After devising the proper methodology for your study, you’re ready to conduct the data analysis to produce the main findings from your research. The key to completing this step is software, and there are many tools available such as NVivo, ATLAS.ti, Dedoose, and MAXQDA for qualitative data as well as statistical programming languages such as SPSS, R, SAS, Stata, C, and Python for quantitative data. This can often be the toughest step to complete for doctoral students, especially if they’re not familiar with programming languages or have yet to acquire the knowledge needed to produce theoretically and conceptually valid results.

For qualitative data, analysis primarily relies on the reading and coding of interview and focus group transcripts, and digging into them to determine the major themes and findings from the research. There is extensive theory available to guide the coding process as well as the interpretation of data for many forms of analysis in qualitative research, from interpretative phenomenological analysis (IPA) or other specific phenomenological protocols, to thematic analysis, to the open, axial, and selective coding method most commonly applied to grounded theory research.

For quantitative data, the analysis often relies on the use of statistical software, and requires both a descriptive (or exploratory) analysis and an inferential analysis to produce statistically valid inferences. A descriptive analysis utilizes tools such as tables and plots to explore the data as well and specific research questions in more depth, prior to conducting inferential analyses, which confirm the significance as well as magnitude of the statistical association between the main variables in your research.

Along with our findings, we also need to present evidence that our research and analysis is academically valid. For qualitative studies, this entails a discussion regarding various facets of trustworthiness in data such as credibility, transferability, dependability, and confirmability. For quantitative studies, this amounts to checking that all of the assumptions of a statistical model (e.g. a regression or correlation) are satisfied, and discussing any potential bias that may have resulted from our methodology.

The final discussion and conclusion needs to focus on correctly interpreting the data analysis results, in addition to providing directions for future study – including any limitations of the study, in addition to any recommendations for future studies as well as implications for social change and theory.

We can assist you with every part of this, and our qualitative analysts, as well as our statistical consultants, are available for consultation every day. For free, customized feedback on how we can guide you through this stage of your dissertation, you can email or call us anytime.