Measuring Gender and Empowerment

Question 1

Which of the following factors should be considered to ensure that measurement tools account for gender dynamics?

Answer 1

Female respondents may give different answers from male respondents. Responses (especially from females) may differ depending on whether the surveyor is a male or female. Women may be available at different times of the day relative to men. Aggregate measures may suggest welfare improvement, however, when disaggregated, suggest increased disparities between genders.

Question 2

Using Naila Kabeer’s categorization of empowerment, a woman’s income would fall into which category? (Select all that apply)

Resources
Agency
Achievements

Answer 2

Resources
Achievements

Income could reflect the resource a woman has available to make decisions: bargaining power, which leads to increased agency, but is not part of the process itself. Changes in income may also suggest in achievement of empowerment (if the woman had been negotiating for the right to earn her own income).

Question 3

Which would be the largest challenge using literacy as an outcome of women’s empowerment in cross-country surveys?

Answer 3

There may be near-100% literacy in some countries, even if there is a measurable gender disparity in other aspects of choice

Literacy may not reflect agency (decision making process), but that does not limit it from being used as an outcome: it can be used to measure achievement—which itself can be a reflection of choice. However, several countries may have full literacy, and still a large variance in what we would consider women empowerment, and using literacy, we wouldn’t be able to detect these differences

Question 4

Assuming there is no social desirability bias, which of the following questions would produce the best measure whether women have the ability to make a meaningful choice?

Do you get to decide how much you earn relative to your husband?
Do you wish you earned the same amount as or more than your husband?
Do you think women should earn the same amount as or more than their husbands?
Do you earn more than, less than, or the same amount as your husband?

Answer 4

Do you get to decide how much you earn relative to your husband?

Income may reflect a woman’s choice or not. Aspirations and attitudes, alone, do not necessarily measure whether a woman has the ability to choose. Decision-making power is a measure of meaningful choice.

Question 5

What are some of the activities we would want to conduct as part of formative research

Answer 5

Review literature (e.g. in anthropology) about the region in which we will work
Conduct qualitative work including one-on-one interviews and focus groups
Direct observation of social interactions (such as meetings)
Collect quantitative data

Question 6

Which of the following indicators was meant to uncover negotiation and decision-making?

Age of marriage
Age of first birth
Use of contraception
Awareness of different forms of contraception
Discussion with spouse about contraception
Desired time-period between births

Answer 6

Discussion with spouse about contraception

Other than the discussion, which measures agency (the process), all other indicators would likely measure resources and achievements.

Question 7

What are some of the benefits of using locally-specific questions to measure empowerment (rather than standardized questions)?

Answer 7

They can be customized to practical day-to-day decisions women make

Locally-specific measures can pick up variation in empowerment that we do not see in standardized measures.

Question 8

Which of the following methods is meant to reduce social desirability bias by ensuring anonymity?

Vignettes
Implicit Association Test
Games
Polling booths
Direct observation
Survey Questions

Answer 8

Polling booths allow respondents to answer questions anonymously, because the response cannot be linked to the respondents individually

Question 9

While piloting a question on mobility in the Bangladesh study, what did Rachel and her co-authors discover?

Answer 9

That “where” women could travel wasn’t as relevant as “for what purpose”

a measure of the mobility suggested that going to school outside of the village might be fine, but going to a fair within the village might not be.`

Question 10

Why should you always consider gender in measurement?

Answer 10

• Even when program is not specifically targeting one gender
• Some considerations – Who is interviewed?
– Who is the interviewer?
• Total impacts may hide important gender dynamics – Program may increase rice production but at expense of woman’s vegetable garden
– Total education may rise, and gender gap in education fall

Question 11

Why is it important to understand local gender dynamics?

Answer 11

– Traditional gender roles may vary across contexts
– How gender roles may interact with program being implemented?
– How gender may interact with attempts to measure outcomes?

• Gender roles may impact practical issues of when and how to collect data

Question 12

Women’s empowerment

Answer 12

“Women’s empowerment is about the process by which those who have been denied the ability to make strategic life choices acquire such an ability.

Question 13

three elements that are required for women to make meaningful choice

Answer 13

Resources - preconditions
Examples of resources would be things like human capital, financial capital, social capital, physical capital.

Agency - process
Examples of agency or the process of decision-making are voice, participation, and decision-making itself.

Achievements - outcomes
And achievements and outcomes are things like education, health and nutrition, income generation and contraceptive use

Note: resources may also be outcomes
Norms & Institutions - can affect this whole process and are likely to vary by context.

Question 14

Challenges to measuring empowerment

Answer 14

• Abstract construct
• Empowerment is a process - so how do we measure a process rather than an outcome? empowerment is not just about final outcomes, but about women’s agency in achieving those outcomes and about preferences that they make.
• Social desirability bias
• Barriers vary by context (so we much alter measurement accordingly) - but not to much, to ensure comparability to globally recognized indicators
• Realistic chance of change
• Specific areas of focus

Question 15

How do we measure the ability to make a choice? How can we measure that a woman had a meaningful choice, when we mainly see an outcome and not the choice process itself?

Answer 15

Well, some outcomes there are so fundamental that we assume that they’re wanted.

For example, if we observe poor nutrition in women and much poorer nutrition in women than men, it likely reflects the lack of real choice. (Only works in very resource-poor settings)

Ask about the decision-making process itself, or ask about preferences and see if outcomes move towards women’s preferences vs. men’s (assuming there is a difference in preferences across genders)

Question 16

How do we think about a women’s ability to make a choice, when they have absorbed the views of society and view patriarchal norms as ‘okay’?

Answer 16

We don’t want to impose our outsider views on what women should want, and yet we also have to be careful that they may be responding reflecting back society’s views of them. So very tricky here.

The ideal, though, is when we can measure the process of making choices and show how an intervention changes resources, including preferences, how it changes agency and decision making, and trace all of that through to a change in important outcomes.

Question 17

Steps in Measurement Strategy

Answer 17

Formative Research –> Theory of Change –> Select indicators and outcomes –> Data collection instruments - development/testing –> Data collection plan

Question 18

Formative research - situation analysis

Answer 18

Systematically gather information about the state of a given issue, problem or opportunity to develop strategies to address it.

Question 19

Formative research - stakeholder analysis

Answer 19

Understand and map the relative interest and power of various people, groups of people and/or orgnaizations’ involved in the issue or objective under study

Question 20

Formative research - Needs assessments

Answer 20

Understand the gap between the current and the desired state of the world for a particular issue to identify the key needs or barriers people face in solving or improving it

Question 21

How do you do formative research?

Answer 21

Read existing literature

Semi-structured interviews with community members, providers, researchers, etc. - one-to-one interviews with open ended questions, goal to understand perceptions

Focus groups - understanding how communities/societies think about a particular issue in general (susceptible to ‘group think’/social desirability bias)

Direct observation

Representative quantitative surveys/statistical analysis of existing data that’s been collected in the area where we’re planning to work

Question 22

Theory of change considerations

Answer 22

• Researchers and implementers work together
• Document the logical chain from program inputs to
intermediate and final outcomes
• Specify indicators that we can measure to track each
major step along the chain
• Specify necessary assumptions to get from one step to another and possible risks
• Get specific about how program will lead to change
and what that change look like

Question 23

Identifying Outcomes and Indicators - considerations

Answer 23

• Be specific
• More than one indicator per outcome may be needed or useful
• Track each major step along the causal chain
• Complement context-specific indicators of empowerment with more standard ones
• When possible, complement indicators that are subject to reporting bias with more objective or proxy indicators

Question 24

Measure process from resource to achievement (4 parts)

Answer 24

– knowledge
– preferences/ attitudes
– negotiation/decision
– Achievements/final outcomes

Question 25

Measuring process - reproductive health example

Answer 25

– Knowledge: different forms of contraception, dangers of early marriage and teenage pregnancy
– Preferences: preferred spacing between births
– Negotiation/decision making: talk to spouse about contraception, say over contraception
– Achievements (final outcomes): age of marriage, age of first birth, maternal morbidity

Question 26

Vignettes

Answer 26

• “Illustration in words” A vignette question describes an event,,or scenario, the wording of which is experimentally controlled by the researcher
• Provide a neutral psychological anchor for validating and revealing non-biased judgments
– More willing to express negative attitudes to fictional people in a fictional place and situation
• Can randomly assign different versions of the vignette to different sets of respondents to see how responses change

Question 27

Implicit Association Tests

Answer 27

• Standard computer-based test to detect the strength of a person’s automatic association between mental
representations of objects (concepts) in memory
• People simplify the world for efficiency
– Use rules of thumb to draw connections
– May not even be aware themselves
• Actually based on response time, not accuracy
• Easier pairings (and faster responses) are interpreted as being more strongly associated in memory than more difficult pairings (slower responses)

Question 28

Implicit Association Tests - Pros

Answer 28

Scores may reflect attitudes that people are unwilling to reveal publicly

Arguably provides a valid measure of deep unconscious belief

Question 29

Implicit Association Tests - Cons

Answer 29

Measurement Issues:
• Extremely sensitive to context (words, images, language, etc)
• Effects reduced with repeated administrations

Conceptual Issues:
• Only measures relative strengths of pairs of associations
• How the test measures association strengths is not well understood
• May measure cultural knowledge as opposed to actual belief

Question 30

Polling booth surveys

Answer 30

• Answers are submitted on a paper and put into a polling booth
– Participants could be open on how they responded
• Used to ask sensitive questions such as sexual behaviour, domestic violence etc.

Question 31

Structured community activities

Answer 31

• Instead of asking people to recall behavior we can observe behavior directly
• Measuring voice, participation, agency: observe a group of people having a discussion, making a decision
• Only a good approach when enumerator presence won’t affect outcomes
• Example: community driven development in Sierra Leone
– Offered community gift of salt or batteries and watch them decide on the gift
– Enumerator recorded whether women speak up and are included/excluded from the decision-making process

Question 32

Games

Answer 32

• Often in games people are asked to make choices about the allocation of resources they are given at the start of the “game”
• Advantage is that they can be used to see whether women allocate resources differently than men in controlled environments
• Can also show whether women allocate resources differently when they are together with men (often their husbands) than when they are on their own
– If different decisions get made when women are with their husbands that suggests their ability to make life choices normally is curtailed, thus they are not fully empowered
• Downside is that games can seem artificial

Question 33

Measuring real choices

Answer 33

• Choices need to be real and appropriate to elicit realistic responses
• Zambia: Study offered women access to contraception with and without their husbands present to see how much the decision differed—it differed a lot!
• Measuring whether and how women’s choices over contraception is constrained by the presence of her husband