Your Quantitative Data Analysis Plan

The objective of a qualitative study is often to answer questions of “why” and “how” regarding a phenomenon. The most common qualitative research designs include phenomenology, case studies, grounded theory, narrative inquiry, ethnography, and Delphi. Data analysis for qualitative studies is frequently conducted via either a content analysis or thematic analysis, with the requirement that qualitative data analyses adhere to predetermined protocols describing step-by-step procedures for managing, coding, and analyzing the qualitative data.  Utilizing software such as NVivo, MAXQDA, ATLAS.Ti, and Dedoose, the objective of a qualitative data analysis is to develop themes and findings relevant to your study.

While every dissertation study is unique with its own requirements, many students undertaking qualitative studies choose a phenomenological research approach with data analysis consisting of interpretative phenomenological analysis or transcendental phenomenology. These phenomenological methods of qualitative analysis are popular because of the relatively small sample size – sometimes as few as 6-12 participants – required to complete a study. When using a phenomenological research design, there is an emphasis on the lived experiences of the participants in the study where individuals meeting the inclusion criteria will be sampled and interviewed for their first-hand, in-depth experiences.

Another popular research approach in qualitative studies is based on a case study research design where you can focus on the professional experiences of the participants in your study. A case study may be the best approach to your qualitative study if you’re gathering overall perceptions of professionals who have previous work experience and deep knowledge of your research topic. While very similar to a phenomenological approach, there is usually a need to obtain a sample of 15-30 participants as the data will be shorter in length and typically require more information until saturation is reached.

Institutions requiring a higher degree of rigor such as Ivy League schools and R1 schools tend to encourage more expansive research designs including grounded theory and ethnography. For studies requiring more complex qualitative methods, more time and resources may be required to conduct the thematic analysis needed to report the results and findings. The coding process can also be much more rigorous, based on relevant theory and structured as open, axial, or selective coding. 

As part of your qualitative data analysis, you are required to justify the validity of your results. Your analysis must include a discussion of any biases that may have affected your interpretations of your qualitative data. This discussion needs to address the trustworthiness of your data, including its credibility, transferability, dependability, and confirmability.

Benefits of Data Analysis

The typical objective of a quantitative study is to determine the existence of any statistically significant association between or among numerically measured variables. The most popular quantitative research methodologies include Analysis of Variance (ANOVA), Linear Regression, Logistic Regression, and Structural Equation Modeling (SEM). Quantitative data analyses begin with a descriptive analysis that explores your research questions in depth using your quantitative data, followed by an inferential analysis designed to produce statistically valid results. Utilizing software tools such as SPSS, Stata, R, and SAS, most students conducting quantitative studies use either primary survey data or secondary data sources to explore significant associations between variables or to provide correlation or regression estimates.  

The descriptive section of your quantitative data analysis utilizes tables and plots to present your data in a manner relevant to the objectives of your study. Tables of descriptive statistics provide summary information including frequencies, means, and medians for the variables of interest. Scatterplots present continuous data while barplots display binary or categorical data. The descriptive analysis section also must demonstrate that any data cleaning or manipulation has been performed correctly and that the data is organized in a format appropriate for the statistical model devised as part of your Research Methodology.

The inferential section of your quantitative data analysis utilizes mathematical models to confirm the statistical significance of your results. Depending on your study’s specific characteristics, many different quantitative inferential methods can be applied to produce the relevant research findings, with each statistical model being grounded in either a correlation or regression based methodology. If you are conducting a study in biostatistics or the health sciences, which may require a more rigorous data analysis plan, then be sure to learn more about our Biostatistics consulting services.

ARC has worked with just about every type of statistical model that exists in academia today. For correlation based methods, the Pearson’s r approach is considered best when the independent variable and dependent variable are both continuously measured; however, the Spearman’s rho approach is the most appropriate when you’re working with ordinal data based on rankings of the observations as opposed to their numerical value. The most popular and simple regression based methods include linear regression, multiple regression, logistic regression, and the family of ANOVA models (including ANOVA, ANCOVA, MANOVA, MANCOVA, etc.), where the choice of which model to use also depends on how the independent and dependent variables in the study are measured. As you work with our consultants to solidify your Research Methodology, we will be able to identify the most suitable Data Analysis Plan for your quantitative study.

As part of your quantitative data analysis, you are also required to justify the validity of your results. For quantitative data, this discussion of validity should address your statistical model’s diagnostic procedures for confirming that all model assumptions have been satisfied.

No matter your data analysis needs, ARC assists students with all aspects of the data analysis process. We can help you to develop and implement a customized data analysis plan that is right for you. Even better, we’ve got access to all the software – NVivo, MAXQDA, ATLAS.Ti, Dedoose, SPSS, Stata, R, SAS, and more!