Stories From Wave 22: Survey Design Q&A
CMIE presented Stories from Wave 22 on June 10, 2021. The presentation was followed by a live question and answer session. Over 150 questions were asked by the audience, with the bulk of them being answered during the session. We have compiled those questions into separate Q&A pieces. This one answers questions that were asked about Survey Design.
Q: Is there documentation available online?
A: Yes, all of our documentation is available online on the How We Do It section of our website. This includes detailed documentation on survey design, survey execution, the questions and indicators in CPHS as well as other information. You can find a list of all of our indicators here. There is no “questionnaire” since the survey is administered electronically through an app. However, you can see each screen used in the app for data collection here. You will require a CMIE user ID and password to access some of this material. It is free to create one. With it, you can read all of our documentation and download sample data without having to be a subscriber. If you need any help in creating a CMIE user ID, we will be happy to help.
Q: How often is the survey conducted?
A: The entire sample is surveyed three times a year, every year since 2014. The first round is from January 1 until April 30. The second round is from May 1 until August 31. The third round is from September 1 until December 31.
Q: Is there a common ID to identify a household across the waves?
A: Yes. All households have a Household ID which remains the same across time and different datasets that contain information from CPHS. In fact, this is true for individuals as well. Every individual in CPHS has a unique Member ID that never changes.
Q: Impressive response rates! What sort of incentives do you provide to the panel members to get such a high response rate?
A: We do not provide any incentives to the sample households. This could insert a bias into the responses. The response rate is impressive because of a strong execution system.
Q: What is the sample composition across the income group / occupation group in CPHS?
Q: What’s the smallest geographic unit available in the unit-level data from the survey?
A: The smallest geographical unit available as identifies in the database is the district. The smallest geographical unit over which reliable estimates can be made over a Wave is the Homogeneous Region which is a collection of neighbouring districts identified in the database.
Q: Given the breadth of issues covered, how do you deal with respondent fatigue in answering the same questions at a reasonably high frequency?
A: Very important question and one on which we have spent considerable time and effort on. A lot has gone into ensuring that the respondents remains engaged and respondent fatigue does not set in. The high frequency actually helps in overcoming fatigue.
Q: Can we generate estimates for million plus cities and smaller cities specifically, given the urban sample is heavy?
A: Yes, the survey design is done specifically to capture cities of different sizes. We have four town size classes in the survey - Small, Medium, Large, Very Large towns. You can create representative estimates for each town-size class separately, even at the state level and below.
Q: Has the reference period for employment survey been 1 day/CDS in the past 4 waves?
A: Yes. The reference period for employment has been 1 day for every wave of CPHS.
Q: You capture information on intention to purchase consumer durables. Isn’t it likely to decline for a static sample of households as the ownership. Isn’t there a downward bias then?
A: The claim you make can be stated for the population as a whole. For example, as more households in India buy fridges, fewer households may intend to buy fridges. In reality, there is still refrigerator demand in India, for a variety of reasons. Our intension in creating the sample is to be representative of India. The trends there, aim to mirror the trends across India. As more households in the sample accumulate an asset, it may be the case that they truly wish to purchase less of that asset. The important question to ask is whether these households can be used to create a representative picture of India, to which our answer is yes.
Q: If a household doesn’t respond for many consecutive waves, do you drop it? If so, how is it reflected in the data? The sample remained stable in 2019 and 2020, so does that mean no household was dropped in those years?
A: If a household does not respond for three consecutive waves, it is marked to be dropped. However, we relaxed this condition during 2020 and will enforce it again once survey operations are back to normal.