One of the greatest difficulties that students and beginning researchers face when conducting research is attempting to link research questions to data analyses. Because of its logical and intuitive appeal, an increasing number of researchers are utilizing the qualitative and quantitative data analysis methods or mixed methods to undertake their studies.
Quantitative research: The most exemplary quantitative methodology is the experimental design or questionnaires, whereby in its simplest form, two groups are measured. When one group is treated, the groups are measured again. Results are then compared. In quantitative research, variables are defined by numbers, such as measuring behaviors or counting instances, which are then used to provide information about our world as a basis for comparisons and conclusions. Knowledge of the type of quantitative research question can help researchers select an appropriate statistical analysis. Quantitative research questions usually represent one of the following three types: descriptive, correlational, or comparative.
Qualitative research: Qualitative researchers use observations, interviews, historical and narrative documentation, and similar subjective data as a basis for discovering and confirming relationships among variables. Many questions probe social situations and use field research techniques, which give considerable attention to describing the context of the social environment. Many qualitative researchers believe that there is only one way to analyze qualitative data, through the method of constant comparative or constant comparison analysis. For example, recently, in an informal poll of school of education faculty at a large university one question was, “how can qualitative data be analyzed?” More than 80% of the participants responded with “constant comparison analysis” . We contend that using a one-size-fits-all approach (i.e., constant comparison analysis) to analyzing qualitative data, at least sometimes, will lead to interpretations that are not consistent with the underlying data—thereby affecting legitimation via components such as interpretive validity and theoretical validity.