In this article we will discuss about:- 1. Meaning of Sampling 2. Why Sampling is Essential? 3. Limitations 4.Criteria 5. Methods.
Meaning of Sampling:
In biological experiment, it is not possible to collect complete information about a population. If the number of pods/plant is to be collected from a field then it is really time consuming and also rarely possible to do it.
Then few plants are taken into account for studying the whole population of plant in that field. The method by which only few items are selected from the population in such a way so that they will represent the population in unbiased way is called sampling.
The size of sample is an important factor in statistical analysis which depends on the number of sampling units selected from a population for investigation. The size should not be too big or too small. Taking only 10 values from a plot of 1000 plants will give erroneous result and also it is difficult to handle with 1000 number of values.
Why Sampling is Essential?
A. Sampling saves time, the data can be collected and summarised more quickly with a sample than a complete count of the whole population.
B. In case of infinite population, sampling is the only method for statistical analysis.
C. Sampling reduces the cost of experiment because only a few selected items are studied in sampling.
Limitations of Sampling:
A. If the sampling is not done properly, i.e., if it is biased then it misleads which results in false, inaccurate interpretation.
B. There may be personal biasness during sampling or choice of method of sampling which may also lead to erroneous interpretation.
Criteria for Good Sampling:
A. Selected samples from the population should be homogenous and should not have any differences when compared with the population.
B. Reasonable number of items is to be included in the sample to make the result more reliable.
C. The selected sample should have the similar characteristics as the original population from which it has been selected.
D. The individual items composing the sample should be independent from each other.
E. The number of observations included in a sample should be more to make the results more reliable.
Methods of Sampling:
The proper method of selection of samples and the relation between the sample and population is the matter which determines the method of sampling. There are various methods of sampling; it totally depends upon a statistician which method will be applicable for proper selection of method of sampling (Fig. 9.1).
A. Simple Random Sampling:
This method is followed where each item of the population has an equal chance of being included in the sample. Random sampling suggests that selection should be done without any biasness.
To ensure the randomness of selection one can adopt either lottery method or refer to table of random numbers. Lottery method is the simplest and most popular method, all the items are numbered and slips of identical size and shape are made. All are shuffled together and selection is done blindly.
Table of random numbers can be used in place of blind selection. For this purpose, Random Number Table (5 digit) prepared by Snedecor and Cochran (1988) can be used either horizontally or vertically for selection of sample without biasness. This method of sampling is more scientific because there are less chances of personal biasness in sampling from the population and chances of selection of every item are equal.
B. Systematic Sampling:
This method is applied when the population is large, scattered and not homogeneous.
Systematic procedure follows to choose a sample by taking every K the individual, where K refers the sample interval calculated by the formula:
K = Total population/Sample size desired
20% sample to be taken from 1000 individual of a population,
K= 1000/20% of 1,000 = 5
So the, First sample will be the 5th individual,
Second sample will be the 10th individual,
Third sample will be the 15th individual, and like this way.
C. Stratified Sampling:
This method of sampling is followed when the population is not homogeneous, hence the population is first divided into several homogeneous groups or strata and the sample is drawn from each stratum at random. This method is useful as it represents the proportionate representative sample from each group and it gives greater accuracy.
D. Cluster Sampling:
A cluster is randomly selected group. This method is used when the units of population are natural groups such as school, hospitals, etc. The technique of cluster sampling allows small number of target population and the data provided statistically valid at 95% confidence limits.
E. Non-Random Sampling:
This method is called as judgement sampling. In this method the selection process of sample is somewhat subjective. The choice of sample items depends exclusively on the judgement of the investigator.
The investigator exercises his judgement in the choice and includes those items in the sample which he considers most typical for his investigation. For example, from a rice field one investigator may select only the healthy plants for artificial inoculation of a pathogen.