The very first task of the researcher, after the collection of data, is to analyse them. The process of data analysis demands a variety of closely related operations, namely:
- The establishment of different categories,
- The application of these categories to the raw data through coding,
- Making tabulations and then drawing statistical inferences.
The unmanageable data should be, necessarily, condensed into a few manageable groups and tables to develop the chances of further analysis. Therefore, for this reason, the researcher should classify the collected raw data into some purposeful and usable categories. Usually at this stage, the Coding operation is conducted, which transforms the categories of data into symbols that may be arranged in a tabular form and then counted.
The quality of the data is improved and polished by the procedure of editing. The thus improved data is then coded, and is made ready for tabulation. Tabulation is that part of the technical procedure where the classified data are arranged in the form of tables. Computers are a great help in this area of research. Especially in the cases of large inquiries, a great deal of data is tabulated by the computers. This procedure not only helps in saving time, but also makes it possible to study a large number of variables affecting a problem simultaneously. After the process of tabulation, the analysis work is generally done by the computation of various percentages, coefficients, etc., by employing various well-defined statistical formulae.
During the process of data analysis, relationships or differences supporting/conflicting with the original or new hypotheses should be subjected to tests of significance, in order to determine with what validity the data can be said to indicate any conclusion/s. For example, take the samples of two monthly wages, each of the samples being drawn from the companies located in different parts of the same town. Now, if the samples give two different mean values, then our problem will be concerned about whether the two mean values are genuinely different, or the difference is just a matter of chance. By using the methods of statistical tests, we can find whether such a difference is an authentic one, or is merely the result of some random fluctuations. If the difference was found to be valid, then it can be concluded that the two different samples came from two different universes. On the other hand, if the difference was found to be due to by chance, then it can be concluded that the two different samples came from the same universe. Similarly, the technique of variance analysis can help us to analyse whether three or more varieties of plants, harvested on specific fields, yield considerably different results or not. Hence, we can conclude that the researcher can analyse his/her collected data with the help of various statistical measures. Now it is up to the researcher, which analysis technique will he/she find most appropriate for his data analysis.