From the Diary of a Novice Researcher

By LEAD Research Team

Currently, we are in the baseline phase of data collection for the project- Affordable Day Care to Empower Indian Women. This is an impact evaluation project, which aims to assess the impact of day care centers on women’s economic and social well-being, in the rural areas of Udaipur district, in Rajasthan. Our sample consists of 160 hamlets, where we have been surveying 3200 women, who have children in the age group of 1-6 years.

The past few months in Udaipur have been an enriching experience for me. I have learnt to deal with various challenges that arise during the implementation process of a research study at the field level. Through this blog, I would just like to recount a few challenges faced by us, and some possible ways to overcome the same.

Photo Courtesy: Seva Mandir, Udaipur

Documenting Age: Tougher than we imagined!

Our criteria for respondent selection is a woman who has a child in the age group of 1-6 years. Also, we are taking anthropometric measurements of the respondent’s kids who are in this age group. Thus, identifying the age of the child as accurately as possible is an imperative step for us.

However, age seems to be something people don’t keep track of in these villages. If you ask a woman here about the age of her child, you might get a response like –‘yeh raha. Aap hi dekh lo kitne saal ka hoga’(Here is he/she. You only see how old he/she might be) or ‘kya pata. Hoga do se chaar saal ka’ (Who knows. He might be two to four years old).

Yet, our height-for-age and weight-for-age charts require us to be very careful about determining the exact age. Placing a child at two years old will create a very different result than four years old. Using the date of birth is the most accurate way to determine a child’s age, but mothers often don’t have this information.

To overcome this challenge, we trained our surveyors on different techniques to determine the date of birth. Some of the techniques are as follows:

Major life events – Using major life events of the respondent helps us get a good estimate of the year the child was born in. For example, our surveyors are trained to ask respondents the year in which the mother was married, and how many years following that the was child born. It is also possible that the respondent remembers the year of birth of the youngest child. Then one could use that information by enquiring about the age-gap ( in years) between both the children, in order to determine the year of birth of the elder child.

Major Festivals – Using major festivals to see if the respondent remembers a festival around which the child was born. We have given the surveyors a local events calendar, which lists the dates of all major festivals, for the past 7 years.  This can help determine either the year or month a child was born in.

Agricultural seasons – Using agricultural seasons to see if the respondent remembers the harvest of crop during which the child was born. We have given the surveyors time periods during which major crops in this region are sown and harvested. This is helpful once we have gotten an estimate of the year the child was born in.

These techniques have worked well in figuring out at least the month, and the year of birth, of the child. These techniques may differ according to the region one is working in. For example, local festivals or harvest seasons will not be the same in every region. The best way to figure out this information is to talk to people who work at the field level. The local field workers of the partner organization, and the village leaders were an excellent source for this kind of information during the course of our project.

Easing them into conversations

Overcoming female respondents’ hesitancy to talk is a major challenge, which many may have faced in India. I believe there is no easy solution to this. It is understandable why women might find it hard to sit with a stranger for an hour answering personal questions. The problem seems to be twofold:

Making sure that the surveyors are surveying the respondent

In a rural setting, people often become curious while encountering outsiders. As a result, neighbours, friends, and relatives might start gathering around the respondent while the interview is being conducted. When this happens, everyone starts responding to the survey questions. It is important that the surveyors are trained to deal with such situations. The surveyors should continuously remind the respondent and the others that they wish to speak to the respondent, and know her opinion. This should be done in a friendly manner, as we do not want to offend the people in the village. It is best if the surveyor can make sure the survey is done in private. For example, an interview in a closed room in the respondents’ house could ensure minimal interferance of others.

Dealing with questions that the respondent does not understand or has never thought about

There might be questions on the survey, which the respondent might have never thought about or the respondent is unable to understand. For example, our survey asks a question “On a ladder from 1 to 10, how much do you agree with the following statement: “I have a lot of control over what will happen to me in my life.” 1 means you agree strongly with the statement, 10 means you disagree strongly “

It could be the case that the respondent has never really thought about this or it could also be that the respondent is unable to understand the concept of rating her agreement with the statement, on a scale of 1 to 10.

For a question like this, one may get a response like – ‘main kya bataun. Aap khud hi laga dijiye’ (What should I tell, choose a number yourself). It is important that the surveyors reiterate that they want to know her opinion, and that there is absolutely no right or wrong answer.

The surveyor should also not lose patience while having to explain the questions more than once.

In such cases, it is crucial that the surveyor is able to build a basic rapport with the respondent, and maintains a calm demeanour when the respondent is having trouble understanding the question, or is not responding.

Village leaders, your first door to knock!

Sometimes people in the villages get suspicious when a team of 4-5 outsiders come to their village, and go door to door, surveying women. They might suspect that we are there to do some harm. Fortunately, we have had very few such situations as our partner organization is well established, and known in these villages. People are usually quite welcoming when we tell them that we are doing a study in collaboration with the partner organization. In other cases, explaining the purpose of the survey to the village leaders, prior to starting the survey may be a good idea.

As we continue our journey through the different phases of the project, I will elaborate more on my experiences.