Following are the sample selection methods :
1. Random sampling
2. Purposive sampling
3. Stratified sampling
I. Random sampling : This method is considered the best method of taking samples because the sample researcher is not affected by biased behaviour. Under this method, every unit of the population has the same opportunity to be chosen as sample. The reason for this is that in this method, the selection of units is done at random. Each unit of the aggregate has an equal probability to be chosen as a sample by the researcher. According to Prof. Yates, Researchers do not select the unit according to their wish, but this task is fortune based. Therefore, in this method, every unit has an equal chance to be either selected in display or as a model.
Following are the main methods of selection of sample through random sampling method :
1. Lottery Method
2. Rotating the Drum
3. Selecting while keeping the eyes closed
4. Organized method of fixed order
5. Selection from tables of random numbers
Merits:
1. There is no possibility of being biased in this method.
2. This method saves time, money and labour.
3. It is a more scientific method of taking a sample from a population. There are fewer possibilities for personal prejudice. There is equal opportunity for every item to be selected in the population.
4. The sampling check of this method can be done by comparison with other models
5. It is economical and saves time, money and labour in investigating a problem.
Demerits:
1. If some special units are to be included in the sample due to their importance then this method is not suitable.
2. If the population size is small or its heterogeniety is high, the samples taken by this method cannot fully represent the population.
3. This method is suitable only when all units of the population are independent.
II. Purposive Sampling : In this method, the researcher selects such units in the form of a sample according to their wishes, which represent the population in their opinion. Which units are to be included in the sample depends entirely upon the wishes of the researcher. This method should be used only if it is necessary to include some special units in the sample. The random sampling method can also be used to select such units. For example, if we have to conduct a survey in the textile industry, the study of Raymond Company and Reliance Company is necessary.
Merits:
1. This method is so easy.
2. This method is less expensive.
3. This method is suitable when it is essential to include some special units in sample.
4. The selection of sample will be correct if the standard of sample is decided previously.
Demerits:
1. It influences the researcher’s biased approach.
2. This is not a scientific method.
3. There are many misleading findings in this method.
4. The authenticity of the samples taken by this method is not guaranteed.
III. Stratified Sampling : In this method, firstly the aggregate is divided into various parts through ideal sampling, and after that, some units are selected from each part through random sampling. Therefore, this method is a mixture of Ideal Sampling Method and Random Sampling method.
For Example- If a school has 400 students, 40 need to be selected for study of their intellectual level, than these 400 students would be divided into 4 strata first category, second category, third category and unsuccessful. If first category contains 100 students, 2nd category has 150, third category has 100, and 50 students failed, So to select 40 students, 10%. from each category must be selected. Thefefore, 10, 15, 10 and 5 students respectively from each category would be selected.
Merits :
1. This method is suitable for the heterogeneous and diversified population.
2. The benefits of both random sampling and ideal sampling are found in this method.
3. Samples represent the population perfectly.
Demerits :
1. It is a complex method.
2. There is a possibility of impureness in this method.
3. It needs more knowledge and expertise.
4. If there is an error in the creation of classes for some reason, then the drawn conclusions can also be faulty.
