This blog will attempt to answer some of the questions that I have received regarding some of the principles of high-quality quantitative studies. As is the case for qualitative researchers, Creswell (2009) is one of the best sources for some best practices of quantitative studies, and he outlines some basic principles of a quantitative study on pages 116-120.

Challenge #1: Too broad of a topic

One of the most common challenges with a quantitative study, from my experience, is that learners often choose too broad of a topic. For example, attempting to derive an algorithm using the attendance, socio-economic, attendance, special need proportions, and other data captured from various educational institutions, and how they affect average yearly progress (AYP) goals is a worthwhile objective, but often too broad for one dissertation. However, studying how one after-school remedial math course might affect standardized test scores in math is doable because it is much more focused.

 Challenge #2: Having open access to data

There are usually two groups in this category - learners that already have the data and are contemplating how to leverage it, and those that have in mind a way to collect it. My advice for the second group is to be cautiously optimistic about the collection of data through Survey Monkey, and other instruments. Often participants are less willing to participate than originally planned, which means that response bias will also have to be something that you discuss in your dissertation. If your research involves a field experiment, successfully encouraging subjects to participate will be an essential component of the success of your research. I also suggest including the actual collected data in your dissertation, as open and transparent access to tables/charts, and the raw data, can be very helpful to the reader beyond just describing it.

Challenge #3: Using the appropriate statistical method

This is one of the easiest challenges to remedy, as your methodologist, and related studies, should guide your selection of statistical methods. Of course, you will need to explain why you used a particular method in your dissertation.

Challenge #4: Expecting the numbers to agree with our anecdotal observations

Finally, what the numbers actually "tell us" can be materially different from what we expected. The results can either be inconclusive, or they may actually reveal that the opposite of our assumptions was actually true. In this case, would it be possible for you to leverage this new, surprising, knowledge in some way? For example, if extra instruction actually lowers test scores in an educational setting, then are the courses worth the extra resources for the organization?

Hopefully, this list of considerations, offered as just a start of some of the challenges of a quantitative dissertation, will generate some thought at just the right time for you in the doctoral process!

Best regards,
Daniel J. Smith, MA, MBA, PhD

480.861.8851 (cell)

[email protected]

Creswell (2009). Research design: Qualitative, quantitative and mixed methods approaches. Los Angeles, CA: Sage Publications.