Scientific method fundamentals in human services research

Scientific method is a systematic method of testing hypotheses in research. The four steps are as follows: Step 1: "Observation and description of a phenomenon or group of phenomena" (Introduction to the scientific method, 2012, University of Rochester). The scientific method begins with existing research, and builds upon the research of other individuals in the field and/or observation of a particular phenomenon. Step 2: " formulation of a hypothesis to explain the phenomena" (Introduction to the scientific method, 2012, University of Rochester). The hypothesis is what is tested over the course of the experiment. It is specific, and not merely descriptive and is limited enough so that other variables that could cause the phenomenon can be eliminated. Step 3: "use of the hypothesis to predict the existence of other phenomena, or to predict quantitatively the results of new observations" (Introduction to the scientific method, 2012, University of Rochester). Step 4: "performance of experimental tests of the predictions by several independent experimenters and properly performed experience" (Introduction to the scientific method, 2012, University of Rochester). A hypothesis is not confirmed by merely one experiment, given that experiments can be affected by unaccounted-for variables and can be also influenced by random, chance events (especially in small samples) and observer bias. After drawing conclusions from the initial experiment, the researcher must retest the results, and, if necessary, further isolate the conditions under scrutiny.

However, when conducting research on human beings, it can be more difficult to create hypotheses that eliminate other influencing phenomenon or to create a perfectly representative control group that mirrors the experimental group, except for the variable under study. When studying human beings, experimenters are limited ethically and logistically to how much they can limit the variables on human beings vs. studying the laws of nature.

For example, one use of the scientific method in human services might entail the question of whether a program emphasizing diet or exercise is more effective in promoting weight loss amongst low-income African-American teenage girls. The experiment might hypothesize that dietary intervention is more effective, based upon research which suggests that there is a limit to how much exercise can make up for a poor diet, particularly with teens' increasingly sedentary lifestyles, when they not actively exercising. For example, one recent study found that "a new study finds that while engaging in high levels of physical activity is a good bet for preventing obesity in white adolescent girls, it does not give their black peers the same benefit. The study, published in the Archives of Pediatrics and Adolescent Medicine, found that among black adolescent girls who moved the most at age 12, obesity at age 14 was nearly as likely as it was for those whose activity rates were far lower," a finding which was attributed to the excessively caloric diet of the girls (Healey 2012). "In African-American households headed by a single-parent struggling to feed kids on a limited income...if you have to eat off the dollar menu, that's what you do...[and] family cooking traditions that emphasize less healthful foods and food preparation can be difficult to change" (Healey 2012).

In a proposed study based on existing research, four groups would be created of African-American teenage girls with a similar range of BMI. One group would receive an intervention primarily based on dietary reform; the other group on exercise. A third group would take an intervention program based upon both dietary and exercise modifications, and a fourth group would be the control group. The girl's BMI would be assessed before and after the program, and one year after the intervention. It would be hypothesized that the difference in BMI in terms of weight loss would be greatest for the dietary intervention group and the dietary and exercise intervention group. If the weight loss was greatest in the diet-only group or the same between the first and third groups, this would support the importance of dietary modifications over exercise as a strategy for weight loss. However, given the many variables that can affect individual levels of weight, it would be essential to retest this hypothesis several times, amongst different populations of low-income African-American girls.

The difficulties of testing a phenomenon as complicated as obesity calls some researchers to question the value of traditional, experimental research in a quantitative and numerical format and instead to favor qualitative approaches. Rather than using the scientific method in a numerically-driven fashion, the qualitative approach attempts to encapsulate the experience of people in the 'real world' using techniques such as ethnographies or participant-observation. The approach tends to be phenomenological, rather than data-driven, in the sense that it observes a phenomenon.