Hi Chaitra,

No - there can never be a null hypothesis in qualitative research. They are purely the domain of certain approaches in quantitative research. Qualitative research uses 'questions' instead. You cannot statistically test the narrative data that emerges from qualitative data collection.

Hi Chaitra,

As Dean has said, you are working within another paradigm when dealing with qualitative research -- one that isn't asking for predictions or that employs (generally) predictive statistics. In quantitative or experimental research, interpretation is based on statistical logic whereby large numbers of dependent variables are collected from a randomly selected sample of subjects, the data are recorded, and predictive statistics are employed to determine how significant the findings are and how appropriate it is to apply these findings to cases outside of the experimental or quasi-experimental setting. In qualitative research, however, the significance and utility of the findings are based on the awareness and operation of underlying principles/mechanisms that give rise to complex observable variables (See the logic from Glaser & Strauss, or Silverman or Morgan's work.  Whitehead has several chapters since 2012 dealing with this issue also).  Instead of statistical representativeness or mathematical (statistical) logic, therefore, the issue of practical or clinical utility or extension (i.e., can the findings be applied to other subjects or populations) are couched in terms of the generalizability of cases to theoretical (or underlying operational) propositions rather than to statistical populations or universes (See Bryman, 1988 among others on this point).  That is, in qualitative research the application of findings is not determined by the number of subjects or the mathematical probabilities regarding the recurrence of behavior but, rather, by an understanding of the underlying themes, mechanisms and bases for human social action --explication (See Silverman, 1993). Once understanding is achieved through qualitative methods, these foundational mechanisms are (at times) extended to other instances and individuals and predictions and generalizations might be made.  This is the ultimate description and understanding of social phenomena as procedural affairs.  Again, the goal is explication not predictability...

Along with everything said above, there is a practical consideration based on the small Ns in qualitative work, which are almost always hopelessly inadequate for statistical tests. 

If you want to do quantitative work, the data source you suggested should work.

Almost by definition, statistical hypotheses cannot be used in qualitative research, but, given the nature of your hypothesis and the data source you are planning to use to test it, I don't understand why you are calling your investigation "qualitative" in the first place. If you have a clear, falsifiable hypothesis and a large, high quality data set, what you are doing is QUANTITATIVE research.