The compilation of this Collection of Data Notes makes students exam preparation simpler and organised.
Sampling Errors and NSSO
This comprehensive lesson on the Collection of Data focuses on the role and functions of the NSSO in India. It also explains point-wise, the meaning of ‘standard error’ and enumerates the factors that affect it and how they can be manipulated to reduce the error.
National Sample Survey Organization (NSSO)
Any lesson on the collection of data is incomplete without studying the National Sample Survey Office (NSSO). The organization is at the centre of the collection of data and is intrinsic to India’s mechanism of collection and interpretation of crucial data on the households of the country. It carries out useful surveys on demographic, agricultural, industrial, social, and economic factors, covering rural and urban households across the country.
It forms the government’s major statistical wing to provide data necessary for planning purposes. The organization was born out of the National Sample Survey Directorate under the Ministry of Finance in 1950. Later, in 1970, it was revamped into the NSSO and came under the Ministry of Planning in the department of statistics.
Objectives of NSSO:
- To make statistical and related information available for purposes of planning and policy prescriptions.
- Then to help in devising statistical techniques to analyze and interpret data and make estimations about future trends.
- To collect and publish relevant information on socioeconomic indicators and demographic parameters.
- Also to analyze information and make such analysis available to those engaged in research in socio-economic fields.
Functions of NSSO:
- Conducting large-scale surveys on relevant indicators such as employment, income, health, expenditure, etc.
- Deciding the topic of the particular round of the survey.
- Conducting an annual survey of industries on a yearly basis.
- Prepare reports on agricultural production in India by devising methods for timely estimation of crop yield.
- Coordinating the results obtained by states on crop estimation surveys and compiling them for overall analysis.
The second part of this lesson focuses on a crucial aspect of statistical estimation, viz. sampling error. As the name suggests, this kind of error arises due to the selection of a sample instead of working with the population. The other kind of error is a non-sampling error, but here we shall look at only the first kind.
Sampling error is the difference between the estimated values obtained from the sample and what would have been the actual values if all units of the population were surveyed. We can calculate the Sampling error from the sample itself. It is important in our study because it is relevant to know the degree of accuracy in our findings from the sample. This gives us an estimate of how accurate our interpretations from the data will be. If sampling is carefully performed, sampling error can be minimized.
Factors that Affect Sampling Error
A number of factors affect the sampling error:
- Sample Size: The more you keep increasing the size of the sample, normally, the sampling error must decline. This is because the sample starts getting more representative of the population by including more units in it.
- Sample Design: The well-designed your sample is to represent the characteristics of the population, the lesser becomes your sampling error.
- Sampling Fraction: This is an extension of (a). The larger the fraction of the sample to the population, the lesser is the value of the sampling error.
- Population Variability: The larger the variability in the population, the larger is the sampling error. Population variability can be reduced by increasing the size of the sample. This helps to reduce the sampling error.
What is the relation between standard deviation and sampling error?
The standard deviation of a set of data depicts how much the individual units differ or deviate from the mean value. In other words, it depicts variability. The greater the variability, the greater will be the standard error. A high value of standard deviation, therefore, corresponds to the high standard error. By increasing the sample size, the standard deviation reduces. As standard deviation reduces, sampling error can be minimized.