Research Methods in Choosing a New Drug - Essay Example

Research Methods Introduction Scenario: You have developed a new drug that you believe is effective in reducing appetite. As a result of some preliminary research, you have been given a small grant to design an experiment that tests the effectiveness of this new drug on morbidly obese individuals. This is a comparison and contrast of the strengths and weaknesses of two different approaches to study the above scenario viz. independent group design and repeated measures design. To test the effectiveness of the new drug on morbidly obese individuals we assume that there is a causal relationship between dispensing the drug (independent variable) and its effectiveness in reducing appetite (dependent variable). To determine the extent of this effectiveness an experimental method will have to be used in which extraneous aspects of the experimental design can be strictly controlled. We have a choice of either administering the drug randomly, which would prevent confounding with other variables or conduct a quasi-experiment in which we select the subjects. The Target Population and Sampling Plan The target population is the group of morbidly obese individuals. To extend the conclusion to the whole population, this experiment must be conducted on a sufficient number of subjects. The number of subjects will further depend on how many are available, and the time, cost and other resources that will be required in relation to the number of subjects chosen for the experiment. Participants can be obtained from hospitals and through advertising in the media. Subjects can either be tested once only (independent group design), or more than once or matched (repeated measures design). The advantages of the latter are that fewer participants are needed (as it may be difficult to find morbidly obese individuals), and this method can provide greater statistical power. On the other hand, there could be practice and carryover effects that would need to be minimised. In the alternative of an independent group design, it may be difficult to randomly assign the subjects and there could be large differences between individuals, which would make it difficult to make any meaningful observations. We would however have the advantage of having a control group so as to control some extraneous factors. Experimental Method Used and its Justification A repeated measures design is preferred due to the expected difficulty in obtaining the required subjects i.e. in finding morbidly obese individuals. They also reduce the effect of individual differences (Schinka, 2003:279). The second justification for a repeated measures design is the greater statistical power relative to the sample size (Amy, 1997). However, subjects will be selected at random and counterbalancing will also be used to offset any practice and carryover effects. As a repeated measures design will be used, this also implies by definition that they will be tested more than once. It is also technically possible to use a combination of the two methods if it is felt that the order effects will be significant i.e. by initially conducting an independent group design then selecting a sub-sample for the repeated measures design. Ethical Issues Ethical issues are always an important consideration in any research study. But medical research of this kind in testing new drugs also poses particular health risks to the participating individuals. They should be made aware of the nature of the research and the possible health risks associated with taking the new ‘under-test’ drugs. It is usual in such medical research to ask for volunteers. This ensures that not only do they give informed consent but also they are well aware of the health risks. The researchers must not be deceptive about any aspect of the drug if it is known beforehand. Apart from this, general ethical guidelines should be followed wherein for example, participants are informed of the benefits of the research, other risks (not associated with the drug) are minimized, and their individual results are kept confidential and the overall results are made available to them. Another ethical concern is the use of placebos in medical research. Expected Results Given that the researcher has developed the new drug there may very well be incentives to quickly ‘prove’ its effectiveness and market the drug in public even to those who may be mildly obese. These are further ethical issues but should in no way influence the study. The preliminary research may very well have shown positive signs for drug but releasing it for wider use by the general public involves great responsibility. Falsification of results must definitely not take place. And, any positive indications must be carefully investigated before publishing the results and promoting the drug. Independent verifications of the study would be advisable to confirm the expected results. The Internal and External Validity of the Study The validity of the conclusions is threatened by factors that have either not been specified or included in the study and which could theoretically therefore influence the dependent variable. The internal validity of the study depends on how well the experiment is designed and how confident the results can be used to make statements. Thus, confounds pose threats to the internal validity of the study, and the fewer that there are of these, the greater the confidence can be in making statements based on the results. How well these results can be generalized to the whole population of morbidly obese individuals is the external validity of the study. To be satisfied with the internal validity we must first of all be satisfied that the defined independent variable i.e. the drug in question is indeed the only factor influencing the outcome. Any hint of other factors is confounds and threaten this validity. For example, it may be that another common factor between the individuals in the sample is reducing their appetite. If so, this would need to be investigated before being certain that the drug is effective. This is a common situation with testing any new drugs. The repeated measures design is also prone to confounds due possible practice and carryover effects. Limitations and Recommendations The number of individuals used in the research could be the greatest limitation for the same reason that a repeated measures design has been chosen. A recommendation therefore would be to conduct a longer-term study in which more morbidly obese individuals can be found and participate in the research. In addition the possibility of another variable causing the expected results must be thoroughly considered to internally validate the research and thereby support for the drug. References Amy, Minke. (1997). Conducting Repeated Measures Analyses: Experimental Design Considerations. Paper presented at the annual meeting of the Soutwest Educational Research Association (Austin, TX, Jan. 23-25, 1997). ERIC. Retrieved June 15, 2009 from: http://www.eric.ed.gov:80/ERICWebPortal/custom/portlets/recordDetails/detailmini.jsp?_nfpb=true&_&ERICExtSearch_SearchValue_0=ED407415&ERICExtSearch_SearchType_0=no&accno=ED407415. Schinka, John and Velicer, Wayne. (2003). Handbook of Psychology: Research methods in psychology. John Wiley and Sons. Read More