Hyderabad: With the state govt all set to carry out a BC caste survey, experts say the focus of the exercise should be to collect precise information. While they agree that ambiguous questions must be avoided, their views differ on how long the survey must be.
While some experts say that any survey should have a maximum of 25 to 30 questions, others point out that asking a limited number of questions could result in an incomplete picture. They say that the number of questions depends on the survey's objectives and add that it is important to list the objectives clearly before surveying to get accurate responses.
‘Too many questions an issue'
"When there are too many questions, there is a risk of people not even attempting to give proper answers. For instance, take the QS World Rankings — that lists the world's top universities — Many are not even willing to take it as they have to answer too many questions," said Pramod K Nayar, a professor of English from the University of Hyderabad (UoH).
He said that in the majority of cases, respondents are unwilling to take lengthy surveys, citing it as one of the major reasons for the failure of some surveys.
The state govt will be conducting the BC survey from Nov 6. It has prepared a questionnaire with 56 questions (a total of 75 if supplementary questions are also included).
Koppisetti Suresh, who will be part of caste survey, said, "Even during the Samagra Kutumba Survey, we saw people becoming impatient as there were around 70 questions. For many questions, we had to put a ‘dash' as they were unwilling to answer. Several questions, especially those related to property or land, education, were incorrectly answered," he said.
He said that even in the case of the BC caste survey there are high chances that people will not give proper answers. Adding that many of the questions seemed irrelevant, he said, "Easily, for 30% of the questions, we are not going to get a proper answer."
‘Length not important'
Several other professors and researchers however said the number of questions is not that much of a priority while preparing a survey.
"A survey's aim is to get the complete picture. So, the number of questions is mostly irrelevant as it needs to cover every aspect to get complete information," said Karunakar Peda from the School of Gender and Livelihoods at Tata Institute of Social Sciences, Hyderabad.
He added, "For a socio-economic and political representation like this, 50 to 60 questions seem normal and anything below that might not work as one needs to gather information about demographics, family member details, education, livelihood, income, property details, access to resources, and political consciousness, among others."
Karunakar, who earlier conducted surveys such as ST livelihood mapping for the Centre, however, pointed out that the survey needs to be meticulously carried out by explaining objectives clearly to respondents.
The findings of the BC caste survey will serve multiple purposes including determining BC reservations in local bodies in tune with population; guiding distribution under welfare schemes; and budget allocations for other communities.