Census and Sample Investigation - Sample - Sample Design - Sampling Process - Advantages and Limitations of Sampling.

Census and Sample Investigation
All items in any field of inquiry constitute a universe or population.  A complete enumeration of all items in the population is known as ‘Census Inquiry’.  It can be presumed that in such an inquiry, when all the items are covered, no element of chance is left and highest accuracy is obtained.  But in practice this may not be true even the slightest element of bias in such an inquiry will get larger and larger as the number of observations increase.
Census type of inquiry involves a great deal of time, money and energy.  When field of inquiry is large this method becomes difficult to adopt because of the resources involved.  Moreover, this method is practically beyond the reach of ordinary researchers.  Further it is not possible to examine every item in the population sometimes and it is not possible to obtain sufficiently accurate results by studying only a part of total population.  When the field of studies or the universe is a small one, it is no use in resorting to a sample survey.  When the field studies are undertaken in practical life, considerations of time and cost almost invariably lead to a selection of respondents ie., a selection of only a few items.
Meaning of Sample:   A sample is a smaller representation of a large unit or value.  In other words a sample is that part of the universe or population which we select for the purpose of investigation.  In the ordinary course we make judgment and take decisions on the basis of samples. For example:
When we go to a shop, we examine a handful of rice to find the quality of rice in the whole bag.
A doctor examines only a few drops of blood to draw conclusions about the blood constitution of the body.
Similarly when field studies are undertaken in practical life, considerations of time and costs almost invariably lead to a selection of respondents ie., a selection of only a few items from the population.  The respondents selected should be a representative of the total population as possible in order to produce a miniature cross section.  The selected respondents constitute what is technically called a sample and the selection process is called ‘Sampling Technique”.  The survey so conducted is called as ‘Sample Survey’.
Sample Design:  A sample design is a definite plan for obtaining a sample from a given population.  It refers to the technique or procedure that enables the researcher for selecting an item for the sample.  Researcher must prepare a sample design for the study ie., he must plan how a sample should be selected and of what size such a sample would be.  Sample design is determined before data collection.


Steps in Sampling or Sampling Process:  Sampling process in research has the following steps:
1.Define the population and determine the sampling unit:  The population is the aggregate of all the elements to be studied.  When the number of unit belonging to a particular universe is certain, it is called ‘Finite Universe’- example: Number of workers in a factory, area of a town etc.  When the number of unit cannot be ascertained, it is called as ‘Infinite Universe’.  Example: number of stars in the sky, listeners of a specific radio program etc.
A properly defined population must be defined in terms of Elements, Sampling units, Extend and Time.  Suppose the research is conducted on the problem of Indian Cotton Textile Export, the element should be different types of cotton and cotton textile products, sampling unit would be the firm’s exporting these, extend is India and the time may specified by 2009 or so on.  The sampling unit should be definite, clear and unambiguous.  It may be a geographical one (example- State, District, Village etc.), a social unit (example- a family, school etc.), or it may be an individual.
2.Identify the Sample frame or Source list:  It is a list which contains the names of the unit of the population or universe.  In the above example the sampling frame would be the lists of the firms that are members of the Council for cotton textile export.  If the source list is not available, the researcher has to prepare such a list, which is comprehensive, correct, reliable, appropriate and representative of the population.
3.Determine the Sample Size:  It refers to the number of items to be selected from the universe to constitute a sample.  The size of the sample shall be optimum while fulfill the requirements of efficiency, representativeness, reliability and flexibility.  The size of the sample depends upon a number of factors like:
a)Homogeneity and heterogeneity of the universe:  If the items in the universe are homogeneous, a small sample is required and vice-versa.
b)Number of classes:  If many classes or groups are to be framed, then a large sample is required so that every class must be of proper size for statistical analysis.
c)Law of statistical regularity:  The law indicates that for obtaining accurate results the sample shall be randomly selected and must represent the whole universe.
d)Other considerations:  Other factors like type of sampling, size of questionnaire, standard of accuracy, availability of finance, training of investigators and time at the disposal of the researcher etc., shall also be considered.
4.Sampling Procedure: Finally the researcher must decide the type of sample he will use ie, he must decide about the techniques to be used in selecting the items for the sample.  There are several sample designs out of which the researcher must choose the best one fit for his study.
Characteristics of a good sample
A good sample must be representative of the universe.  It should contain majority of the characteristics of the universe, if not all.  In other words, a good sample should posses all the basic characteristics of the universe.
Another requirement pf a good sample is that it is free from bias.  Bias may c reap in the selection of the sample for reasons like convenience, substitution, mistake in the adoption of sampling method and failure to cover the whole of the chosen sample.
The third essential of a good sample is its adequacy.  Adequacy in the sense that the sampling size must be adequate to yield reasonable conclusions regarding the population from the sample drawn.
The sample must enable the measures of sampling error in fact the representativeness of the sample and the validity of conclusions depend to a large extent on these errors even if a proper selection is employed, a sample may not be exactly representative of the population owing to sampling errors.  As such the researcher must be in a position to estimate the influences of these errors and declare to the readers the percentage of accuracy of his survey results.
Finally, the results of a good sample should be applicable to all items of the universe with a reasonable level of confidence.  Unless the sample has this characteristic there is no use in conducting a research study.
Advantages of sampling:  The sampling technique is used in survey method of data collection.  The following are the advantages of sampling techniques:
1.When the size of population or universe is very large, sampling technique is best suited for the collection of data as it economizes money, time and effort.
2.When the lesser percentage of accuracy is sufficient in any study, the sampling technique is much suitable for the  collection of data or information through the sample survey.  A higher percentage of accuracy can be ensured only through census survey.
3.The sampling technique enables the investigator to collect the required information from relatively a large size of population or the availability of data is unlimited in character.
4.There are certain types of study where census method cannot be adopted at all.  For instance, in a study relating to the availability of mineral resources in the country sampling method is used to make an attempt to evaluate its availability below the earth in the country.
5.When the items of universe or population is more homogeneous in nature, sampling technique is more feasible and useful.
6.Sometimes the sampling units are get exploited while making the study.  In such situations census study is not possible and thereby sampling study is advantageous.  For instance, a study conducted for measuring the life period of electric bulbs produced by an organization.
Limitations of Sampling:
1.The conclusions or generalizations derived from a sample survey are always less accurate and liable for more errors compared to the conclusions arrived through census method.
2.When the various units of the survey population are not alike and liable to change frequently, the sampling technique will be very difficult.  The conclusions derived from one set of units are not comparable with another set of units which are frequently liable to change in their nature.
3.If due care is not taken in conducting a sample survey, the conclusions will be much misleading and erroring in nature.
4.The adoption of sampling techniques can be successful only if due care is taken to select the sample by competent persons in the field.
5.When the population is heterogeneous and the time available are very short, it is not possible to make a selection of a best sample.  Moreover, if we need 100% accuracy, the sampling techniques cannot be used.