regarding the expected outcomes of the new technology. Moreover, they found that the

structure of the seed market, the organization of the seed industry, and the cost of

37

technology generation and development were key determinants of the profitability of
supplying hybrid maize seed. One of the major conclusions of this study was that public
research plays a vital role in insuring that the needs of small-scale and subsistence-
oriented farmers are met in zones where the welfare of the population is entirely
dependent on maize production.

Exploring the key determinants of the adoption of technologies by farmers
growing upland rice and soybeans in Central—West Brazil, Strauss er al., (1991), reported
that farmer adoption of technology is an economic decision based upon discounted
expected marginal benefits and costs. Furthermore, the major findings of this study
included: (i) farmer’s education level contributed positively to the probability of soybean
farmers performing soil sample analysis to'determine the quantity of fertilizer that they
should apply on their rice fields; (ii) time of residence in the region was positively related
to rice growers’ adoption of blast control and to the probability of adopting certain
planting techniques for soybeans; and (iii) the use of certified rice seed, soil sample
analyses, type and quantity of fertilizer application per hectare, and inoculation of
soybean seeds were positively related to availability and quality of the extension and
research.

Similarly, a study by Rahm and Huffman (1984), designed to evaluate the role of
human capital and factors that affected the adoption of reduced tillage in corn production,
found that farmers’ education and experience play a crucial role in enhancing the
efficiency of the adoption decision. Moreover, they concluded that the probability of
farmers adopting reduced tillage practices was strongly depended on soil characteristics,

the cropping systems, and farm size.

38

   

-.'l
C'..

’7' I bq'w‘w

“11h;
adOFE

thCP‘

.(D
m
l 1'

ddfnl
ӣ15017
ndmfi
theUN¥
Dune?

inJHO\

KeHye
remnn:
mabet
hafibu
human
damni

cereals

 

 

Various studies reported that risk aversion is likely to be negatively associated
with adoption and for any level of farmer’s risk aversion, the likelihood of technology
adoption is positively correlated with the availability and accuracy of information about
the performance Of the specific new technology (Feder and Slade, 1984, F eder er al.,
1985, Kristjanson, 1987).

Recent studies conducted in Eastern Africa, designed to better understand key
determinants of farm-level technology adoption/disadoption, revealed that the major
reasons for technology non-adoption were: (1) farmer’s unawareness of the improved
technologies or a lack of information regarding potential benefits accruing from them; (2)
the unavailability of improved technologies; and (3) unprofitable technologies, given the
farmer’s agro ecological conditions and the complex set Of constraints faced by farmers
in allocating land and labor resources across farm and off-farm activities (Doss, 2003).

In recognition of two decades of on-going market reforms in Sub-Saharan Africa,
Kelly etal., (2003), suggested the need to assess the profitability of input use and
potential explanations of non-profitability before drawing any conclusion regarding
market failure or the limited impact of government intervention, such as input
distribution programs oriented towards smallholders. Profitability studies carried out in
Mozambique and Ethiopia revealed that, after adjusting for the effects of market
distortions (i.e., taxes, subsidies and currency overvaluation), application of fertilizers on
cereals was unprofitable--given existing input and output marketing costs.

Notwithstanding the limited availability of empirical data on the potential role
played by farmer associations in absorbing some Of the transaction’s costs of input

procurement, Kelly er al., (2003) reported that farmer associations in the irrigated rice

39

 

2011':
pTOCi
10 SU
outp:

farm:

tarme
growl
Seque:
pidnti:
are ea:
“ere n

Tami l }

 

prCen:

head 0 .

(1 -~' fl ,
«1ng £1)

33in, :11

 

zone of Mali have reduced costs for their members by using transparent bidding
procedures for sourcing inputs and by securing bank loans to guarantee timely repayment
to suppliers. Additionally, to decrease the transaction costs of inputs acquisition and
output marketing, some NGOs have promoted the establishment of and consolidation of
farmer associations.

A study by Adesina and Sanders (1991), designed to evaluate the adoption and
farm-level effects of cereal technologies in Niger, documented the ability of peasant
farmers to adapt their cropping strategies to the low and erratic rainfall patterns by: (i)
growing a mix of varieties with different maturity lengths and (ii) making appropriate and
sequential decisions based on rainfall expectations, such as applying several different
planting densities, planting in different dates, and/or applying fertilizers when the rains
are early. They concluded that, in general, small farmers in Sub-Saharan Africa (SSA)
were more likely to adopt labor intensive technologies, as they used relatively more
family labor which had a low opportunity cost. Furthermore, Kristjanson (1987), using an
econometric model to determine the relationship between rainfall variables and crop
yields, found similar farmer behavior in Burkina Faso.

A study designed to estimate the effect of key exogenous sociO-economic
characteristics on the adoption of animal traction technology in Maseru, Lesotho (Mbata,
2001) showed that while two-thirds of female farmers adopted animal traction, only a low
percentage of male farmers did so. This was explained by the recognition that men, as the
head of household, bear responsibility for Obtaining income from Off-farm activities--
such as working in South African mines. Consequently, women in rural areas relied on

animal traction to alleviate the burden Of the heavy farm workload. Due to the high

40

 

 

in:

$01

prc

CID

ado

 

interest rates charged by private moneylenders and the non-eligibility of many small
farmers for receiving loans from the Lesotho Agricultural Development Bank, the main
source of capital for adopters Of animal traction was personal savings.

Feder et al. , (1985) argued that off-farm income might affect adoption by
providing a source of cash flow to buffer the risk associated with the introduction of new
crop management practices. Moreover, the existence of effective extension services,
adequate provision Of inputs, timely credit availability, transportation, and functional
marketing channels are Of paramount importance to foster the adoption of new
technologies (F eder eta]., 1985, Mbata, 2001, Rauniyar and Goode, 1992).

A study designed to assess the role of agricultural technology in improving food
security and providing more employment in Bangladesh evaluated the welfare
implications of the irrigation-induced technological change (i.e. modern varieties,
fertilizers, and pesticides) in rice production. This study found that irrigation-induced
technological change could substantially reduce the percentage of the rural population
below the poverty line by increasing the income of the poor population through
augmented employment opportunities. However, rapid expansion of irrigation was
deemed as a condition sine qua non for farmer adoption and achieving high productivity

(Akhter and Sampath, 1992).

41

 

 

COD

bun

«I '~
Ldft

CDUH

Turti'

FIUBT

 

 

 

3.4 Chapter Summary

Several studies highlighted the importance Of B-carotene as predominant source of
pro-Vitamin A in preventing VAD in developing countries. However, the bioavailability,
conversion Of B-carotene into Vitamin A, and effective utilization Of this Vitamin A by
humans is conditioned by a complex set of factors--including type and amount of
carotenoid in a meal, matrix in which the carotenoid is incorporated, presence of
absorption and conversion modifiers, health status of the target population, and the
presence of other micronutrients.

Studies aimed at assessing the impact of food-based interventions in developing
countries revealed that these initiatives yielded different results from country-to-country.

Furthermore, the seasonal availability of enriched sources of B-carotene, the low

bioavailability and conversion of B-carotene into Vitamin A, and the prevalence of
infectious diseases hampers the efficacy Of home-gardening initiatives to mitigate VAD
in developing countries.

Currently, a multidisciplinary study, which is being conducted under the
auspicious of VITAA (Vitamin A for Africa), is evaluating the potential for reducing
VAD and inadequate caloric intake through dietary diversification, improved feeding
practices, and improved serum retinol status--particularly in young children under five
years of age in Zambezia province. This study will provide more insight regarding the
impact of this food-based intervention in mitigating VAD.

Recent studies have identified many socioeconomic and policy related factors that
affect farmers’ technology adoption decisions, including access to farm labor, farm

machinery, storage and distribution facilities. inputs and output markets, credit

42

" “‘90 . #3

V1: mu. "guy-'4.-.“—

 

 

 

 

availability, farm size, level of education, land tenure, profitability of the technology,
farmer’s awareness of existing technologies, and government policies. However, the
effect of those explanatory variables and interactions among them on farmer’s technology
adoption varies from region to region. Thus, extrapolation of results from one farming
system to a completely different one should be avoided. Rather, country-specific studies
need to be carried out to understand factors associated with farmer adoption of a specific

technology in a specific location.

43

i ' h A . .
I.“ 3" r“. " (“my-+—
' _

 

3d0f

inthi

41

house!
adopti

Rubnn

[ethncd
link-j l‘; \‘

Hti‘l‘m _

uvilt’re

 

I
AF}! ,
““1301.

 

CHAPTER IV

RESEARCH METHODS AND DESIGN

This chapter presents the theoretical and empirical model used to analyze farmers’
adoption decision and describes the sampling approach and data collection methods used

in this study.

4.1 Technology Adoption Decision (Objective 2)

4.1.1 Theoretical Model

Qualitative response models--also called binary-choice, discrete or dichotomous
models--are Often used to evaluate the farmer’s decision-making process concerning the
adoption of agricultural technologies. Those models are based on the assumption that
households and farmers are faced with a choice between two alternatives (adoption or no
adoption) and the choice depends upon identifiable characteristics (Pindyck and
Rubinfeld, 1997). Based on the assumption that the decision made by farmers, regarding
adoption of a particular technology, is guided by a utility maximization objective, a
technology 2 (t2) is preferred to technology 1 (t1) as long as the utility derived from
technology 2 is greater than the utility derived from technology 1.The utility function
ranking the in, farmers’ preference for technologies is represented as follows (Rahm and
Huffman, 1984):

U (Ru; Au)

Where utility U depends on a vector Ru , describing the distribution of net returns for
technology tj and a vector A“ ,corresponding to other attributes associated with the
technology tj. The variables Ru and A, are not observable, but a linear relationship is

postulated for the i", farmer between the utility derived from the tj technology and a

44

\‘e’C’t

 

distt

quaii

 

a, .
‘ a... Y. _' -
[y ks‘uw -‘n_—._.- _

rt r._ _‘

L. .V

and f.

Pi

Where

 

1:31.311,10 I

 

 

vector of Observed farm and farmer characteristics X, and a zero mean random
disturbance term pt:

U" = X" + p, wheret=l,2and i=1,2,....n. (1)

As mentioned previously, the i", farmer adopts t; if Utz is greater than U”. A
qualitative variable Y can represent the farmer’s adoption decision.

Y =1 if U12 > U” and new technology t; is adopted replacing t1 and

Y = 0 otherwise (2)

The probability that Y, is equal to one is expressed as a function of specific farm
and farmer characteristics:

Pi = Pr (Y = 1) = Pr (UH < Ut2)

Pf (Xi Oil + llii<X10i2 + H21)

Pr [ 1111 - 112i < X1012 - (11)]

Pr ( Yr < Xi B) = F (X113) (3)
Where

Pr (.) is a probability function,

yi = p“ - p2, is a random disturbance term

B = (1.2 - or] is a coefficient vector and;

F (X43) is a cumulative distribution function for 7, evaluated at X10.

The marginal effect Of a variable X, on the probability Of adopting new

technology can be calculated by differentiating P, with respect to X]:

6Pi / 6Xij =f(XiB) . Bj, (4)

45

. , . 1 .
Ohm-

agar:

:11

P .
in!“ 1122. “'6 .m—

\Vhe

 

nunfl

THO-(l:

techn

pannr

f.)

 

(As
(I;

 

 

Where f (.) is the marginal probability density function of y, and j = 1, 2, ...... J is the
number of explanatory variables. The general form of the univariate dichotomous choice
model is expressed as:

P, =P,(y,=1) =G(X,, 0)wherei=a,2,....n. (5)

Equation (5) states that the probability that the im farmer will adopt a specific
technology is a function Of the vector of explanatory variables X, and the unknown
parameter vector 0.

Three alternative functional relationships are commonly used by researchers to
specify G: Linear Probability (LP), Probit, and Logit models. A Linear probability model
(Yi = or + B X, + p,) has been used extensively in econometrics applications. However,
its specification has caused estimation problems and the non-normality of the disturbance
terms makes the use Of traditional tests of significance (t-test and F-test) inadequate.
Pindyck and Rubinfeld (1997) summarize the limitations of the LP functional form as
follows:

1. It gives a heteroscedastic regression model and its variance-covariance matrix

varies systematically with the independent variables;

2. The predicted value of X 0 is not restricted to lie between 0 and 1, which is
inconsistent with the definition of Y, as a conditional probability;

3. Some studies have revealed that adoption decision functions are curvilinear
rather than linear. Thus, Ordinary Least-Squares (OLS) would produce inefficient
parameter estimates.

Given the problems associated with the linear probability model, economists have

developed alternative functions that confine the estimated probabilities between 0 and 1.

46

Th

C111

Takir

 

le”;

Ptoc ,

Till} ,

 

The two most common functions used in econometric applications are the logistic and the
cumulative normal distributions, creating the logit and probit models, respectively. Thus,
the probability that a farmer will adopt a new technology is expressed as a function of:
P(Y= l)=F(XB)

According to the logit model, the probability of an individual household (farmer
or firm) adopting a new technology t2, given a well-defined set of socio-economic and
physical characteristics (X). is represented as:

P(tz | X)=exp‘xl’+“) /[l +exp‘xBT“)]

Likewise. the probability Of not adopting the new technology t2 (continuing with
technology t,) is given by:
P(t, IX) =1 - P(Tz IX) = 1 - {exp (“H“) /[1 +exp (””0”

=1 /[1+exp"x”*“’] (6)

The relative Odds of adopting versus not adopting a new technology are given by
P (t2 | X) /P (t, | X) = [exp WW] {1+ exp (”WM [1 + exp WW]
: exp1XB+Hl (7)

Taking the logarithm Of both sides: In [P (t2 1 X) / P (t, | X)] = X B + p (8)

In a logit model, the parameter estimates are linear and, assuming a normally
distributed disturbance term (p), the logit maximum likelihood (LML) estimation
procedure is used to get efficient, consistent, and asymptotically normal estimators.

Those estimates will represent the effects and statistical significance of the explanatory

47

\‘aria't
logit r
mean
illit’t‘lw

takes 1

x- v 's’. .
_ .‘v rim-rur- -

 

._
D
f“)
C)
«a

I - warm}—
'51
(‘D
“:1

deter:

abiding

modfl)

 

 

abill'f i1

 

 

variables on the adoption of a particular technology (Pindyck and Rubinfeld, 1997). In a
logit model, the marginal effect of the explanatory variable is generally computed at the
mean value of the independent variable for continuous variables; while the marginal
effects of categorical variables are estimated by the difference before and after the change
takes place.

The probit model is also more appealing than the linear probability model, since it
incorporates nonlinear maximum-likelihood estimation. Probit analysis accounts for
heteroscedasticity of the error terms and restricts predictions to lie between 0 and 1 range.
The probability of a farmer adopting or not improved technology in the probit model is
defined in terms of an index that may have any value between -00 and + 00. This index is
converted into probability values by using a standard cumulative normal distribution and
this transformation guarantees that all corresponding probability values are confined
between 0 and 1(Pindyck and Rubinfeld, 1997, Maddala G., 1983). The functional form
is represented as follows:

P, = F (2,) = 1 / (2 n )0-5 1 exp W dp

Where Z, = X, B + p,

An estimated [3 value in a logit or probit model does not give the change in the
dependent variable, due to a unit change in the explanatory variable. This effect is
obtained by computing the partial derivative of the Prob (Y, = 1) with respect to [3.

Since logit and probit models yield similar results in the case of binary choice
models (Nayga and Capps, 1992, Maddala, 1983, Amemiya, 1981), the choice of one

above the other is a matter of convenience. This study uses the probit model to evaluate

48

V v‘.
irmr‘" '

i v

”w.\"d-.~c-—
I ' '_
a?! ‘-

 

£ vein

factc

 

\Vlé

l'ariet}
Pr

llhere

 

 

 

factors associated with a sweetpotato grower’s decision to grow OFSVS (rather than

WFSVs) in the 2003/04 wet season.

4.1.2 Empirical Model
Data gathered, regarding household cultivation of orange-fleshed varieties in the
last cropping wet season (2003/04), were used to define two categories of sweetpotato
farmers:
o Adopter, a farmer who planted OF SVs in his/her field in the 2003/04 wet season
(the period when the survey took place), and
o Disadopter, a farmer who previously received vines of one or more improved
orange-fleshed variety, but did not grow any of them during the 2003/04 wet
season. However, all the interviewed respondents grew and harvested improved

orange variety at least once.

The probit procedure for estimating key determinants of orange-fleshed sweetpotato

variety adoption is specified as follows:

Pr =13, + [3, x,, + 32x2, + .................... +13, xm
Where:
Pr = 1 if the farmer cultivated any orange-fleshed variety in 2003/04 and 0 otherwise
I = 1, 2, 3 ................ t are observations
Xm = the nth explanatory variable for the ith observation
n = 0,1, 2 ................. , n
0,. = an unknown parameter; n = 0,1 ........... , n

49

The variables included in the probit model were selected based on both adoption
literature, regarding factors that have influenced technology adoption in several countries,
and location-specific factors that key informants hypothesized would affect the adoption
of OFSVS. The explanatory variables, including the direction of their expected effects,
are described as follows:

0 Respondent age ( - ): The units of this variable are years (age).

0 Level of education (+): This variable is measured as the number of years of
formal education completed by the respondent (educ).

0 Young children (+): This variable is specified as a dummy, taking a value of one
if the respondent has young children (under five years of age) and zero otherwise
(young). Since the message delivered to farmers focused on young children as the
target population to be benefited from consumption of OFSVS, it is hypothesized
that household with young children are more likely to adopt OFSVS.

0 Extension services ( + ): This variable is measured as the number of times the
farmer visited or was visited by an extension technician regarding any
technological issue related to orange-fleshed sweetpotato farming during the last
two years (extvis).

0 Field days/demonstrations ( + ): This variable is specified as a dummy, taking a
value Of one if the respondent participated in field days/demonstration activities
regarding OF SVs processing techniques and zero otherwise (fdemo). It is
hypothesized that farmers who attended these events would be more likely to
adopt OF SV in order to make OFSVS processed products.

0 Livestock ownership ( - ): This variable, a proxy for income, represents the
number of cattle owned by the respondent (tbov). Key informants hypothesized
that wealthier farmers are likely to adopt new technologies since they are more
able to assume risk.

50

Frequency of vines distribution ( + ): This variable is measured as the number of
times the farmer received vines of OFSVS (vinrec). Key informants hypothesized
that farmers who received vines more times would more likely to be growing
them in 2003/04 wet season.

Experience with varieties ( + ): This variable is measured as the number of
different OFSVS that the respondent received (varexp). It is hypothesized that
farmers who received several varieties would be more likely to identify a
preferred variety and continue to grow it.

Off-farm activity ( + )2 This variable is specified as a dummy, taking a value of
one if the respondent and/or spouse participated in off-farm activities in 2003 and
zero otherwise (offarm).

Gender of the respondent ( + ): This variable is specified as a dummy, taking the
value of one if the respondent was female and zero otherwise (female). Key
informants hypothesized that since sweetpotato is a woman’s crop, they would be
more likely to adopt OFSVS.

Family size ( + ): This variable is measured as the number of people who lived in
the household at least nine months during 2003 (famsize). Key informants
hypothesized that larger households would be more likely to adopt a technology
that increases subsistence food production.

Sweetpotato cultivated area in the 2003/04 wet season ( + ): This variable is
expressed in hectares (spared). Key informants hypothesized that households with
a larger sweetpotato area considered it to be an important crop, so they would be
more interested in new OF SVs.

Awareness of sweetpotato processing techniques ( + ): This variable is specified
as a dummy, taking the value of one if the respondent is aware and zero otherwise
(praware). Key informants hypothesized that farmers who were aware Of
sweetpotato processing techniques would be more likely to adopt OFSVS in order
to make processed products out of them.

51

 

Average annual precipitation ( + ): This variable is specified as a dummy”,
taking a value of one if Xai-Xai district and zero if Chokwe district. This variable
captures the difference in average annual precipitation between the two districts
covered by this study (distpr). Rainfall varies greatly between the two districts.
Ceteris paribus, as sweetpotato production is less constrained by rainfall in Xai-
Xai district, farmers are more likely to have been able to maintain vines received
in 2000 and in subsequent years.

Sweetpotato root consumption ( + ): This variable is measured as the number of
times per week the household consumed sweetpotato roots during the months
when roots are available (motes). Key informants hypothesized that farmers who
were more dependant on sweetpotato as a food source would be more interested
in growing Vitamin A-enriched OF SVS.

Commercialized sweetpotato in wet season ( + ). This variable is measured as the
quantity (kg) of orange-fleshed sweetpotato roots sold in the 2003 (rsold). Key
informants hypothesized that farmers who sold OFSVS roots, as a source of cash
income would be more interested planting higher-yielding OFSVS.

Respondent experience in processed products ( + ): This variable is specified as a
dummy, taking the value of one if the respondent has actually made at least one
sweetpotato processed product and zero otherwise (proexp). Key informants
hypothesized that farmers who had made processed products from OFSVS would
be more likely to continue to grow them.

Total household farmland ( + ): This variable is measured as the total area
(hectares) cultivated by the household in the 2003/04 wet season (tland). Key
informants hypothesized that since households with larger farms were more
dependant on agriculture, and more able to assume risk, they would be more
likely to adopt a new agricultural technology.

 

17

It is recognized that this dummy variable might capture the effect of all other no included factors on
farmer adoption of OFSVS. However, given the fact that most respondents reported having lost their
planting materials due to flooding, it seems that the difference in average annual precipitation
influenced the rate of adoption of OF SVs in both districts.

52

4.2 Prices Traders Offer for Roots (and leaves) Produced from Orange-fleshed

Varieties (Objective 3)

There is an increasingly demand for products with improved safety (e.g. lower or
no contamination levels of pesticides or other pathogens) and high nutritional attributes
or characteristics (6. g. Vitamin-812, Vitamin A, Vitamin-B6, phosphorus, magnesium,
iron, lower fat content), since their availability is strongly related to improved social
health status and consequently lower health care expenses. However, economists have
only recently attempted to measure the benefits accruing from food safety and nutritional
attributes. Also, these benefits are difficult to evaluate because safety and nutrition
characteristics are often deemed as nonmarket goods associated with market failures, due
to a lack of sufficient or asymmetric information for market participants.

The nonmarket valuation research of greatest relevance to food safety and
nutrition emphasizes estimating the value of reduced morbidity and the value of
preventing deaths or disease outbreak in a specific target population or region (Fisher et
al., 1989). Among the methodologies used to value food safety and nutritional quality,
contingent valuation has been widely employed by economists (Caswell, 1995).

A formal willingness to pay analysis is beyond the scope of this study. However,
to gain insights regarding consumers’ willingness to pay for the nutritional characteristics
of OFSVS, the researchers planned to interview sweetpotato traders in urban markets to
investigate whether they offer sellers the same price, a lower price (price discount), or a
higher price (price premium) for orange-fleshed varieties, compared to white-fleshed
varieties. However, due to the lack Of sweetpotato (especially OFSVS) in the market, only

three interviews were conducted in Xai-Xai and Chokwe.

53

4.3 Description of Data and Sampling Approach

4.3.1 Key Informant Interviews

Both primary and secondary data were collected. Interviews with key informants
(e. g. extension officers, researchers, farmer’s association representatives, NGOS, and
staff of input and output marketing agencies and traders operating in the study area were
conducted to gain insights about the history of the orange-fleshed sweetpotato program
and what factors govern farmers’ decision-making process and affect the cropping

patterns of the study area. These data were used to redesign the structured questionnaire.

4.3.2 Sweetpotato Growers Survey18 (Objectives 1, 2 and 5)

A multi-stage sampling technique was used to select a sample of 150 sweetpotato
growers. First, Gaza province was selected, due to its importance as a major sweetpotato-
producing area. Second, two districts (i.e. Xai-Xai and Chokwe) were chosen, using the
criterion mentioned above. Third, from the list of NGOs operating in selected districts,
five NGOs were selected, based on their participation in the first massive distribution of
OFSVS planting materials. At INIA headquarters, lists of farmers who received vines of
OF SVs in early 2000 from the NGOS were obtained and a random systematic sample of
15 farmers per village was selected. Thus, the sample size Of this survey-based study was
150 farmers, all of whom received planting materials of improved OFSVS from the

multiplication and distribution program.

 

'8 The survey carried out in November 2003 focused on the supply-side of sweetpotato production it

collected neither some data required to explain the key factors of adoption or disadoption of the new
orange-fleshed varieties nor information regarding consumers’ taste and preferences for roots and/or
leaves of these new orange-fleshed varieties.

54

A structured questionnaire was used to collect data from the farmers by means of
face-to-face interviews. The farmer questionnaire was used to collect data on the socio-
economic characteristics of the household, (e.g., age, gender, family size, household total
farmland, level of education, off-farm activities, livestock ownership, etc.), agronomic
practices followed (e.g., manual land preparation versus animal traction, intercropping or
monoculture, planting, etc), labor used in sweetpotato production (e. g. family labor
and/or hired labor), farmer’s criteria for preferring all or some of the new orange—fleshed
varieties, access to market for outputs, and access to extension services (see Appendix C
for a copy of the questionnaire).

In addition, secondary data from previous household surveys and agricultural

censuses were collected from relevant governmental and non-govemmental institutions.

4.3.3 Sweetpotato Traders (Objective 3)

Three open markets, located in the study area, were selected for data collection. In
each market, the researcher expected to select and interview 15 vendors who sold
sweetpotato tubers and leaves in order to collect data regarding the sweetpotato
marketing system, type of sweetpotato tubers and leaves sold, consumer preferences for
cream/white versus OFSVs, and prices at which they sold tubers and leaves of each type
of variety. However, due to widespread flooding in the lowland areas, all the cultivated
crops were destroyed, including sweetpotato. Thus, since few vendors were found in the
markets of Chokwe, Lionde and Xai-Xai, it was only possible to conduct three traders
interviews. In addition, four traders interviews were conducted at formal and informal

markets of Maputo. These data are used to provide a basic description of the sweetpotato

55

I —j

marketing system and to tentatively assess whether orange-fleshed varieties are being
sold at the same, a lower (price discount), or a higher (price discount) price than

white/cream-fleshed varieties.

4.4 Chapter Summary

This chapter presents the conceptual framework and literature regarding binary-
choice models, describes the sampling approach used to implement this study, and
describes the research instruments that were used to collect data needed to evaluate
farmer decision-making process regarding the adoption Of improved orange-fleshed
varieties.

Using a multi-stage sampling approach, a sample size of 150 sweetpotato growers
in Chokwe and Xai-Xai districts was selected from the lists of farmers who benefited
from the massive distribution program of vines of OFSVS that took place in 2000. First, a
structured questionnaire was used to collected data on the sociO-economic characteristics
of the household, agronomic practices followed, labor used in sweetpotato production,
fanners’ assessment of the traits of some of the new orange-fleshed varieties, access to
market for outputs, and access to extension services. Second, a structured questionnaire
was used to collect data on the commercialization of sweetpotato (both white/cream and
orange-fleshed), consumer tastes and preferences, prices traders offer to sellers, and
consumer prices. The cross-sectional data gathered through the farmer questionnaire were
used to describe the characteristics Of sweetpotato producers and their production system;

and to determine the factors affecting farmer adoption of OFSVS.

56

CHAPTER V
DESCRIPTIVE ANALYSIS OF SWEETPOTATO PRODUCERS, PRODUCTION

AND FARMERS’ EVALUATION OF OFSVS.

5.1 Socio-Economic Characteristics of Sweetpotato Growers

5.1.1 Education Level, Age and Gender

Respondents in the study area had little formal education. The average
sweetpotato grower had completed only 1.7 years of schooling-44.7% were illiterate,
49.3% had completed primary school (4 years), and only 6.0% had completed more than
4 years of school (Figure 5). Similarly, in Gaza province, the level of illiteracy is
estimated at 38.6%, being 63.3 for women and 24.3% for men (INE/MADER, 2001).

The average age of the sweetpotato producers was 42.3 years old. However,
55.4% were between 26 and 45 years old, 29.3% were between 46 and 65 years of age,

and only 6.0% were older than 66 years of age (Figure 6).

Figure 5: Distribution of Years of School Completed,

Xai—Xai and Chokwe, Mozambique (N=150)

 

Frequency

 

 

Years

Source: Survey 2004

57

Figure 6: Respondents’ Age Distribution, Xai-Xai

and Chokwe, Mozambique (N=150)

 

Frequency

 

 

   

Source: Survey 2004

In terms of gender, of the 150 randomly selected respondents, 144 were female
(96%). Although some men’s names were selected from the NGOs’ lists of vine
recipients, most of them reported that while they received the vines as the household

head, their wives were responsible for growing sweetpotato.

5.1.2 Household Size and Number of Children (less than 18 years of age)

Household size--defined as the total number of people living in the household at
least 9 months per year--across all 10 villages averaged 7.1 (ranging from 1 to 15)
(Figure 7). These households had an average of 3.4 children younger than 18 years of age
(ranging from 0 to 8), 2.0 young children under 5 years of age (ranging from 0 to 7, as
shown in Figure 8), 3.6 adults (ranging from 1 to 11), and 0.8 household member living
away from home (ranging from 0 to 10). The respondents had lived in the same village

for an average of 25.8 years (ranging from 3 to 72 years).

58

Frequency

Frequency

Figure 7: Respondents' Family Size Distribution,

Xai-Xai and Chokwe, Mozambique (N=150)

 

 

 

2 4 6 8 10 12 14 16
Number

Source: Survey 2004

Figure 8: Distribution of Young Children in Households
Xai-Xai and Chokwe, Mozambique (N=150)

 

 

 

 

0 1 2 3 4 5 6 7

Number

Source: Survey 2004

 

 

1111 if.
‘JIEVEIQu‘htfu i. an”? i ... .r .. . w

Few household members participated in off-farm activities. In 2003, only 3.3% of
respondents engaged in Off-farm activities. In addition, 25.3% of respondents’ spouses
participated in off-farm activities in 2003. Most of the respondent’s spouses were
primarily working in South Africa, Maputo and/or Xai-Xai city. However, it is important
to point out that 28.7% of the respondents repOrted that they did not have a spouse, as

they were either widowed or divorced.

5.1.3 Livestock Ownership and Household Total Farmland

Livestock (cattle and small ruminants) is a proxy Of household wealth, a source of
draft power for land preparation and transportation, and a source of income”. In most
developing countries, particularly in Southern Africa, livestock ownership is strongly
associated with a privileged social status. According tO the Agriculture and Livestock
Census (CAP, 1999-2000), about 72% of cattle keepers use draft power, either for
ploughing or for transporting agricultural products, fuelwood, and water. Approximately,
38% of cattle owners in the Gaza province commonly use animal traction for ploughing
and/or transportation, followed by Inhambane and Tete provinces with 24 and 18%,
respectively.

Among the survey households, 48% Of the respondents owned livestock--27.3%
owned cattle and 39.3% owned small ruminants. Among households owning cattle, about
32% had fewer than 5 bovines and 46% owned between 5 and 10 bovines (ranging from

1 to 57, as shown in Figure 9). Among households owning small ruminants (pigs, goats

 

The geographical distribution of cattle population and number of holdings is strongly determined by
two major factors: (ii) culture (in the southern region, cattle ownership is traditionally associated with
social status and welfare) and; (ii) the predominance of tsetse fly, in most areas of central and northern
Mozambique, people cannot raise cattle.

60

and sheep), 61% raised fewer than 5 animals and 32% owned between 5 and 10 small

ruminants (ranging from 1 to 25).

Figure 9: Distribution of Households’ Total Cattle,

Xai-Xai and Chokwe, Mozambique (N=150)

 

Frequency

 

 

 

 

Number

Source: Survey 2004

Apparently, in the study area land is not a constraint. On average, household
planted 2.4 hectares to annual crops (Figure 10). Approximately 67% of the respondents
prepared their fields manually, using hand tools. Of the households (33%) who used

animal traction, 9% used it on a rental basiszo.

 

According to the Agriculture and Livestock Census 1999-2000, of the total of producers in
Mozambique using animal traction for land preparation and/or transportation, 32% are in Gaza
province (INE/MADER, 2001).

61

Figure 10: Distribution of Households‘ Total Farmland,

Xai-Xai and Chokwe, Mozambique (N=150)

 

Frequency

 

 

9.0 11.0

Hectares

Source: Survey 2004

5.1.4 Awareness and Extension Services Provided to Farmers

Taking into account the importance of technical knowledge as a key determinant
of technology adoption (see the literature review chapter), questions were included in the
questionnaire to determine the frequency, source, and type of technical assistance that
farmers had received during the last two years, with respect to orange-fleshed
sweetpotato farming. Firstly, the respondents were asked how often during the last two
years they participated in field days and/or on-farm demonstration events and to name the
institution (3) that organized or funded those awareness activities. Secondly, the
respondents were asked how often any technician from public or NGO extension services
had Visited them during the last two years and what type of technological message was
delivered. Finally, they were asked whether they were aware of the nutritional benefits
associated with the consumption of OFSVS (OFSPV) and who provided that information

(Table 7).

62

Table 7: Frequency, Sources, and Type of Technical Assistance Received by Farmers,
Chokwe and Xai-Xai, Mozambique (% of Respondents, N= 150)

 

 

 

 

Source of Field Days & Extension Visits Nutritional
Technical Demonstration Related to Not Related to Benefits of
Assrstance Activmes Sweetpotato Sweetpotato Consumption of
Production Production OFSPV
Public extension 0 16.0 4.0 1.3
SARRNET 6.0 NA NA NA
NGO 8.7 4.0 1.3 78.7
Health Institutions NA NA NA 3.4
TOTAL 14.7 20 5.3 83.4
Source: Survey 2004 NA = Not applicable

Field days events were mainly devoted to exposing the farmers to the various
products that could be made out of OF SVs (e.g., juice, cakes, biscuits, flour, improved
porridge, and chips). Approximately 15% of the respondents reported having participated
in an orange-fleshed processed product demonstration event organized by SARRNET
(6%), Lutheran World Federation (8%) and Save the Children (0.7%) during the last two
years (Table 7).

Regarding the number of times extension technicians visited farmers during the
last two years, 17.3% reported having been visited once, while 8% received two visits.
About 20% of the farmers reported having received technical information about OFSP
production. The technological messages included the benefits of planting in ridges
(versus the traditional and common method of planting in mounds) and planting 1-2 vines
per hole (versus the traditional number Of 8-10 vines per hole). Messages not related to
sweetpotato production included information on planting methods for maize, beans, and
vegetable crops, the use of inputs in rice, maize, beans, and vegetable crops farming, and

pest control for major cash crops (vegetables, rice, and maize).

63

 

 

 

I1

 

11111}

1111;]

“N
“r

‘ “11

 

 

A surprisingly large percentage of the respondents (83%) were aware of the
nutritional benefits of orange-fleshed sweetpotato (OFSP) varieties to young children and
pregnant and lactating women. Respondents reported that they had learned about these
benefits from several different sources, including extension services (public and NGOs),
rural health institutions, and the media.

In recent years, the development literature has highlighted the role of social
capital in facilitating the exchange of information among farmers (Huysman M., et al.,
2004). Considering farmer associations as a proxy for social capital, respondents were
asked whether they and/or their spouses were members of any kind of farmer associations
and to what extent they had benefited from membership. Only 18% of the respondents
were members of a farmer association”. These farmers and/or their spouses belonged to
ten different farmer associations in the study area. All the farmer associations were
directly or indirectly supported by NGOS, which assisted farmers by providing inputs and
technical assistance or sponsored food-for-work projects. Nine out of the ten farmer
associations mentioned were primarily engaged in crop production, while one promoted
animal husbandry on a rotational basis”.

The 18% of respondents who were members of a farmer association identified the
following benefits from their participation:

(i) Agricultural products (derived from collective production) for

household consumption (56%);

 

2' According to the Agricultural and Livestock Census, 2.8% of the total households in Gaza province

were members of a farmer association in 2000/01 cropping season.

22 The beneficiary of distributed cow is obliged to give one calf to other member of the farmer association.

64

 

 

(ii) Planting materials of vegetable crops (produced under irrigation in

collective plot) for transplanting in their individual fields (18%);

(iii) Profit from sales of products derived from collective plot (11%);
(iv) Access to animals (on a rotational basis) (8%), and
(v) Use of association’s animal traction for land preparation in their

individual fields (7%).

Clearly, only a small share of the farmers was served by a farmer association. The
sustainability of these farmer’ associations is questionable, since continued provision of
NGO-supported technical assistance and input depends on continued access to funds
from the international development community. Furthermore, other bottlenecks threaten
the functionality and sustainability of these farmer associations”, including the absence

of a crop marketing system and the non-availability of credit opportunities for farmers.

5.1.5 Land Tenure and Cropping Systems

All respondents cultivated family-owned land. Approximately 40% cultivated
three plots, 28% cultivated two plots, 20% grew crops in four plots, and 7% cultivated
only one plot. The median number of plots per household was three. About 97%
intercropped maize; beans, groundnuts, roots and tubers, and 3% grew rice, maize, and/or
vegetable crops (e. g. tomato, onion) in monoculture.

Farmers cultivated plots in two different environments. Depending upon the

season, from 69 to 81% of the respondents cultivated plots in lowland alluvial soils, 4 to

 

23 The ten farmer associations found in the study area are as follows: Voz da F relimo, Fidel Castro,
Samora Machel, Graca Machel, Gandlhazi, Vamos a Frente, Grupo de Agricultores (Macunene
village), Tsakhane, Mapimo Manene, and Agro-Pecuaria (Muzumuia village).

65

 

 

8% farmed lowland peat soils, and 14 to 26% cultivated upland light-textured soils plots,
either as their first, second, or third household field.

Typically. farmers cultivated upland and lowland fields in order to reduce crop
failure risk due to drought or flooding. However, across the three cropping seasons,
farmers gave priority to planting in high fertility alluvial soils, as these soils are likely to
produce the highest yields and thereby to contribute to smallholder food self-sufficiency

(Table 8).

Table 8: Sweetpotato Plots Cultivation Across Different Soil Types (% of Plots),
Chokwe and Xai-Xai, Mozambique (2002/03 and 2003/04 Seasons).

 

 

Cropping Season
Soil Types
2002/03 Wet 2002/03 Dry 2003/04 Wet
Lowland alluvial soils (bilene) 70.3 80.7 69.1
Upland sandy soils (serra) 25.2 13.5 26.4
Lowland peat soils (machongos) 4.5 5.8 4.5

 

Source: Survey, 2004

5.2 Descriptive Statistics of Sweetpotato Production

5.2.1 Identified Varieties Grown by Households

The respondents planted several different sweetpotato varieties, including 17
white/cream-fleshed and 5 OFSVS, namely: Jonathan, Resisto, Japing Selecto, Taimung
64, L0 323, and Cordner. The four varieties most frequently grown in the 2002/03 wet

season, the 2003/03 dry season, and the 2003/04 wet season are presented in Table 9.

66

 

iii-Ii

I I! V. .7 . r 11
C. ... . bu .. I . a . r x... H .
EL. .. . .

h

   

 

Table 9: Four Most Frequently Grown WFSP and OFSP Varieties, Chokwe and

Xai-Xai Districts, Mozambique (N=150)’

 

 

Cropping Season
Variety Name Flesh Color
2002/03 Wet 2002/03 Dry 2003/O4 Wet
Sector White 67.1 61.2 68.2
Mudiliva White 55.0 49.3 43.7
Jonathan Orange 33.1 35.5 35.9
Resisto Orange 28.6 37.0 30.3

 

Source: Survey 2004

The total percentages of respondents are greater than 100% because most farmers planted more than

one variety.

5.2.2 Sweetpotato Cultivated Area

In order to gain insights regarding the evolution of their sweetpotato areas,

farmers were asked to estimate their sweetpotato areas in the 2003/03 wet season, the
2002/03 dry season, and the 2003/04 wet season (Table 10), and the proportion of the
area they planted to white/cream-fleshed relative to orange-fleshed varieties.

In all three seasons, more than 50% of the respondents cultivated less than 80 m2
of sweetpotato and approximately 10% of the respondents cultivated between 250 and
2,500 m2. In the 2002/03 dry season, the percentage of respondents who planted less than

80 m2 of sweetpotato increased, since some of them did not cultivate sweetpotato during

the dry season due to the low rainfall and no water for irrigation.

67

Table 10: Sweetpotato Areaa Cultivated by Smallholders in Different Cropping Seasons
in Chokwe and Xai-Xai Districts, Mozambique.

 

 

Statistics 2002/03 Wet 2002/03 Dry 2003/04 Wet
Season Season Season
(N=150) (N=123) (N=150)
Area (m2) i :
Mean 206 107 215
Median 80 30 80
Mode 80 0 125

% of Respondents with

Area less than 80 m2 55.3 78.7 50.7
81 - 150 24.7 10.0 25.3
151 — 250 10.0 4.6 12.7
251 — 500 6.0 4.7 7.3
501 — 2,500 2.7 1.3 2.7
2,501 — 5,000 1.3 0.7 1.3

 

WFSVs = White-fleshed Sweetpotato Varieties
Source: Survey 2004

Sweetpotato areas planted in the 2002/03 wet and dry seasons were estimated based on recall data
provided by respondent. Data for the 2003/04 wet season are based on the researcher estimates. In
cases when the respondent was not able to provide information in hectares or square meters, she was

asked to compare the magnitude of her sweetpotato-cultivated area with the standard size household
residential plot and this estimate was converted to square meters.

In all three seasons, the respondents planted a far greater share of their fields to
white versus orange-fleshed varieties, ranging on average from 83 to 86% in the 2003/04
wet season, and 2002/03 dry season. In general, more than 70% of the respondents
planted more than 75% of their sweetpotato fields to WFSVs (Table 11).

Presently, farmers do not apply the entire technological package for orange-
fleshed sweetpotato farming, which includes fertilizers, and pesticides. Moreover, there is

no evident difference in terms of agronomic practices and production costs between white

and orange-fleshed sweetpotato farming.

68

Table 11: Proportion of Sweetpotato Area Smallholders Planted to WFSVs in
Different Cropping Seasons in Chokwe and Xai-Xai Districts, Mozambique.

 

 

Statistics 2002/03 Wet 2002/03 Dry 2003/04 Wet
Season Season Season
(N=150) (N=123) (N=150)

Proportion of WFSVs * ' ' ' _ 7 7

Mean 84 86 83

Median 90 100 90

Mode 90 100 100

% of Respondents with

Proportion less than 50 1 1.3 7.3 5.3
50 4.7 0.7 2.0

51 - 75 14.0 15.4 18.0

76 - 100 70.0 76.6 74.7

 

WFSVs = White-fleshed Sweetpotato Varieties
Source: Survey 2004

Farmers reported several reasons for planting different proportions of their farms
to white/cream versus OFSVS (Table 12). The main reasons (which varied by season) the
respondents reported for cultivation a higher proportion of white/cream varieties included
the limited availability of OF SP vines, limited off-season propagation capacity of
OFSVS, abundant availability of planting materials of white/cream varieties, and the
relatively low drought-tolerance of OF SVs. Conversely, the major reasons respondents
gave for planting an equal or greater proportion of OF SVs included perceived high-yield,
the better taste of orange-fleshed varieties, and a desire to experiment with new varieties.

Considering that there is no difference in terms of production costs the
incremental gross margin revenue between WFSVs and OF SVs would be exclusively

determined by the yield differential between these two types of varieties.

69

Table 12: Farmers’ Reasons for Planting Different Proportions of Sweetpotato Varieties
(°/o of Respondents), Chokwe and Xai-Xai, Mozambique.

 

2002/03 2002/03 2003/04
Farmers’ Reasons wet season dry season wet season

(N=106) (N= 60) (N=77)
Planting greater proportion to WFSVs

 

—- .. -v- _—-——

Limited quantity of OFSV vines 79.2 15.0 5.2
Limited off—season propagation capacity of OFSVS 4.7 46.7 45.5
OFSVS not tolerant to water stress on upland 1.9 3.3 1.3
Abundant planting materials of WFSVs 1.9 0 0
Respondent not familiarized with OF SVs 0.9 5.0 5.2
Time constraints to increase area of OFSVS 0.9 3.3 0
WFSVs are more appropriate for animal feeding 0.9 1.7 1.3
OFSVS are watery and too sweet 0 3.3 3.9
WFSVs are more productive than OFSVS 0 1.7 5.2
Planting equal or greater proportion to OFSVS

High-yield 3.7 11.7 18.2
Experimentation of new varieties 1.8 3.3 0
High—yield and better taste 1.9 5.0 13.0

—-————-—-

 

WFSVs = White-fleshed Sweetpotato Varieties
OFSVs = Orange-fleshed Sweetpotato Varieties
Source: Survey 2004

It is important to point out that the percentage of respondents who grew OF SVs
differed from one cropping season to another. In the 2002/03 wet season, approximately
71% of the respondents grew at least one OFSV, while in 2002/03 dry season and
2003/04 wet season the adoption rate was only 38% and 51%, respectively (Table 13).

The lower percentage of respondents who grew OF SVs in the dry season is
associated with erratic and insufficient rainfall in the upland areas or a lack of water for

irrigation in the lowlands during the dry season.

70

Table 13: Adoption (%) of OFSVs in Each Zone Covered by NGO, and by Season in
Chokwe and Xai-Xai Districts, Mozambique.

 

NGO-Covered Areaa
Total
Household Cultivated OFSPV LWF CRC WRS WV STC Adoption

 

In 2002/03 wet season 100 50.0 86.7 40.0 76.7 70.7
In 2002/03 dry season 60 16,7 36.7 6.7 70.0 38.0
In 2003/04 wet season 96.7 40.0 40.0 6.7 70.0 50.7

 

LWF = Lutheran World Federation CRC = Caritas Regional Chokwe WR= World Relief Services
STC = Save the Children WV = World Vision
Source: Survey 2004

a . . . .
Two v1llages covered by each NGO were selected, and 15 farmers per Village were 1nterv1ewed.

Furthermore, the highest percentage of respondents who grew OF SVs lived in
Cuamba and Punguine villages, where vines of OFSVS were recently distributed to
farmers (February 2004) by the Lutheran World Federation (Appendix D, Table D-l).
Farmers in these villages mainly planted vines in the uplands, since the lowlands were
waterlogged due to “excessive rainfall” which occurred during this year in the region.
However, the heavy rains contributed substantially to better crop establishment and

development of OFSVS in the uplands.

5.2.3 Planting Methods of Sweetpotato Varieties
In addition to distributing vines, the program’s extension services promoted two
improved agricultural practices, namely: planting in ridges (versus the traditional method

of planting in mounds) and planting one to two vines per hole (versus eight to ten per

71

mound). Approximately 55% of the respondents planted sweetpotato in ridges. Among
the respondents who planted in mounds (45%), their justification included their habit of
doing so (82%) and that they did not know any planting method other than planting in
mounds (13%). Among the respondents who reported using different planting methods
for white versus orange-fleshed varieties (42%), 51% planted 1-2 vines of OF SVs per
hole, 44% planted OF SVs in ridges while planting WFSVs in mounds, and 5% reported
that planting in mounds was more appropriate for upland growing conditions,
characterized by erratic rainfall and sandy soils with low water holding capacity.

Soil type and topography influence whether farmers plant sweetpotato on
mounds, in ridges, or on flat land. Particularly in lowland peat soils with a higher water
table and poor drainage, cultivation on mounds is recommended since it increases both
soil drainage and the topsoil layer (Woolfe, 1992). Most farmers who cultivated lowland
alluvial (heavy clay) soils planted sweetpotato along the borders delimitating their fields
or along drainage canals on manually formed mounds. In contrast, farmers who ploughed
using animal traction mostly used ridges. In the uplands, characterized by light texture
and well-drained soils in which deep cultivation is required to provide room for roots to
grow, farmers often plant sweetpotato on flat ground (Woolfe, 1992a). However, in both

districts surveyed, very few cases of planting on flat were observed in uplands.

5.2.4 Varieties Crew and Vines Received by Households
Households in the selected villages were exposed to OF SVs in different ways and
the frequency of vine distribution varied among NGOs operating in the sampled villages.

Most respondents received vines of OFSVs once (69%), although 8% received them

72

twice, and 23% received them three times. About 73% of the respondents grew one and
23% grew at least two different OF SVs.

Table 14 shows the reasons reported by the respondents who received vines of
OF SVs more than once (31%). More than 22% justified having lost the varieties they
had previously received due to adverse agroclimatic conditions--mainly due to either

drought or flooding in the lowlands.

Table 14: Major Reasons Reported by Farmers Who Received Planting Materials of
OFSVs More than Once (N= 46), Chokwe and Xai-Xai Districts,
Mozambique.

 

Major Reasons Frequency

 

First distributed materials lost due to drought and the second due to flooding

(poor soil drainage in lowlands) 52.0
Planting materials of OF SP lost due to severe drought 28.3
Distribution is free of charge, even if they previously received an OF SPV 1 1.0
Different varieties in each distribution and willing to increase OFSPV area 6.5
Other 2.2

 

Source: Survey 2004

As mentioned above, characteristics of lowland soils and the lack of an
operational drainage system contributes substantially to frequent flooding in the region.
However, despite the fact that farmers experienced devastated flooding, if they actually
valued those nutritionally improved OFSVS, they should have been able to save these
varieties by transferring them to an upland field or planting them in their backyard.

Interestingly, farmers who lost their orange-fleshed varieties were able to save

their WFSVs. This suggests that, notwithstanding the adverse climatic and edaphic

73

growing conditions, farmers may have failed to save their orange-fleshed varieties
because they were pretty sure that they would be able to receive free of charge more
planting materials from an NGO. As mentioned previously, the frequency of vine

distribution and the number of OFSVs distributed varied by NGO (Table 15).

Table 15: Frequency of OFSPV Vines Distribution, and Number of Varieties Each
Household Received from 2000/01 and 2003/04 Wet Season by NGO, Chokwe
and Xai—Xai, Mozambique.

 

NGO-Covered Areaa

 

 

 

Variable Number pr CRC WRS wv STC Total
One NA 80 100 100 66.7 69.3
Frequency of
OFSP Vines Two NA 20 0 0 20.0 8.0
Distribution
Three 100 0 0 0 13.3 22.7
Number of OFSP One 80 80 100 80 23.3 72.7
Cultivars
Received Two 20 20 0 20 76.7 27.3
LWF = Lutheran World Federation CRS= Caritas Regional Chokwe WRS= World Relief Services
STC = Save the Children WV = World Vision NA = not applicable

Source: Survey 2004

a Two villages covered by each NGO were selected and 15 farmers per village were interviewed.

All respondents in both villages covered by the Lutheran World Federation
(Cumba and Punguine, C hokwe district) reported receiving vines three times. In contrast,
in villages covered by World Relief Services (Macunene and Malhazene, Chokwe
district), Save the Children (24 de Julho, Xai-Xai district), and World Vision (Fidel
Castro and Siaia, Xai-Xai district), respondents reported receiving vines only once

(Appendix D, Table D-2).

74

Both Caritas Regional Chokwe and World Relief Services covered almost all
villages in Chokwe district during the massive distribution effort in 2000 and later their
staff carried out more restricted distributions to specific targeted groups of farmers in
some of these villages.

Save the Children distributed vines directly to farmers once, without any
involvement of local authorities. In addition, this NGO provided vines two more times to
a farmer association for multiplication and further distribution to farmers under the
responsibility of the association leadership. Unfortunately, these multiplication efforts
failed twice, due to severe drought in 2002 and flooding in 2004.

World Relief Services carried its first massive distribution in 2000 and later
provided orange-fleshed planting materials to groups of farmers working with its village
activists.

Compared to the other NGOS, World Vision followed a slightly different
approach for its massive distribution of vines to farmers just after the devastating flood of
2000. Under the auspices of the Food Security project, this NGO established and
managed several conventional and rapid multiplication plots. However, due to a lack of
financial resources and a desire to promote sustainability, responsibility for managing
those conventional multiplication plots was handed over to the farmer association-~but
unfortunately without success.

The overall performance of the distribution program raises an issue regarding the
sustainability of these kind of promotional activities, since their longevity is determined
mainly by the duration of project fundinguas soon as project funding phases out, they are

likely to collapse.

75

5.3 Producers’ Evaluation of OFSVS

5.3.1 Producers’ Perceptions about the Performance of OFSVS.

Producers were asked their opinions regarding the OF SVs they had planted (Table
16). In terms of the weighted average, 40% of the responses indicated that OFSVS had
relatively low drought-tolerance (compared to WFSVs), particularly when planted in

upland sandy soils, but also under lowland growing conditions without irrigation.

Table 16: Respondents’ First and Second Opinions Regarding the Performance of
OFSVS, C hokwe and Xai-Xai, Mozambique.

 

Percentage of Responses

 

First Second Weighteda
Opinions about OFSVS Comment Comment Average
(N= 11 1) (N= 60)
Not tolerant to drought conditions 50.5 20.0 39.8
High-yield under favorable growing conditions 32.4 3.3 22.2
Nutritionally better for young children 6.3 25.0 12.9
Bigger root size than WFSP varieties 1.8 26.7 10.5
Low off-season propagation capacity 4.5 20.0 9.9
Others 4.5 5.0 4.7

 

Source: Survey 2004

a All the responses received proportionally the same weight regardless their order.

Their opinion is consistent with the reason why farmers reported losing their
orange-fleshed materials and their justification of the need for receiving vines more than
once. Other farmers reported that orange-fleshed varieties were high-yielding under
favorable growing conditions (22%), nutritionally better for young children (13%),

produced bigger roots (11%), and have a low off-season propagation capacity (5%).

76

Considering that any good embodies both desirable and undesirable
characteristics, producers were asked to evaluate several traits of orange-fleshed varieties
(i.e. crop yield, crop duration, propagation capacity, taste of fresh and boiled roots,
consistency, etc.), compared to “local” white/cream sweetpotato varieties. Generally, in
making their assessment, the farmers compared the OFSVS with the most widely
cultivated white-fleshed varieties such as Sector and/or Mudiliva.

Farmers” assessment of two of the orange-fleshed varieties-Jonathan and Resisto-
-reported in Table 17. Both of these orange-fleshed varieties where ranked positively
with regard to total yield, root size, taste of both fresh and boiled roots, taste of the leaves
and cooking time.

In contrast, these two OFSVS were ranked negatively in terms of off-season
propagation capacity, dry matter content, and tolerance/resistance to drought, pests, and
diseases.

The low figures for taste of leaves is associated with a high number of
respondents who reported not consuming leaves--either because they were not used to
eating them or they did not like the reddish color of the leaves of some OFSVS, such as
Jonathan and Japon Selecto.

According to Collins et al., (1995), throughout the world where people eat
sweetpotato as a staple food, the most preferred sweetpotato root type is high in starch,
white colored, and moderately sweet tasting, while orange-fleshed roots are acceptable as
a dessert. Although tastes and preferences vary from region to region, results from a

survey conducted by the Asian Vegetable Research and Development Center (AVRDC)

77

Table 17: Farmers’ Evaluation24 of the Performance of OFSVs Versus “Local” Cultivated
Varieties”, Chokwe and Xai-Xai Districts, Mozambique.

 

Jonathan Resisto
Variety Traits (N=128) (N=40)
(+) (-) (+) (-)

 

Production-Related Traits

Total yield (Higher/ Lower) 64.1 23.4 85.0 7.5

Crop duration (Shorter / Longer) 57.8 26.6 50.0 12.5
Off-season propagation capacity (Better/ Worse) 1.6 86.7 7.5 92.5
Root size (Bigger /Smaller) 75.0 10.2 82.5 7.5

Consistency (dry matter content) (Mealy / Watery) 28.9 62.5 20.0 65.0
Tolerance/Resistance to drought (Higher / Lower) 5.5 84.4 5.0 72.5
Tolerance/ Resistance to pests (Higher / Lower) 9.4 55.5 2.5 45.0
Tolerance/Resistance to diseases (Higher / Lower) 4.7 56.3 7.5 45.0

Consumption-Related Traits

 

Taste of fresh root (More sweet / Less sweet) 74.8 7.8 75.0 15.0
Taste of boiled root (1) (More sweet / Less sweet) 80.5 12.5 80.0 10.0
Taste of boiled root (2)a (More sweet / Less sweet) 47.7 6.3 57.5 7.5
Cooking time (Shorter / Longer) 66.4 3.9 72.5 0
Taste of leaves (More sweet / Less sweet) 35.9 5.5 37.5 5.0
(1) Respondent’s comment (2) Opinion of respondents’ spouses.

' Approximately 23.4% of respondents do not have spouses, and 17.2% of respondents do not know

their spouses’ Opinion.

 

2‘ Some respondents evaluated the traits as being equal (no difference between white and orange

varieties). Thus, the sum of (+) and (-) do not sum up to 100%.
25 Other orange-fleshed varieties (Japon Selecto, Taimung 64 and L0 323) were grown by farmers in the
selected villages. However. since few responses were obtained (8, 6 and 2 respectively), they are not
reported in Table 16.

78

   

found that 77% of the respondents preferred dry and moderately dry textured varieties,
whereas moist textured varieties were only preferred in developed countries.
Furthermore, more than 50% of the respondents surveyed revealed a preference for white

or cream-fleshed varieties, while 30% preferred orange-fleshed ones (Lin et al., 1992).

5.3.2 Major Constraints to Sweetpotato Production

Producers were asked to identify the three most important constraints to
sweetpotato production (Table 18). Among water/drainage related factors, respondents
reported (in terms of weighted average) drought (33%), flooding (21%), and poor/sandy
upland soils (4%). These constraints contribute to a high risk of crop failure. As
elaborated in Chapter 2, most of the flooding events occur as a result of poor drainage of
the peat and alluvial heavy-clay soils in the lowlands. Consequently, while the irrigated
land has a high potential for agriculture production and development, it is not being fully
exploited due to the poor quality of the drainage systems, which contributes to flooding,
salinity, and sodicity-related problems. This is especially a constraint to sweetpotato,
since the crop is sensitive to alkaline and saline growing environments (Woolfe, 1992a).

Under the prevalent conditions of low, erratic, and irregular precipitation in the
region, lowlands are extremely important for crop production. Therefore, rehabilitation of
drainage systems in lowlands is extremely important to increasing the sweetpotato
production in general, particularly high-yield OFSVS that have relatively higher water

crop and nutrient requirements.

79

Table 18: Constraints of Sweetpotato Production, Chokwe and Xai-Xai, Mozambique.

 

Percentage of Responses

 

Identified Major Constraints First Second Third Weighteda
(N=l45) (N=127) (N=46) Average
Water/drainage related constraints , ,. , _ -. __ _'____,__,,;
Frequent severe drought 58.6 13.4 12.5 33.9
Frequent flooding 14.5 30.7 15.0 21.0
Poor sandy soils not suitable for SP growing 2.1 4.7 7.5 3.9
Low off-season propagation of OFSV 1.4 3.9 0 2.2
Institutional related constraints
Lack of animal traction 1 1.0 22.8 30.0 18.5
No market for output surplus in the village 1.4 16.5 25.0 10.8
Lack of agricultural tools 2.8 4.7 7.5 4.2
Pests (rodents, snails, grasshoppers, etc) 5.5 0.8 0 2.8
Others 2.8 2.4 2.5 2.6

 

Source: Survey, 2004

a

All the responses received proportionally the same weight regardless their order.

While sweetpotato is a perennial crop, it is often grown as an annual. Thus,

having sufficient rainfall during the non-growing season (dry) is extremely important,

since farmers must maintain vines during the dry season as planting material for the main

cropping season (wet). Both districts are characterized by extremely low rainfall during

dry season, averaging less than 150 mm and 300 mm in Chokwe and Xai-Xai,

respectively. In contrast, optimum rainfall during the dry season is estimated at 250—500

mm (Woolfe, 1992a). Therefore, low rainfall during the dry season is a constraint to off-

season propagation of sweetpotato varieties in general, and particularly orange-fleshed

varieties. While not reported by farmers as a major water/drainage related constraint, they

assessed OFSVS as much worse than white/cream in terms of off-season capacity during

the dry season.

80

Furthermore, farmers reported that ‘local’ WFSVs are relatively more tolerate to
drought conditions, since they can withstand longer period of water stress and they easily
propagate as soon as rain falls during wet season.

Among institutional related constraints, respondents reported lack of animal
traction (19%), lack of market for sweetpotato surplus in the villages (11%) and limited
provision of agricultural tools (4%). These constraints provided no incentive to increasing

area for sweetpotato and adoption of OFSVS.

5.3.3 Proposed Actions Needed to Improve OFSVS.

In order to improve sweetpotato production, including the adoption of OF SVs,
producers were asked to propose the most important actions that needed to be undertaken.
Most farmers gave only two responses (Table 19).

The most important actions that farmers proposed were: (i) providing access to
credit for animal traction (33%); (ii) distributing vines of orange-fleshed varieties that

are more tolerant to drought (27%), and (iii) improving the drainage system in the
lowlands (19%).

The high percentage of responses regarding credit for animal traction is justified,
since most of women interviewed were either widows or their husbands worked away
from home--making labor the major constraint for agricultural production. Thus, animal
traction could play an important role in removing this bottleneck, particularly in lowland

heavy clay soils.

81

Table 19: Main Actions Proposed by Respondents for Improving Sweetpotato
Production, Chokwe and Xai-Xai Districts, Mozambique.

 

Percentage of Respondents
Proposed Actions First Second Weighted ‘
(N=132) (N =60) Average

Establish credit system for animal traction 31.1 35.0 32.6
Distribute vines of OFSPV more tolerant to drought 25.8 28.3 26.6
Improve the drainage system in lowlands 20.5 16.7 19.3
Organize transport of SP roots to the market 1 1.4 5.0 9.4
Control pests (rodents, snails. locust, etc) control 2.3 8.3 4.2
Improve roads 4.5 3.3 4.1
Disseminate production and processing techniques 1.5 1.7 1.6
Others (H20 pump, agric. tools, market in the village) 3.0 1.7 2.6

 

Source: Survey 2004

All the responses received proportionally the same weight regardless their order.

Finally. given the current frequency of flooding in the lowlands, improving the
drainage system is obviously a priority action that is needed to improve agricultural

production in all villages covered by this survey, including sweetpotato production.

5.4 Consumption and Commercialization of Sweetpotato
5.4.1 Sweetpotato Consumption Patterns
Fresh sweetpotato roots are typically boiled or roasted and consumed at breakfast,
usually served with tea”’. However, sweetpotato is not deemed to be a major staple food

in the study area”.

 

26 According to Woolfe, (1992b), the sweetpotato consumption pattern in the study area suggests that this

commodity is fulfilling the role ofa special food (1-7 meals per week, but only at one particular meal).

27 . . . . . . . .
Sweetpotato are consumed in a variety of ways 1n other countr1es, Including baked, b01led, served w1th

meats, in soups. candied, in salads, desserts, cereals, cakes, and are used to make various cold and
alcoholic beverages. Also, they can be used to produce syrup and flour (Cockerham et al., 1921;
Woolfe, 1992a, IPC, 2002).

82

To assess the importance of sweetpotato in the diet, the respondents were asked how
often they consumed sweetpotato roots and about their knowledge of processed products
derived from sweetpotato (Table 20). Approximately 64% of the respondents consumed
sweetpotato four or more months per year and 68% ate sweetpotato roots at least four
times a week during the months they consumed them. While a high percentage of the
respondents reported being aware of sweetpotato processed products (70%), few made

any processed sweetpotato product.

Table 20: Household Consumption of Roots, Awareness, Experience in Processing
Products Out of Sweetpotato.

 

 

Months HH Times per Week HH Knowledge Experience in
Consumed Roots Consumed Roots of Processed Processing
6 4 3 3 4 5 Products Products
Percentage of
Respondents 41.3 22.7 17.3 47.3 20.7 18.0 70.0 4.0

 

Source: Survey 2004

Most of the respondents reported that while they had been told about those
products (juice, cakes, flour, chips), they had not had an opportunity to see how to
prepare them in practice. Respondents who participated in field days/demonstrations
reported they did not make processed products because they lacked the necessary
ingredients. Others smartly argued that since they began planting OFSVS, they have not

yet obtained good yields. Thus, they did not have a sufficient quantity of roots to use for

83

any purpose other than for traditional consumption (i.e., boiled, roasted, or porridge with
“mafurrazg”.

Processing of sweetpotato into diversified, which contributes to nutritional value
and convenience of use, could play an important role in promoting and increasing
sweetpotato consumption among different segments of society. However, it is important
to point out that in general, an inverse relationship has been observed between levels of
consumption of sweetpotato with income levels (Woolfe, 1992; Omosa, 1997),
suggesting that sweetpotato is deemed to be an inferior good.

Clearly, changing consumption food habits by incorporating new products into
the diet or by introducing changes in the way food is traditionally prepared is a
challenging and time-consuming task. Success will require a more consistent and
comprehensive strategy that not only targets the rural area, but also the urban people--
assuming that the latter are more educated and thus, more likely to have a greater
“potential” demand for these new value-added and nutritious sweetpotato products. For
example, a study conducted in Kenya found that preference for new sweetpotato
processed products--biscuits, bread, chips, crisps, porridge and cakes--was dependent
upon income and staple food habits. Moreover, greater awareness about and interest in

sweetpotato products was found among households with low and middle-income levels

(Omosa M., 1997).

 

28 A tree specie (Trtchilia emetica) found in Sub-Saharan Africa commonly in open riverine forests and

open savannah woodlands. From the seed aril is made a milky suspension used for cooking while seed
oil is used to manufacturing soap and cosmetics and producing candles (F A0, 1986).

84

Scientists in developed countries are also trying to promote sweetpotato products.
For instance, in the US, the North Carolina Sweetpotato Commission Foundation
(N CSCF) has been strongly involved in identifying new value-added sweetpotato
products, in order to achieve its target of increasing sweetpotato consumption by 25%.
Presently, American per capita sweetpotato consumption is estimated at only 4.1 pounds,
compared to 140 pounds of white potatoes.29 New value-added sweetpotato products that
are being tested include: (i) sweetpotato casserole pudding and bread pudding; (ii)
scalloped sweetpotato; (iii) sweetpotato bran muffins; (iv) sweetpotato beverages; and (v)
sweetpotato tarts.

In terms of sweetpotato leaves, 63% of the respondents reported that their
households consumed leaves. Among households which did not consume sweetpotato
leaves, the main reasons, included: (i) not being used to consuming sweetpotato leaves
(20%); (ii) disliking eating leaves of sweetpotato (8%); and (iii) disliking the reddish
color of some orange-fleshed varieties (6%).

Scientists have documented the superior nutritional value of sweetpotato leaves--
compared to other leafy vegetables, such as, spinach, amaranth, cabbage and head
lettuce--in terms of Vitamin B2 content (Villareal et al., 1995). Furthermore, one study
found that among various leaf qualities (i.e., tenderness, flavor, stern and leaf color, and
hairiness) accounted for 64% of the variation in overall consumers’ leaf acceptability
(V illareal et al., 1979). Although, leafy vegetables are widely consumed in the study

area--including cowpea, amaranth, pumpkin and other legumes as a vegetable-based

 

29 North Carolina Sweetpotato Commission Foundation, 2004. Value-Added Sweetpotato Products Now

on Market: Success May be a Taste Away. http://www.ncsweetpotatoes.com/foundation.htm, August
2004.

 

85

complement to maize porridge--sweetpotato leaf color is a factor that is likely to affect
adoption.

While OF SVs are promoted as being especially good for children, approximately
79% of the respondents fed both (white/cream and orange-fleshed varieties) to their
children and 13% fed them only WFSVs. Some of the latter respondents justified this by
saying that their children preferred WFSVs. However, one possible alternative
explanation for this behavior is that these respondents may not be aware of nutritional
value of OFSVS (approximately 13%) or they may be disadopters who only produce and
consume WFSVs in their households.

It was hypothesized that there might be an association between the NGO that
covered the villages and the level of awareness and household use of processed
sweetpotato products. Presumably, greater knowledge regarding processing techniques
should lead to a greater percentage of households making these products.

As shown in Table 21, in the villages covered by Lutheran World Federation,
World Relief Services. and Save the Children, higher percentages of respondents were
aware of processed sweetpotato products. However, the relationship between level of
awareness and adoption is relatively weak. For example, in the villages covered by World
Relief Services, 100% were aware, but only 3.3% had made processed products,
compared to the villages covered by Save the Children (73.3% aware, 6.7% adopted).
More information regarding levels of awareness and adoption of processed products by

village is presented in Appendix D, Table D-3.

86

Table 21: Respondents’ Levels of Awareness and Use of Processing Techniques of OFSVs
in Each Zone Covered by NGOs in 2003/04 Wet Season, Chokwe and Xai-Xai
Districts, Mozambique (% of Respondents).

 

 

 

NGO Covered Areaa
Variable LWF CRC WRS wv STC Total
Respondent is aware of
OFSV processed products 100 46.7 100 30.0 73.3 70.0
Respondent made at least one
processed product from OF SV 10 0 3.3 0 6.7 4.0

 

LWF = Lutheran World Federation CRC = Caritas Regional Chokwe WR= World Relief Services
STC = Save the Children WV = World Vision
Source: Survey 2004

' Two villages covered by each NGO were selected and 15 farmers per village were interviewed.

5.4.2 Commercialization of Sweetpotato

Since farmers grew sweetpotato in two seasons, questions were included in the
questionnaire to investigate the existence and volume of sweetpotato commercialization
in both the wet and dry seasons (Table 22). The results indicate that sweetpotato is not
widely commercialized in the study area. The respondents unanimously justified this on
the grounds that a surplus of sweetpotato was not available. However, during group
interviews, key informants mentioned that sweetpotato roots are sometimes used for in-
kind payment of labor hired for agricultural activities.

Some respondents expressed a concern regarding the nonexistent market for
sweetpotato, noting that while traders visited some of the villages to buy cassava and
other agricultural products, they seldom bought sweetpotato roots and leaves.
Particularly in the villages located far away from the main primary and secondary roads,

poor tertiary roads result in high transaction costs, which limit the commercialization of

87

sweetpotato and thereby limit the potential contribution of sweetpotato to increasing

households’ cash income.

Table 22: Indicators of Sweetpotato Commercialization, Chokwe and Xai—Xai,

 

 

 

Mozambique
Cropping Season
2002/03 Wet 2002/03 Dry

Respondents (%) who sold sweetpotato roots 18.7 2.0
Respondents (%) who sold orange-fleshed roots (OF SPV) 6.0 0.0
Respondents (%) who sold less than 100 Kgs of OFSPV “ 5.4 0.0
Maximum (Kg) sold of OFSPV b 120 0.0
Mean (Kg) 3.6 0.0
Source: Survey 2004 NA = Not applicable

3 Few respondents (9%) sold less than 100 Kg of sweetpotato roots (both color-fleshed varieties) in the

2002/03 wet season and 1% in the 2002/03 dry season.

b The maximum quantities of sweetpotato roots sold were 600 Kgs in 2002/23 wet season, and 475 Kgs in
2002/03 dry season.

Depending on the size and quality of sweetpotato roots, the retail price ranged
from approximately 2,500 to 3,800 Meticais30 per Kg in Xai-Xai and from 6,600 to
10,000 Meticais per Kg in Maputo city. Despite the extensive awareness campaign
regarding the superior nutritional value of OF SVs, this aspect is not yet reflected in price
differences between white/cream-fleshed and orange-fleshed varieties, since both types
were sold at the same price.

Furthermore, while sweetpotato vendors in Chokwe and Xai-Xai reported that
consumers preferred WFSVs on the grounds of habits and custom, vendors interviewed

in Maputo reported that consumer’s preferences did not show a preference for either

 

3° Meticais is the Mozambican currency. 1n average, the exchange rate in 2002 was 23,180.40 Mt = 1

USD (INE, 2002)

88

type. Because the sample size of sweetpotato vendors in Xai-Xai and Chokwe was small
(N =3), they may not be representative. However, the vendors revealed that some
consumers bought OF SVs when WFSVs roots were not available. Moreover, they
hypothesized that some consumers’ preference for WFSVs roots was related to higher
dry matter content and good taste, while the preference for OFSVS was associated with
sweeter taste and higher nutritional value in Vitamin A. No vendor was found selling
leaves of sweetpotato, which they argued was due to a limited demand for leaves.
However, some vendors reported that when they occasionally sold leaves, they sold
leaves of both types at the same price per bunch.

Since sweetpotato is seldom traded in world markets, its price is endogenously
determined within producer countries. The major constraints to exporting are related to
the difficulty of maintaining high fresh root quality, due to the high perishability of
roots, and the high cost of post-harvest handling methods that are required to preserve
the roots, including curing, cooling/heating, and ventilation (Collins, W. et al., 1995).

Technological change contributes significantly to increasing aggregate output,
which in turn leads to lower production costs per unit of output. Assuming that there are
no market constraints and the demand is relatively elastic, farmers can maintain or
increase their profits--even if they sell the product at a cheaper price. However, the
extent to which consumers benefit from lower production costs depends largely on the
food marketing and distribution system. The food marketing and distribution system
must be sufficiently competitive in order to ensure that cost savings at the farm gate are
passed up through the marketing chain and not exclusively captured (as additional

profits) in the marketing chain (Hazell et al., 2001). Moreover, in areas where an

89

imperfect market system leads to a large discrepancy between farm and retail prices for
food crops, a bias toward food crop production to ensure food security is likely to
encourage farmers to give priority to meeting home consumption needs at the expenses
of income generation (Goldman R. et al., 1981).

For instance, in Xai-Xai district, the farm gate prices of sweetpotato roots varied
from 1,300 to 1,500 Meticais (MT) per Kg, while at Xai-Xai market sites, roots were
sold for 2,500 to 3,800 MT per Kg. At formal and informal markets in Maputo city,
sweetpotato roots were sold for 6,600 to 10,000 MT per Kg. Notwithstanding the limited
number of farmers and vendors who sold roots, the estimated transportation costs31 range
from 150 to 200 MT per Kg to Xai-Xai market and 2,380 MT per Kg to Maputo city.
Thus, the difference between farm gate and retail prices in Xai-Xai market
(transportation costs included) ranges from 1,000 to 2,100 MT per Kg, while the
difference ranges from 2,920 to 6,120 MT per Kg in Maputo city.

Despite the high price in city markets, a low percentage of the respondents sold
sweetpotato roots during the last two seasons. This may indicate that sweetpotato
growers used sweetpotato roots largely for household consumption. In addition the high
consumer price in the city markets could impair the sweetpotato consumption by low-

income people in urban areas.

 

3 ' Transportation costs per Kg of sweetpotato roots were computed, assuming that the farmer paid for the

round trip to the market site and transported an average of four sacks of sweetpotato roots.

90

 

5.5 Chapter Summary

In the study area, sweetpotato is mostly grown by middle-aged women. Due to the
lack of job opportunities in most of villages covered by this study, few households (27%)
participated in off-farm activities in 2003. Most respondents’ spouses were working away
from home primarily in South Africa or in Xai-Xai and Maputo cities.

The average family size was 7 people with an average 2 young children (under 5
years of age) and 3 children (under 18 years of age). While about 27% of households
owned cattle, livestock ownership was highly skewed in both districts. Apparently land
tenure is not a major constraint, since all respondents own family land averaging 2.4 ha
of total farmland per household. One of the major constraints farmers reported was a lack
of draft-power for land preparation, particularly the lowland heavy-clay soils since only a
third of respondents used animal traction in 2003/04 wet season. In addition, lowland
alluvial and peat soils play an important role in the overall smallholder crop farming.
However, due to the lack of investment in poor drainage improvement, these land
resources have been frequently flooded.

Public and NGO extension services have focused on providing information to
farmers about orange-fleshed sweetpotato cropping practices, and nutritional education
regarding the benefits of consuming the roots and leaves of OFSVs. However, only 20%
of respondents had received technical information related to sweetpotato production from
an extension agent in the past two years. Surprisingly, a huge percentage of respondents
(83%) were aware of the nutritional benefits of consuming OFSVS for young children

and pregnant and lactating women.

91

Notwithstanding the small percentage of respondents (15%) who participated in
field days and demonstration events, most of respondents (70%) reported having heard
about the products made out of OFSVs (cakes, juice, flour, chips) from SAARNET,
extensionists, health workers and the media. Unfortunately, a weak relationship was
observed between awareness and adoption of these products at the household level
allegedly due to the lack of know-how and/or insufficient production of orange-fleshed
roots for purposes other than the traditional methods (boiled, roasted or porridge). Thus,
changing food habits by incorporating new products into the diet or by introducing
changes in the way food is traditionally prepared is a long-term task and still to be
addressed. Despite the fact that the majority of respondents (63%) reported having
consumed sweetpotato leaves, the reddish stern color of some of OFSVS is likely to affect
the adoption of these varieties specifically for leaf consumption.

Few respondents (18%) are members of the area’s ten farmer associations.
However, the functionality and sustainability of these farmer associations are threatened
by the high dependence on external funding from the international development
community in addition to the limited crop marketing system and lack of credit
opportunities for farmers.

Farmers’ assessment of two OF SVs3 2 (Jonathan and Resisto) revealed that these
varieties were rated positively in terms of total yield, root size, taste of both fresh and
boiled roots, taste of leaves, and cooking time. Conversely, the same varieties were rated
negatively with respect to dry matter content of roots, tolerance to rainfed growing

conditions (drought), and tolerance to pests and diseases. In addition, the major

 

’2 Other OF SVs were assessed. However, few responses were gathered regarding Japon Selecto (8),

Taimung 64 (6), and LO 323 (2).

92

constraints to increasing sweetpotato production as identified by the respondents, include
frequent drought and flooding, a lack of animal traction, no market for sweetpotato roots
in the villages, and low off-season propagation capacity of OFSVS due to the low rainfall
during the dry season.

Sweetpotato roots were rarely commercialized. Few respondents (19%) reported
having sold sweetpotato roots (only 6% sold orange-fleshed roots) in 2003. Poor road
infrastructure linking the remote sweetpotato producing areas to the nearest (semi)-urban
markets, increases the transaction costs for sweetpotato commercialization, and
consequently hampers the potential contribution of this commodity to overall household
cash income.

The commercialization chain is characterized by a huge gap between farm-gate
and retail prices. For instance in Xai-Xai district, the retail price (3,150 Mt) was on
average 125% higher than the farm-gate price (1,400 Mt). This gap increased
proportionally as the distance from producer area to the market site increases. For
example, at the Maputo markets, the consumer price (8,300 Mt) was on average 493%
higher than the producer price in Xai-Xai district. Despite the superior nutritional value
of OFSVS, there is no difference in price between white/cream and orange-fleshed roots.

Presently, farmers do not apply the entire new technological package for orange-
fleshed sweetpotato farming, which includes fertilizers, and pesticides. Few respondents
(10%) reported having given vines of orange-fleshed sweetpotato to neighbors and/or
relatives and there is no evidence of vines being sold in the study area. Moreover there is
no difference in terms of agronomic practices and production costs between white and

orange-fleshed sweetpotato farming. Thus, the difference in terms on gross margin

93

revenue between white and orange-fleshed varieties is exclusively determined by the
yield differential between these two types of varieties.

Assuming that nutritional education is successful in increasing the consumption of
orange-fleshed varieties, there are additional health benefits associated with prevention
of Vitamin A deficiency that should be taken into account in evaluating the impact of

orange-fleshed sweetpotato varieties.

94

CHAPTER VI

FACTORS INF UENCIN G F ARMER ADOPTION OF OF SVs

6.1 The Adoption of OFSVS.

All respondents received vines of OFSVS in 2000 and some again received vines
in later years. However, during the last three to four years, the study area alternately
experienced frequent drought and excessive rain, especially in March 2004, which
flooded the lowland cultivated fields. As a result, most respondents lost their OFSVS
planting materials--mainly due to drought, floods or both, but also due to pests (rodents,

locust and snails) and goats eating the vines.

6.1.1 Definitions of Explanatory Variables

The specified probit model used in this analysis includes one dependent variable
and 18 explanatory variables. As previously mentioned, the dependent variable takes the
value Y =1, if the farmer cultivated orange-fleshed variety in the 2003/04 wet season, and
Y=0, if the farmer did not cultivate orange-fleshed varieties in the 2003/04 wet season.

The explanatory variables, including the direction of their expected effects, are
described as follows:
X. Respondent age ( - ): the units of this variable are years (age).

X2 Level of education ( + ): the number of years of formal education completed by
the respondent (educ).

X3 Young children ( + ): a dummy variable taking a value of one if the respondent
had young children (under five years of age) and zero otherwise (young).

95

X4

X5

X6

X7

x8

X9

X10

X13

X14

Extension services ( + ): the number of times the farmer visited or was visited by
an extension technician regarding any technological issue related to OF SV
farming during the last two years (extvis).

Field days/demonstrations ( + ): a dummy variable taking a value of one if the
respondent participated in field days/demonstration activities regarding OFSVs
processing techniques and zero otherwise (fdemo).

Livestock ownership ( - )2 the number ofcattle owned by the respondent (tbov).

Frequency of vines distribution ( + ): the number of times the farmer received
vines of OF SVs (vinrec).

Experience with varieties ( + ): the number of different OF SVs that the
respondent received (varexp).

Off-farm activity ( + ): a dummy variable taking a value of one if the respondent
and/or spouse participated in off-farm activities in 2003 and zero otherwise

(offarm).

Gender of the respondent ( + ): a dummy variable taking the value of one if the
respondent was female and zero otherwise (female).

Family size ( + ): the number of people who lived in the household at least nine
months during 2003 (famsize).

Sweetpotato area cultivated by the household in the 2003/04 wet season ( + ): the
sweetpotato area in hectares (sparea).

Awareness about sweetpotato processing techniques ( + ): a dummy variable
taking the value of one if the respondent was aware and zero otherwise @raware).

Average annual precipitation ( + ): a dummy variable taking a value of one if Xai-
Xai district and zero if Chokwe district. This variable captures the difference in
average annual precipitation between the two districts covered by this study

(distpr).

96

X15 Sweetpotato root consumption ( + ): the number of times per week the household
consumed sweetpotato roots during the months when roots are available (motes),

X”, Commercialized sweetpotato in wet season ( + ): the quantity (kg) of orange-
fleshed sweetpotato roots sold in the 2003 (rsold),

X17 Respondent experience in processed products ( + ): a dummy variable taking the
value of one if the respondent actually made at least one sweetpotato processed
product and zero otherwise (proexp).

X13 Total household farmland ( + ): the total area (hectares) cultivated by the
household in 2003/04 wet season (tland).

The summary statistics for these explanatory variables are presented in Table 23.
The correlation matrix (Appendix D) shows that these explanatory variables are not

highly correlated.

97

Table 23: Summary Statistics of Explanatory Variables in Xai-Xai and Chokwe
Districts, Mozambique (N=150)

 

 

 

Variable Mean Std. Error Minimum Maximum
No Code Type

X, age C 42.3 12.4 19 79
X 2 educ C 1.70 1.88 0 8
X3 young D 0.83 0.38 0 7
X4 extvis C 0.31 0.60 0 2
X5 fdemo D 0.14 0.35 0 1
X6 tbov C 2.40 6.52 0 57
X7 vinrec C 1.53 0.84 1 3
X3 varexp C 1.27 0.45 l 2
X9 offarm D 0.03 0.18 0 1
X10 female D 0.96 0.20 0 l
X“ famsize C 7.13 3.11 l 15
X); sparea C 0.02 0.06 0.008 0.5
X” praware D 0.71 0.45 0 1
X” distprec D 0.40 0.49 0 l
X15 rootcs C 3.60 1.15 1 7
X“, rsold C 3.13 15.37 0 120
X17 proexp D 0.04 0.20 0 1
X18 tland C 2.41 1.72 0.1 11.5
Source: Survey 2004 C = Continuous D = Dummy

6.1.2 Results of the Probit Model

Based on the probit procedure, the maximum likelihood parameter estimates for
the unrestricted (full) model were computed (Table 24). First, the coefficients (Bi) of the
explanatory variables in the full (unrestricted) model were examined and assessed with
respect to a priori expectations of the signs (according to adoption of technology theory)

and the statistical significance of the coefficients. Second, a restricted model (Table 25)

Was estimated, which excluded four variables from the full model, which had

98

Table 24: Maximum Likelihood Estimates of the Unrestricted (full) Model for Factors
Influencing Adoption of OFSVs in Xai-Xai and Chokwe districts (N=150),

 

 

Mozambique
Coefficient Standard Error p-value

Variable
intercept -2.8028 1.4208 0.049
age —0.0145 0.0130 0.264
educ 0.0139 0.0823 0.866
young -0.6485 0.1474 0.110
extvis 0.3870 0.3408 0.233
fdemo 1.2989 0.7185 0064*
tbov -0.0245 0.0185 0.185
vinrec 0.8071 0.2608 0.002***
varexp 0.9732 0.4579 0.034"
offarm " -0.5440 1.8229 0.765
female " 0.4748 1.051 1 0.651
famsize 0.0734 0.0634 0.247
sparea 0.0030 0.0013 0.023"
praware 0.1994 0.3507 0.570
distpr -1.0659 0.5020 0.034"
rootcs " 0.001 1 0.1144 0.992
rsold -0.0078 0.001 1 0.482
procexp " 0.0139 0.9468 0.988
(land 0.1653 0.1022 0.106

 

Likelihood Ratio test x2( 19 df) = 89.68
Pseudo R2 = 0.4313
** significant at 5% *** significant at 1%

Log Likelihood = -59.1180

Prob > x2 = 0.0000

* significant at 10 %

Source: Survey 2004
Excluded in restricted model

:1
signs inconsistent with adoption theory and/or were not jointly significant likely due to
the absence of variability among the respondents, including offarm (household off-farm
activities in 2003), female (respondent gender), motes (frequency of household
consumption of sweetpotato roots), and procexp (representing the experience of the

respondent in processing sweetpotato products).

99

Table 25: Maximum Likelihood Estimates of the Restricted Model for Factors

Influencing Adoption of OF SVs in Xai—Xai and Chokwe districts (N=150)

 

 

Coefficient Standard Error Marginal33 p-value
Variable Effect
intercept -2.7868 1.3628 NA 0.041
age -0.0145 0.0129 —0.0051 0.262
educ -0.0151 0.0818 -0.0053 0.854
young -0.6659 0.4642 -0.2028 0.151
extvis 0.3722 0.3363 0.1307 0.268
fdemo 1.3005 0.7071 0.3211 0066*
tbov -0.0247 0.0183 -0.0087 0.176
vinrec 0.7977 0.2584 0.2801 0.020**
varexp 0.9877 0.4476 0.3468 0.027M
famsize 0.0766 0.0610 0.0269 0.209
sparea 0.0030 0.0013 0.001 1 0.023"
praware 0.2049 0.3458 0.0734 0.553
distpr -1.0746 0.4808 -0.3806 0.025”
rsold -0.0076 0.0109 -0.0027 0.485
tland 1.6489 0.1019 0.0579 0.106

 

Log Likelihood for restricted model= -59.1658

Prob > x2 = 0.0000
* significant at 10 %
Source: Survey 2004

In order to check whether the coefficients of the excluded explanatory variables

were indeed zero (jointly are not statistically significant), a subsequent likelihood ratio

** significant at 5%

Likelihood Ratio {(14 dt) = 89.59
Pseudo R2 = 0.4309

*** significant at 1%
NA = Not Applicable

statistic was computed. This likelihood ratio statistic has an asymptotic chi-square

distribution with degrees of freedom equal to the number of restrictions being tested

(Wooldridge 1., 2003). The test indicated that the restricted model performed similarly to
the full model. The likelihood ratio test of the restricted model shows that the coefficients

are jointly significant at the 5% level.

 

33

100

The marginal effects were computed using the average values of the explanatory variables.

6.2 Discussion of the Results

Five variables included in the restricted model were statistically significant. The
dummy variable fdemo (participation of respondent in field days and demonstrations
activities regarding processed products derived from OFSVS roots and leaves) is
statistically significant at the 10% level. Its positive coefficient indicates that farmers
who participated in these events were more likely to have planted OFSVS in the 2003/04
wet season.

The variable varexp (number of OF SVs grown by the respondent) is statistically
significant at the 5% level. Its positive coefficient indicates that farmers who planted
more than one OFSPV were more likely to have grown nutritionally improved varieties
in the most recent season. This suggests that the probability of a farmer obtaining an
OF SPV that meets her/his tastes and preferences increases with the number of alternative
varieties at his/her disposal.

The variable vinrec (number of times the respondent received vines of OF SVs) is
statistically significant at the 5% level. Its positive coefficient indicates that farmers who
received OF SVs more times were more likely to have grown them in the most recent
season. This result could be partially explained by the high risk of crop failure in the
study area. For instance, most of the respondents in Punguine and Cuamba had recently
received vines (August 2003 and/or February 2004) and grew them in the 2003/04 wet
season.

The variable sparea (sweetpotato-cultivated area) is statistically significant at the
5%. Its positive coefficient indicates that farmers who planted a larger area of

sweetpotato were more likely to have planted OF SVs in the most recent season. This

101

result suggests that farmers who considered sweetpotato an important crop were more
interested in improved sweetpotato technologies.

The distpr (difference in terms of average precipitation between the two districts)
is statistically significant at the 5% level. Its negative coefficient (i.e., lower adoption in
Xai-Xai versus Chokwe) is consistent with farmers’ observation that the consequences of
erratic precipitation (i.e., drought, flooding) are problematic for sweetpotato farming. As
mentioned in Chapter II, sweetpotato is mostly grown in lowland alluvial and peat soils
where drainage is a serious constraint. The occurrence of high and relatively intense
rainfall during the wet season, combined with poorly drained soils, increases the risk of
flooding in the region. This risk is exacerbated in the coastal district of Xai-Xai, which is
characterized by low altitude, extensive lowland alluvial and peat soils, and poorly
functioning drainage systems. However, in both districts, farmers reported having lost
their OF SVs planting material due to flooding in the lowlands—which is due to a lack of

investments in rehabilitation of the drainage systems.

6.3 Goodness-of Fit

Using STATA, the estimated probability of each respondent being an adopter
was computed. Using 50% as the cut-off probability34 of being an adopter, a comparison
of the observed numbers of adopters and disadopters versus the estimated numbers was
made (Table 26). The model correctly predicted 78% of the total outcomes, correctly
predicted 71% of adopters and 85% of disadopters). This measure of goodness-of fit
indicates that the model has relatively good explanatory power, although disadopters are

more accurately predicted than adopters.

 

3" Estimated probability of adoption less than 50% is considered as being disadopter.

102

Table 26: Comparison of Observed and Predicted Outcomes of Adoption of OFSVS,
Xai-Xai and Chokwe, Mozambique (N=150)

 

 

Predicted Adopters Predicted Disadopters Total
Observed Adopters 54 (71%) 22 (29%) 76 (100%)
Observed Disadopters 1 1 (15%) 63 (85%) 74 (100%)
Total 65 85 150

 

Correctly Classified (l 17 out of 150) = 78%
Source: Survey 2004

6.4 Chapter Summary

The restricted model identified five factors that influenced the adoption of OF SVs
among farmers in Xai-Xai and Chokwe districts. These factors include: (1) participation
in field days and demonstration activities, (2) the number of OF SVs received by the
respondent, (3) the number of times the farmer received vines, (4) the household’s
sweetpotato cultivated area, and (5) the district where the respondent resided.

These results suggested that in order to increase adoption, there is a need to: (1)
provide farmers with several different varieties; (2) provide vines several times to
farmers; (3) continue to conduct field days and demonstration activities, and (4) invest in

improving drainage in lowland areas.

103

CHAPTER VII

SUMMARY AND POLICY IMPLICATIONS

7.1 Summary

In Mozambique, the majority of the population relies on subsistence agriculture.
Thus, the low levels of production and productivity, in conjunction with the relatively
high population growth rate, have contributed substantially to a high unemployment rate,
extensive poverty, and widespread VAD--especially among children and pregnant and
lactating women.

In early 2000, Mozambique experienced a severe flood that devastated cultivated
land throughout the southern and central provinces of Maputo, Gaza, Inhambane, Manica
and Sofala. Soon after the floods subsided, the government launched a program of to
multiply and distribute OFSVS. The program, which targeted the flood-affected rural
people, was aimed at mitigating hunger and food insecurity, and reducing VAD. In
addition, SARRN ET, in close collaboration with its partners, conducted training activities
to teach farmers about improved sweetpotato agronomic practices, the crop’s nutritional
value and agro-processing techniques for sweetpotato; developed teaching materials on
nutritional concepts for the primary school curriculum; and sponsored field
days/demonstration activities at which food products made out of the roots of OF SVs
were displayed.

Several studies have highlighted the importance of B-carotene as the predominant
source of pro-Vitamin A in preventing VAD in developing countries. However, the

bioavailability (conversion of B-carotene into Vitamin A and effective utilization of this

104

Vitamin A by humans) is conditioned by a complex set of factors, including the type and
amount of carotenoid in a meal, the matrix in which the carotenoid is incorporated, the
presence of absorption and conversion modifiers, the health status of the target
population, and the presence of other micronutrients. Studies aimed at assessing the
impact of food-based interventions in developing countries revealed different results from
country-to-country. Furthermore, the seasonal availability of enriched sources of [3-
carotene, the low bioavailability and conversion of B-carotene into Vitamin A, and the
prevalence of infectious diseases hamper the efficacy of food-based initiatives to mitigate
VAD in developing countries.

The willingness of farmers to grow OFSVS will greatly affect the program’s
impact on reducing VAD. Recent studies have identified many social and economic
factors that affect farmers’ technology adoption decision--including access to farm labor,
farm machinery, storage and distribution facilities, inputs and output markets, credit
availability, farm size. level of education, land tenure, profitability of the technology,
farmer’s awareness of existing technologies, and government policies. However, since
the effect of those explanatory variables (and interactions among them) on farmer’s
technology adoption varies from region to region, location-specific studies are required to
understand factors that affect adoption of a new technology in a given country.

In Gaza province--one of the provinces targeted by the OFSVS multiplication and
distribution program and the focus of this study--small-scale farming is by far the most
important sector. The absence of a commercial cash crop sector reflects the recent shift
from large concession companies towards contract farming, in which small-scale holders

cultivate primarily cotton and tobacco. The main role of the companies is to provide

105

inputs and a secure market for cash crops. One of the major constraints to crop
production (including sweetpotato) in Gaza province is the low and irregular rainfall
during the growing season, resulting in high risk of crop failure. Thus, the development
of short-duration (early-maturing) varieties is crucial to increasing crop production under
rainfed growing conditions. Furthermore, the lack of investments in drainage systems,
which are required to farm lowland fertile soils, discourages the use of these soils for
agricultural development.

To assess the adoption of OF SVs in Gaza province, a multi-stage sampling
approach was used to select a sample of 150 sweetpotato growers in Chokwe and Xai-Xai
districts. Respondents were drawn at random from lists of farmers who received OF SVs
vines in 2000. A structured questionnaire was used to collect data on the socio-economic
characteristics of the sampled household, agronomic practices followed, labor used in
sweetpotato production, farmers’ assessment of traits of some of the new OFSVS, access
to output markets, and access to extension services.

In the study area, middle-aged women mostly grow sweetpotato. Due to the lack
of local opportunities. few households engaged in off-farm activities in 2003. However,
most of respondents’ spouses worked away from home, primarily in South Africa or in
Xai-Xai and Maputo cities.

All respondents cultivated family-owned family land averaging 2.4 ha per
household. On average, farmers planted approximately 200 m2 of sweetpotato. Most
farmers cultivated lowland alluvial and peat soils. However, due to the lack of investment
in drainage improvement, farmers reported that their fields were frequently flooded

resulting in crop losses.

106

Only 27% of households owned cattle and the livestock distribution was highly
skewed in both districts. Thus, while farmers needed draft power for land preparation,
especially when cultivating heavy-clay lowland soils, only one third of the respondents
used animal traction in 2003/04 wet season.

While public and NGO extension services have promoted orange-fleshed
sweetpotato cropping practices, few farmers (20%) received technical information related
to OFSV cropping. However, most respondents (83%) were aware of the nutritional
benefits of consuming OFSVS for young children and pregnant and lactating women.

Although few respondents (15%) participated in field days/demonstration
activities, most of the respondents (70%) reported having heard about the products made
out of OF SVs (e. g., cakes, juice, flour, chips) from SAARN ET staff, extensionists, health
workers and the media. Unfortunately, there was little relationship between awareness
and adoption of these products by households--likely due to a lack of know-how and/or
insufficient production of OFSVS roots for purposes other than the traditional methods
(boiled, roasted or porridge). Farmers assessed two OF SVs3 5 (Jonathan and Resisto)
varieties positively (compared to their WFSVS) in terms of root size; taste of fresh and
boiled roots, and cooking time. Conversely, the same varieties were ranked negatively
with respect to dry matter content of roots, tolerance to rainfed growing conditions
(drought), pests and diseases.

The farmers reported several constraints to increasing sweetpotato production,

including frequent drought and flooding, no market for sweetpotato roots in the villages,

 

35 Other OF SVs were assessed. However, few responses were gathered regarding Japon Selecto (8),

Taimung 64 (6), and L0 323 (2).

107

a lack of animal traction, and low off-season propagation capacity of OF SVs due to the
low rainfall during the dry season.

While most respondents reported having consuming sweetpotato leaves, they did
not like the reddish stem color of some of OFSVs--which is likely to affect the adoption
of these varieties for leaves consumption.

Sweetpotato roots were rarely commercialized. Only 19% of the respondents sold
sweetpotato roots (mostly white-fleshed roots) in 2003. Poor road infrastructure linking
the remote sweetpotato-producing areas to the nearest semi-urban and urban markets
increases the transaction costs for sweetpotato commercialization, which hampers the
potential contribution of this commodity to overall household cash income.

A huge gap existed between farm-gate and retail prices. For instance, in Xai-Xai
district the retail price averaged 125% higher, compared to the farm-gate price; while in
Maputo markets, the consumer price averaged 493% higher than the producer price in
Xai-Xai district. Despite the superior nutritional value of OFSVS, there is no difference in
price between white and orange-fleshed roots.

All farmers received one or more OF SV in the 2000 wet season from an NGO and
some again received OFSVS in a subsequent season. However, the percentage of farmers
who continued to plant OF SVs varied by season, ranging from 71% in 2002/03 wet
season to 51% in 2003/04 wet season. Most farmers planted only a small percentage of
their sweetpotato area to OFSVs--ranging from 10% in 2002/03 dry season to 20% in
2002/03 and 2003/04 wet seasons. Reasons farmers reported for not planting a larger area
to OF SV in 2003/04 wet season low off-season propagation capacity and low drought-

tolerance of OFSVS, and abundant planting materials of WFSVs. While many farmers

108

lost their OF SV to flooding and/or drought, they were able to rescue their WFSVs
suggesting that they valued their WFSVs more than OFSVS.

A probit model was used to identify factors affecting farmer adoption of OFSVS.
Five factors were statistically significant and positively associated with adoption--
participation in field days and demonstration activities, the number of OFSVS received by
the respondent, the number of times the farmer received vines, the household’s
sweetpotato cultivated area, and the district where the respondent resided (e. g. higher

adoption in Chokwe due to relatively less frequent flooding).

7.2 Policy Implications
The findings of this study have policy implications for agricultural research
institutions, rural extension and development, agricultural and rural development

directorates, NGOS, and health institutions.

7.2.1 Agricultural Research & Development36

Firstly, this study revealed a significant positive relationship between the number
of different OF SVs received by farmers and adoption of improved varieties. Sweetpotato
is used for two purposes «root and leaf consumption. Farmers reported preferences for

varieties based, on both their leaf and root characteristics, and likely continued to plant

 

36 . . . . . .
By enhancmg the nutrlent quality of sweetpotato, agricultural research can play an important role in

improving the diet diversification. However, breeding programs face several challenges, including
developing varieties with: (i) high nutrient-density with little or no yield trade-off to guarantee their
widespread adoption by farmers, and (ii) traits desired by consumers (e.g., storability, cooking time,
appearance and taste (Hazell et al,. 2001).

109

the varieties that met their preferences. Thus, crop scientists should develop OF SVs with
traits that better meet farmers’ tastes and preferences for both roots and leaves.

Secondly, many respondents reported that they received vines more than once
because they lost the OFSVS, which they originally received, due to drought. Thus,
research institutions should develop OFSVS that are both suitable37 for farmers’ rainfed
growing conditions (characterized by erratic and irregular rainfall) and meet
farmers/consumers’ tastes and preferences.

This suggests the need for on-farm research trials in which farmers participate in
evaluating and selecting OFSVS38 for multiplication and distribution. To implement this
initiative, the program could utilize its existing partnership with researchers,
extensionists, and NGOs to establish a network of on-farm research trials. Farmers’
active participation at an early stage in assessing the performance of promising improved
OFSVs would provide useful feedback to researchers, regarding relevant traits that need
to be incorporated into OFSVS to insure that released varieties are suitable for the
region’s agroclimatic conditions and have other traits that are desired by
farmers/consumers. Furthermore, the participation of farmers at the early stage of
technology generation and dissemination would give farmers a greater sense of

ownership of future improved varieties--and thereby increase the likelihood of adoption.

 

37 Goldman R. et al., 1981, argued that evidence from the Green Revolution revealed that the key
determinant of new technology adoption is the adaptability/compatibility of the new seed
variety/planting material and associated cultivation practices with the site-specific agroclimatic
environment.

38 Some studies reported that orange flesh color and dry matter content are negatively associated. Thus,
the development of nutritionally enriched pro-vitamin A orange-fleshed sweetpotato variety that meets
the consumers’ preference for high dry matter will be a challenge to research institutions (Jones, 1977;

Collins, et al., 1995).

110

Thirdly, most of the surveyed sweetpotato growers reported that the market for
orange-fleshed sweetpotato was very limited, particularly in villages far away from
markets. Notwithstanding the limited resources available for research, a post-harvesting
research program on sweetpotato should be incorporated into the national research
agenda. The research should focus on identifying potential orange-fleshed sweetpotato-
based products for agro-enterprise development. The creation of new market for value-
added orange-fleshed sweet potato-based products would raise sweetpotato from
subsistence to a cash crop, and thereby play an important role in promoting increased
adoption and production of OFSVS.

Finally, most of the respondents reported only consuming sweetpotato roots
during 3 to 5 months per year. Thus, seasonal availability of sweetpotato roots constrains
the potential of this commodity to mitigate VAD. Post-harvesting research should
investigate and test appropriate methods of storing sweetpotato roots in order to extend

the availability of sweetpotato roots for human consumption.

7.2.2 Rural Extension & Development

First, since the majority of sweetpotato growers are women, who have a low level
of literacy (average of 1.7 years of school), extension materials written in Portuguese are
not appropriate for them. Thus, there is a need to produce good quality and
understandable extension materials, which are written in local languages. Furthermore,
the rural extension network (provincial level, in close collaboration with NGOs and
community-based associations) should disseminate those materials to sweetpotato

growers. While interpersonal contacts (e.g. field days and demonstrations) should be

111

relied on as much as possible to promote OFSVS, there is also a need to continue to
produce and broadcast awareness programs via radio using local languages (e. g.,
nutritional value of OFSVS, market opportunities in urban areas).

Second, women are both the decision-makers regarding orange-fleshed
sweetpotato farming and are responsible for child feeding practices in their households.
To date, food-based interventions have focused on women. However, men’s tastes and
preferences affect what women feed the family and men strongly influence the household
purchase of food and/or health services”. Thus, there is a need to target messages about
the nutritional value of OFSVs towards men in order to increase their awareness about
nutritional issues.

Third, few respondents reported having made any of the disseminated processed
orange-fleshed sweetpotato products that were promoted through field days and
demonstration events. In order to enhance the adoption of processing techniques, special
attention should be paid to assessing the practicability and feasibility of making these
products-- taking into account locally available ingredients, specific local food
preparation knowledge, and consumer’ preferences. In addition, at demonstration
sessions, the consumption of alternative sources of micronutrients that are locally
available and accepted should be equally promoted. For instance, some of the villages
covered by this study are suitable for the production of pumpkins, pineapple, papaya,
banana, citrus and mangos. Thus, the integrated promotion of production, agro-

processing, and consumption of these commodities, as alternative sources of

 

39 . . . . . . .
Results from Kenyan and Tanzanian studies on food-based interventlons showed that mvolvrng men in

the process enhanced the adoption of orange-fleshed sweet potato varieties, solar drying technology,
and related nutritional messages (JohnsomWelch, 1999).

112

micronutrients, should be emphasized in order to ensure that households have access to
an improved and balanced diet, throughout the year given the seasonal availability of

different crops40 in the farming systems.

7.2.3 DDARDS and NGOs

First, due to the lack of investment in drainage systems, the frequent occurrence
of flooding is a serious bottleneck for sweetpotato production. Thus, the District
Directorates for Agriculture and Rural Development (DDARDS) and NGOs should
search for financial resources to improve drainage systems in the lowlands. This
investment would make more fertile land available for agricultural development in
general, and particularly for sweetpotato farming.

Second, the heavy texture of alluvial soils limits the effectiveness of manual land
preparation. Thus, DDARDS and NGOs should facilitate the formation of farmers’
associations, aimed at promoting animal husbandry (mainly cattle) by providing credit to
members on a rotational basis among members. In the long run, greater access to draft
power would contribute substantially to increasing the average area cultivated per
household. This would contribute to supporting an integrated food-based intervention by

providing rural households access to a more diversified and balanced diet.

40

According to Agriculture and Livestock Census 1999/2000, of the total number of holdings (219,541)
in Gaza province. 64.5% had mango trees, papaya (50.6%), orange (35.1%), banana (31.7%),
pineapple (25.8%), and mandarin (23.9%). In addition, approximately 6%, 5.0%, and 4% of the total
holdings sold mango. oranges. and pineapples, respectively.

113

7.2.4 Health Institutions

First, from the nutritional standpoint, there is a strong relationship between the
health status of a person and the bioavailability and conversion of pro-Vitamin A. Thus,
food-based interventions should be complemented by the provision of health services to
the rural people--particularly to young children and pregnant and lactating women, as
well as improved sanitation. These health services should address reducing bacterial
infections, intestinal infections, and parasitism in order to avoid negative impact of these
health disorders on the bioavailability and efficacy of utilization of B-carotene provided
by food—based nutritional interventions.

Second, in accordance with the Food Security and Nutritional Strategy of
Mozambique, food-based interventions should be encouraged and enhanced to prevent
VAD. However, since the effectiveness of food-based intervention is still uncertain,
health institutions should continue to implement complementary strategies, including

curative supplementation of Vitamin A and other micronutrients.

7.3 Limitations of the Study

The findings reported in this study strictly apply to the study area and should not
be generalized to other regions with different agroclimatic and socio-economic
characteristics.

The marketing system for sweetpotato roots was not analyzed in detail due to the
existence of few traders involved in sweetpotato marketing during the survey period. This
fact was partially due to the limited sweetpotato supply, as a result of flooding that

devastated most lowland fields, including sweetpotato plots.

114

Farmers” assessment of two out of nine OF SVs was carried out and reported in
Chapter V. While a few respondents assessed three other varieties, these results were not
reported because they might not be representative, due to the small sample size.

Most of data gathered in this survey were based on recall information provided by
the sweetpotato growers. In addition, most of the respondents were illiterate and
improved OFSVS were quite new in their farming systems. All of these factors may have

affected the accuracy of the data collected.

7.4 Further Research Needs

First, this study used cross-sectional data to analyze the factors influencing
adoption of OFSVS in a certain point in time (2003/04 wet season). However, the
technology adoption process is inherently dynamic--decisions made in one period are
strongly dependent on the consequences of decisions made in previous periods. Thus,
there is a need to select and periodically survey a panel of sweetpotato growers in order
to monitor changing patterns in the use of nutritionally improved orange-fleshed
varieties--including trends in the partial or total replacement of white-fleshed varieties,
area cultivated, and the rate of adoption of new varieties by farmers over time.
Additionally, these panel data could be used to assess the impact of new OFSVs on the
well-being of farmers and whether or not the benefits of being an early adopter are
preserved, even when many farmers have adopted the technology.

As discussed in the previous chapter, poor marketing is a major bottleneck to
increasing sweetpotato production, particularly OF SVs. Therefore, post-harvesting

research aimed at assessing ways to establish linkages between sweetpotato-growing

115

farmers and sweetpotato processors4| is vital for identifying potential markets, especially

for farmers in the remote and isolated production zones that are disconnected from

potential urban markets.

To address the limited demand and seasonal availability of OF SVs, studies should

be carried out to investigate the present and potential demand for fresh and processed

sweetpotato products. The focus of these studies should include assessing the following:

‘1)

Potential processed sweetpotato products that are highly acceptable for
human consumption and are likely to be traded on a commercial basis;
Attributes that determine product quality, as judged by consumers (i.e.
consumer’s tastes and preferences);

Feasibility of utilizing sweetpotato as a raw material in existing or

142 processes and appropriateness of available food

potential industria
processing technology to produce these products and;

Assessment of the potential fresh sweetpotato supply, relative to the

projected demand.

 

41

For instance, the processing industry in Taiwan revealed that up to 15% of wheat flour can be replaced

with sweetpotato without altering its physical properties. Thus, the development of an effective
industry of sweetpotato flour to satisfy the demand was considered as a vehicle of increasing
sweetpotato production and consumption (Tsou and Hang, 1992, Espinola et al., 1998, Hagenimana et

aL,l999)

42

It is hypothesized that the successful development of a processing industry is likely to contribute to
increased utilization of sweetpotato by reducing the losses during storage and marketing. Although the
relative importance of key factors conducive to achieving this goal is not well known yet, it seems that
economic and cultural constraints are of the same magnitude or greater than the technical limitations
towards development of successful processing industry (Bouwkamp .1. 1985).

116

The results of these studies would provide important insights regarding the
constraints to and opportunities for developing a sweetpotato-based agro-industry, which
would serve to increase the consumption of orange-fleshed processed products

throughout the year, increase farmers’ incomes, and improve household nutrition.

117

APPENDIX A

PHYSIOGRAPHY AND SOCIOECONOMIC FEATURES OF MOZAMBIQUE

118

ill ll fl..- Diplnslo M. .
2‘! .0. .

A.l Geography and Agroclimatology

The Republic of Mozambique is located 10° 27 N, 26 52 S, 40 51 E and 30° 12 E
on the eastern coast of Southern Africa, occupying an area of 799,380 square kilometers.
The country shares its 4,571 Km of land borders with Tanzania to the north; Malawi,
Zambia, Zimbabwe, and the South African province of Mpumalanga to the west; and
Swaziland and the South African province of Kwazulu Natal to the south. Mozambique’s
Indian Ocean coastline stretches over approximately 2,470 Km (IN E, 2002).
Mozambique is divided into ten agroecological zones-- from mostly coastal lowland and
uplands of 200-600 meters of altitude in the south and center regions to high plateaus and
mountains more than 1,000 meters in the northwest and west. The highest point is Mount
Binga (2, 436 meters) located in Manica province (MAF, 1996).

The climate of Mozambique is strongly influenced by the warm Indian Ocean
currents (moving south from the Equator) and the altitude of the Mozambican plateau.
The climate of Mozambique is predominantly tropical, with a cool (dry) season from
May to August. However, the climate in Maputo, which lies just south of the Tropic of
Capricorn, is subtropical with a substantial maritime influence. Most rainfall normally
falls between November and February and the humidity is high for most of the year and n
averaging 70%. Although the annual average mean temperature is 235° C, temperatures
above 43° C are common (Pereira L. and Omar F., 2001).

Mozambique is very exposed to natural disaster caused by climatic irregularities,
as nine of the main rivers in the Southern African region flow to the Indian Ocean

through the country. However, none of the previous droughts, floods or cyclones was as

119

rapid and severe as the heavy rains and floods that occurred across Southern
Mozambique during early 2000 and caused widespread destruction.

Although the spatial and temporal precipitation patterns-- modulated by latitude,
topography and coastal shape --are highly irregular, there are some identifiable
homogeneous regions (e. g. semi-arid, moist, and intermediate regions. Except for Tete
province, south of Zambeze River, annual precipitation averages above 1,000 mm north
of parallel 20° S and below 1,000 mm south of this parallel. Rainfall is highest in the
coastal zone and decreases rapidly towards the interior of Gaza province-- attaining a
maximum of 400 mm in Chicualacuala (Pereira L. and Omar F., 2001).

The temporal distribution of precipitation across the year is governed by the
motion of two wind patterns, one coming from the south and other from the north. The
former, which begins in September, reaches its peak in the second week of February as
the rainfall moves from the coastal zone to the interior. This phenomenon leads to a very
short cropping season in the south interior. The north wind, which is very active,
persistent and extends inland to 20° S. Consequently, the rainfall in the northern region of

Mozambique has a more regular pattern (Pereira L. and Omar F., 2001).

120

A.2 Population Growth

According to the national population census of 1997, the population of
Mozambique was estimated at 16.1 millions --corresponding to an average population
density of approximately 20 inhabitants/kmz. Almost 44% of the Mozambican Republic’s
population is less than 14 years of age, and 53 % is between 15 and 64 years of age.
Approximately 70% of population lives in rural areas (INE, 1999).

Population in Mozambique increased at an average national growth rate of 2.3 %
annually during the early 19803 to late 19905. Approximately 45% of the population is
currently economically active. Of this total, approximately 75% carry out subsistence
agriculture, 19% are employed as waged workers, and 5% are self-employed outside of
agriculture. Internal migration from the countryside to the three major cities of southern
provinces of Mozambique (Maputo, Gaza and Inhambane) was 32%, 18% and 17 % in
1997, respectively (INE, 1999). Although population growth declined to 1.9 % in the late
1990, population continued to grow to 18.8 millions in 2003 (World Bank, 2004).
Unfortunately, the high rate of rural and urban population growth leads to a high demand

for food and increasingly higher rates of unemployment and poverty.

A.3 Macroeconomy

According to recent macroeconomic indicators, the performance of the
Mozambican economy has been satisfactory over the recent post-war years. The GDP
growth rate was 12.6 % in 1998, 13.0 % in 2001 and 7.7% in 2002 compared to an annual
rate of only 0.1 % from 1980-1990 and 6.9 % from 1990-2001. At constant prices of

1996, Mozambique’s annual GDP growth rate averaged 11.6 % from 1996 to 2002 (INE,

121

2004). However, inflation is an issue of concern, as the GDP implicit price deflator
averaged 4.6 % in 1998, 10.6 in 2001 and 11.2 % in 2002. Furthermore, Mozambique
faces a substantial trade deficit of goods and services, estimated at 17.4 % of GDP in
1998, 13.6% in 2001, and 14.7 % in 2002 (World Bank, 2004).

Mozambique’s economy is dominated by three sectors -- Manufacturing and
Mining Industry, Commerce and Repair Services, and Agriculture and Fisheries. In 2002,
these three largest economic sectors contributed 22.3 %, 17.2 and 17.6 % respectively, to
the Mozambican GDP (Fig. 1). In contrast, from 1996-2002, the fastest growing sectors
(in constant prices of 1996) were Electricity and Water (49.2%), Construction (26.8%),
Public Administration and Defense (23.8%). During the same period, growth in
Manufacturing and Mining Industry, Commerce and Repair Services and Agriculture and

Fisheries averaged only 15.4%, 7.3%, and 5.2 %, respectively (INE, 2004).

Figure A: Major Economic Sectors and their Contribution to Mozambican

   
 

GDP in 2002
.P-ubl1c. Other Serv1ces Agriculture &
Admmlstratlon & 14% . .
D f ‘1 /,. Flsherles
e ense ‘ 18%
5%

Transportation & /

Communications 1 .
Manufacturlng and

9V
0 i \W‘ Mining Industry
Construtlon / Commerce & Repair 22%
15% Services
17%

Source: INE, 2004

122

A.4 Agriculture Sector in Mozambique
A.4.l Agricultural Policy & Strategies
In 1995, the Government of Mozambique (GOM) developed a policy framework -

- Agricultural Policy and Strategy for Implementation-- which identified the main
agricultural development policy “the recovery of agricultural production in accord with
self-sufficiency and food reserve, and the promotion of increases in the levels of
commercialization of export products” (GOM, 1995). In the context of GOM agricultural
policy, agricultural development is defined as the progressive transformation of
subsistence agriculture into agriculture more integrated into the functions of production,
distribution and processing towards provision of surplus to the market, food self-
sufficiency and primary materials for national industry and its contribution to the balance
of payments.

Mozambique is still one of the poorest among the group of least developed countries -
- characterized by a food deficit and a huge dependency on emergency food aid while
local production satisfies only 80% of minimum daily calorie requirements (MAF, 1996).

The predominance of subsistence agriculture -- characterized by traditional
technology, frequent drought, diseases and pests, and the lack of inputs and improved
technology-- have contributed substantially to low levels of production and productivity.
However, improving the agricultural sector is vital to the country’s economic
development and recovery, as it is mainstay of food security, economic growth, and

employment.

123

Consequently, the GOM has launched initiatives to improve its agricultural policies
and the institutional setting which are aimed at creating an enabling environment for
agricultural growth, including (GOM, 1995):

o Restructuring of commercial state farms through privatization and joint ventures;

0 Implementing a balanced phasing-out of food aid in order to promote and boost
domestic production;

0 Promoting rural marketing through liberalization of domestic trade and
improvement of access to credit and to agricultural inputs;

0 Developing adequate land tenure policies and systems aimed at protecting
propriety rights and to provide incentives for national and foreign investment in
commercial agricultural related activities; and

0 Providing timely and accurate information on input and commodity prices to

support the decision-making process of producers.

These agriculture policy reforms, implemented within the broader context of the
Structural Adjustment Program (SAP), have been largely completed and maintained
through a broad and more comprehensive National Agriculture Development Program
(PROAGRI). Implemented since the beginning of 1999, PROAGRI is envisaged to
strengthen capacity within MADER to provide efficiently and effectively core public
services and to develop improved institutional arrangements for the financing and
delivery of agricultural services for the family sector. Agricultural input and output

markets have been liberalized and prices have been decontrolled for most crops.

124

Since the mid 19903, the GOM has introduced several incentives to agriculture
development, including eliminating import taxes on agricultural inputs and the
establishment of special fiscal and customs regions with extensive exemptions from
import duties and consumption and circulation taxes on building materials, and equipment
deemed necessary for implementing an approved investment project. In zones endowed
with minerals and high agriculture potentials, the GOM has established specialized
entities to coordinate the resource development-related activities in these regions. For
instance, in order to boost the harmonious development of the Zambezi valley, the GOM
established the Gabinete do Plano do Zambeze (GPZ). Investors are assured security and
legal protection of property and rights in connection with the investments made, export of
foreign investors‘ profits, repatriation of capital invested upon sale, as well as freedom to

import equity capital or secure loans locally.

A.4.2 Agricultural Research & Development Institutions

Increases in agricultural production and land productivity are the key components
for stimulating economic growth and improving the welfare of the present population and
future generations in Mozambique. This will require increasing farm-level production
and productivity through the development and transfer of improved, and profitable

technologies.

125

A.4.2.1 Agricultural Research
Under the auspicious of the Ministry of Agriculture and Rural
Development, there are four research institutesl, namely:
0 National Institute for Agronomic Research (INIA)
0 National Institute for Veterinary Research (INIVE)
0 Institute for Animal Production (IPA) and,

0 Center for Forestry Experimentation (CEF)

The overall goal of agricultural research is “to develop, test, and disseminate
agricultural technology to producers that enhances national and farm level production,
provides food security and self-sufficiency, maintains the natural resource base, provides
a surplus for industry and export, and generates employment and income in an equitable
fashion” (MAF, 1996).

The agricultural technology improvement and socio-economic research are two of
the six major categories given priority in the Agricultural Research Strategy for
Mozambique. The later should investigate problems and constraints at the farm level,
improving farm management, and identify ways to increase rates of technology adoption,
as well as emergent sectoral opportunities and threats for producers. Although this
agricultural research strategy was elaborated and approved in 1996, little socio-economic
research has been carried out, due to lack of qualified personnel and funds. It is important
to point out that GOM investments in agricultural research have been decreasing over

time and accounting for less than 4% of its agricultural sector budget (MAF, 1996). This

 

' Recently the GOM Council of Ministers approved the creation of the Instituto de Investigacao Agraria de
Mocambique (11AM), which amalgamates these four research institutes, Agricultural Training Center
(CFA), and Agricultural Documentation Center (CDA) (GOM, 2004).

126

low level of investment in agricultural research—compounded by the frequent turnover
of top-level scientists and research program managers, as well as year-to-year
fluctuations in the budget allocated for research» has substantially hampered the research

institutes in fulfilling their mandates (MAF, 1996).

A.4.2.2 Extension Services

Due to its importance and the contribution of smallholders’ agricultural
production in the country’s economy, public sector extension targets smallholders.
Developed under the auspicious of the National Agricultural Development Program
(PROAGRI), the National Extension Master Plan (1999-2003) provided for the
development of an Integrated National Extension System (SINSE). The National
Directorate of Rural Extension (DNER) is expected to provide public services to the
priority districts and to promote the active participation of other stakeholders potentially
interested in providing extension services, such as private commercial farmers, NGOS,
farmers’ organizations, and other private sector institutions.

Due to limited availability of financial, human, and material resources to meet the
extension demand throughout the country, the National Extension Master Plan
accommodates multiple financial and delivery arrangements, including outsourcing and
cost sharing with private extension entities. Through outsourcing, extension services are
provided to some of the targeted population by private sector extension institutions, but
these services are coordinated and regulated by the public sector. Thus, in addition to

being one among several extension providers, the GOM acts as regulator, evaluator, and

127

facilitator of a more comprehensive and integrated extension service network (Gemo H.

et al, 2001).

A.4.2.3 Market Information Systems

In recognition of the importance of timely market information organization to
agricultural growth and development, the Ministry of Agriculture and Rural Development
has developed the Agricultural Market Information System (SIMA), to collect, process,
and disseminate weekly information on prices of basic commodities at the major markets
to potential users including, producers, NGOs, traders, governmental officers, and policy
makers. Since SIMA it has developed a price database and has published numerous
bulletins and papers annually (Mabota A. et al, 2003).

Presently, the market systems for export and industrial crops are relatively better
developed than those for food crops. Thus, farmers have a strong incentive to devote their
scarce resources in generating marketable surpluses of these crops.

While it is imperative that farm gate prices are high enough to create incentive for
farmers to increase production, there is also a need to guarantee that the urban population
has access to an adequate and inexpensive food supplies. Thus, as markets are being
liberalized, the GOM should address the institutional and marketing constraints in order
to prevent an eventual disastrous erosion of urban purchasing power until increased

supply from the rural areas contributes to price stabilization.

128

A.5

Trade Policy and Strategy

Improving access to output markets (for the majority of rural people engaged in

subsistence agriculture) is important for increasing agricultural production and rural

incomes. Key trade policy interventions that are relevant for agricultural development are

as follows:

Rehabilitation and further expansion of the commercial network, including: (i)
establishment of appropriate physical market facilities and provision of fiscal
incentives for investments in market infrastructure, particularly in the remote rural
areas; and (ii) institutional co-ordination for rehabilitation of the road network in
most populated and major agricultural production areas.

Creation of an enabling environment for the private sector to progressively
increase its contribution in developing the agro-industry sector in order to create
employment opportunities and stabilize the food supply.

Improvement of agricultural marketing systems, including (i) promotion of
farmers or groups association which are based on common and shared interests;
(ii) gathering, processing, and dissemination of information on major agricultural
commodities in domestic and external markets on a periodical basis; (iii)
promoting investments in improved handling, storage, and road/rail transportation

services, and (iv) stimulating export of agricultural commodities.

These strategies are aimed at developing a market-friendly policy environment for

agriculture and rural development, whilst recognizing and emphasizing the role of civil

society, the private sector, and NGOs in meeting the challenging tasks of stimulating

129

more equitable and sustainable growth in the rural areas. In addition, agriculture research
can be expected to contribute substantially to economic growth and thereby to long-term
poverty reduction in countries with liberalized markets and highly developed rural
infrastructure. Furthermore, farmers can benefit from greater opportunities derived from
diversifying their agricultural production into higher-value products, if they are properly
connected to larger and efficient domestic and international markets (Hazell P. and

Haddad L., 2001).

A.6 Land Use

Mozambique has 36.1 million hectares of land suitable for agriculture, of which
only 5 million are presently cultivated, and 46 million hectares of forest. While
Mozambique’s resource-base for agriculture development includes 3.3 million hectares of
irrigable land, which presently only 50,000 are exploited for irrigated agriculture (GOM,
1995). The smallholder sector is by far the largest, accounts for 97% of total cultivated
area in 2000/01 (3.9 million of hectares), of which 83% were for basic food crops. The
average farm size varies from one agro-ecological region to another, but 84% of all
holdings have an area below two hectares (INE/MADER, 2002). Although the
commercial agricultural sub-sector is expanding as a consequence of new government
agricultural policies envisaged to promote private investments in agriculture sector, it is
still small.

The large public state farms, which produced all of the county’s cash crops (e.g.
sugar cane, tea, sisal, copra and cotton) in the 19803, have been completely privatized

under the IMF and World Bank- promoted Structural Adjustment Program.

130

According to Agriculture and Livestock Census 1999-2000, the average cultivated
are per household is 1.12 ha (Table 3). Gaza province had the highest average cultivated
area per household (2.0 ha) in 2000/01 cropping season, followed by Inhambane (1.6),
and Manica province (1.5 ha). However, Nampula province had the highest share of total

cultivated area (19.2%), followed by Zambezia (14.1%), and Gaza (12%).

Table A: Households, Cultivated Area and Average Cultivated Area in
2000/01 Cr0pping Season.

 

Households Cultivated % of Total Cultivated Area

 

 

Province Area Area per Household

(1000) (1000 ha) Cultivated (ha)

Gaza 228.3 460 11.9 2.01
Inhambane 259.6 414 10-7 1.59
Manica 201.9 307 7-9 1.52
Tete 268.0 396 10-2 1.48
Niassa 189.9 234 6-1 1.23
Cabo Delgado 336.5 376 9.7 1.12
Sofala 275.8 278 7.2 1.01
Nampula 794.4 741 19-2 0.93
Zambezia 726.3 546 14.1 0.75
Maputo 174.8 115 3.0 0.66
TOTAL 3,455.5 3,867 100 1.12

 

Source: INE (2002)

131

APPENDIX B

MICRONUTRIENT DEFICIENCY AND NUTRITION POLICY IN

MOZAMBIQUE

132

B.l Micronutrient Deficiency

Several studies conducted in Mozambique revealed that micronutrients deficiency
is a public health concern, especially among young children and woman at reproductive
age (MISAU, 1999; MISAU, 2003).

The recent Demographic and Health Survey (1997) indicated that approximately
36 % of young children (less than three years of age) suffered from chronic micronutrient
malnutrition. Furthermore, seven out of Mozambique’s 11 provinces exhibited levels of
chronic malnutrition that were greater than the average national: Cabo Delgado (57%),
Niassa (55%), Tete (46%), Manica (41%), Sofala (39), Nampula (38), and Zambezia
(37%). In addition, data collected through the Questionnaire on Welfare Basic Indicators
(QUIBB) showed that chronic micronutrient malnutrition among children increased to
44% in 2001, and that Zambezia province had the highest level of micronutrient
deficiency (INE, 2002).

Unfortunately, from the food security perspective, results of the QUIBB 2002
revealed that 38% of households felt that their economic status had deteriorated relative
to previous year, 35.1% reported no changes at all, and 26% felt that had improved.

Shrimpton R. (2002), identified three main causes of malnutrition in Mozambique
as follows:

0 Household food insecurityl

0 Limited availability of health care and basic sewage and

 

' “Food insecurity exists when the availability of nutritionally adequate and safe foods, or the ability to
acquire acceptable foods in socially acceptable ways, is limited or uncertain.”(United States of America
General Accounting Office, 1999:1-2).

“Food security means access by all people at all times to the food needed for a healthy life. Sustainable
food security aims to achieve this goal without compromising the productive capacity of natural
resources, the integrity of biological systems, or environmental quality.”(FAO and UNDP, 1994).

133

0 Limited availability and provision of health care to infants and pregnant

and lactating women.

Furthermore, a study of micronutrient deficiency carried out in 1998 indicated
that there is a high probability of VAD in Maputo, Gaza, Manica, and Cabo Delgado
provinces, due to a low level of consumption of both animal and vegetal enriched sources
of Vitamin A (Fidalgo et al, 1999; MISAU, 1999).

In early 2002, a nation-wide survey was carried out in 21 randomly selected
districts to determine the serum retinol levels in children 6 to 59 months and their
mothers. This survey revealed that 69% of children and 11% of their mothers suffered
from VAD, and approximately 14 % of children and 1% of their mothers evidenced
severe VAD. In addition, this study indicated a strong relationship between VAD and the
risk of Plasmoa'iumfalciparum infection (Odds Ratio = 3.19; CI 95%; 2.2 —4.5) (Ismael

et al, 2003).

B.2 Nutrition Policy & Strategy

Malnutrition is a major contributor to the alarmingly high infant and child
mortality in developing countries. Governments, international agencies, and research
institutions have increasingly recognized that malnutrition has multiple causes and
complex interrelationships with health, economic, environmental and social conditions
(Austin I. et al, 1981, Sommer A. et al, 1996). Thus, there is a need for greater and more
concerted nutrition intervention that channels additional resources more quickly and

effectively to nutritionally at-risk groups (Austin I. et al, 1981).

134

Designing and implementing a comprehensive national nutrition plan is a
complex and organizationally difficult task, since many institutions and disciplines are
involved. In some countries, there are strong nutrition planning units. However, it seems
more feasible to have individual governmental institutions incorporating nutritional
considerations into their activities rather than establishing a totally integrated national
nutrition plan with power concentrated in a single and separate nutrition planning unit
(Austin 1. et al, 1981). Seven types of nutrition programs are envisaged to mitigate
malnutrition in developing countries, namely:

1) Supplementary feeding

2) Nutrition education

3) Fortification2

4) Formulated foods

5) Consumer food price subsidies

6) Agricultural production, Technical change and Nutritional goals

7) Integrated nutrition and Primary Health Care programs.

These nutrition programs are viewed as component of a broader and more
comprehensive effort, aimed at adjusting basic socioeconomic infrastructures to remove
inequities and poverty. These interrelated types of nutrition programs are directed at
addressing the consequences of different causal factors at different stages, as depicted in
Figure 2. The interventions, which are indicated in dotted lines in Figure 2, should be

implemented simultaneously, due to their complementarities. For instance, the

 

2 Fortification is defined as being “the process whereby nutrients are added to foods to maintain or improve
the quality ofthe diet ofa group, a community or a population” (FAO/WHO, 1971).

135

introduction of formulated foods into the diet, requires an intervention to educate people
about their importance and /or how to prepare them. By the same token, formulated foods
could be used as commodities in supplementary feeding programs or they might be
subsidized in order to increase their accessibility to the lower—income segments of the
population (Austin I. et al, 1981).

The National Policy for Micronutrient Supplementation is a crucial component of
the more comprehensive National Strategy for Mitigation of Micronutrients Deficiencies,
including: (1) the promotion of production and consumption of micronutrients enriched
and fortified foods: (2) the promotion of appropriate child feeding practices (provision of
complementary foods to young child and duration of breastfeeding during the first two
years); and (3) the control of levels of infectious and parasitic worms diseases (MISAU,
2003; Martins, H., 2004).

Based on the recommendation of the World Health Organization, International
Vitamin A Consultative Group, and the United Nation Children’s Fund (1988), the
Ministry of Health of Mozambique adopted a Vitamin A supplementation program
envisaged to prevent and cure Vitamin A deficiencies in the short and medium run,

especially among young children and women.

136

.52 Na 3 ._. eumz< ”8.50m

     
   

cog—zoom

       

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

:osfimcem 530m >595 3:530
meotoomfi wEmEPSm coon voom
OHU<m A<mD<U
+ w
2:5on eoeaozwm memeoom momEmnsm meoom eouosnoi
uBEwES 835:2 Eco—coagm ootm A1 coca—agom cease—brow _w§_:otw<
ZO§Zm>MmHE

 

t
2922828 Alll 20552.55 Al, oEmmmooE All zofiosooma

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

mO<Hm

6:339:85 .855: Z we momma. 2.9.9.59 ween—a 3:28:23— ": 95»:—

137

The goal of the preventive supplementation program was to distribute Vitamin A
capsules through the existent health system network to all children 6 to 59 months every
six months and to distribute a single dose of 200,000 IU3 of Vitamin A to women within
two months after parturition. Children 6 to 1 1 months of age were targeted to receive a
dosage of 100,000 IU. and children over one year and under six years old, were to receive
a double dosage.

The curative Vitamin A supplementation for xerophthalmia in young children and
women follows the same dosage mentioned above, but it is administrated three times,
with the first dose given just after the disease is diagnosed, the second on the following
day, and the last dose a month later. In addition, the curative Vitamin A supplementation
for controlling measles in young children follows the same pattern of three doses--
100,000 IU to children six to eleven months and 200,000 IU to children over one year
and under six years old. Finally, children suffering from acute malnutrition, persistent
diarrhea, and pneumonia, are given a single Vitamin A supplement at the same dosage
mentioned previously for both children’s age groups (MISAU, 2002; MISAU, 2003).
Vitamin A supplementation to children age 6-59 months has been carried out and
achieved high coverage. However, in countries with clinical VAD, it can take up to 15-20
years to successfully replace a universal high-dose Vitamin A supplementation program
with a food fortification program and/or improved dietary practices (Underwood, B.
1998). The main micronutrient deficiencies addressed through fortification are: iodine
(which causes goiter, cretinism and deaf mutism), Vitamin A (causing sight impairment
or blindness), iron or folic acid (which causes anemia) and B vitamins (thiamin,

riboflavin and niacin). The major advantage of this nutritional intervention is that

 

3 1 lntemational Unit (1U) = 0.30 11g Retinol Equivalent (RE) = 0.60 11g B-carotene

138

additional nutrients (fortificants) are incorporated into existing diets without requiring
major changes in consumption behavior.

According to Underwood B., (2000), agriculture could play an important role in
developing more micronutrient-dense staple crops, while not neglecting continued
research to identify crops (varieties) qualitatively more balanced in their amino acid

content, including vegetables and legumes or livestock.

139

APPENDIX C

QUESTIONNAIRE ON THE ECONOMICS OF SMALLHOLDER

PRODUCTION AND ADOPTION OF OF SVS IN GAZA PROVINCE.

140

Respondent Number

NATIONAL INSTITUTE OF AGRICULTURE RESEARCH

DEPARTMENT OF AGRICULTURE ECONOMICS (MSU)

Questionnaire on the Economics of Smallholder Adoption and
Production of Orange-fleshed SweetPotato Varieties in Gaza Province of

Mozambique.

INSTRUCTION TO THE ENUMERATOR

First, ask who is regponsiblg for sweetpotato production and interview this

Qerson.

Please introduce yourself before starting to interview the farmer by reading
the consent of statement. If the farmer agrees to be interviewed begin the
interview. Please ask each question patiently until the farmer gets the point.
Please fill up the questionnaire accordingly. For open—ended questions, fill
the farmer ’8 answers and for close-ended questions, please indicate by

ticking or encircling.

141

Respondent Number

 

District: 1- Xai-Xai 2-Chokwe DIST:

 

Posto Administrativo:

Localidade: Date: / / 04

Enumerator’s Name:

 

I. Land and Livestock Inventory

1.1 In 2002/03 dry season, how many fields did you plant?

 

 

 

 

Field Location Tenure Area Mono/ Main crops planted

Number (up/low) (house mixed (list most important crop
plots) first)

1

2

3

 

Land tenure: 1 = Own 2 = Rent 3 = Family land 4 = Share cropping
5 = Borrowed land

1.2 In which fields did you plant sweetpotato? Field #3

 

142

Respondent Number

1.3 In 2003/04 wet season, how many fields did you plant?

 

 

 

 

Field Location Tenure Area Mono/ Main crops planted

Number (up/low) (house mixed (list most important
plots) crop first)

1

2

3

 

Land tenure: 1: Own 2 = Rent 3 = Family land
5 = Borrowed land

1.4 In which fields did you plant sweetpotato?

1.5 Do you currently have livestock?
IF YES: Go to the table below.

4 = Share cropping

Field #3

 

0=Ym O=N®

 

Type Cows Calves Oxen Donkeys

Pigs Goats Sheep

 

Number

 

II. SweetPotato Production

2.1 How did you prepare the fields where you planted sweetpotato in the:
0 2002/03 wet season? (l= manual 2= animal traction 3= tractor)
0 2003/04 wet season? (1= manual 2= animal traction 3= tractor)

2.2 In the 2002/03 wet season, did you hire labor for any sweetpotato operation?

IF YES: For what operations?

 

 

143

Respondent Number
2.3 In the 2002/03 dry season, did you hire labor for any sweetpotato operation?

IF YES: For what operations?

 

2.4 In the 2003/04 wet season, did you hire labor for any sweetpotato operation?
(0 = No 1 = Yes)

IF YES: For what operations?

 

 

2.5.1 Which sweetpotato varieties did you grow in the 2002/03 and the 2003/04 wet

 

 

 

 

 

 

 

 

 

 

 

seasons?
Year/ Name Type Leaf Growth Location Production
Season shape habit planted problems
(up/low)
2002/03
Wet
Season
2002/03
Dry
Season
2003/04
Wet
Season
Type: Flesh color: 0 = White/cream flesh 1 = Orange flesh
Leaf shape: 0 = round 1 = narrow
Growth habit: 0 = prostrate 1 = erect

144

Respondent Number

2.6 In what year/season did you first planted OFSP vines?
Year: Season

 

2.7 Did you plant OFSP varieties in the 2002/03 wet season?
(0= N0 = Yes)
IF NO: Why not?

 

IF YES:

How large of an area did you plan to sweetpotato?
(house plot comparison)

How much area did you plant to OF SP varieties, compared to WFSP varieties?

o More area to WFSP varieties (circle 1]) Yes Relative proportion %
0 Same areas to WFSP varieties as WFSP varieties (circle if) Yes
0 More area to OFSP varieties (circle if) Yes Relative proportion %

Why did you plant this proportion?

 

2.8 Did you plant OF SP varieties in the 2002/03 dry season?
(0= N0 = Yes)
IF NO: Why not?

 

IF YES:

How large of an area did you plan to sweetpotato?
(house plot comparison)

How much area did you plant to OF SP varieties, compared to WFSP varieties?

o More area to WF SP varieties (circle if) Yes Relative Proportion %
0 Same areas to WFSP varieties as WFSP varieties (circle if) Yes
0 More area to OF SP varieties (circle if) Yes Relative proportion %

Why did you plant this proportion?

 

145

 

Respondent Number

2.9 Did you plant OFSP varieties in the 2003/04 wet season?
(0 = No 1 = Yes)
IF NO: Why not?

 

IF YES:
How much area did you plan to sweetpotato?
(house plot comparison)

How much area did you plant to OFSP varieties, compared to WFSP varieties?

o More area to WFSP varieties (circle if) Yes Relative proportion %
0 Same areas to WFSP varieties as WFSP varieties (circle if) Yes
0 More area to OFSP varieties (circle if) Yes Relative proportion %

Why did you plant this proportion?

 

2.10 Do you plant your sweetpotato in mounds, ridges, or on flat ground?

(1=mounds 2=ridges 3=flat land)__

IF NOT RIDGES: Why not?

 

 

2.11 Are there any differences in the way you plant OF SP varieties, compared to

WF SP varieties? (0 = No 1 = Yes)

IF YES: What are the differences?

 

146

 

 

3625 :55

 

 

 

 

 

 

 

>33 see 22.. ea: 2 >320 :6 a;
>33 as: 22.. 2:: as. e >33 no 33.332424:
>mmn§9 :2: 0.8.: 083 32 mm >mmmo .«o 3:39.82. :95
>mmn§9 :2: b3:— oacn: 2:3 32 mm >mmmo mo 5%
>mmm3 :2: .3893 2:3 E23» 9.2: 2 >mmm0 m0 323. .3 23h.
>mmm>9 :2: .3qu 2:3 «093 39. mm ~3me mo 03m 33—
:23 55 twee. 25.... 5:3... e >33 :6 3
Pawn? 55 .5393 2:3 .522? mm >mmmo mo SEE: 3.3—8% 25>
>mmm3 :2: .5833 2:3 893 32 mm >mmmO mo :22: 25>
«33¢. 32 28.. 69:2— ..e 22¢.
>mmm>> :2: .5893 2:3 333 32 m: >mmm0 do any... .3 22;.
>mmm>> :2: 3.83 2:3 .539: a >mmm0 mo 3
>mmn§> :2: Luuzfi 2:3 22.83 m_ >mmm0 .«o .5333. no.5
>33 55 =ng 25... are. 3 >333 “aisles
Beacons? 3:32.832: as;

 

39.6 33.5

3832.5.» 5mm? 3 €22.83 3.2.2:: awn—O 22:56 :9» 3. 3:: «id

147

 

3625 3:5

 

 

 

 

 

 

 

148

>33» 55 22: «a: a >330 :0 g
>mmn§> :2: 9:2: 2:3 3:— mm >mmmO :0 32339.28;
>mmn§> :2: 2::— 023 30. mm >mmm0 mo 3:322-..— :95
>mmm>> :2: b3:— o..::_ «E3 32 mm ~3me mo 3%
>mmn§> :2: :2023 983 E23» 9.2: 2 >mmmO :0 32.3.
>mmn§> :2: .535 2:3 323 3:. mm >mmmO mo 02m «:3—
>33 5.: .522 as: .233 : >38 :0 3
>mmn§> :2: :2093 :23 :2::3 mm >mmmo :0 52:3: 3.3.53 28>
>mmm3 :2: :2333 2:3 893 32 fl >mmm0 a: 523:: 25>
89.3 32 2:9. 3:2. a: 22¢.
>mmn§> :2: :2023 2:3 823 32 fl >mmm0 mo 22: :9: .2: 22?
>mmhr$ :2: 3:95 2:3 :28: 2 >mmm0? :cznwnmch: ma
>25; :2: 32:2 :83 :2::3 mm >mmmo mo a
>3; 55 .233 2:: :32 : >35 :0 3
3309223 :oamouéwfizO 33,—.

 

AQ—Ohmu OmNQT—v thmHOmhfl> mmha Ow fiuhfln—EOU £030MLG> gumbo Oua=_fl>0 Ba?» 63 Bay: "£N~.N

Respondent Number

2.13 IF THE F ARMER GREW MORE THAN ONE OFSP variety:

Ask which one was the best in terms of:

 

Name of OFSPV First Second

 

Higher total yield

 

Shorter crop duration

 

Preferred root size (Circle): small or (Circle): small or
big roots? big roots?

 

More root consistency

IFS——

 

Better taste of fresh roots

 

Better taste of boiled roots

 

Shorter coming time

 

Better taste of leaves

 

Other (specify)

 

2.14 Who decides whether or not to grow a OFSP variety?

 

(1 = male 2 = female 3 = decide together)

149

Respondent Number

2.15 How did you obtain your OFSP vines in each season that you planted OFSPs?

 

 

 

Year Season Variety Name Source
2000/01 WS
DS

 

2001/2002 WS

 

 

 

 

 

DS
2002/03 WS
DS
2003/04 WS
ws= Wet Season DS= Dry Season

Sources

1 = Received from Agriculture Office 2 = Received from INIA/SARRNET 3 = Purchased
4 = Received from NGOs 5 = Retained from own harvest 6 = Received from relatives

7 = Received from neighbors 8 = Village multiplication plot

2.16 IF THE FARMER RECEIVED VIN ES MORE THAN ONCE FROM A NGO:
Why?

 

 

Why didn’t he/she save OFSP planting material?

 

 

2.17 Did you sell any sweetpotato roots from your 2002/03 or your 2003/04 wet season
crop? (O=No l = Yes)

150

Respondent Number

 

IF YES:Ask
Name of Variety Type Amount Sold
OF W
( or F) Quantity Unit

 

 

 

 

IF NO SALES: Why not?

 

2.18 If you wanted to sell sweetpotato roots, where/how would you sell them?

1 = Take to the District market and sell to consumers
2 = Take to District market and sell to market vendors
3 = Sell to traders who come to this village

4 = Other (specify)

Selling Method:

2.19 Have you ever sold or given OF SP vines to anyone? (0 = No 1= Yes)

IF YES: Go to the table below

Year Season To whom Person’s village Quantity Given or Sold

 

 

2.20 Why did she/he/they want OF SP vines?

 

 

151

Respondent Number

111. Sweet Potato Consumption

3.1 How many months during the year does your family eat/consume sweetpotato
roots?

 

3.2 When sweetpotato roots are available, how many times a week do you eat them?

3.3 How do you prepare/eat sweetpotato roots for adults?

Most common methods:

 

 

Other ways:

 

3.4 Have you heard of preparing food from sweetpotato, such as:

 

Product Have you heard of it (No/Yes) Have you made it (N o/Y es)

 

Juice

 

Cake

 

Flour

 

 

3.5 When sweetpotato roots are available, how many times a week do you feed them
to your children?

 

152

Respondent Number
3.6 How do you prepare sweetpotato roots for young children to eat?

Most common methods:

 

 

Other ways:

 

3.7 What type of sweetpotato varieties do you feed your children? 1
(0=WF 1 = OF 2 = both)

3.8 Have you heard that OFSP varieties are nutritionally better for I
women/children, than WFSP varieties? (O = N0 1 =Yes)

IF YES:
From whom did you hear this information?

 

Why are they nutritionally better?

 

 

3.9 Do you eat the leaves of OFSP varieties (O = NO 1 =Yes)

IF NO: Why not?

 

IV. Institutional Factors

4.1 During the past m years, have you attended an agricultural field day or an on-

 

farm demonstration (0 =No l =Yes)
IF YES:
Who sponsored the event What did you learn

 

 

 

153

Respondent Number

4.2 During the past _tw_o years, how many times has an extension technician assisted
you?

IF AT LEAST ONCEI Ask:

Agency of the technician What type of assistance did he/she provide

 

 

 

 

 

4.3 During the past two years, have you participated in an agricultural training

 

course? (0 =No 1 =Yes)
IF YES:
Who was the sponsor What did you learn

 

 

 

4.4 Are you or your spouse, a member of any association? (0: No 1=Yes)

IF YES:

 

Name of the Association How does the association benefit you or your spouse?

 

 

 

154

Respondent Number

V. Farmer Characteristics

5.1 Respondent=s gender (0 = Male, 1= Female)

5.2 How old are you?
5.3 How many years of school have you completed?
5.4 How many years have lived in this village?

5.5 How many years have you grown sweetpotato?

VI. Household Members

6.1 How many people live in your household at least 9 months of the year?
6.2 How many adults (18 or older) live in your household?

6.3 How many adult members of your family live away from home?

6.4 How many children do you have?

6.5 What are the ages of your children?

 

 

 

 

Age Age Age Age Age Age
6.6 In 2003, did you work off farm? (1 =Yes O =No)
6.7 In 2003, did your souse work off farm? (1 =Yes 0 =No)

155

Respondent Number
VII. Comments/Suggestions

7.1 Do you have any comments/suggestions about the OFSP varieties that you
have grown?

 

 

7.2 What are the main problems that limit your potato production?

 

 

7.3 Do you have any suggestion about what needs to be done to help farmers
increase their sweetpotato production?

 

 

Thank you very much for your participation.

156

APPENDIX D

ADOPTION (%) OF OFSVS, FREQUENCY OF VINES DISTRIBUTION AND

AWARENESS OF PROCESSED OFSVS PRODUCTS BY VILLAGES IN

CHOKWE AND XAI-XAI DISTRICTS.

157

Sea .825 HooSom

 

8:5 253 u >3 56:6 as 23 n 05

822% E3. 233 "ea 25.26 359x 8:5 u omo 8:808“ 233 £553 u “:5

:Ommom

SW 98 OS 3 3 3:. 3:. 3m 98 a? 2: a; 588
COmflOm

3m 35 98 S 3 3m 3:. o 3.... 2m Ex 5 $88
COWMOm

2: Se 3” on 2:. S» S” 2... Se 2: 2: as 388

 

 

2:3 0:50; 9580
on vm mu ~o> 22m 36E 2.3252 ococsom—z omen—com £35322 2.59:5 3:50 >mmm0
cease/x an>E=U
58. Ohm >3 mm? Umb n5: 223.ch
no.2. @2300 002
«53-83 633533: 53.55 2633" 1.3 252.0
a naueao 5oz 32am: 3 3.28 83:3 a 9.5%: 832$ 332583. assuage co ex; 8:33.. 3-: «35.

158

058=qo< 8 2 "<2
000_>._0m o0=0m 2.83 MM?

53> 253 u >3
0.5—2.0 Ecomwom mgr—«O n OMO

voom >0>Sm ”028m

coho—EU 05 0>mm u Ohm
coup—03m 2.83 5305:.— " m3.—

 

 

msm mdn oo_ odv o o o o odv mem— hdm N
32003.
>mmm0
6Q. ode o odo oo_ oo_ oo_ oo_ odo mom 0mm _ no .0952

fimm o ndm o o o o o o oo_ oo_ m
5:35.05
od o odv o o o o o odv <2 <2 N 85>
>mmm0oo
5532.;

mdo oo_ Wm». oo_ oo. oo_ oo_ oo~ ode <2 <2 _

2:3 08:05 0580
0d «N no ~o> «Em .02» 0:39:02 0:02:82 coonEom «383:2 2.59:5 35.6
.88. 033:;
Ohm >3 mm? OzO m3;
m0c< @2025 002

 

.aaazaanaz ._ax-_ax E: 05.20 6oz E 8.35 «35> 3 5.30 a? 2:83
a €82 as. 3333. 22.3.3: and seat; a: .3252 E... 5.5332: 85> Emma a. 32...»: ad 2...;

159

6030.285 0:03 03:? :2. 30:5,. m. 0:0 60.00.00 0:03 002 :80 3 00.030 m0w0=_> 03 h a

voom >0>Bm “0050m

82> 233 u >3 5.220 20 25m u 05
82200 00:3. 253 ":3 23.26 .2200: £50 u 0:0 5:203: 233 52053 u :3:

uOSUOhnm UOmmOUOum

 

oé o md. o o o hd o o M... :0 >mm...0 0:0 .000. .0
0.00.). 0:020:00”.

08:00:.

odh odo hdm hdv md . oo. oo. odv mdm oo. oo. 0030095 >mm...0
..o 033<

m. 0:005:03.

 

2:3. 0:50.". obmaO
0: cm 0: No> 0.0.m .02”. 0:38:02 05:30.2 owonEom 0.:E:~:$. 0:_=w::n. BEsO
.Soh 030E>

Ohm >3 mm? OMO ".34
.00..< :0:0>oO 002

01.5533). £00.30:— .aX-.aX 6:: 0.5.0.5 .::0a0m 00>? vofioon
a. 002 .3 @2025 0::N seam 5 00.00.05» .520 a: mosamuagh w:.00009.m a: 00.002.— 15 000:0..ak< a: «.054 "MA— 030,—.

160

APPENDIX E

CORRELATION MATRIX OF VARIABLES IN THE PROBIT MODEL

161

 

 

 

mde dived- mom—d- wmmmd mvvmd mi ~d Sm _ d- vwm d d 2.x d d- meg d anNd and:
23d 3m _d mumdd 33d wde wddmd mmodd omddd nmmdd Emdd- 32d 989a
wvmvd- dedd- 33d mowed- 38d- Snood- Swmd :hdd- 38d domed 33d- EOE
3 8d- mandd- Nde mmcdd 53d- 83d Rodd fifld wvwdd duodd- dhddd- 25808
ddmmd- 3m _d nvwmd mmdmd- 33d- mmmdd thmd mnmdd- gdmd m _ 8d- Edmd- 029me
moddd SS d- vmmmd omwmd dandd mm _ Nd mvw _ d- E. d dd- 3 _ Nd- 33 d Knmd 0838a
mdomd- wmomd owodd- 32d Smfio omddd dowdd vwmdd- a? _d good 32d 83%
thdd odmdd 33d vo_~d «vomd d2 dd 38.0 mwdmd ddddd dedd- dommd 368$
82: o d m _ d- caved dam _ d thdd vmwod mamvd- Non ~ d 0mm _ d- mm Rd- wdhdd £95m
ddddd cdmdd- Smdd ommdd- Smdd mmmmd dmwdd omfid nmwdd- Smdd 83cc
ddddd adw _ d m d mod- 5mm d d d? _ d Rudd 35d mmmdd mow _ d 985»
dddd._ Rmmd owned nod _ d- m _ mod ”3 _d wooed- 30m d 88$
82: moved doc _ d- 3d _ d m Sod- mmmdd- m Edd >9:
dddd._ vm 8d mmmod Seed 35d 3mmd 0808
dddd._ SE d- a _ dmd add d d Ndvdd- £33
88._ 38¢ ammo- 223 3.5%
code; w mm m d- dded. .60
88; $8.? own
ddddd >

0225 Santa 985$ 02:3 >08 0803 33x0 395% :3 am“ >
.352 .595 2: fi 833...; .3 5.5:: 5:29.39 "H 033—.

162

 

 

 

884 385 3:3 $85 ~83- Rad 88.0 983 as:
884 M230- M: :3 ”Rod- #85 8mg so; 325
88; $55 325 MES- 88d- $8.? 20a
88; ES $8.? 285 £86 £888
88; Emma- 38.? ~23- 029%
88; $85 523 Egg
884 S85 83%
88; 8E3
was: 988m 202 30308 02966 oSBEQ 33% 368$
.8288 a 2...;

163

REFERENCES

Adesina A.A., and Sanders J.H., 1991. Peasant Farmer Behavior and Cereal
Technologies: Stochastic Programming Analysis in Niger. Agricultural Economics Vol.
5(1): 21 —38.

Adesina A.A., and Moses M. Z., 1993. Technology Characteristics, Farmer's Perceptions
and Adoption Decisions: A Tobit Model Application in Sierra Leone. Agricultural
Economics Vol. 9(4): 297-311.

Akter U.A., and Rajah KS, 1992. Effects of Irrigation-Induced Technological Change in
Bangladesh Rice Production. American Journal of Agricultural Economics Vol. 74(1):
144-15 7.

Amemiya, T., 1981. Qualitative Response Models: A Survey. Journal of Economic
Literature 19(4): 483-536.

Andrade M. I., and Ricardo J., 1999. Results of First Round Provincial Trials on the
Evaluation of Nineteen Orange-Fleshed Sweetpotato Clones Across Fourteen Different
Environments of Mozambique. National Institute of Agriculture Research (INIA) and
Southern Africa Crops Research Network (SARRN ET). Maputo, Mozambique.

Andrade M. I., Ricardo J. and Gani A., 2002. “Combating Vitamin A Deficiency in Rural
Mozambique with Orange-Fleshed Sweet Potato: Results of a Survey on Nutritional
Status and Two Round Provincial Trials on the Evaluation of 19 Orange-Fleshed Sweet
Potato Across 21 Different Environments Over 8 Provinces in Two Seasons.

IN IA/SARRN ET, Maputo, Mozambique.

Austin E. J. and Zeitlin F. A., 1981. Nutrition Interventions: A Conceptual Framework. In
Nutrition Intervention in Developing Countries. An Overview. Harvard Institute of
lntemational Development. Oelgeschlager, Gunn & Hain, Publishers , Inc. Cambridge,
Massachusetts.

Bleeker, P., 1983. Soils of Papua New Guinea. CSIRO and ANU Press. Carnberra.

164

Bloem. M.W., Huq N., Gorstein J ., Burger S., Kahn T., Islam N., Baker S, Davidson F.
1996. Production of Fruits and Vegetables at the Homestead is an Important Source of
Vitamin A Among Women in Rural Bangladesh. European Journal of Clinical Nutrition
1996; 5 0(Suppl 3): 62-6 7.

Bloem M.W., de Pee S., and Damton—Hill I., 1998. New Issues in Developing Effective
Approaches for the Prevention and Control of Vitamin A Deficiency. Food and Nutrition
Bulletin, Vol. 19 (2), 1998, The United Nations University.

Bowen, M.L., 1993. Socialist Transitions: Policy Reforms and Peasant Producers in
Mozambique, Ch. 12 in Bassett, Thomas J. and Donald E. Crummey (eds.), Land in
African Agrarian Systems, University of Wisconsin Press, Madison, Wisconsin.

Bouwkamp, J. C. 1985. Sweet Potato Products: A Natural Resource for the Tropics. CRS
Press, Inc., Florida.

Bouwkamp, J. C., 1985. Processing of Sweet Potatoes - Canning, Freezing, Dehydrating.
In Sweet Potato Products: A Natural Resource for the Tropics. CRS Press, Inc., Florida.

Capps, 0. Jr. and Schmitz, D. J ., 1991. A Recognition of Health and Nutritional Factors
in Food Demand Analysis. Western Journal of Agricultural Economics 16 (I): 21-35.

Caswell A. J ., 1995. Valuing Food Safety and Nutrition. Westview Press, Inc. Boulder,
Colorado.

Catignali G. L., and Bieri J. G., 1983. Simultaneous Determination of Retinol and
Alpha—Tocopherol in Serum or Plasma by Liquid Chromatography. Clinical Chemistry
1 983; 29: 708- 712.

Chug-Ahuja J .K., Holden J .M., Forman M.R. et al., 1993. The Development and
Application of a Carotenoid Database for Fruits, Vegetables, and Selected Multi-
Component Foods. Journal of A merican Diet Association 1993; 93 :3 1 8-3 23

CIP/CIAT/lFPRI/IITA and IPGRI. 2000. Roots and Tubers in the Global Food System: A
Vision Statement to the Year 2020. Report to the Technical Advisory Committee (TAC)
of the Consultative Group on lntemational Agricultural Research (CGIAR) by the
Committee on Inter-Centre Root and Tuber Crops Research (CICRTCR), Lima, Peru.

165

 

Cockerham, L. K., and Hand, E. T., 1921. The Sweet Potato: A Handbook for the
Practical Grower. The Macmillan Company, New York.

Coelho A. e Barreiros J ., 1994. Praticas Culturais para Transferencia ao Sector Familiar
do Sul de Mocambique na Campanha 1994. Instituto Nacional de Investigacao
Agronomica, Maputo, Mocambique.

Collins, W. W. and Walter, M. W., Jr. 1995. Fresh Roots for Human Consumption. In
Sweet Potato Products: A Natural Resource for the Tropics. Bouwkamp, C. J ., Editor.
CRC Press, Inc., Boca Raton, Florida.

Collinson, M. P., 1972. Farm Management in Peasant Agriculture: A Handbook for Rural
Development and Planning in Africa. Praeger, New York.

Huysman M., and Wulf V., 2004. Social Capital and Information Technology. The MIT
Press, Cambridge, Massachusetts, London, England.

Davidson, S. S.; Passmore, R.; Brock, J. F ., and Truswell, A. S. 1979. Human Nutrition
and Dietetics. Seven Edition. Churchill Livingstone Inc. New York.

de Pee S., West C. E., Muhilal K. D., Hautvast J. G., 1995. Lack of Improvement in
Vitamin A Status with Increased Consumption of Dark-Green Leafy Vegetables. Lancet
1995; 346:75-81.

de Pee S., West C. E., 1996. Dietary Carotenoids and their Role in Combating Vitamin A
Deficiency: A Review of the Literature. European Journal of Clinical Nutrition 1996; 50
Suppl 3:838—53.

De Pee S. West C.E, Perrnaesih D., Martuti S., and Hautvast J., 1998. Orange Fruit is
More Effective than are Dark-Green Leafy Vegetables in Increasing Serum

Concentrations of Retinol and B-Carotene in Schoolchildren in Indonesia. American
Journal of Clinical Nutrition 1998; 68(5): 1 058-1 06 7.

Dent, J.B., Harrison, S.R., Woodford, K.B., 1986. Farm Planning with Linear
Programming: Concept & Practice. Butterworths Pty Limited.

166

Dinar, A. and D. Yaron, 1990. Influence of Quality and Scarcity of Inputs on the
Adoption of Modern Irrigation Technologies. Western Journal of Agricultural Economics
15(2): 224-233.

District Development Profiles. 1997. http://www.unsvstemmoz.org/, August 2004.

Doss, C. R., 2003. Understanding Farm Level Technology Adoption: Lessons Learned
from CIMMYT’s Micro Surveys in Eastern Africa. CIMMYT’s Economics Working
Paper 03-07. Mexico, D. F .: CIMMYT.

Dykshoom, J .A., Marques, M.R., Semo, G., Gomes, F., and Brito, R. 1988. Estudo Pedo-
Hidrologico do Vale do Infulene - Parte'Sul da Via Rapida da Moamba. Parte A: Solos;
Parte B: Agrohidrologia. Série Terra e Agua do INIA, Comunicacao N° 58.

Espeut D., Fidalgo M.L., Ismael C.H., Johnson R. and Mukuria A. 2001. Estado
Nutricional e de Saude das Criancas e Maes em Mocambique. Conclusoes do Inquerito
Demografico e de Saude em Mocambique em 1997. Macro lntemational, Calverton,
Maryland.

Espinola N., Creed-Kanashiro H., Ugaz M., van Hal M., Scott G. 1999. Development of a
Sweetpotato-Based Instant Weaning Food for Poorly Nourished Children Six Months to
Three Years Old. In Impact on a Changing World: Program Report 1997-98.
lntemational Potato Center, Lima, Peru.

Faber M., Phungula M., Venter S., Dhansay M., and Benade A. S., 2002. Home Gardens
Focusing on the Production of Yellow and Dark-Green Leafy Vegetables Increase the

Serum Retinol Concentrations of 2-5-Year-old Children in South Africa. American
Journal of Clinical Nutrition 76 (5): 1048-1054.

F AO/WHO Expert Group, 1967. Requirements of Vitamin A, Thiamine, Riboflavine and
Niacin. World Health Organization Technical Report Series, No. 362. Geneva.

FAO, 1986. Some Medicinal Forests Plants of Africa and Latin America. FAO Forestry
Paper 67. FAQ, Rome.

167

FAQ/UNDP, 1994. Joint Declaration Signed by James Gustave Speth (UNDP
Administrator) and Jacques Diouf (FAO Director General), September 21th, 1994. Rome,
Italy: FAO.

F eder G., and Slade R., 1984. The Acquisition of Information and the Adoption of the
Technology. American Agricultural Economics Association Vol. 66(3): 312-320.

F eder, G; Richard E. J. and Zielbermam D. 1985. Adoption of Agricultural Innovations in
Developing Countries: A Survey.” Economic Development and Cultural Change, Vol.
33(2): 255-298.

Fidalgo L. Ismael C., and Khan S.G., 1999. Relatorio da Avaliacao de Deficiencia em
Micronutrientes a Nivel da Provincias de Cabo Delgado, Manica, Gaza e Maputo.
Ministerio da Saude- Reparticao de Nutricao, Maputo, Mozambique.

Fisher, A. Chestnut, G. L. and Violette M. D.1989.”The Value of Reducing Risks of
Death: A Note on New Evidence’. Journal of Policy Analysis and Management Vol. 8
(1): 88-100.

Gemo, H. and Rivera W., 2001. Mozambique’s Move Towards a Pluralistic National
System of Rural Extension. Agriculture Research and Extension Network, Network Paper

No. 1 10, January 2001. http://www. odi. org. uk/agren/papers/agrenpapeLI 1 0. pd}:
September 2004.

Goldman H. R. and Overholt C. A. 1981. Agricultural Production, Technical Change, and
Nutritional Goals. In Nutrition Intervention in Developing Countries: An Overview.

Harvard. Austin J. and Zeitlin M. Editors. Harvard Institute for lntemational
Development. Oelgeschlager, Gunn & Hain, Publishers, Inc. Cambridge, Massachusetts.

GOM, 1995. Agricultural Policy and Strategy for Implementation (PAEI). Maputo,
Mozambique.

GOM 1996b. Rural Poverty Profile. Ministry of Planning and Finance. Poverty
Alleviation Unit, April 1996, Mozambique.

168

GOM, 2001. Action Plan for the Reduction of Absolute Poverty (2001-2005)- PARPA:
Strategic Document for the Reduction of Poverty and Promotion of Economic Growth.
Final Version Approved by the Council of Ministers, Maputo, Mozambique.

GOM, 2002. Mozambique Famine Early Warning System Network. 18 March 2002.

GOM, 2003. Emergency Vulnerability Final Report. Mozambique Vulnerability
Assessment Committee. Mozambique.

GOM, 2004. Criacao do Instituto de Investigacao Agraria de Mocambique (IIAM):
Fundamentacao. Republica de Mocambique, Ministerio da Agricultura e
Desenvolvimento Rural, Maputo, Mozambique.

Gomes F ., Mafalacusser J ., Marques M. and Brito R., 1998. Mozambique Country Paper:
Wetlands for Agricultural Development. In Wetland Characterization and Classification
for Sustainable Agricultural Development. Food and Agriculture Organization of the
United Nations (FAO), Sub-Regional Office for East and Southern Africa (SAF R),
Harare.

Hagenimana V., K’osambo L., and Carey E., 1999. Potential of Sweetpotato in Reducing
Vitamin A Deficiency in Africa. In Impact on a Changing World: Program Report 1997-
98. lntemational Potato Center, Lima Peru.

Hagenimana V. and Oyunga M., 1999. The Effect of Women Farmers’ Adoption of
Orange-Fleshed Sweet Potatoes: Raising Vitamin A Intake in Kenya. lntemational Center
for Research on Women (ICRW/OMNI Research Program), Research Report Series No
3. HarvestPlus, 2003.” Breeding Crops for Better Nutrition”.
http://Ww.hmestplus.org/sweetpotatohtml, February 2004.

Halbrendt, C. ,Stertling L. ,Snider S., and Santoro G., 1995. Contingent Valuation of
Consumer’s Willingness to Purchase Pork with Lower Saturated Fat. In Caswell A. J.
(Ed): “Valuing Food Safety and Nutrition”. Westview Press, Inc. Boulder, Colorado.

Hazell, P. and Haddad L., 2001. Agricultural Research and Poverty Reduction. Food,
Agriculture, and Environment Discussion Paper 34, International Food Policy Research
Institute, Washington , DC, USA.

169

 

Hermele, K. 1988. Land Struggles and Social Differentiation in Southern Mozambique:
A Case Study of Chokwe, Limpopo 1950-1987”, Scandinavian Institute for Afiican
Studies, Uppsala.

Hutchings, J. B. 1994. Food Colour and Appearance. Blackie Academic & Professional,
Wester Cleddens Road, Glasgow, UK.

INE/MADER, 2002. Censo Agro-Pecuario 1999-2000: Apresentacao Sumaria dos
Resultados: Quadros e Graficos. Instituto Nacional de Estatistica (INE), Ministerio da
Agricultura e Desenvolvimento Rural (MADER), Maputo, Mocambique.

INE/MADER, 2002. A Thematic Analysis of the Agricultural and Livestock Census in
Mozambique, 1999-2000: Land Tenure and Use. Instituto Nacional de Estatistica (INE),
Ministerio da Agricultura e Desenvolvimento Rural (MADER), Maputo, Mocambique.

INE/MADER, 2002. Analise Basica do Censo Agro-Pecuario (CAP) de Mocambique,
1999-2000: Producao De Culturas Alimentares Basicas. Instituto Nacional de Estatistica
(INE), Ministerio da Agricultura e Desenvolvimento Rural (MADER), Maputo,
Mocambique.

INE, 2002. Anuario Estatistico 2002-Mocambique. Instituto Nacional de Estatistica,
Maputo, Mocambique.

INIA/SARRN ET. Novas Variedades de Batata Doce. Maputo, Mozambique

INIA-IITA/ SARRN ET, 2003.Terminal Report on the Accelerated Multiplication and
Distribution of Healthy Planting Materials of the Best High Yielding Varieties of Cassava
and Sweetpotato in Mozambique: A Drought and Flood Mitigation Activity. lntemational
Institute of Tropical Agriculture, Ibadan, Nigeria.

Instituto Nacional de Estatistica, 1999. Censo 97: II Recenseamento Geral da Populacao e
Habitacao. Resultados Definitivos.

Instituto Nacional de Estatistica, 2002. Questionario de Indicadores Basicos de Bem-
Estar (QUIBB, 2002): Relatorio Final. INE, Maputo, Mozambique.

170

 

Instituto Nacional de Estatistica, 2004. Producao 1996-2002 a Precos Constantes de
1996. www.ine.gov.mz, February, 2004.

lntemational Potato Center, 2003. Fueling Growth, Health, and Prosperity. Annual
Report 2002. Lima, Peru.

Ismael C., Khan, S.G., Thompson R., Meershoek S., and van Streirteghem, V., 2003.
Inquerito Nacional sobre a Deficiencia de Vitarnina A e Prevalencia de Anemia e Malaria
em Criancas dos 6-59 Meses e Respectivas Maes- Versao Preliminar. Ministerio de
Saude- Reparticao de Nutricao, Instituto Nacional de Saude, Helen Keller lntemational e
UNICEF, Maputo, Mozambique, Outubro 2003.

Jamison, J. R., 2003: Clinical Guide to Nutrition & Dietary Supplements in Disease
Management. Churchill Livingstone.

Jones, A. 1977. Heritabilities of Seven Sweet Potato Root Traits. Journal of American
Society for Horticultural Science 102 (4): 440-442.

Johnson-Welch C., 1999. Focusing on Women Works: Research on Improving
Micronutrient Status Through F ood-Based Interventions. Opportunities for Micronutrient
Intervention (OMNI) Research, Field Support and Research, Office of Health and
Nutrition.

Kosarek, L. J ., Garcia P. and Morris L. M., 2001. Factors Explaining the Diffusion of
Hybrid Maize in Latin America and the Caribbean Region. Agricultural Economics Vol.
26(3): 26 7-280.

Kotler, P., Zaltman, G. 1971. Social Marketing: An Approach to Planned Social Change.
Journal of Marketing Vol. 35(3): 3-12.

Kristjanson, P. M., 1987. The Role of Information and Flexibility in Small Farm
Decision-making and Risk Management: Evidence from West African Semi-Arid
Tropics. University of Wisconsin, Madison, WI.

Lathan C. Michael, 1979: Human Nutrition in Tropical Africa. A Textbook for Health
Workers with Special Reference to Community Health Problems in East Africa. FAO,
UNICEF and WHO, Rome.

171

 

 

Lin, S. S., Creighton P. C. Chen, D. and Lo, H. 1992. Sweet Potato Production and
Utilization in Asia and the Pacific. In Sweet Potato Products: A Natural Resource for the
Tropics. Editor Bouwkamp, J. C. CRC Press, Inc., Boca Raton, Florida.

Low J ., Walker T. and Hijmans R., 2001. The Potential Impact of Orange-fleshed Sweet
Potatoes on Vitamin A Intake in Sub-Saharan Africa”. Centro Intemacional de la Papa.
Paper Presented at a Regional Workshop on Food-based Approaches to Human
Nutritional Deficiencies. The VITAA Project, Vitamin A and Orange-Fleshed Sweet
Potatoes in Sub-Saharan Africa. Nairobi, Kenya.

http://www.cipotato.oggjvitaa/paper vitaamdf” May 2004.

Mabota A., Arlindo P., Paulo A. and Donovan C, 2003. Market Information. A Low Cost
Tool for Agricultural Market Development? Flash No 37e. Research Results from SIMA-
DEST and Department of Policy Analysis. MARD-Directorate of Economics. Maputo

Mozambique. http://www.aec.msu.edu/agecon/st/mozambique/flash37e.pdf., July 2004.

Maddala, G. S. 1983. Limited-Dependent and Qualitative Variables in Econometrics.
Econometric Society Monographs No 3. Cambridge University Press. Cambridge. UK.

Maredia K. M. and Minde J .I., 2002. Technology Profitability and Agricultural
Transformation: Concepts, Evidence, and Policy Implications. In Perspectives on

Agricultural Transformation: A View from Afiica. Nova Science Publishers, Inc. New
York.

Martins H., 2004. Proposta de Componente Estrategica dum Plano de Desenvolvimento
Nutricional em Mocambique- Relatorio de Consultoria. Ministerio da Saude, Helen
Keller lntemational, Maputo, Mocambique.

Mbata N. J. 2001. Determinants of Animal Traction Adoption in Traditional Agriculture:

An Application of the Multivariate Probit Procedure to the Case of Lesotho. Development
Southern Africa Vol. 18 (3): 309-325.

Mejia L.A., Chew F ., 1988. Hematological Effect of Supplementing Anemic Children
with Vitamin A Alone and in Combination with Iron. American Journal of Clinical
Nutrition 1998; 48:595-600.

Ministry of Agriculture and Fisheries, 1996. Agricultural Research Strategy for
Mozambique. May 1996, Maputo, Mozambique.

172

 

Ministerio da Agricultura e Pescas, 1996. Zonas Agro-Ecologicas e Sistemas de
Producao. Programa de Investimento em Extensao Agraria. Documento de Trabalho N°
213, Maputo, Mocambique.

Ministerio de Saude (MISAU), 1999. Avaliacao das Deficiencias em Micro-Nutrientes a
Nivel das Provincias de Maputo, Gaza, Manica e Cabo Delgado. Reparticao de Nutricao,
Maputo, Mocambique.

Ministerio da Saude (MISAU), 2002. Programa de Suplementacao com Vitamin A na
Rotina (Consulta da Crianca Sadia). Direccao Nacional de Saude, Reparticao de
Nutricao, Maputo, Mocambique.

Ministerio da Saude (MISAU), 2003. Proposta de Politica Nacional para a Utilizacao dos
Suplementos de Nutrientes nos Programas de Saude. Direccao Nacional de Saude,
Reparticao de Nutricao, Maputo, Mocambique.

Ministerio da Saude (MISAU), 2003.1nquerito Nacional sobre a Deficiencia de Vitamina
A, Prevalencia de Anemia e Malaria em Criancas dos 6—59 meses e Respectivas Maes.
Reparticao de Nutricao, Instituto Nacional de Saude. Maputo, Mocambique.

Monteiro J .F.S. 1957. Resgate dos Machongos do Sul do Save. Reparticao Te'cnica de
Agricultura, Seccao de Hidraulica Agricola. Provincia de Mocambique.

Mucavele F., and Mlay G. 1998. “2020 Vision Network for East Africa: Country Note for
Mozambique”. lntemational Food Policy Research Institute.

Nagao, A., 2000. Oxidative Conversion of Carotenoides to Retinoids and Other Products.
Journal of Nutrition 2004; 134:23 7S-240S.

Nagao, A. Maeda, M. Lim, B.P. Kobayashi, H. and Terao, J ., 2000. Inhibition of B-
Carotene-IS, 15’-Dioxygenase Activity by Dietary F lavonoids. Journal of Nutritional
Biochemistry 2000; 1 1 :3 48-355

National Research Council (US). SubCommittee on the Tenth Edition of the RDAs
(1989). Recommended Dietary Allowances/ Subcommittee on the Tenth Edition of the
RDAs, Food and Nutrition Board, Comission on Life Sciences, National Research
Council. Washington, DC, National Academy Press.

173

 

Neill, McKee, 1992. Social Mobilization & Social Marketing in Developing
Communities: Lessons for Communicators. Southbound, Malaysia.

Nayga, J. N. and Capps, O., 1992. Determinants of Food Away from Home
Consumption: An Update. Agribusiness Vol. 8 (6): 549-559.

Nyamuno, C.; Langa, P. and Chivambo, B., 1995. Planeamento de Uso da Terra do
Distrito de Xai-Xai- Volume 3: Proposta de Uso da Terra. Instituto Nacional de
Investigacao Agronomica. Comunicacao N° 84c, Maputo, Mocambique.

Olson, J .A. and Hayaishi, O., 1965. The Enzymatic Cleavage of B-Carotene into Vitamin
A by Soluble Enzymes of Rat and Intestine. Proceedings of National Academic Science,
USA 54: 1364-1369.

Olson J .A. 1994. Needs and Sources of Carotenoids and Vitamin A. Nutrition Reviews
1994; 52:67-73

Omaye T. S., Krinsky I. N., Kagan, E. V., Mayne T. S., Liebler C. D., and Bidlack R. W.,
1997. Symposium Overview. Beta-Carotene: Friend or Foe? Fundamental and Applied
Toxicology 1997 Vol. 40: 163-1 74.

Omosa, M. 1997. Current and Potential Demand for Fresh and Processed Sweetpotato
Products in Nairobi and Kisumu, Kenya. Social Science Department Working Paper No.
1997-1. Posharvest Management, Marketing Program, lntemational Potato Center (CIP).
Lima, Peru.

Pereira M. L., and Omar F., 2001. Preliminary Comparison Between Natural Water
Resources and Poverty in Mozambique. Part I: Rainfall. Universidade Eduardo
Mondlane. 2nd WARFA/ Watemet Symposium: Integrated Water Resources
Management: Theory, Practice, Cases. Cape Town, South Africa.

Pindyck Robert and Rubinfeld Daniel, 1997. Econometric Models and Economic
Forecasts”. Fourth Edition, The McGraw-Hill Companies, Inc.

Pyror A. W., Stahl W., Rock LC, 2000. Beta-Carotene: From Biochemistry to Clinical
Trials. Nutrition Reviews 2000; 58(2): 39-5.

174

 

Rahm R. M., and Huffman E. W., 1984.” The Adoption of Reduced Tillage: The Role of
Human Capital and Other Variables”. American Agricultural Economics Association Vol.
66(4): 405-413.

Rahman M.M, Wahed M.A., Fuchs G.J., 2002. Synergetic Effect of Zinc and Vitamin A
on the Biochemical Indexers of Vitamin A Nutrition in Children. American Journal of
Clinical Nutrition 2002; 75 (1): 92-98.

Rarnakrishnan, U., Martorell, R., Latham C. L and Abel R., 1999. Dietary Vitamin A
Intakes of Preschool-Age Children in South India. Journal of Nutrition 1999, 129: 2021-
2027

Rauniyar, P. G. and Goode, M. F., 1992. “Technology Adoption on Small Farms”. World
Development Vol. 20(2): 275-282.

Reddy, S.J., 1986. Agroclimate of Mozambique as Relevant to Dry-land Agriculture.
Serie Terra e Agua do Instituto Nacional de Investigacao Agronomica, Comunicacao N0
47. Maputo, Mocambique.

Russell, M.R., 2004. The Enigma of Beta-Carotene in Carcinogenesis: What Can Be
Learned from Animal Studies. Journal of Nutrition 134; 262S-268S, 2004.

Salunkhe, KB. and Kadarn, SS, 1998. Handbook of Vegetable Science and
Technology: Production, Composition, Storage and Processing. Food Science and
Technology. Marcel Dekker, Inc.

Sanders, J., Nagy J.G. and Ramaswamy S., 1990. Developing New Agricultural
Technologies for the Sahelian Countries: The Burkina Faso Case”. Economic
Development and Cultural Change Vol. 39(1): 1-22.

Sanders, J., Shapiro B. and Ramaswamy S., 1996. The Economics of Agricultural
Technology in Semiarid Sub-Saharan Africa. Baltimore: Johns Hopkins University Press.

Sankar A.V., Gittelsohn J., Pradhan E. K., Dhungel C., and West K.P, Jr. 1998. Home
Gardening and Access to Animals in Households with Xerophthalmic Children in Rural
Nepal. Food and Nutrition Bulletin Vol. 19(1): 34-41.

175

 

Schutjer, W. A., and Marlin Van der Veen. 1977. Economic Constraints on Agricultural
Technology Adoption in Development Countries. US. Agency for International
Development, Occasional Paper No. 5, Washington, D. C.

Shrimpton R., 2002. Um Plano Estrategico para a Nutricao em Mocambique. Helen
Keller lntemational, Maputo, Mozambique.

SNAP/DINA/MADER, 2002. Relatorio: Situacao da Campanha Agricola 2001/2002,
Estirnativas da Superficie Cultivada e Producao das Culturas Alimentares, Areas Perdidas
e Familias Afectadas pela Estiagem. Maputo. Sistema Nacional de Aviso Prévio (SNAP).

Snijders, F.L., 1986. Land Use Inventory of Mozambique. Série Terra e Agua do IN IA.
Comunicacao N° 43. Maputo. Mocambique.

Sommer, A. 1982. Nutritional Blindness: Xerophthalmia and Keratomalacia. New York,
NY: Oxford University Press, Inc.

Sommer, A. and West K. Jr. 1996. Vitamin A Deficiency: Health, Survival, and Vision.
Oxford University Press, Inc. New York.

Steinmetz K.A., Potter J.D., 1996. Vegetables, Fruit and Cancer Prevention: A Review.
Journal of A merican Diet Association 1996; 96:102 7-1 03 9

Strauss, J ., Barbosa M., Teixeira S., Thomas D., and Gomes R. Jr., 1991. Role of
Education and Extension in the Adoption of Technology: A Study of Upland Rice and
Soybean Farmers in Central-West Brazil. Agricultural Economics, Vol. 5 (4): 341 -359.

Tanumihardjo S.A., Cheng J .C., Permaesih D., Rustan E., Karyadi D., and Olson J .A.,
1996. Refinement of the Modified-Relative-Dose-Response Test as a Method for
Assessing Vitamin A Status in a Field Setting: Experience with Indonesian Children.
American Journal of Clinical Nutrition 1996; 64(5): 966-971.

Tsou, S. S., and Hong, LT 1992. The Nutrition and Utilization of Sweetpotato. In
Sweetpotato Technology for the 21St Century. Edited by Hill A. W. et al, Tuskegee
University, Alabama.

176

 

 

Underwood, B., 1998. Prevention of Vitamin A Deficiency. In “Prevention of
Micronutrient Deficiencies: Tools for Policy Makers and Public Health Workers.
Committee on Micronutrient Deficiencies, Board on lntemational Health, and Nutrition
Board, Institute of Medicine; Howson, C.P, Kennedy T.T., and Horwitz A. Editors.
Washington D.C., National Academy Press.

Underwood, B., 2000. Overcoming Micronutrient Deficiencies in Developing Countries.
Is there a Role for Agriculture? Food and Nutrition Bulletin, Vol. 21 (4), 2000. The
United Nations University. United States of America General Accounting Office. 1999.
Report on Food Security. Washington, D.C., March.

Valerie A. Kelly, Eric W. Crawford, and T. S. Jayne, 2003. Agricultural Input Use and Market
Development in Afiica: Recent Perspectives and Insights. Policy Synthesis for
Cooperating USAID Oflices and Country Missions No 70.
(http://www.aec.msu.edu/agecon/f52/psvnindx.htrn), January 2004.

Van Jaarsveld P. J ., Faber W. M., Tanumihardjo S. A., Lombard C. J., and Benade S.J.A.,
2003. The Efficacy of Orange-fleshed Sweet Potato to Improve Vitamin A Status of
Children 5-10-Years of Age: Final Report to the Micronutrient Initiative.

Van het Hof, K.; West C.E., Weststrate J ., and Hautvast J ., 2000. Dietary Factors that
Affect the Bioavailability of Carotenoids. Journal of Nutrition, 2000; 130:503-506

Van Stuijvenberg M. E., Kruger M., Badenhorst C. J ., Mansvelt E. P. G., Laubscher J .A.,
1997. Response to an Iron Fortification Programme in Relation to Vitamin A Status in 6-
12-Year-old School Children. International Journal of Food Sciences and Nutrition Vol.
48:4 [-4 9.

Van Stuijvenberg M. E., Kvalsvig J. D., Faber M., Kruger M., Kenoyer G.D and Benade
S.J.A., 1999. Effect of Iron-, Iodine-, and B-Carotene-Fortified Biscuits on the
Micronutrient Status of Primary School Children: A randomized Controlled Trial.
American Journal for Clinical Nutrition 1999; 69(3): 497-503.

Villareal, L. R.; Tsou, C. S.; Lo, F. H. and Chiu, C. S. 1995. Sweet Potato Vine Tips as
Vegetables. In Sweet Potato Products: A Natural Resource for the Tr0pics. Bounwkamp,
C. J ., Editor, CRC Press Inc., Boca Raton, Florida.

177

 

 

Von Elbe, H. J. and Laborde, F. L. 1989. Chemistry of Color Improvement in Thermally
Processed Green Vegetables. In Quality Factors of Fruits and Vegetables: Chemistry and
Technology. Edited by Jen, J. J. American Chemical Society, Washington, D.C., 1989.

West C. E, 2000. Meeting Requirements for Vitamin A. Nutrition Review 58 (11): 341 -
345.

West K.P., Jr, and Damton-Hill I., 2001. Vitamin A Deficiency. In Nutrition and Health
in Developing Countries. Edited by Semba RD, and Bloem M.W. Humana Press Inc.
Totowa, NJ.

WHO/UNICEF/IVACG, 1988. Vitamin A Supplements: A Guide to their Use in the
Treatment and Prevention of Vitamin A Deficiency and Xerophthalmia.World Health
Organization, Geneva.

Wooldridge, J ., 2003. Introductory Econometrics: A Modern Approach. Michigan State
University. Thomson, South-Westem.

Woolfe A. J ., 1992a. Sweet Potato: An Untapped Food Resource. Cambridge University
Press, Cambridge.

Woolfe A. J ., 1992b. The Contribution of Sweetpotato and its Products to Human Diets.
In Sweetpotato Technology for the 21St Century. Edited by Hill A. W. et al, Tuskegee
University, Alabama.

World Bank, 1985. Mozambique: An Introductory Economic Survey. Washington DC:
Eastern and Southern Africa Region, Country Programs 1.

World Bank, 1996. The World Bank Participation Sourcebook.
http://www.worldbank.org/wbi/sourcebook/sbpdf.htm, May 2004.

World Bank Group. 2004. World Development Indicators 2004.

Wuyts Marc, 1978. Peasants and Rural Economy in Mozambique. Universidade Eduardo
Mondlane. Centro de Estudos Afiicanos.

178

 

Yaron D. Dinar A. and Voet H., 1992. Innovations on Family F arms: The Nazareth
Region in Israel. American Journal of Agricultural Economics, Vol. 74 (2): 3 61 -3 70.

Zeitlin, M.F., Megawangi, R., Kramer, E.M., and Armstrong HQ 1992. Mothers’ and
Children’s Intake of Vitamin A in Rural Bangladesh. American Journal of Clinical
Nutrition Vol. 56: 136-1 4 7.

179