1.3 Conducting Research in Social Psychology

Because social psychology concerns the relationships among people, and because we can frequently find answers to questions about human behavior by using our own common sense or intuition, some people think that it is not necessary to study it empirically. How good is our common sense when it comes to predicting behavior?

For each statement in Table 1.1, respond with either “True” or “False” and provide a percentage of how confident you are. For example, if you are absolutely certain you are right, your confidence percentage would be 100%, but if you are only somewhat confident, your confidence percentage might be, say, 45%. (To review the answers and a brief description of the scientific research supporting each of these topics, refer to Chapter 1, Section 4 “Chapter Summary”.)

One of the reasons we might think that social psychology is common sense is that once we learn about the outcome of a given event (e.g., when we read about the results of a research project), we frequently believe that we would have been able to predict the outcome ahead of time. For instance, if half of a class of students is told that research concerning attraction between people has demonstrated that “opposites attract,” and if the other half is told that research has demonstrated that “birds of a feather flock together,” most of the students in both groups will report believing that the outcome is true and that they would have predicted the outcome before they had heard about it. Of course, both of these contradictory outcomes cannot be true. The problem is that just reading a description of research findings leads us to think of the many cases that we know that support the findings and thus make them seem believable. The tendency to think that we could have predicted something that we probably would not have been able to predict is called the hindsight biasThe tendency to think that we could have predicted something that we probably would not have been able to predict..

The hindsight bias is one example of a very pervasive tendency in psychology—we are strongly influenced by information that we have already received, and that information changes how we think about the events that occur around us. In this case, our knowledge of the results of existing research findings blinds us to other possibilities, making us overestimate the likelihood that we could have predicted the event before we heard about it.

Our common sense also leads us to believe that we know why we engage in the behaviors that we engage in. In fact, we do not always understand the causes of our own actions. When we think about a behavior before we engage in it, we believe that the thinking guided our behavior, even when it did not. People also report that they contribute more to solving a problem when they are led to believe that they have been working harder on it, even though the effort did not increase their contribution to the outcome. These findings, and many others like them, demonstrate that our beliefs about the causes of social events, and even our own actions, do not always match the true causes of those events.

Social psychologists conduct research because it often uncovers results that could not have been predicted ahead of time. Putting our hunches to the test exposes our ideas to scrutiny. The scientific approach brings a lot of surprises, but it also helps us test our explanations about behavior in a rigorous manner. Accurate knowledge about social behavior also improves the ability of science to make changes that benefit humans globally. Having a strong understanding of the research methods used in psychology helps us evaluate the validity of the research discussed in this book, in other courses, and in our everyday lives. Being informed consumers and informed users of psychological science means being able to use the results of scientific research to guide us in making good decisions.

We will discuss the empirical approach and review the findings of many research projects throughout this book, but for now, let’s take a look at the basics of how scientists use research to draw overall conclusions about social behavior. Keep in mind, however, that although social psychologists are pretty good at understanding the causes of behavior, our predictions are not perfect. They cannot be. We are not able to control the minds or the behaviors of others or to predict exactly what any one person will do in any given situation. Human behavior is complicated because people are complicated and because the social situations that we find ourselves in every day are also complicated. The more factors we know about, the better our predictions. For example, later in this book, we will discuss many of the factors that contribute to the likelihood of behaving aggressively. The more of those factors that are present, the greater the likelihood we will lash out. However we cannot predict if any particular individual will behave aggressively in any particular situation. “Psychology does not deal in certainties; it deals in probabilities.”

One important aspect of using an empirical approach to understand social behavior is that the concepts of interest must be measured (refer to Figure 1.2). If we are interested in learning how much Skyler likes Riley, then we need to have a measure of liking. But how, exactly, should we measure the broad idea of “liking”? In scientific terms, the variables that we are interested in are known as conceptual variablesThe variables that we are interested in., and how a conceptual variable is defined for a particular study is called an operational definitionHow a conceptual variable is defined for a particular study..

For anything that we might wish to measure, there are many different operational definitions, and which one we use depends on the goal of the research and the type of situation we are studying. To better understand this, let’s look at an example of how we might operationally define “Skyler likes Riley.”

One approach to measurement involves directly asking people about their perceptions using self-report measures. Self-report measuresA measure in which individuals are asked to respond to questions posed by an interviewer or on a questionnaire. are measures in which individuals are asked to respond to questions posed by an interviewer or on a questionnaire. Generally, because any one question might be misunderstood or answered incorrectly, in order to provide a better measure, more than one question is asked, and the responses to the questions are averaged together. For example, an operational definition of Skyler’s liking for Riley might involve asking Skyler to complete the following measure:

The operational definition would be the average of Skyler’s responses across the three questions. Because each question assesses the attitude differently, and yet each question should nevertheless measure Skyler’s attitude toward Riley in some way, the average of the three questions will generally be a better measure than would anyone question on its own.

Although it is easy to ask many questions on self-report measures, these measures have a potential disadvantage. As we have seen, people’s insights into their own opinions and their own behaviors may not be perfect, and they might also not want to tell the truth—perhaps Skyler really likes Riley but is unwilling or unable to tell us so. Therefore, an alternative to self-report that can sometimes provide a more valid measure is to measure behavior itself. Behavioral measuresA measure designed to directly measure an individual’s actions. are measures designed to directly assess what people do. Instead of asking Skyler how much they like Riley, we might measure Skyler’s liking by assessing how much time they spend with Riley or by noting how often Skyler smiles at Riley. Some examples of behavioral measures that have been used in social psychological research are shown in Table 1.2.

Still, another approach to measuring our thoughts and feelings is to measure brain activity. One approach, known as electroencephalography (EEG)A technique that records the electrical activity produced by the brain’s neurons through the use of electrodes that are placed around a volunteer’s head., is a technique that records the electrical activity produced by the brain’s neurons through the use of electrodes that are placed around a volunteer’s head. An electroencephalogram (EEG) can show if a person is asleep, awake, or anesthetized because the brainwave patterns are known to differ during each state. An EEG can also track the waves that are produced when a person is reading, writing, and speaking with others. A particular advantage of the technique is that the volunteer can move around while the recordings are being taken, which is especially useful when measuring brain activity in children who may have difficulty keeping still. Furthermore, by following electrical impulses across the surface of the brain, researchers can observe changes over very fast time periods.

Although EEGs can provide information about the general patterns of electrical activity within the brain, and although they allow the researcher to see these changes quickly as they occur in real-time, the electrodes must be placed on the surface of the skull, and each electrode measures brainwaves from large areas of the brain. As a result, EEGs do not provide a very clear picture of the structure of the brain.

Other techniques exist to provide more specific brain images. Functional magnetic resonance imaging (fMRI)A neuroimaging technique that uses a magnetic field to create images of brain structure and function. is a neuroimaging technique that uses a magnetic field to create images of brain structure and function. In research studies that use the fMRI, the volunteer lies on a bed within a large cylindrical structure containing a very strong magnet. Nerve cells in the brain that are active use more oxygen, and the need for oxygen increases blood flow to the area. The fMRI detects the amount of blood flow in each brain region and thus is an indicator of which parts of the brain are active.

Very clear and detailed pictures of brain structures (refer to Figure 1.4) can be produced via fMRI. Often, the images take the form of cross-sectional “slices” that are obtained as the magnetic field is passed across the brain. The images of these slices are taken repeatedly and are superimposed on images of the brain structure itself to show how activity changes in different brain structures over time. Normally, the volunteer is asked to engage in tasks while in the scanner, for instance, to make judgments about pictures of people, to solve problems, or to make decisions about appropriate behaviors. The fMRI images show which parts of the brain are associated with which types of tasks. Another advantage of the fMRI is that it is non-invasive. The volunteer simply enters the machine, and the scans begin.

Although the scanners themselves are expensive, the advantages of fMRIs are substantial, with scanners available in many university and hospital settings. The fMRI is one of the most commonly used methods for learning about brain structure, and it has been employed by social psychologists to study such topics as social cognition, political attitudes, morality, emotions, responses to being rejected by others, and racial prejudice.

Once we have decided how to measure our variables, we can begin the process of research itself. There are three major approaches to conducting research that are used by social psychologists—the observational approach, the correlational approach, and the experimental approach. Each approach has advantages and disadvantages.

The most basic research design, observational researchResearch that involves making observations of behavior and recording those observations in an objective manner., is research that involves making observations of behavior and recording those observations in an objective manner. Although it is possible in some cases to use observational data to draw conclusions about the relationships between variables (e.g., by comparing the behaviors of older versus younger children on a playground), in many cases, the observational approach is used only to get a snapshot of what is happening to a given set of people at a given time and how they are responding to the social situation. 

One advantage of observational research is that, in many cases, it is the only possible approach to collecting data about the topic of interest. A researcher who is interested in studying the impact of a hurricane on the residents of New Orleans, the reactions of New Yorkers to a terrorist attack, or the impact of a pandemic on, well, everyone cannot create such situations in a laboratory but must be ready to make observations in a systematic way when such events occur on their own. Thus observational research allows the study of unique situations that could not be created by the researcher. Another advantage of observational research is that the people whose behavior is being measured are doing the things they do every day, and in some cases, they may not even know that their behavior is being recorded.

One early observational study that made an important contribution to understanding human behavior was reported in a book by social psychologist Leon Festinger and his colleagues. The book, called When Prophecy Fails, reported an observational study of the members of a “doomsday” cult. The cult members believed that they had received information, supposedly sent through “automatic writing” from a planet called “Clarion,” that the world was going to end. More specifically, the group members were convinced that the earth would be destroyed as the result of a gigantic flood sometime before dawn on December 21, 1954.

When Festinger learned about the cult, he thought that it would be an interesting way to study how individuals in groups communicate with each other to reinforce their extreme beliefs. He and his colleagues observed the members of the cult over a period of several months, beginning in July of the year in which the flood was expected. The researchers collected a variety of behavioral and self-report measures by observing the cult, recording the conversations among the group members, and conducting detailed interviews with them. Festinger and his colleagues also recorded the reactions of the cult members, beginning on December 21, when the world did not end as they had predicted. This observational research provided a wealth of information about the indoctrination patterns of cult members and their reactions to disconfirmed predictions. This research also helped Festinger develop his important theory of cognitive dissonance.

More recently, researchers in Poland conducted an observational study of mask-wearing during the COVID pandemic on three separate dates spaced about a week apart in May 2020. Their eighty-two medical student research assistants fanned out across thirteen different regions. Each assistant observed ten consecutive passersby in a public area. They used a checklist of what to look for—such as noting the type of mask being worn if the mask was worn properly, and, if not, what a he incorrect practices were, such as leaving noses uncovered. Observations were made once a week for three weeks. The observers found that women were more likely to wear masks than men and mask-wearing was more common in closed spaces than open spaces. In general, mask-wearing gradually declined over the three-week period. Additionally, the observers found that people wearing N95 masks or face shields were more likely to use them correctly over time, whereas those wearing cloth masks or medical masks were less likely to use them over time.

Despite their advantages, observational research designs also have some limitations. Most importantly, because the data that are collected in observational studies are only a description of the events that are occurring, they do not tell us anything about the relationship between different variables. It is exactly this question that correlational research and experimental research are designed to answer.

Because social psychologists are generally interested in looking at relationships among variables, they begin by stating their predictions in the form of a precise statement known as a research hypothesis. A research hypothesisA specific and falsifiable prediction regarding the relationship between two or more variables. is a statement about the relationship between the variables of interest and the specific direction of that relationship. For instance, the research hypothesis, “People who are more similar to each other will be more attracted to each other” predicts that there is a relationship between a variable called similarity and another variable called attraction. In the research hypothesis, “The attitudes of cult members become more extreme when their beliefs are challenged,” the variables that are expected to be related are the extremity of beliefs and the degree to which the cults’ beliefs are challenged.

Because the research hypothesis states that there is a relationship between the variables and the direction of that relationship, it is said to be falsifiableWhen the outcome of the research can demonstrate empirically either that there is support for the hypothesis (i.e., the relationship between the variables was correctly specified) or that there is actually no relationship between the variables or that the actual relationship is not in the direction that was predicted.. Being falsifiable means that the outcome of the research can demonstrate empirically either that there is support for the hypothesis (i.e., the relationship between the variables was correctly specified) or that there is actually no relationship between the variables or that the actual relationship is not in the direction that was predicted. Thus the research hypothesis that “people will be more attracted to others who are similar to them” is falsifiable because the research could show either that there was no relationship between similarity and attraction or that people we see as similar to us are seen as less attractive than those who are dissimilar.

The goal of correlational researchResearch that involves the measurement of two or more relevant variables and an assessment of the relationship between the variables. is to search for and test hypotheses about the relationships between two or more variables. In the simplest case, the correlation is between only two variables, such as that between similarity and liking or between income and helping.

In a correlational design, the research hypothesis is that there is an association (i.e., a correlation) between the variables that are being measured. For instance, many researchers have tested the research hypothesis that a positive correlation exists between the use of violent video games and the incidence of aggressive behavior, such that people who play violent video games more frequently would also display more aggressive behavior.

A statistic known as the Pearson correlation coefficient (symbolized by the letter r) is normally used to summarize the association, or correlation, between two variables. The correlation coefficient can range from −1 (indicating a very strong negative relationship between the variables; as one variable increases, the other variable decreases) to +1 (indicating a very strong positive relationship between the variables; as one variable increases, the other variable also increases). Research has found that there is a positive correlation between playing violent video games and engaging in aggressive behavior and that the size of the correlation is about r = 0.30.

One advantage of correlational research designs is that, like observational research (and in comparison with experimental research designs in which the researcher frequently creates situations in a laboratory setting), they are often used to study people doing the things that they do every day. And correlational research designs also have the advantage of allowing prediction. When two or more variables are correlated, we can use our knowledge of a score on one of the variables to predict a likely score on another variable. Because high school grade point averages are correlated with college grade point averages, if we know a person’s high school grade point average, we can predict their likely college grade point average. The higher the correlations, the more confident we can be in our prediction. And it is a prediction, not fate. If a student’s high school grade point average is an A, that does not guarantee a college grade point average of an A. Similarly, if we know how many violent video games a child plays, we can predict how aggressively they will behave. Again, the higher the correlation, the more confidence we have in our prediction. These predictions will not be perfect, but they will allow us to make a better guess than we would have been able to if we had not known the person’s score on the first variable ahead of time.

Remember: “Psychology does not deal in certainties; it deals in probabilities.” No researcher will say that if you play Call of Duty or Grand Theft Auto every day for thirty days straight, you will murder someone on day 31. Instead, we think in terms of probabilities. Here is another way to think about it. If we have a room of 100 people and we know that they each play an average of, say, five hours of violent video games a day, we can predict that many of them will aggressively behave when provoked. We will not be able to identify which among the 100 will behave aggressively, just that many will. In contrast, if we have a room of 100 people and we know that none of them play violent video games at all, we can predict that fewer of them will aggressively behave when provoked. 

Despite their advantages, correlational designs have a very important limitation. This limitation is that they cannot be used to draw conclusions about the causal relationships among the variables that have been measured. An observed correlation between two variables does not necessarily indicate that either one of the variables caused the other. Although many studies have found a correlation between the number of violent video games that people play and the amount of aggressive behaviors they engage in, this does not necessarily mean that viewing video games causes aggression. While one possibility is that playing violent games increases aggression,

another possibility is that the causal direction is exactly the opposite. Perhaps increased aggressiveness causes more interest in, and thus increased playing of, violent games. There is no way to rule out this possibility on the basis of the observed correlation.

Still, another possible explanation for the observed correlation is that it has been produced by the presence of another variable that was not measured in the research. Third variablesIn a correlational design, a variable that is not part of the research hypothesis but that causes the variables of interest to be correlated, thus producing a correlation between them. are variables that are not part of the research hypothesis, but that cause the variables of interest to be correlated, thus producing a correlation between them (refer to Figure 1.5). It has been observed that students who sit in the front of a large class get better grades than those who sit in the back of the class. Although this could be because sitting in the front causes the student to take better notes or to understand the material better, the relationship could also be due to a common causal variable, such as the interest or motivation of the students to do well in the class. Because a student’s interest in the class leads them both to get better grades and sit nearer to the instructor, seating position and class grade are correlated, even though neither caused the other.

The possibility of third variables must always be taken into account when considering the results of correlational research designs. For example, in a study that finds a correlation between playing violent video games and aggression, it is possible that a third variable is producing the relationship. Some possibilities include family background, diet, and hormone levels of the children. Any or all of these potential third variables might be creating or contributing to the observed correlation between playing violent video games and aggression. Higher levels of testosterone, for instance, may cause children to both have a greater interest in playing violent video games and behave more aggressively.

Think of third variables in correlational research designs as “mystery” variables since their presence and identity are not usually known to the researcher because they have not been measured. Because it is not possible to measure every variable that could possibly cause both variables, it is always possible that there is an unknown third variable. For this reason, we are left with the basic limitation of correlational research: Correlation does not imply causation.

The goal of much research in social psychology is to understand the causal relationships among variables, and for this, we use experiments. Experimental research designsResearch that includes the manipulation of one or more variables followed by the measurement of one or more other variables. are research designs that include the manipulation of one or more variables followed by the measurement of one or more other variables.

In an experimental research design, the variables of interest are called the independent variables and the dependent variables. The independent variableIn an experiment, the conceptual variable manipulated by the researcher. refers to the conceptual variable manipulated by the researcher, and the dependent variableIn an experiment, the conceptual variable that is measured after the manipulation of the independent variable or variables has occurred. refers to the conceptual variable that is measured after the manipulation of the independent variable or independent variables has occurred. In an experimental research design, the research hypothesis is that the manipulated independent variable (or variables) causes changes in the measured dependent variable (or variables). We can diagram the prediction like this, using an arrow that points in one direction to demonstrate the expected direction of causality:

Consider one experiment designed to directly test the hypothesis that playing violent video games would cause increased aggressive behavior. In this experiment, college students were given a chance to play either a violent video game (operationally defined as playing Wolfenstein 3D for fifteen minutes) or a nonviolent video game (operationally defined as playing Myst for fifteen minutes). During the experimental session, the participants played the video game that they had been given for fifteen minutes. Then after the play, they participated in a competitive task with another student in which they had a chance to deliver blasts of white noise through the earphones of their opponent. The operational definition of the dependent variable (aggressive behavior) was the level and duration of noise delivered to the opponent. The design and the results of the experiment are shown in Figure 1.6.

Experimental designs have two very nice features. For one, they guarantee that the independent variable occurs prior to measuring the dependent variable. This eliminates the possibility of reverse causation. Second, the experimental manipulation allows ruling out the possibility of third variables that cause both the independent variable and the dependent variable. In experimental designs, the influence of common-causal variables is controlled and thus eliminated by creating equivalence among the participants in each of the experimental conditions before the manipulation occurs.

The most common method of creating equivalence among the experimental conditions is through random assignment to conditionsThe most common method of creating equivalence among the experimental conditions before the experiment begins., which involves determining separately for each volunteer which conditions they will experience through a random process, such as drawing numbers out of an envelope or using a random number generator. In our example, researchers first randomly assigned about 100 participants to each of their two groups. Let’s call them Group A and Group B. Because they used random assignment to conditions, they could be confident that before the experimental manipulation occurred, the students in Group A were, on average, equivalent to the students in Group B on every possible variable, including variables that are likely to be related to aggression, such as family, peers, hormone levels, and diet—and, in fact, everything else.

Then after they had created initial equivalence, the researchers created the experimental manipulation—they had the participants in Group A play the violent video game and the participants in Group B the nonviolent video game. Then they compared the dependent variable (the white noise blasts) between the two groups and found that the students who had viewed the violent video game gave significantly longer noise blasts than those who had played the nonviolent game. Because they had created initial equivalence between the groups, when the researchers observed differences in the duration of white noise blasts between the two groups after the experimental manipulation, they could draw the conclusion that it was the independent variable (and not some other variable) that caused these differences. The only thing that was different between the students in the two groups was which video game they had played.

When we create a situation in which the groups of participants are expected to be equivalent before the experiment begins, when we manipulate the independent variable before we measure the dependent variable, and when we change only the nature of independent variables between the conditions, then we can be confident that it is the independent variable that caused the differences in the dependent variable. Such experiments are said to have high internal validity, where internal validityThe extent to which changes in the dependent variable in an experiment can confidently be attributed to changes in the independent variable. refers to the confidence with which we can draw conclusions about the causal relationship between the variables.

Despite the advantage of determining causation, experimental research designs do have limitations. One is that the experiments are usually conducted in laboratory situations rather than in the everyday lives of people. Therefore, we do not know whether the results that we find in a laboratory setting will necessarily hold up in everyday life. To counter this, in some cases, experiments are conducted in everyday settings—for instance, in schools or other organizations. Such field experimentsExperimental research that is conducted in a natural environment, such as a school or a factory. are difficult to conduct because they require a means of creating random assignment to conditions, and this is frequently not possible in natural settings.

A second and perhaps more important limitation of experimental research designs is that some of the most interesting and important social variables cannot be experimentally manipulated. Suppose we want to study the influence of the size of a mob on the destructiveness of its behavior, or to compare the personality characteristics of people who join suicide cults with those of people who do not join suicide cults. In that case, these relationships must be assessed using correlational designs because it is simply not possible to manipulate mob size or cult membership.

Social psychological experiments are frequently designed to simultaneously study the effects of more than one independent variable on a dependent variable. Factorial research designsExperimental research designs that use two or more independent variables. are experimental designs that have two or more independent variables. By using a factorial design, the scientist can study the influence of each variable on the dependent variable (known as the main effects of the variables) as well as how the variables work together to influence the dependent variable (known as the interaction between the variables). Factorial designs sometimes demonstrate the person-by-situation interaction.

In one such study, researchers tested the hypothesis that exposure to aggression-related words would increase aggressive responses toward others. Although they did not directly manipulate the social context, they used a technique common in social psychology in which they primed (i.e., activated) thoughts relating to social settings. In their research, half of their participants were randomly assigned to see words relating to aggression, and the other half were assigned to view neutral words that did not relate to aggression. The participants in the study also completed a measure of individual differences in agreeableness—a personality variable that assesses the extent to which the person sees themselves as compassionate, cooperative, and high on other-concern.

Then the research participants completed a task in which they thought they were competing with another student. Participants were told that they should press the space bar on the computer as soon as they heard a tone over their headphones, and the person who pressed the button the fastest would be the winner of the trial. Before the first trial, participants set the intensity of a blast of white noise that would be delivered to the loser of the trial. The participants could choose an intensity ranging from 0 (no noise) to the most aggressive response (10 or 105 decibels). In essence, participants controlled a “weapon” that could be used to blast the opponent with aversive noise, and this setting became the dependent variable. At this point, the experiment ended.

As you can see in Figure 1.7, there was a person-by-situation interaction. Priming with aggression-related words (the situational variable) increased the noise levels selected by participants who were low on agreeableness, but priming did not increase aggression (in fact, it decreased it a bit) for students who were high on agreeableness. In this study, the social situation was important in creating aggression, but it had different effects for different people.

No matter how carefully it is conducted or what type of design is used, all research has limitations. Any given research project is conducted in only one setting and assesses only one or a few dependent variables. And any study uses only one set of research participants. But we can only have confidence in the relationships between variables when those or similar relationships are found when tested using other research designs, with other operational definitions of the variables, with other participants, by other experimenters, and in other times and settings.

External validityThe extent to which the results of a research design can be generalized beyond the specific settings and participants used in the experiment to other places, people, and times. refers to the extent to which the results of a research design can be generalized beyond the specific settings and participants used in the experiment to other places, people, and times. Science relies primarily upon replicationThe repeating of research.—that is, the repeating of research—to study the external validity of research findings. Sometimes, the original research is replicated exactly, but more often, replications involve using new operational definitions of the independent or dependent variables or designs in which new conditions or variables are added to the original design. And to test whether a finding is limited to the particular participants used in a given research project, scientists may test the same hypotheses using people from different ages, backgrounds, or cultures. Replication allows scientists to test the external validity as well as the limitations of research findings.

In some cases, researchers may test their hypotheses not by conducting their own study but rather by looking at the results of many existing studies, using a meta-analysisA statistical procedure in which the results of existing studies are integrated to draw new conclusions about a research hypothesis.—a statistical procedure in which the results of existing studies are combined to determine what conclusions can be drawn on the basis of all the studies considered together. In one meta-analysis of twenty-one longitudinal studies, researchers found that violent game-playing teenagers between the ages of fourteen and sixteen were most likely to exhibit aggressive behavior. The older violent video game players were, the less likely they were to behave aggressively. The summary information gained through a meta-analysis allows researchers to draw even clearer conclusions about the external validity of a research finding.

It is important to realize that the understanding of social behavior that we gain by conducting research is a slow, gradual, and cumulative process. The research findings of one scientist or one experiment do not stand alone—no one study “proves” a theory or a research hypothesis. Rather, research is designed to build on, add to, and expand the existing research that has been conducted by other scientists. That is why whenever a scientist decides to conduct research, he or she first reads journal articles and book chapters describing existing research in the domain and then designs his or her research on the basis of the prior findings. The result of this cumulative process is that over time, research findings are used to create a systematic set of knowledge about social psychology (refer to Figure 1.8).