1. Good social work research, like all good social research, can not tell us exactly what to do but it can contribute to social work practice and helps us to make good decisions as social workers. Research can help us understand what is ( i.e., facts) ; it can not by itself tell us what ought to be done (i.e., values).
2. Social work practice remains more an art than a science (more a craft than a set of technical procedures); but it is an art, when practiced well, that relies heavily on research for creating a knowledge base for making informed judgments about what to do.
3. Research contributes to practice by providing empirical information. Empirical information is information about the world as it is.
4. Empirical information takes a variety of forms from qualitative to quantitative. Qualitative information is more in narrative than numerical form and is geared to providing intensive, in-depth information on what some aspect of reality subjectively means to people. Qualitative information is especially useful for understanding subjective points of view. Quantitative information is more numerical than narrative and is designed to provide extensive or generalizable information that objectively indicates what is happening across many people, cases, situations, etc. Quantitative information is particular useful for specifying broad patterns or trends across the populations being studied.
5. Good social research, including social work research, need not be described as scientific. Science is often claimed to produce objective, factual, causal explanations of reality that are definitively true. No research probably ever does that.
6. Science as a distinctive form of inquiry grows out of the desire to know about the world around us based on our own experience and observations rather than relying on other sources of knowledge. As the Rubin and Babbie remind us, people seek a general understanding of the world around them. Much of what we know we know by agreement rather than experience. Tradition and authority are important sources of understanding that we often defer to unreflectively. Yet, when we understand through experience, we make observations and seek patterns of regularities in what we observe. Science starts out as nothing more than the attempt to systematize and formalize learning about the world through such observations. Yet it is not an infallible method.
a. In the 1950s, Karl Hempel claimed there was one scientific method that all scientists used to prove theories to be true by testing hypotheses against facts. According to this verificationist theory of science, science verified causal theories according to objective facts.
b. In the 1960s, Karl Popper claimed that if there was a scientific method, it involved not proving theories but disproving them. We could never have all the facts that could definitively prove anything. The scientific method was no more than being open-minded, doubting everything, stating all claims to truth as hypotheses, trying to disprove them, and accepting the ones that survived this process as tentative truths. According to this falsificationist theory of science, science falsified theories according to available facts.
c. In the 1970s, Thomas Kuhn claimed that all research was done according to a paradigm--an overarching theoretical perspective that framed the subject matter, defined the facts and specified how to study it. According to this paradigmatic theory of science, science was affected by the values of its paradigmatic perspective that colored all it did and could never be said to be an activity where independent, object facts proved or disproved theories. Good research needs to be sensitive to how paradigms color perspective, frame research, lead values to influence what we take to be fact, and lead us to produce less than perfect truth.
7. We therefore need to be sensitive to how in research it is unavoidable that values will affect facts. Values are subjective preferences often implying what is right or wrong. Facts are objective information about the world. We probably never get to know facts in their pure form. Personal values as well as paradigm values lead us to consciously as well as unconsciously construct a partial view of the world.
8. Values affect research in a variety of ways starting with how we frame a research problem. We need to be sensitive to how biases in our perspective of a problem affect how we develop a research focus, phrase a research question.
9. While remaining sensitive to the issue of bias, we often develop a research question by placing ourselves in a research community , finding out how others have studied the topic, how they have framed their research, what methods they have used, and how our research can build on and add to what has already been done. Using library search tools can us find the studies we need to consult. As W. L. Newman points out in Social Research Methods (Allyn and Bacon, 1997), there are a variety of goals to be served by conducting a literature review (demonstrating familiarity with a field, tracing the path of prior research, integrating what is known, learning from others how to do your own research, stimulating new ideas).
10. A good research question is one that is researchable or, in other words, is a question that can be answered with empirical information. While we need to be wary of value bias, we need to focus on trying to acquiring factual information whether it is qualitative or quantitative. Research is all about acquiring sound empirical information about the world as it is.
11. A good research strives to achieve controlled observation . Controlled observation is where we establish conditions such that we can isolate how certain selected factors affect other factors. If we are using a causal perspective as is often the case when doing quantitative research, then controlled observation is focused on trying to establish conditions where we can isolate a causal factor's effect on some other factor--such as length of unemployment on self-esteem. Good controlled observation would allow us to isolate a causal factor’s affect while controlling for all other factors that might affect what we are studying--e.g., what is the effect of length of unemployment on self-esteem controlling for other factors that might affect self-esteem, such as family history, relationships with others’, income, gender, age, etc.
12. When trying to acquire quantitative empirical information, research is often framed in terms of an hypothesis. An hypothesis is a statement about what is likely to be true. It is like a bet or a wager. It is a predictive statement. Hypotheses are often stated in a way that suggests that one factor tends to be related to another as in the example--the longer someone is unemployed, the lower their personal self-esteem. Hypotheses often are stated in a way that specifies an independent factor (such as length of unemployment) and a dependent factor (self-esteem), where the independent factor (length of unemployment) is specified to affect the dependent factor (self-esteem). If data collected for instance show that levels of self-esteem vary with length of employment even after taking other factors into account, then we might want to suggest that the independent factor (such as length of unemployment) tends to affect the dependent factor (self-esteem).
13. Hypothesizing highlights the links between theory and research. It suggests that both induction (moving from the specific to the general or from facts to theories) and deduction (moving from the general to the specific or from theories to facts are involved in conducting good research. This can be represented by the "Wheel of Science" depicted on page 50 of Rubin and Babbie where good research is shown to involve a continual process of moving from theories to facts as in induction that are then used to generalize by moving from facts to theories. Good research can start anywhere on the wheel but it keeps moving around it. You might first develop a theory and deduce from that particular facts which you would predict to be true. After collecting data to test the theory and see if those facts were indeed present, you would then try to generalize by engaging induction and moving from the facts to a general statement about that subject. This would give rise to further theoretical propositions about the topic, new questions, new hypotheses, more deductions, new tests, additional data, leading to inductively make more generalizations. For instance, you might theorize that child who have been abused are more likely than other children to grow up to abuse their own children. You might develop a survey and interview a sample of parents in treatment to see if they were abused as children and compare those results to what other parents interviewed say about their childhood experiences. You might find that while there is some evidence for this hypothesis, it applies only to certain children who were repeatedly abused. This might lead to new theories and even a new research project.
14. Therefore, we can say that the research process converts theories into hypotheses. To do this it must also then convert concepts into variables. This process is called operationalization. Operationalization is taking a theory with concepts and converting it to an hypothesis with variables. An hypothesis states a theory in a way that makes it testable. In order for an hypothesis to be testable, we need variables that are measurable. Variables are measures or indicators of the concepts. Variables are called variables because they vary across the people we study. To operationalize a concept is to make it measurable by converting it into a variable. Some concepts are easy to operationalize, such as age, earnings, years of formal schooling, etc. Others might be more controversial such as race. We might operationalize self-esteem by measuring it according to responses people give to survey questions about how they perceive themselves. Once we operationalize concepts into variables, we can collect data on them and see how they vary across the population we are studying.