INVESTIGATORS

Deborah Johnston, SOAS/LCIRAH

Elizabeth Hull, SOAS/LCIRAH

Hazel Malapit, IFPRI Main Team

Sara Stevano, LCIRAH/SOAS

Suneetha Kadiyala, LCIRAH/LSHTM

Agnes Quisumbing, IFPRI

Birte Snilstveit, 3ie

Imraan Valodia, Wits Advisory Group

Indira Hirway, CFDA

Naila Kabeer, LSE

Valeria Esquivel, UNRISD

CONTEXT AND OVERVIEW

What are the

intended and

unintended

nutritional

consequences

of agricultural

practices and

interventions?

Exclusion criteria

High income country

Language

Urban focus

Reverse causality

Study type

No time use

No agriculture or no

nutrition

Non human

11 30

404

2

Latin America

Sub-Saharan Africa

Asia

MENA

Oceania

2 studies were conducted on more than one country, across different regions

AGRICULTURAL PRACTICES AND TIME USE

Studies (47) look at:

• Time allocation or its determinants in a given setting, with a more or

less detailed description of the prevailing agricultural systems and/or

practices;

• Gendered division of labour in a given setting and/or in relation to

particular agricultural systems and/or practices;

• Participation – especially women’s participation – in specific

agricultural activities.

Widespread participation of women in agriculture

AGRICULTURAL INTERVENTIONS AND TIME USE

Study Agricultural

Intervention

Time use outcomes

Farming Waged

agricultural

work

Non-

agricultural

work

Domestic

work

Leisure

Admassie, A.

and Bedi, A.S.

(2003)

Introduction of

agricultural

technology:

machinery,

improved seeds;

Ethiopia

↓

children

(girls and

boys)

↑

girls

Not reported Not reported Not reported Not

reported

Dammert,

A.C. (2008)

Anti-coca policies;

Peru

↑

children

Not reported ↑

men

↑

children

Not

reported

Riley, P.J. and

Krogman,

N.T. (1993)

Irrigation projects to

promote vegetable

production;

Lesotho

↑

women

Not reported Not reported ↓

women

↓

women

Rubin, D.S.

(1990)

Sugarcane out-

growers scheme;

Kenya

Not reported Not reported ↑

women

↑

women

↑

women

Shirajee, S.S.

et al. (2010)

Aquaculture

extension project;

Bangladesh

Not reported ↑

women

↑

women

→

women

Not

reported

KEY FINDINGS (1)

• Women play a key role in agriculture and this is reflected

in their time commitments to these activities, whether as

farmers or farmworkers

• Women are important actors in the uptake and response

to agricultural interventions

• Agricultural interventions tend to increase women’s,

men’s and children’s time burdens.

AGRICULTURE, TIME USE AND NUTRITION

• Indicators of food and nutrition could worsen

due to time burdens of women

• But non-maternal care givers are important

• Sometimes the income effect dominates

No clear-cut nutritional impact

Study Agricultural

Intervention

Time use Nutrition

Farming Waged

Agricultura

l work

Non-

agricultur

al work

Domestic

work

Leisure Child care/

feeding

Calorie

intake

Nutrient

intake

Child

Nutrition

Adults BMI

Paolisso,

M.J. et al.

(2002)

Vegetable

and Fruit

Cash Crop

Programme;

Nepal

↑

women

and men

in hhs

with one

preschool

er

Not

reported

Not

reported

Not

reported

Not

reported

↓

hhs with one

preschooler

→

hhs with

more than

one

preschooler

Not

reported

Not

reported

Not

reported

Not

reported

Bellin, F.

(1994) in

von

Braun, J.

and

Kennedy

E.T.

(1994)

Bo-Pujehun

Development

Project;

Sierra Leone

↑

women

and men

Not

reported

Not

reported

Not

reported

Not

reported

Not reported →

hh

expendit

ure

Not

reported

↓

especially

in poorer

hhs

(anthropo

metric

indicators

)

Not

reported

Kumar, S.

K. (1994)

Adoption of

hybrid maize;

Zambia

↑

women

and men

Not

reported

Not

reported

↑

women

Not

reported

Not reported ↑

hh

expendit

ure

↑

hh

expenditur

e

↓

seasonally

(stunting,

wasting)

↓

seasonally

women and

men

Quisumb

ing, A.R.

et al.

(2013)

Strengthenin

g the Dairy

Value Chain

Project;

Bangladesh

↑

women

and men

Not

reported

Not

reported

Not

reported

Not

reported

↓

in relation

to control

group C1

Not

reported

Not

reported

Not

reported

Not

reported

KEY FINDINGS (2)

• An increase in time commitments has a complex impact on nutritional outcomes, and outcomes depend on how it is managed

• If we consider the varied responses, new and innovative approaches for policy makers can be identified

Differentiating factors:

• Seasonality

• Income and socio-economic status

• Household composition and household members

• Nature of our focus on nutrition and food consumption/security

IMPLICATIONS

Season

ality

Socio

-econ

om

ic status

STRENGTHENING THE EVIDENCE BASE

• Improve primary data collection

• Improve quality of time use data, National Time Use Surveys

• Improve employment indicators

• Combine different indicators of food consumption/security and nutrition

• Bring men back into the picture

Limitations of review

• Limitations in the quality of included studies

• Impossible to conduct meta-analysis

• More studies on critical gaps to explore impact pathway: men, nutritional change in rural areas, energy intensity

QUALITY APPRAISAL: TIME USE DATATime use data

What are the overarching

objectives of time-use data?

o Comprehensive investigation of time-use patterns

o Detailed investigation of particular activities

o Unclear Are activities described in a

way appropriate to the

objective?

o Yes o No o Unclear

Are simultaneous activities

considered?

o Yes o No o Unclear

How is time recorded?

o Interview - Fixed intervals - Open intervals - Unclear

o Observation o Unclear

Is contextual or background

information recorded?

o Yes o No o Unclear

Are literacy levels taken into

consideration?

o Yes o No o Unclear

QUALITY APPRAISAL: QUALITATIVE EVIDENCE

Qualitative evidence

Is the study peer reviewed? o Yes o No o Unclear

QUALITY APPRAISAL: QUANTITATIVE EVIDENCEQuantitative evidence

Randomised studies

Were there attempts to

control for selection bias?

o Yes and appropriate o Yes but inappropriate o No o Unclear

Were there attempts to

control for confounding?

o Yes and appropriate o Yes but inappropriate o No o Unclear

Were there attempts to

control for motivation bias?

o Yes and appropriate o Yes but inappropriate o No o Unclear

Were there attempts to

control for performance bias?

o Yes and appropriate o Yes but inappropriate o No o Unclear

Were there attempts to

control for reporting bias?

o Yes and appropriate o Yes but inappropriate o No o Unclear

Were there any attempts to

control for any other form of

bias?

o Yes and appropriate o Yes but inappropriate o No o Unclear

Non-randomised studies o

Is the model specification

appropriate?

o Yes o No o Unclear

Are the methods of inference

appropriate?

o Yes o No o Unclear

Is the study peer reviewed? o Yes o No o Unclear

OVERALL QUALITY ASSESSMENT

Time use

16% 20% 64%

Qualitative evidence

76% 6% 18%

Quantitative evidence

36% 19% 45%

Quantitative evidence (peer-review only)

65% 9% 26%

High

quality: Medium

quality: Low

quality:

SOURCES OF TIME USE DATA

National Time Use Surveys: 1

Nationally-representative household surveys: 15

Small-scale household surveys and interviews: 45

Observation: 13

Mixed methods: 5

Unclear: 10