Social Work Research Methods. Reginald O. York
is a method of inquiry that has the purpose of discovery, not justification. You do not engage in scientific research to prove a point. Instead, you do it to find out the truth.
2 Science is systematic, objective, and comprehensive.
3 Science relies on objective measurement, leading to data from which conclusions are drawn.
4 Science is designed to reduce human error in decision making, as evidenced by a reduction in decision regret.
5 Science is founded on common principles with critical thinking, which emphasizes each of the critical aspects of scientific inquiry, including the purpose of reducing decision error.
6 Science as a basis for knowledge is different from experience, authority, or tradition, each of which may be useful for the social worker but are based on methods that are different from the methods of scientific study.
7 Science is distinctively different from pseudoscience as a way of discovery because pseudoscience, unlike science, tends to rely on unsubstantiated claims, anecdotal evidence, authority, and cherry-picking of facts to support a preconceived conclusion.
8 Science, as a method of inquiry, can be viewed in light of some commonsense phrases such as the following:Don’t put the cart before the horse!Two heads are better than one!Some things happen just by chance!Don’t reinvent the wheel!
Chapter Discussion Questions
1 Suppose someone said to you, “My 20 years of practice experience is my scientific evidence.” What do you think of this statement?
2 Under what circumstances have you engaged in service activities (or treatment approaches) for which you had no scientific evidence of effectiveness. Did you think about evidence in this regard? Did you seek evidence by asking others or going to the literature?
3 What is one of the signs that an assertion about practice may be based on pseudoscience?
4 Suppose you found that your clients had a measured gain of 20% on an anxiety scale between the beginning and end of treatment but the differences between the pretest and posttest were not found to be statistically significant. Would you conclude that your clients had a modest gain with regard to anxiety? Explain.
5 What do you think about the question of whether the full moon is associated with unusual behavior?
6 What kinds of professional decisions do you believe are most in need of assistance from scientific studies?
7 Suppose you are using a practice method that has little evidence of its effectiveness with the target population on whom you are using it. Do you believe that the burden on you to evaluate your practice (with data) is greater because of this lack of evidence?
Chapter Test
1 A researcher has decided that he wants to conduct a research study of the people in his fishing club, using interviews of each member of the club. He has not decided on a purpose or general research question that he should pursue with this group of people. He asks others for advice on what research question or issue he should pursue in this study. What is the error in this situation, in reference to the various phrases that connect common sense with science?Don’t put the cart before the horseTwo heads are better than oneDon’t throw the baby out with the bathwaterSome things happen just by chance
2 Suppose you have decided to get feedback from your clients. What step would logically come first in the research process?Selecting the questions to put on the questionnaireSelecting the sample of persons to receive the questionnaireDetermining the purpose of the study and the general question to be answeredSelecting the statistical measures to employ in the analysis of the data
3 Which of the following statements of the purpose of a research study is consistent with the spirit of scientific inquiry?To demonstrate that dialectical behavioral therapy is effective in the treatment of persons with borderline personality disorderTo challenge the criticism that human services are not worth the costBoth of the aboveNone of the above
4 Which of the following is demonstrated by the above quiz question?Don’t put the cart before the house!Don’t reinvent the wheel!Scientific research is a process of discovery, not a justificationTwo heads are better than one
5 Suppose you have decided to conduct a study of whether the full moon makes people act strange, or different. Many people have asserted that the full moon makes people “crazy” or “strange” or “weird” and so forth because they have witnessed such behavior during the full moon. Is this information sufficient to draw a scientific conclusion about the effect of the full moon?Yes, because the perceptions of people of the behavior of others should be considered trustworthy; otherwise, there would be no way to do scientific studies of social behaviorYes, because this information provides a clear connection between this kind of behavior and the presence of the full moon; if people act this way during the full moon, it is logical to assert that the full moon has an effectNo, because you would need to employ only instruments published in books to measure your variablesNo, because you need also to analyze information about the presence of this behavior when the moon is full and when it is not full as well as the absence of this behavior when the moon is full and when it is not full
6 Suppose you conducted a study with the purpose of determining if those who regularly engage in aerobic exercise have less stress than others. You collect data on exercise and stress for a group of people and find that those who exercise have less stress but the difference in stress scores between those who did and did not exercise was not statistically significant. What commonsense phrase is related to this situation?Don’t reinvent the wheelSome things happen just by chanceDon’t put the cart before the horseYou can’t make a silk purse out of a sow’s ear
7 A researcher gave a questionnaire to those who attended his dinner party and found that the eight men at the party had an average salary that was higher than the average salary for the 13 women in attendance. These differences were found to be statistically significant. What would be the appropriate conclusion?Salaries for men are higher than for womenSalaries for men are higher than for women in the city where this party took placeOf those who attended this party, salaries for men were found to be higher than for womenOf those who attended this party, there were not enough people to draw any conclusions
8 A police officer investigated an auto accident and needed to decide who was at fault. There were three people on the street who were witnesses. The officer interviewed all these three people. What basic saying in common sense does this situation illustrate?Some things happen just by chanceTwo heads are better than oneDon’t put the cart before the horseDon’t throw the baby out with the bathwater
9 Are there any steps you take before you draw conclusions in a scientific study?Yes, you develop the basic research questionYes, you develop the research methods (who will be in the study, how the variables will be measured, etc.)Yes, you do both of the aboveNo, you draw your conclusions from the literature review you conducted
10 A researcher has decided to conduct a study of whether people act differently during the full moon than at other times. She has decided to compare the number of disciplinary incidents in school during the full moon and the new moon (when the moon is least full). A fellow researcher has reported that dozens of many similar studies have been done and almost all show that there is no difference in behavior during the full moon. When this fellow researcher suggests that it does not make sense to do such a study, what is the saying that is most appropriate for this?Don’t reinvent the wheelTwo heads are better than oneSome things happen just by chanceDon’t put the cart before the horse
ANSWERS: 1 = a; 2 = c; 3 = d; 4 = c; 5 = d; 6 = b; 7 = c; 8 = b; 9 = c; 10 = a
Chapter Glossary
Chance. A possibility. The absence of any cause of events (actualities) that can be predicted. In scientific inquiry, you typically have a theory being tested by systematically collected data, but you realize that data can occur by chance, so you test to see if your data can too easily be explained by chance to be taken seriously with regard to the research question being examined.