Quota Sampling and Indian Official Statistical System

Quota sampling technique is a method for selecting survey respondents from a population. In quota sampling, a population is first segmented into mutually exclusive sub-groups (or two or more strata). Then judgment is used to select the units from each segment based on a specified proportion. For example, a surveyor may be asked to sample x males and y females between the certain age groups. This means that individuals can set a demand on who they want to sample.
This second step of selection makes the technique non-probability sampling. In quota sampling, the selection of the sample is non-random sample and thus can be unreliable for making inferences. It is just possible that interviewers might be tempted to interview only those people who look to be most helpful, or may choose to use accidental sampling to question those closest to them, for time-saving sake. The problem is that such samples may be biased because not everyone gets a chance of selection. This non-random element is a source of biasness in the actual sample. Quota is normally confused and is advocated to give some probability of selection of units/individuals in the sample.
Quota sampling is useful when time is limited, a sampling frame is not available, the research budget is very tight or when detailed accuracy is not important. Subsets are chosen and then either convenience or judgment sampling is used to choose people from each subset. The researcher decides how many of each category is selected.
Quota sampling is clearly the non probability version of stratified sampling. In stratified sampling, subsets of the population are formulated so that each subset has a common characteristic, such as gender, age. Random sampling chooses a number of subjects from each subset with, unlike a quota sample, each potential subject having a known probability (normally equal probability) of being selected. Fixing of quota or sample size of each strata (or group) on the basis of reliability of estimates one wish to have is fine. But the second stage selection should then be on basis of some random selection as otherwise we are assuming every one is same (homogeneous) within the various groups and this can’t be true. One can’t have convenience sampling at the second stage of sampling.
The Indian corporate sector is divided into two segments namely, Public Limited Companies and Private Limited Companies and some quota of companies to be selected from each segment is fixed. But as the full sampling frame for both the segments is not available even with the Department of Company Affairs, Government of India as there are large many number of respective companies who do not file their annual reports with the Ministry of Corporate Affairs. Many a times it has been noted that even Top Indian Companies do not bother of filing their annual reports with the Ministry regularly. Many companies get themselves registered with the Ministry at the time of its constitution, but quite a few Private Limited Companies close their operations and do not report this important event to the Ministry. In the absence of correct sampling frames, random selection can’t be done. Thus, the estimates generated by using Quota Sampling are under question on account of their reliability. Many such examples are exiting and are crippling the Indian Official System.