Conducting ResearchThis section, "conducting research," covers activities that take place during the experimental phase of the research process. Thus, the cases in this section have been developed to draw the reader's attention to a few of the many dilemmas that may arise during formulation of a research protocol and the collection and selection of data. As is true of all the cases in the handbook, these are intended to present dilemmas and uncertainty and not present simple situations in which behaviors are blatantly wrong. For that reason, none of the cases describes a scenario in which data are simply fabricated or falsified.
Case A1, for example, illustrates the importance of selecting an appropriate experimental range, and how varying that range can dramatically influence the conclusions drawn from a research project. In addition, the case presents the dilemma of a trainee who must decide how important this variance is in her work and how to respond to her mentor, who dismisses her concerns.
Cases A2 and A3 explore the difficulty in determining when data points are truly outliers (data that are erroneous, stemming from extraneous causes) and when they reflect an actual experimental effect. The reader must also consider the appropriate way to handle outliers when compiling research results.
Although nonscientists in particular may think of data as only sets of numbers, data in fact come in many physical forms. Tissue samples, videotapes, sound recordings, and photographs may all serve as the sole tangible evidence of research findings. Yet, unlike numerical representations of a research phenomenon, which most trainees would know not to modify for the purposes of enhancing their results, the same might not be true of these other physical representations of experimental findings. These issues are dealt with in Cases A4 and A5.
The issues worth discussing in the context of the conduct of research are numerous. One salient issue, for example, relates to the objective of the research endeavor. Too often scientists set out to prove what they already believe to be true. However, in an objective search for truth—the essence of the scientific undertaking—the scientist instead should test a hypothesis fairly and carefully and appreciate that a negative outcome can be just as fruitful and edifying as a positive one. The instructor may wish to develop new cases on this or other pertinent topics, which include:
- formulating a hypothesis
- selecting appropriate controls
- conducting and interpreting statistical analyses, and
- replicating experiments
The suggested readings for this section deal with issues of experimental design, data selection, and data management. Readers may also wish to refer to the Selected Guidelines on the Conduct of Research found at the back of this handbook, as many of those references deal to some extent with topics pertinent to the experimental phase of research.