Social Work Research Methods. Reginald O. York
clients for the purpose of proving that the program is effective. You conduct it for the purpose of finding out whether it is effective.
Second, your examination of information should be objective and comprehensive. You should not cherry-pick information so that you only consider data that support your expectations and you systematically ignore data that fail to do so.
Third, you follow a logical set of procedures when you conduct scientific research. This means that you do not start the process by deciding on the questions you want to put on the questionnaire. Instead, you start the process by deciding on the purpose of your study.
Fourth, your conclusions should be consistent with the data you analyzed. You should keep your opinions out of the conclusions that you draw from a scientific study. If your data were found not to be statistically significant, then you have data that can be explained by chance, so you cannot take them seriously with regard to the study conclusions. Your statement should say that you failed to find improvement in anxiety in your clients or you failed to find that those who engage in regular aerobic exercise had less stress than others. This may seem counterintuitive, because you did find that scores were higher at the end of treatment than before treatment. But if these differences were not found to be statistically significant, you cannot rely on them to do a good job of predicting what you would find if you repeated this study with another sample of people drawn from the same study population.
Suppose you undertook a study of the relationship between various sources of social support and social functioning (school grades, health behaviors, etc.) for a sample of adolescent mothers. You measured support with regard to four sources: partners, mothers, fathers, and grandmothers. Suppose you found that those with high support from partners had better outcomes. Suppose further that you found the same relationship between support from mothers and outcomes for the adolescent subjects. What should be your conclusions? Should you include the variable of grandmother support as a key element of a service plan for this population? See Figure 1.1 for a graphic display. If you failed to find that grandmother support was related to social functioning for this population, why would you conclude that this be included in a service plan? It does not make sense.
Fifth, the scientist assumes that people sometimes err in making observations. The questionnaire may not accurately measure what it was intended to measure. For this reason, methods are used in science to test the accuracy of a measurement tool. Furthermore, we continue to conduct research on a given theme to see if the results are consistent with what was found previously.
Figure 1.1 ■ Consistency: What’s Wrong With This Picture?
Science and Critical Thinking
According to two of the prominent authors of work on this subject, critical thinking is a “process in which the thinker improves the quality of his or her thinking by skillfully taking charge of the structure inherent in thinking and imposing intellectual standards upon them” (Paul & Elder, 2004, p. 1). The focus is on the structure of thinking rather than the conclusions that you draw about life. There is no liberal or conservative bias in critical thinking; so if you want a structure that supports your bias, you will not find it in critical thinking.
You probably have already concluded that there is an inherent compatibility between the concept of critical thinking and the concept of science. According to Eileen Gambrill (2015),
Critical thinking and science go hand in hand. In both, there is an openness—even an eagerness—to learn by discovering that one has been wrong, a desire to accurately understand, present, and learn from other viewpoints, a deep curiosity about the world, and a willingness to say “I was wrong.” (p. xvi)
Steps in the Scientific Research Process
Several comments have been offered on the process of scientific research. Here we will examine four simple steps in this process: (1) the determination of the research question and the knowledge base that supports your inquiry, (2) the determination of the methods to be used to find a sample of study participants and to measure the study variables, (3) collection and analysis of data, and (4) the drawing of conclusions.
The first logical step is to determine the basic research question to be pursued. In other words, you are determining the purpose of your study. Let’s say your purpose is to determine if your tutoring service is having the intended effect of improving the grades of your at-risk students in your tutoring program. The second logical step is to select your study methods. In the tutoring example, you have your current clients as your study subjects. You will measure their grades to answer your research question. You will compare the grades of these clients at the end of the tutoring program with their grades at the beginning. The third step is the collection and analysis of data. In your example, let’s suppose the grades at the end of the program were statistically significantly better than the grades at the beginning. Your final step is to draw conclusions. With your data, you can conclude that the tutoring program was effective in the improvement of the grades of your clients.
Summary of What Makes It Scientific
We have reviewed the nature of science with regard to both its spirit and its processes. Several principles have emerged, such as the following:
1 If it is scientific, it is an investigation into the natural world, not the supernatural.This places emphasis on things that can be observed (measured). The Salem witch trials were an illustration of a supernatural investigation based on an unintelligent examination of things that could not be observed in the natural world.
2 If it is scientific, it starts with knowledge that has emerged from the work of other scientists, not the idiosyncratic curiosities of an individual.While we are all at liberty to examine idiosyncratic curiosities, we would not usually refer to this as scientific because it fails to be based on the existing knowledge. It leaves you vulnerable to the mistake of reinventing the wheel.
3 If it is scientific, it employs systematic procedures, not haphazard ones.In scientific inquiry, we do not put the cart before the horse. We do not, for example, decide that we want to conduct a study using interviews of social work research students before we decide on the research purpose. We do not just “go with the gut” about how to conduct a research study.
4 If it is scientific, it has objective methods of inquiry rather than biased ones.If we undertake a study to prove a point, we are not adhering to the spirit of scientific inquiry, no matter how important we believe the point is and how much the world will be better off if all believed it. Such endeavors may be found in political maneuvers or other methods of data collection, but it is not based on the scientific process, which is objective.
5 If it is scientific, it employs methods of measurement that reduce human error in observation (measurement error).People sometimes err in making observations, like responding to a questionnaire in a way that is not totally honest. Science uses multiple ways of observation (two heads are better than one) to reduce the effect of human error.
6 If it is scientific, it has an analysis of data that appreciates the idea of chance as one of the possible explanations.Scientific analysis of data tests for the likelihood that chance is an appropriate explanation for what has been tested. If we appreciate what chance means, we will avoid drawing conclusions based on nonsignificant data.
7 If it is scientific, it presents conclusions that are restricted to the results of the data analysis.The scientist avoids the temptation to offer opinions as study conclusions when these opinions are outside the bounds of the data analyzed. The scientist also avoids drawing conclusions that go beyond (exaggerate) the study results, keeping in mind the limitations of the study methods. Consequently, the scientist presents conclusions from a study that are tentative in nature, not given as the final answer to a question.
Social Work Practice and Science
Social