Transcript Gender-disaggregated data - Homepage | APMAS Knowledge …

Karabi Baruah Ph.D
Gender, HIV & Development Specialist
For Training Course on “Gender Equitable
development Projects”
APMASS & WAP, AIT:
27th June 2012 Danang, Vietnam
SESSION OUTLINE



Basic concepts and issues related to the
collection, analysis and use of genderdisaggregated data
Common understanding of terms useful in gender
disaggregated data
Approaches to collecting and analysing gender
disaggregated data
2
GDD AND ANALYSIS
Gender-disaggregated data the collection of
information, from a sample group that includes both
male and female participants, on the different
experiences, needs, interests, and access to
opportunities and resources of men and women so
as to establish an accurate picture of the local
context.
 Gender analysis the examination of relationships
between men and women and the factors that create
and influence differential opportunities and
constraints for men and women at the local, regional
and global level.

Source: Hovorka, IDRC, 1998.
GENDER-DISAGGREGATED DATA (GDD)



Collection of data disaggregated by sex (strictly,
this is sex-disaggregated data)*
However GDD is more than simply collecting data
FROM women and men**
GDD requires a gender-sensitive data collection
process to reveal hidden or untold information
4
4
WHY THE NEED FOR GENDER
DISAGGREGATED DATA?





Capture real need, contribution,
benefits
GDD needs to be accompanied by
disaggregated data on other
variables (age, race, etc) to reflect
gender dynamics
To improve project effectiveness and
sustainability (project is then more
responsive).
Better information lead to better
performance (fish harvest, income,
etc)
The benefits are to both women and
men
GENDER DISAGGREGATED DATA
 Not
only about what men and women do,
or their characteristics
 Need
data to understand differentiated
impacts, vulnerabilities, opportunities*
 Especially
evaluation
important for monitoring and
GDD AT DIFFERENT PHASES
 GDD
in baseline (so we can measure changes
from a gender perspective)
 Difference in labor, income, control and
access to resource, perceptions
 GDD in process (both within project and on
target population)
 Difference in participation
 GDD on outcome and impact
 Focus on long-term effects of the project,
measuring against baseline data
7
COMMON UNDERSTANDING OF
TERMS USEFUL IN GENDER
DISAGGREGATED DATA
8

•Data:
Sex disaggregated data, and
 Gender disaggregated data


Statistics

•Indicator & Gender sensitive Indicators:

Qualitative & quantitative
9
SEX & GENDER-DISAGGREGATED
DATA (GDD)
 Data:
Unprocessed” information that can be
quantified
 Sex-disaggregated
data: Collection of data
disaggregated by sex (strictly by physical
attributes )
 Gender
Disaggregated Data: are
Analytical indicators derived from sexdisaggregated data on socioeconomic
10
attributes
10
STATISTICS:


“Processed” data from a sample
Numerical information answering the question,
“how much”, “how many” that are usually
presented in aggregate form as numbers or
proportions in tables and graphs (Hedman, Perucci and
Sundstrom 1996


Quantitative descriptions of some aspect(s) of
the study population (Fowler 1992)
Characteristics of the study sample (Blalock 1979) )
11
GENDER- SENSITIVE INDICATORS




“An indicator is a pointer. It can be a
measurement, a number, a fact, an opinion or a
perception that points at a specific condition or
situation, and measures changes in that condition
or situation over time.” (CIDA)
“Gender-sensitive indicators have the special
function of pointing out gender-related changes in
society over time.” (ibid)
Gender-sensitive indicators should be developed
alongside other indicators measuring progress or
achievements
Who develops indicators? Need for a participatory
approach
12
12
EXAMPLES OF INDICATORS
Quantitative indicators
Qualitative indicators
13




Level of income generated
from agricultural activities
for both male- and femalecontrolled crops.
Levels of women’s and men’s
inputs, by socio-economic
grouping, in terms of labor,
tools, etc.
Number (or %) of women
and men in key decisionmaking positions, by socioeconomic grouping.
Average household
expenditure of female/male
headed households on
education/health.




Respondent attitude towards
[new project component],
disaggregated by sex.
Level of satisfaction by
women and men with degree
of participation in project
implementation.
Perception of change in
gender equality within the
community since the project
started.
Feedback in relation to the
usefulness of training
sessions and gender training
material.
APPROACHES TO COLLECTING AND
ANALYSING GENDER
DISAGGREGATED DATA
14
QUALITATIVE & QUANTITATIVE METHODS
OF GDD
 Structural
questionnaire surveys for
quantitative GDD
 Qualitative
data collection : in-depth
interviews, Survey & structured
interviews, Focus group discussion;
Narratives, case-studies, life stories etc…
15
EXAMPLES OF FORMAT FOR QUANTITATIVE
QUESTIONS

Question 1 (yes/no, several options, multiple choice)
Option a
o Option b
o Option c
o
o
Question 2 (no specific number: how many
things/years/children/etc?; income?
o
o
_________
Question 3 (Rating questions)
Do you agree?
Strongly disagree


o

Strongly agree


1
6
QUANTITATIVE INDICATORS MAY FAIL TO
CAPTURE GENDER
Examples:
 an
increased income may hide an
increase in dependence towards a
spouse; an equal sex-ratio may hide
that an activity is not tailored to
women’s needs
1
7
QUALITATIVE INDICATORS
Qualitative indicators can be defined as
people's judgments and perceptions about a
subject, such as the confidence those people
have in sewing machines as instruments of
financial independence.” (CIDA)
 Hence qualitative indicators are crucial to
participatory methods, since they don’t
measures ‘things’ or ‘numbers’ but people’s
views.

“
1
8
HOW GENDER-SENSITIVE ARE THE SURVEY
QUESTIONS?
Issues to avoid


Question that don’t generate
gender-disaggregated data
(household income, or
respondent income)
Questions that only cover
waged labor or cash-crop (since
these will be male dominated)
to measure livelihoods.
Assume the respondent knows
better than other family
members (access to training,
resources, decision-making). A
husband and a wife may give a
different view on their level of
decision-making, or on
domestic violence)

Questions designed to cover
differentiated task




Who collects water (or fuelwood,
fodder, foodstuff) in your household?
How far do you [respondent] or this
person have to travel to collect water?
Or different crop cycles


19

Issues to include
Plowing, planting/transplanting,
weeding, picking, grinding, etc.,
which may better represent both men
and women’s economic activity
Questions that ask about intrahousehold dynamics
Question on time-use (to pick up
what specific questions don’t)
Informal work when asking about
labor activity
EXAMPLES OF BAD AND GOOD DATA
COLLECTION METHODS IN TERM OF GENDER




In a household survey, use HH as
respondents (most HH are men,
responses will reflect their views)
In-depth interviews with women are
conducted by men interviewers
(contextual: possible in some, not in
others)
Selection of respondents based on
village leaders, or available list
(leaders tend to be men, so are their
networks; list often use designated
HH)
Depending on context, mixed Focusgroup discussion where men talk
and women remain silent (or men
sit in the middle, women on the
outside)
Good




20
Bad
Respondents are alternated between
W and M, or both W and M
(father/mother), (husband/wife) are
chosen
In-depth interviews with women are
conducted by women interviewers
(opposite may also be true in some
context, men should interview men)
Random selection with equal
number of women and men, or
separate selection methods in some
contexts (may take into account
division of labor; where are M and
W)
Male and female only FGD.
However, whenever possible mixed
FGD can be very useful to show
contrasting or common views
QUALITATIVE ANALYSIS


“Qualitative analysis is used to understand social
processes, why and how a particular situation that
indicators measure came into being, and how this
situation can be changed in the future. Qualitative
analysis can and should be used at all stages of the
project cycle, and should be used alongside quantitative
and qualitative indicators.” (CIDA)
Gender M&E should use qualitative analysis to measure
the ‘quality’ of a change and to understand barriers not
revealed by quantitative analysis
2
1
MEASURING CHANGES IN GENDER ROLES







Productive
Reproductive
Community
How have these three spheres been influenced by
the project?
Are gender roles changing towards more gender
equity
Can positive process indicators (no. of women
participating) lead to change in gender roles
outcome (distribution of reproductive work)
Qualitative indicators and analysis may explain
obstacles towards more equitable gender roles
(stereotypes, perceptions, etc)
2
2
22
GENDER DISAGGREGATED DATA,
ESPECIALLY COLLECTED THROUGH
QUALITATIVE METHODS, REQUIRE
GENDER AWARE DATA COLLECTION
TOOL DESIGNER AND DATA COLLECTOR
23
23
WHICH METHOD TO CHOOSE?
 Is
the method appropriate to the evaluation
exercise (what kind of data needed, are we
looking at numbers or processes, etc)
 Can the method best measure what one wants
to measure (assets vs perceptions)
 Does the data generated need to be
comparable?
 Or are we looking for visual representation of
the evidence?
 Is it feasible, within cost, scope and limitations
2
4
FOCUS-GROUP DISCUSSION






Good technique to understand attitudes and
behavior of a target group
Questions are usually open-ended
Answers can add details to motives, why no or yes,
can be useful to understand data collected in a
survey
One can judge if a certain behavior or attitude is
shared by the group
However, one cannot extrapolate data to a general
or other population (may not be representative)
Risk of having the group interviewer provide
personal opinion that may affect results
2
5
IN-DEPTH INTERVIEWS




In-depth and semi-structure interviews are a less rigid method
to acquire data than than structured interviews
Respondents are allowed to answer at length, sometimes
bringing in related information that was not asked by the
interviewer
Mostly open-ended questions though close-ended ones can also
be added
Can use random sampling (probability sampling) if large sample
to be generalizable, unless the research is focused on a specific
and small target-group, or that respondents are hard to find. In
this case purposive (only disabled people for instance) or
convenience sampling, such as snow-ball techniques, both nonprobability sampling, can be used. However you have to mention
how bias can be introduce (such as too many old people, or rich,
etc) or whether the sample acquired is representative of a crosssection of the population
2
6
IN-DEPTH INTERVIEWS




In-depth and semi-structure interviews are a less rigid method
to acquire data than than structured interviews
Respondents are allowed to answer at length, sometimes
bringing in related information that was not asked by the
interviewer
Mostly open-ended questions though close-ended ones can also
be added
Can use random sampling (probability sampling) if large sample
to be generalizable, unless the research is focused on a specific
and small target-group, or that respondents are hard to find. In
this case purposive (only disabled people for instance) or
convenience sampling, such as snow-ball techniques, both nonprobability sampling, can be used. However you have to mention
how bias can be introduce (such as too many old people, or rich,
etc) or whether the sample acquired is representative of a crosssection of the population
2
7
NARRATIVES, CASE-STUDIES, LIFE STORIES
Underutilized in M&E
 Provide a broader view of one’s experience,
including changes over time
 Allow us to understand better social costs and
benefits from a personal standpoint
 Allow closeness with subject of research which
may help the interviewer gather information
that she or he wouldn’t find otherwise
 However, may not be representative as every life
is different. Can be cross-checked with other
stories or triangulated with other forms of data
collection

2
8
COLLECTING SENSITIVE INFORMATION







Gender related information can be about difference and
similarities, inequalities, control and in more extreme case about
abuse and violence
They tend to require a level of trust that is hard to build from
survey questions
Women may be dependent of their partners, influencing
responses
In some cases, similarities with the data collector may help (the
peer approach)
On the other hand, one may not want to divulge some
information with a local
So there is a need to assess the potential benefits (trust,
relations) with shortcomings (fear of being ostracized)
Hence GDD require an ethical data collection process
2
9
GENDER-SENSITIVE LOCATION
Should conduct the interview/survey where
respondents feel safe, comfortable and
open-; Should consider:
Location
Timing
Distance
3
0
CONCLUSION
 GDD
requires both quantitative and
qualitative methods, it is collected on the
project (input and process) as well as on the
community (output and outcome)
 Quantitative measurements are often limited
to capture the quality of change in gender
outcomes, hence the need in gender M&E to
use qualitative methods of assessment
 GDD requires gender-sensitive survey design
as well as trained data collectors
3
1
GROUP WORK
How would you generate gender differentiated
resource use pattern?
 Community in which you are working has
implemented a project on access to energy by
providing solar lamps and bio-digesters to
households; How would you collect data on
women and men are affected by implementing
the project.
 In the community you are working you have
established an eco-tourism project which brings
about xxx number of visitors to the community
every week. How would you collect gender
sensitive indicators?.

32