"... investigating the possible influence of bottom screens and hive entrance design on small hive beetle populations in honey bee colonies." Research papers reviewed in this column typically follow a format that includes the author(s) description ofan experimental design to test a hypothesis, To understand the hypothesis that is being tested can sometimes be difficult, but it can usually be deciphered from the experimental design. For example, if a researcher decides to test whether clover honey orsoybean honey is equally adequate for overwintering in Illinois, the design used would likely compare the performance of a number of colonies fed equal amounts of the two types of honey over the Winter. Let's assume that the researcher measures some parameter(s) of overwintering performance in the Spring and finds no measurable difference between the two groups following statistical analysis. In this case, the researcher is unable to reject the hypothesis that both foods are equally usable by the honeybees. Thus, the researcher infers that both honeys are suitable Winter food. Now let's take the other case - a researcher decides to test whether bees can survive the winter eating used motor oil as well as honey. The researcher designs his/her experiments with multiple colonies set up to contain either motor oil or honey in the frames. By Spring, it is obvious that motor oil does not support overwintering, as all the colonies die. A statistical test is employed to compare resulting survivorship numbers and the results are found to be statistically significant. Thus, the hypothesis (that motor oil and honey are equally adequate for overwintering) is rejected and the researcher needs to go on to develop an alternative hypothesis that can be tested (orin this case, if she/he actually tested the motor oil hypothesis, perhaps the next step is to look for another job). It is an interesting exercise to examine research reports and look for the underlying hypotheses that are being tested, how the reportedresults relate to the hypotheses and how the researchers use those findings to either tentatively accept the hypothesis (in the case of the soybean/clover honey experiment) or reject the hypothesis and go on to discuss alternatives (as in the motor oil experiment). Finally, it is useful to recognize that hypotheses can only be tentatively accepted. Why? Because, as the result of inductive reasoning (the particular tested case or cases) they are always open to retesting and possible refutation.
展开▼
