Good results speak for themselves

NUMBERS & GRAPHS

A  few years ago, using numbers and computation was a problem for many researchers, especially in the non-science stream. Social scientists had to spend a lot of time in manual computation. With the advent of calculators and user-friendly statistical packages, these have become relatively easier. Data is available much faster today when compared with earlier days. This has led to an important requirement of organising data, to prepare a database by filtering the collected information and to classify and tabulate data. Easily available data, generally published by various government agencies offer challenges to social scientists to develop and test their models and add to their knowledge in the context of changes happening in political and economic scenario.

The use of quantitative methods helps in categorising and measuring phenomena. One of the early achievements of social scientists using this technique was to understand causes of poverty through a new technique, namely, the social survey and to determine how, through samples, huge populations could be described and measured. Information technology has redefined the amount and quality of data, as well as made analysis simpler and faster.

Though computers and calculators help in the mechanical part of analysis, it is the researcher’s task to choose the right tools and techniques and to select the right models, which have to be developed and tested under various conditions. Since political, economic and other social conditions are rapidly changing, the social researcher has to be equipped well to deal with them and use the appropriate tools with caution.

This requires that the researcher has enough exposure and familiarity with different tools and techniques as the onus of selecting and implementing the right one rests with the researcher. In order to fulfil this responsibility, the researcher has to pay attention to every small detail of the study, right from data collection, measurement of data, using the right tool for different types of data and making the correct inferences from the data using the correct and valid techniques of analysis. A very important aspect in this is that the researcher should become conversant with the available tools and the situations in which they can be used.

She or he has to be familiar with statistical models and tools in order to organise data so that one can draw the right type of inference. A well-planned programme is essential for collecting the data. Knowledge about alternative statistical methods, their assumptions, limitations and their usability is very important for a researcher.

There are a large number of user-friendly packages available to aid research. SPSS (renamed as PASW) is probably the most commonly used software by social scientists. STATA, SAS, ATLAS-ti, Ethnograph, SPSS-Amos etc are some other software that are used in social sciences. However, usage of these require caution as they operate, like other software packages on the principle of “garbage-in, garbage-out”. A researcher in any field will surely benefit by taking up a course in basic statistics to know the usage and appropriateness of various concepts.

However, depending on software, and in turn, depending on experts to use these, will not constitute a proper research. It has been found that many research scholars use various measures without paying attention to data, its scale and its validity.

This may produce some result, but it may turn out to be a skewed analysis. A wrong tool may lead to erroneous conclusions that may affect the credibility of the researcher. These statistical tests applied in data can be understood by learning the theory behind it. Only when the concepts are clear, can the appropriate tests be determined.

The most important point to be noted is that experiments or research have to be designed to use statistics. Many researchers embark on a study without pausing to think of how they can use statistics right from the design stage. There is a story about an eminent Professor at Cambridge who gave a paper at a scientific meeting and was asked by a participant, “What statistical test did you use to verify your results?” The Professor explained that he used his own statistical test: “In our Department, we have a long corridor with a notice board at one end. I draw a histogram of my results, pin it to the notice board, then walk to the other end of the corridor. If I can still see a difference between the treatments, then it’s significant.” The relevance of this story lies in what it does not say!

If an experiment is designed and executed properly, as we would expect of an eminent scientist, then the results often speak for themselves. A good experimental design involves having a clear idea about how we will analyse the results when we get them. That's why statisticians often say “Think about the statistical tests we will use before we start an experiment.”