2.2.4 Interpreting research results : the role of statistics : Once the research has been conceptualised and conducted
the major treatment to be given is statistical analysis. In the
beginning of any research we can think of possible statistical
analysis, however, it is only after collection of the data one can
decide which statistical test can be conducted on the basis on the
tendency of the data. For example, if you have collected a data
which do not meet the assumptions of the normal distribution then,
you will have to go for non-parametric statistical tests rather than
parametric tests. It is also important to see that what is your
research question, whether you want to check the relation, the
difference among the variables or predict a variable on the basis of
other variables is a deciding factor for statistical treatment. Further,
it is essential to know that the findings of the study are not by
chance. To determine this test of significance of the statistical
values can help us, which will state whether the findings are by
chance or not.
In any stream of social sciences it is less likely that a finding
will be accepted without the replication of the study in variety of
situations and by different authors. Replications of the study with
similar findings help to establish the theory. It is observed that there
are studies conducted on similar concepts, but there are
differences in the measurement tools or the parameters used.
These types of problems are dealt by conducting Meta analysis.
Meta analysis is a statistical technique for combining data from
various studies to determine that specific variables have significant
effects across studies.