Research is essentially about problem-solving, and humans are very good problem solvers. Relatedly, Kuhn (1962) describes normal science as puzzle-solving. What makes humans good at these kinds of activities? In addition to common sense, we can imagine objects used in a variety of ways. In essence, seeing potential or flexibility is a form of creativity. This kind of problem-solving creativity we have as humans was described by Hubert Dreyfus (1972) when he said that humans don’t necessarily see the function of an object as fixed. Consider using a turkey baster to fill a sports car running low on transmission fluid or a door as a desk surface. The artist Marcel Duchamp used found objects, called readymades, as art; his bicycle wheel mounted upside down on a stool from 1913 is a well-known example. Nevertheless, we shouldn’t take this flexibility for granted, for either objects or processes. For example, we may apply the same process (procedure) when it is no longer appropriate to solve a problem. This is essentially a problem-solving set effect, meaning that we approach a problem using an established (repeated) procedure. In other words, we don’t recognize that there might be a more efficient way of solving the problem. This repeated procedural approach is a problem for researchers because we might settle in on a particular approach to evaluate a hypothesis because that is what other researchers have done (the tradition). Until Sperling introduced the partial report technique, scientists’ estimate of the capacity of immediate visual memory was limited by the procedure used (the whole report) to measure it. We need to stop and ask ourselves how else we might go about investigating that particular issue. Can we improve on the tradition?
In the case of work on bias and discrimination, for example, researchers have been limited by using scales that directly ask questions about beliefs and attitudes. For example, an item from the Modern Sexism Scale is “It is rare to see women treated in a sexist manner on television” (Swim et al., 1995). Participants who see such scale items are likely to self-monitor and answer with social desirability, presenting themselves in a good light (see Chapter 5 on measures).
Social desirability: Responding to experimental stimuli and/or scales in a way that presents the respondent in a positive (socially appropriate) light.
A procedural breakthrough in addressing these kinds of problems with self-report measures has come in the form of Implicit Association Tests (IATs; Greenwald et al., 2003), which use reaction time to measure people’s associations to topics (e.g., race, sex, obesity, and age) where your explicit response might be different than your implicit response (see https://implicit.harvard.edu/implicit/takeatest.html if you want to try out an IAT for yourself). If the pairing of “fat people” with the adjective “good” takes longer to react to than the pairing of “thin people” with the adjective “good,” then we, as well as the individual taking the IAT, have learned about whether the individual’s biases are congruent with the explicit positions that person expresses about weight. In all likelihood, if we had only explicitly asked about people’s attitudes toward people who are thin versus heavy, we would not see differences. Generally, people do not want to appear biased, in this case, against those in a particular weight category.
The challenge of research is to appreciate what previous studies have shown us (tradition) without becoming limited by them in the questions we can ask (innovation). But with experience, our thought processes become routinized, regularized, and less likely to see the new in the old, to think outside the box. All too soon we are unwilling to break out of the box. Are you up to the challenge?!
Theories: What They Are and Why They Matter
The chapter has devoted quite a bit of space to a description of how information can be packaged in manageable ways to support the research process. There are three important terms that reflect different kinds of “packaging”: hypotheses, theories, and laws. In The Dictionary of Psychology (Corsini, 2002), a hypothesis is defined as “a testable proposition based on theory, stating an expected empirical outcome resulting from specific observable conditions” (p. 463). This dictionary defines a theory as “a body of interrelated principles and hypotheses that explain or predict a group of phenomena and have been largely verified by facts or data” (p. 994). A law is defined in this dictionary as “a theory accepted as correct, that has no significant rivals in accounting for the facts within its domain” (p. 536). Thus, a theory is the pivotal link between hypotheses, which are generated from theories at the least-verified end of the spectrum, and laws, on the other end of the spectrum, which emerge when a theory is viewed as having been consistently verified and operates without challenge (Figure 1.6).
Theory: “A body of interrelated principles and hypotheses that explain or predict a group of phenomena and have been largely verified by facts or data” (Corsini, 2002, p. 994).
Law: “A theory accepted as correct, that has no significant rivals in accounting for the facts within its domain” (Corsini, 2002, p. 536).
A question you might ask is whether the social and behavioral sciences have any laws. Three-and-a-half pages of laws are listed in Corsini’s (2002) dictionary, including the law of association, the law of belongingness, the law of common fate, the law of effect, the law of mass action, the law of proximity, and the law of vividness (pp. 537–540). These “laws” may conform to the idea of a theory being accepted as correct and lacking significant rivals, but many social scientists might have some difficulty easily coming up with an example of a “law” in their discipline. In contrast, we could all probably think of some of the laws of thermodynamics from high school; at the very least, we would know that there were laws of thermodynamics. Many of us think of laws as those referred to as natural laws, in terms of aspects of nature that predictably occur under specific conditions.
As is evident in that description, one challenge in the social and behavioral sciences is that human behavior doesn’t conform to this idea of invariable occurrence. As a result, in research, we spend most of our time testing hypotheses; if we are fortunate, these hypotheses are generated within the context of a theory. Thus, hypotheses and theories and their interrelationships are important to understand.
Let’s further consider the importance of theories. One of the fundamental goals of research is to provide information that allows us to predict what will happen in future situations without having to retest the assumptions. For example, it would be useful to know whether a given antibiotic, such as penicillin, works on a variety of bacteria, and not just one type. Perhaps you remember taking amoxicillin, a type of penicillin, for childhood ear infections. Amoxicillin, sometimes described as a broad-spectrum antibiotic, has been demonstrated to be effective in treating tonsillitis, bronchitis, and pneumonia, among other bacterial infections. Thus, scientists could claim the effectiveness of this drug over a range of situations; it has generality.
One common definition of research is systematic investigation with the goal of generalizable knowledge. This definition comes from the federal regulations known as the revised Common Rule that most institutions follow to regulate how research is conducted (this rule is discussed in Chapter 4 of this book). Generalizable knowledge refers to the idea that the knowledge gained from research applies beyond the circumstances and characteristics of the particular investigation (i.e., we should be able to infer what will happen beyond the circumstances at hand).
Figure 1.6 Venn Diagram of Link Between Law, Theory, and Hypothesis
How does this idea of generalizability relate to theory? At its core, a theory provides an interrelated set of principles demonstrated to explain not just the situation in question but a group of phenomena. Such principles have been tested and verified. Without theories, we would continually examine situations as if they were unique, without appreciating what they have in common with other