How to determine the appropriate sample size for conducting a study?

Posted by Ronald | July 27, 2012  |  1 Comment

A sample size is an important component of a research study whereas the objective is to make inferences about a population from the sample. While there is no standard sample size, the size needed for a study is entirely dependent on the nature of the research questions. For collecting quantitative data, a person can use a statistical formula to give a rough estimate of the sample size compared to a “rule of thumb” approach from a qualitative data perspective.

On Friday, July 27, “The Holmes Education Post Talk Show” on WTAL 1450AM Tallahassee continued its segment on interviewing individuals who successfully fulfilled all the requirements for the Doctor of Philosophy (Ph.D.) or Doctor of Education (Ed.D.) regarding their dissertation work. Our special guest was Dr. Terry Shoemaker who earned a Ph.D. in statistics from Southern Illinois University – Carbondale and completed a dissertation entitled “A Monte Carlo Study of the Problem of Sample Size in Factor Analysis.” An excerpt from the interview with Shoemaker conducted by Constance Holmes follows:

Q: What motivated you to complete a doctoral degree?

A: I had already earned a bachelor’s and master’s degree in criminal justice and was teaching at a university at the time. Since I wanted to continue teaching, obtain tenure and increase my employability, I decided the doctorate would be required for me to meet these personal objectives.

Q: How did you select your dissertation topic?

A: I had the great fortune of working in the university’s statistics lab, a service provided to graduate students and faculty across the entire campus. People who worked in the lab provided consulting services on research methods, designs, statistics and measurement. While working on my doctorate and in the Lab, I learned that one of the most frequently asked questions of statisticians was, “How many subjects are needed for a study?” During a conversation with my major professor regarding the issue of sample size in factor analysis, the idea begin to unfold about how I might be able to address this issue and gain greater insight to how many cases one needs to have in a factor analytic study in order to obtain reliable results. People doing research realize that if their sample size isn’t large enough, their results are not going to be reliable.

Q: How did you select your dissertation committee?

A: I observed that people working on a dissertation sometimes got stuck or encountered numerous problems. In some cases, I thought it was due to their own making, but at other times I thought it was because of the dissertation committee members they selected. I concluded that the first and most important thing I needed to do was decide who was going to be my dissertation chair. Ideally, I wanted someone who had a great depth of expertise and was well-respected at the university. Once I had the chairperson selected, the next thing I wanted was committee members who had the necessary expertise, were supportive of doctoral students, had a good work ethic and could work effectively with both my dissertation chairman and each other.

Q: What was the significance of your dissertation study?

A: The purpose of my study was to provide an empirical basis for determining an appropriate sample size for conducting a study so that one wouldn’t have to rely on “rules of thumb” about sample size. I conducted a Monte Carlo study that involved creating computational algorithms, setting and modifying selected parameters, and then generating repeated random samples so that outcomes of interest could be observed and recorded across samples and conditions.

Q: What challenges did you encounter in completing your dissertation?

A: The most significant challenge for me was in writing all the computer code to generate the data for a Monte Carlo study. On occasions, I encountered problems with the code which meant I had to audit the code and correct any programming errors. One night I could not find the problem so I called my dissertation chair to seek his advice. He was a very different person with me when I was preparing for my comprehensive exams as compared to when I was working on my dissertation. When I looked for guidance in preparation for the exam, he told me to read and know all the content in the department’s list of references and that I would do fine—thus, not so much help. After I passed the comprehensive exams and was working on the dissertation, he became very supportive and helpful. While on the phone talking to him about my issue, he said, “Put on a pot of coffee and I will be right over to your apartment.” He came over and we worked into the middle of the night until we found and corrected the problem.

Q: What tips can you offer to students for completing a dissertation?

A: I observed that students who left the university before finishing their dissertations often had a difficult time finishing, due to all sorts of competing demands on their time. As such, I would recommend to students to stay at the university until the dissertation is complete, if possible. Second, I recommend that students employ good time management skills. You should have a plan, set realistic timelines and do your best to stick to your self-imposed deadlines. Lastly, students must have patience and persistence with the process. Finishing the dissertation does not happen overnight. So, commit yourself to getting something done on the dissertation every day.

We sincerely thank Dr. Terry Shoemaker for sharing his dissertation experience, knowledge and expertise as an avenue to improve public education. You can listen to the entire interview under the resource section of TheHolmesEducationPost.com. If you would like to be considered a guest on the talk show, please submit your name, email address, telephone number and a copy of your doctoral dissertation to
[email protected].

Dr. Ronald Holmes is the author of two books, “Education Questions to be Answered” and “Current Issues and Answers in Education.” He is president of “The Holmes Education Post,” an education focused Internet newspaper. Holmes is the national superintendent of education for the National Save the Family Now Movement, Inc., a former teacher, school administrator and district superintendent. He can be reached at [email protected].

One Comment

  1. Bukola O. says:

    I’m excited I came across this website and interview questions and response. It answered some of the questions I have concerning my dissertation although my approach will be qualitative. Thanks.

Leave a Comment