A "Baker's Dozen" of Issues Concerning Theses & Dissertations

Preparation Phase

1.  Doable projects do not spring from a student’s personal experience or imagination. Thesis and dissertation topics must be based in and developed from the research literature. However, experience determines personal preferences which often direct the search for interesting topics in the research literature.

2.  Students should not choose a topic of marginal interest. A topic must be interesting to sustain the student through the rigors of a thesis or dissertation.

3.  Too often students’ literature reviews are not exhaustive. Follow the trail to the beginning; your study may have already been done.

4.  Too often the literature review is made up of conclusions of the authors of cited studies rather than a critical analysis of the complete documents.  An annotated bibliography is not a literature review.

5.  The literature review should be a balanced presentation of the current state of the literature on and controversies about the phenomenon of interest.

6.  The experimental designs and analytic methods of all cited studies should be carefully and critically reviewed. Students should understand that the introduction to a research paper is the author’s interpretation of the cited literature, and the discussions and conclusions are the author’s interpretations of his/her findings. Only the research design and methods sections of research papers are relatively objective in their presentation.

Methodology Phase

7.  Students should understand the history of development and assumptions of the methods they use. They should be experts on the methods used in their thesis or dissertation.

8.  Students should understand that every theory and method has a range of application, and within that range they may be most appropriately applied to a narrower range of phenomena.  Intelligence tests were designed to predict future learning based upon prior learning. For example, the Wechsler might be a good predictor for middle class persons from the USA, but it would not be as useful in predicting future learning for persons from cultures on which it has not been normed. There is much discussion on cultural bias and standardized testing.

9.  The methods used in a research project should be justified as the most appropriate approach to answer the question under examination and should be informed by the designs and methods  used in prior research on the subject of interest.

10.  Research design and statistical tests are different aspects of a research project. Proper design permits valid interpretation. Statistical analysis provides a way to make decisions about effects.

11.  The words “proof” or “prove” are not part of the research vocabulary. Research questions are studied, examined or explored.  Hypotheses are supported, accepted, or rejected. Statistical results are significant or not. There is neither the logic nor sufficient evidence to declare a hypothesis proved or a research question answered.  

Completion Phase

12.  Overstating an effect is an ethical violation. Interpretation of results should be conservative.

13.  Good science is self-correcting and knowledge is cumulative. The processes of research are such that no matter how able the researcher, in the long run he or she will be shown to be wrong.

Paul Zelhart, PhD  9/16/15