Sampling

SAMPLING

Research involves the collection of data to obtain insight and knowledge into the needs and wants of any problem or matter under consideration such as market research. In nearly all cases, it would be very costly and time-consuming to collect data from the entire population of a market. Accordingly, in market research, extensive use is made of sampling from which, through careful design and analysis, Marketers can draw information about the market.

Sampling theory has helped in reducing the cost and time of business and social research. Efficient sampling designs have been developed with confidence limits and class intervals. The study of sampling methods is not essential for researchers, but the user of research should also have understanding about the value and limitations of sample data.

Implications of Sample data:

A sample is a representation of the universe or population and if drawn on random basis can provide accurate data about universe. Majority of research studies, today, collect data through samples. To keep the cost, time and quality of research under control, reliable sampling designs are prepared.

Sample design must be such that data analysis conducted on the basis of a given sampling technique, proves helpful in achieving the objectives of the research project or study. When data is collected through ‘sample’ as opposed to census, the user of the data should know the following implications:

1.     Sampling error will be there.

2.     Sampling errors can be calculated from the summary information or means value, percentages and proportions or any other type of similar calculations.

3.     ‘Sampling error’ provides manager/ administrator additional information. This in turn can sharpen the judgment of the intelligent user of sample data.

4.     Size of sampling error and size of sample are inversely related. Also it should mention that increase in the size of sample leads to increase of cost.

5.     Data collected from different types sampling designs would not bring exactly the same analytical conclusion. Some sample designs, within almost the same cost and time would bring higher degree of accuracy and reliability, because of better knowledge of the researcher-cum-statistician in designing probability sample. In exceptional cases even in non-probability sample design, experts would improve the efficiency of sampling plans.

6.     The theory of sampling is well advanced today. The use of computers is helping statisticians, analytical expert and user of managerial information in making greater use of sampling research.

Important characteristics of Sampling and Sample data:

1.     A Sample is a subset of a total population and sampling data is taken for the purpose of learning more about the population in less time and at lower cost. Because the sole purpose of sampling is to know about the characteristics of a population.

2.     If sampling is done carefully, it can provide good knowledge of universe. The size of sample should be adequate and the sample design should reflect the traits of the population.

3.     Social scientist must recognize the risk associated with the sampling along with the merits of sampling.

4.     In many business situations, it is either impractical or impossible to examine the entire population; sample information is the only choice.

5.     The analysis of a sample can also be much more than it would be practical, if the entire population were questions about its relevant characteristics.

6.     An adequate degree of precision is required for satisfactory conclusion in making business decision on the basis of sample.

7.     It should be clear to managers that perfect information is rare and cannot be obtained within reasonable time. However, the management should insist on sample size, which would be adequate for decision-making. This implies that management seeks the greatest possible accuracy.

8.     Sufficient knowledge about the population or universe is essential in drawing sample on random basis. Important factors must be considered when sampling is to be carried out from population.

9.     The population in which management is interested is the one, which would provide the desired data. The researcher should be clear as to who are the persons, what are their qualification and characteristics. 

Once this is clear, only then sample can be drawn to represent the particular population, which becomes universe for the purpose of research study.

Random Sampling:

There are two ways of selecting a sample, either random or non-random basis. Random sample can be stratified, systematic or area sample. In probability sampling each element that makes up the population should have a known chance of being selected as part of sample. Special conditions in the selection of random sample units:

1.     Each item in the population has an equal chance of being selected    and the       probability of being selected can be determined.

2.     Each sample combination has an equal probability of occurrence.

  

Non-Random Sample:

In the contrast of random sampling, non-random sampling and the procedure to be included in the in the sample are not determined by chance. Their selection is based on the choice or judgment of the interviewers. Those units are selected which the field worker feels would serve as a good representative sample.

Sampling design is concerned with method of sampling and the procedure to be adopted in selecting sample-units from universe, which are basic source of data. Sampling units may be households, head of the family, housewife, business organization and any individual constitution the universe under study.

The sampling units can be drawn on random/probability or on non-random      basis.

The most significant difference between the two is that statistical inference can be drawn only from probability sampling methods. Probability sampling methods have been developed on the basis of a number of theories and concepts of mathematics and statistics.

Statistical inference is used to infer the characteristics of the universe from the data collected from the sample units, which are representative of universe. Whenever we apply random sample techniques for drawing conclusions or generalization, we have to calculate standard error and state confidence limits. This is done to make our statements meaningful in risk-involving situations

Sample Design

Sample design covers the method of selection, the sample structure and plans for analyzing and interpreting the results. Sample designs can vary from simple to complex and depend on the type of information required and the way the sample is selected.

Sample design affects the size of the sample and the way in which analysis is carried out. In simple terms the more precision the market researcher requires, the more complex will be the design and the larger the sample size.

The sample design may make use of the characteristics of the overall market population, but it does not have to be proportionally representative. It may be necessary to draw a larger sample than would be expected from some parts of the population; for example, to select more from a minority grouping to ensure that sufficient data is obtained for analysis on such groups.

Many sample designs are built around the concept of random selection. This permits justifiable inference from the sample to the population, at quantified levels of precision. Random selection also helps guard against sample bias in a way that selecting by judgment or convenience cannot.

Defining the Population

The first step in good sample design is to ensure that the specification of the target population is as clear and complete as possible to ensure that all elements within the population are represented. The target population is sampled using a sampling frame. Often the units in the population can be identified by existing information; for example, pay-rolls, company lists, government registers etc. A sampling frame could also be geographical; for example postcodes have become a well-used means of selecting a sample.

Sample Size

For any sample design deciding upon the appropriate sample size will depend on several key factors

(1)    No estimate taken from a sample is expected to be exact: Any assumptions about the overall population based on the results of a sample will have an attached margin of error.

(2)    To lower the margin of error usually requires a larger sample size. The amount of variability in the population (i.e. the range of values or opinions) will also affect accuracy and therefore the size of sample.

(3)    The confidence level is the likelihood that the results obtained from the sample lie within a required precision. The higher the confidence level, more it is certain to wish to be that the results are not atypical. Statisticians often use a 95 per cent confidence level to provide strong conclusions.

(4)    Population size does not normally affect sample size. In fact the larger the population sizes the lower the proportion of that population that needs to be sampled to be representative. It is only when the proposed sample size is more than 5 per cent of the population that the population size becomes part of the formulae to calculate the sample size.

Simple Random Sample

Simple random sample is the simplest for of probability sampling. Since all probability samples must provide a known nonzero chance of selection for each population element, the simple random sample is considered a special case in which each population element has a known and equal chance of selection.

Simple random sample is often impractical, as it requires a population list that is often not available. The design may also be wasteful because it fails to use all the information about a population. This type of sampling is also time and money bound. So it is only feasible where there is a complete list of population is available and enough time and money is available to carry forward the case study on the basis of Simple Random Sampling technique.

Cluster Sample

The whole population can be divided into groups of elements with some groups randomly selected for study. This type of sampling is called Cluster Sampling. It provides an unbiased estimate of population parameters. When we need economic efficiency (time and money) a Cluster Sample is better choose than Simple Random Sample. Also in case of frequent unavailability of practical sampling frame for individual elements.

Although the statistical efficiency for cluster samples is usually lower than for simple random samples as clusters are usually homogenous, but the economic efficiency is greater enough to overcome this weakness. So the net efficiency result from a trade-off between economic and statistical factors. It may take 690 interviews with a cluster design to give the same precision as 424 simple random interviews, yet the cost may be double on interviewing process of simple random sample design.

Stratified Sample

Most populations can be segregated into several mutually exclusive subpopulations, or strata. The process by which the sample is constrained to include elements form each of the segments is called stratified random sampling. With the ideal stratification, the subpopulation groups are mutually exclusive. Also each stratum is homogeneous internally and heterogeneous with other strata. A simple random sample is taken after dividing population into different groups. It is choose of the researchers due to increased sample’s statistical efficiency and also its characteristic to provide adequate data for analyzing the various subpopulations within a huge size population.

It is useful when researchers are interested to study characteristics of certain population subgroups. When different methods of data collection are applied in different parts of population it is best choice is stratified sampling. If the subgroup information correlate the variable under consideration it is good to do stratified sampling otherwise a simple random sample is a better choice.