Module 2 - Methodology: How Social Psychologists Do Research PDF

Module 2 - Methodology: How Social Psychologists do research Site: New Era University Virtual Learning Environment Printed by: AUGUSTO M. RAMOS Course: PSY315-18 - Social Psychology Date: Thursday, 19 September 2024, 6:24 PM Book:...

Module 2 - Methodology: How Social Psychologists do research Site: New Era University Virtual Learning Environment Printed by: AUGUSTO M. RAMOS Course: PSY315-18 - Social Psychology Date: Thursday, 19 September 2024, 6:24 PM Book: Module 2 - Methodology: How Social Psychologists do research  Description Lesson 1: Title  Table of contents 1. Introduction/Overview 1.1. Learning Outcomes 2. Lesson 1: How do social psychologists go about testing their theories? 3. Lesson 2: Forming and Testing Theories 4. Lesson 3: Research Designs 4.1. The Observational Method 4.2. Correlational Research Methods 4.3. Experimental Research Methods 4.4. Field Experiements 4.5. Replications and Meta-Analysis 5. Lesson 4: Ethics of Experimentation 6. Concepts to Remember 7. Discussion Forum  1. Introduction/Overview Before we discuss how social psychological research is done, we begin with a warning: The results of some of the experiments you encounter will seem obvious because social psychology concerns topics with which we are all intimately familiar—social behavior and social influence (Richard, Bond, & Stokes-Zoota, 2001). This familiarity sets social psychology apart from other sciences. From the first module, we learned about the hindsight bias or the I-knew-it-all-along a phenomenon which explains that we tend to deceive ourselves into thinking that we know and knew events which are predictable or obvious more than we do and did. And that is precisely why we need science to help us sift reality from illusion and genuine predictions from easy hindsight, and the reason why social psychologists conduct research.  1.1. Learning Outcomes At the end of the module, student should be able to: realize how social psychologist test their theories familiarize the characteristics of a good theory understand the concept of laboratory and field research and its methodologies recognize the importance of ethics in conducting social psychology experiment  2. Lesson 1: How do social psychologists go about testing their theories? Social psychologists are of course interested in “big” phenomena. What causes intergroup conflict? Why do people stereotype members of other groups? How do people form impressions of other people? Why an individual behave differently when they are in a group? What leads people to change their attitudes? What factors influence whether close relationships succeed or fail? To answer these questions, social psychologists develop theories. For example, a social psychologist might want to develop a theory about the causes of intergroup conflict. In the first instance, this involves identifying constructs (abstract concepts, such as ‘threat’ or ‘prejudice’) or variables (a measurable representation of a construct, such as scores on questionnaire measures of threat perceptions or intergroup hostility) which might be relevant to the question, and speculating about how these relate to one another. Crucially, theories typically consist of propositions about the causal relationships between constructs. In this way, social psychologists are not content with simply describing these ‘big’ phenomena; rather, they seek to explain them by identifying their antecedents. In developing a theory of intergroup conflict, the interest here is not only about what conflict is like, but also in what causes it and how it might be reduced. For example, we might theorize that intergroup conflict is caused by feelings that the interests or well-being of one’s own group are threatened by another group (see Branscombe, Ellemers, Spears, & Doosje, 1999). We might initially base our theories on observation of real – life events, on intuition, or on existing theories. But coming up with theories is only part of the story. Many other disciplines – such as philosophy, sociology, and anthropology – are concerned with the same issues and phenomena that interest social psychologists. What helps to distinguish social psychology – and psychology as a whole – from other disciplines mentioned earlier is the commitment to the scientific method through testing theories against evidence. The essential characteristic of a theory is “it must be testable”. This means that we should be able to derive specific predictions (or hypotheses) from the theory concerning the relationship between two or more constructs, and to gather evidence that could support or contradict those predictions.  3. Lesson 2: Forming and Testing Theories We might initially base our theories on observation of real – life events, on intuition, or on existing theories. But coming up with theories is only part of the story. Many other disciplines – such as philosophy, sociology, and anthropology – are concerned with the same issues and phenomena that interest social psychologists. What helps to distinguish social psychology – and psychology as a whole – from other disciplines mentioned earlier is the commitment to the scientific method through testing theories against evidence. The essential characteristic of a theory is “it must be testable”. This means that we should be able to derive specific predictions (or hypotheses) from the theory concerning the relationship between two or more constructs, and to gather evidence that could support or contradict those predictions. A theory is an integrated set of principles that explain and predict observed events. In everyday conversation, “theory” often means “less than fact” – a middle rung on a confidence ladder from guess to theory to fact. Thus, people may dismiss Charles Darwin’s theory of evolution as “just a theory.” Indeed, notes Alan Leshner (2005), chief officer of the American Association for the Advancement of Science, “Evolution is only a theory, but so is gravity.” People often respond that gravity is a fact—but the fact is that your keys fall to the ground when dropped. Gravity is the theoretical explanation that accounts for such observed facts. To a scientist, facts and theories are apples and oranges. Facts are agreed upon statements about what we observe. Theories are ideas that summarize and explain facts. “Science is built up with facts, as a house is with stones,” wrote the French scientist Jules Henri Poincaré, “but a collection of facts is no more a science than a heap of stones is a house.” Theories not only summarize but also imply testable predictions, called hypotheses. Hypotheses serve several purposes. First, they allow us to test a theory by suggesting how we might try to falsify it. Second, predictions give direction to research and sometimes send investigators looking for things they might never have thought of. Third, the predictive feature of good theories can also make them practical. A complete theory of aggression, for example, would predict when to expect aggression and how to control it. As pioneering social psychologist Kurt Lewin declared, “There is nothing so practical as a good theory.” Consider how this works. Suppose we observe that people who loot, taunt, or attack often do so in groups or crowds. We might therefore theorize that being part of a crowd, or group, makes individuals feel anonymous and lowers their inhibitions. How could we test this theory? Perhaps we could ask individuals in groups to administer punishing shocks to a hapless victim without knowing which member of the group was actually shocking the victim. Would these individuals, as our theory predicts, administer stronger shocks than individuals acting alone? We might also manipulate anonymity: Would people deliver stronger shocks if they were wearing masks? If the results confirm our hypothesis, they might suggest some practical applications. Perhaps police brutality could be reduced by having officers wear large name tags and drive cars identified with large numbers, or by videotaping their arrests—all of which have, in fact, become common practice in many cities. But how do we conclude that one theory is better than another? A good theory: effectively summarizes many observations makes clear predictions that we can use to confirm or modify the theory generate new exploration suggest practical applications When we discard theories, usually, it is not because they have been proved false. Rather, like old cars, they are replaced by newer, better models.  4. Lesson 3: Research Designs Social psychology is a scientific discipline with a well-developed set of methods for answering questions about social behavior. There are three types of methods: the observational method, the correlational method, and the experimental method. METHOD FOCUS QUESTION ANSWERED Observational Description What is the nature of the phenomena? Correlational Prediction From knowing X, can we predict Y? Experimental Causality Is variable X a cause of variable Y? Any of these methods could be used to explore a specific research question; each is a powerful tool in some ways and a weak tool in others. Part of the creativity in conducting social psychological research involves choosing the right method, maximizing its strengths, and minimizing its weaknesses. Here we discuss these methods in detail and try to provide you with a firsthand look at both the joy and the difficulty of conducting social psychological studies. The joy comes in unraveling the clues about the causes of interesting and important social behaviors, just as a sleuth gradually unmasks the culprit in a murder mystery. Each of us finds it exhilarating that we have the tools to provide definitive answers to questions philosophers have debated for centuries. At the same time, as seasoned researchers, we have learned to temper this exhilaration with a heavy dose of humility, because there are formidable practical and ethical constraints involved in conducting social psychological research.  4.1. The Observational Method Let’s now see how social psychology is done. Understanding the logic of research can also help you think critically about everyday social events and better understand studies you see covered in the media. There is a lot to be learned by being an astute observer of human behavior. If the goal is to describe what a particular group of people or type of behavior is like, the observational method is very helpful. This is the technique whereby a researcher observes people and records measurements or impressions of their behavior. The observational method may take many forms, depending on what the researchers are looking for, how involved or detached they are from the people they are observing, and how much they want to quantify what they observe. ETHNOGRAPHY This is the method by which researchers attempt to understand a group or culture by observing it from the inside, without imposing any preconceived notions they might have. The goal is to understand the richness and complexity of the group by observing it in action. This is the chief method of cultural anthropology, the study of human cultures and societies. This is increasingly being used to describe different cultures and generate hypotheses about psychological principles (Fine & Elsbach, 2000; Flick, 2014; Uzzel, 2000). Consider this example from the early years of social psychological research: In the early 1950s, a small cult of people called the Seekers predicted that the world would come to an end with a giant flood on the morning of December 21, 1954. They were convinced that a spaceship from the planet Clarion would land in the backyard of their leader, Mrs. Keech, and whisk them away before the apocalypse. Assuming that the end of the world was not imminent, Leon Festinger and his colleagues thought it would be interesting to observe this group closely and chronicle how they reacted when their prophecy was disconfirmed (Festinger, Riecken, & Schachter, 1956). To monitor the hour-to-hour conversations of this group, the social psychologists found it necessary to become members and pretend that they too believed the world was about to end. On the fateful morning of December 21, 1954, with no flood waters lapping at the door and no sign of a spaceship, they observed a curious thing: Rather than admitting that she was wrong, Mrs. Keech “doubled down” on her beliefs, announcing that God had spared Planet Earth because of the Seekers’ faith, and that it was now time for the group to go public and recruit more members. Based on his observations of Mrs. Keech’s tenacious adherence to her beliefs. The key to ethnography is to avoid imposing one’s preconceived notions on the group and to try to understand the point of view of the people being studied. Sometimes, however, researchers have a specific hypothesis that they want to test using the observational method. An investigator might be interested, for example, in how much aggression children exhibit during school recesses. In this case, the observer would be systematically looking for particular behaviors that are concretely defined before the observation begins. For example, aggression might be defined as hitting or shoving another child, taking a toy from another child without asking, and so on. The observer might stand at the edge of the playground and systematically record how often these behaviors occur. If the researcher were interested in exploring possible sex and age differences in social behavior, he or she would also note the child’s gender and age. How do we know how accurate the observer is? In such studies, it is important to establish interjudge reliability, which is the level of agreement between two or more people who independently observe and code a set of data. By showing that two or more judges independently come up with the same observations, researchers ensure that the observations are not the subjective, distorted impressions of one individual. ARCHIVAL ANALYSIS The observational method is not limited to observations of real-life behavior. The researcher can also examine the accumulated documents, or archives, of a culture, a technique known as an archival analysis (Mullen, Rozell, & Johnson, 2001; Oishi, 2014). For example, diaries, novels, suicide notes, popular music lyrics, television shows, movies, magazine and newspaper articles, advertising, social media, and the ways in which people use the Internet all tell us a great deal about human behavior. One study, for example, analyzed millions of Twitter messages sent in 84 countries to examine daily rhythms in people’s mood. Judging by the content of the messages they send, most people’s positive moods appear to peak at two different times of the day: In the morning, soon after they get up, and late in the evening, before they go to bed (Golder & Macy, 2011). LIMITS TO OBSERVATIONAL METHOD The study that analyzed Twitter messages revealed interesting daily patterns, but it did not say much about why moods peak in the morning and at night. Furthermore, certain kinds of behavior are difficult to observe because they occur only rarely or only in private. You can begin to see the limitations of the observational method. Had Latané and Darley chosen this method to study the effects of the number of bystanders on people’s willingness to help a victim, we might still be waiting for an answer, given the infrequency of emergencies and the difficulty of predicting when they will occur. And, archival data about pornography, while informative about who is accessing it, tells us little about the effects on their attitudes and behavior of doing so. Social psychologists want to do more than just describe behavior; they want to predict and explain it. To do so, other methods are more appropriate.  4.2. Correlational Research Methods A goal of social science is to understand relationships between variables and to be able to predict when different kinds of social behavior will occur. With the correlational method, two variables are systematically measured, and the relationship between them—how much you can predict one from the other—is assessed. People’s behavior and attitudes can be measured in a variety of ways. Just as with the observational method, researchers sometimes make direct observations of people’s behavior. For example, researchers might be interested in testing the relationship between children’s aggressive behavior and how much violent television they watch. They too might observe children on the playground, but here the goal is to assess the relationship, or correlation, between the children’s aggressiveness and other factors, such as TV viewing habits, which the researchers also measure. Researchers look at such relationships by calculating the correlation coefficient, a statistic that assesses how well you can predict one variable from another—for example, how well you can predict people’s weight from their height. A positive correlation means that increases in the value of one variable are associated with increases in the value of the other variable. Height and weight are positively correlated; the taller people are, the more they tend to weigh. A negative correlation means that increases in the value of one variable are associated with decreases in the value of the other. If height and weight were negatively correlated in human beings, we would look very peculiar; short people, such as children, would look like penguins, whereas tall people, such as NBA basketball players, would be all skin and bones! It is also possible, of course, for two variables to be completely unrelated, so that a researcher cannot predict one variable from the other. SURVEYS The correlational method is often used in surveys, research in which a representative sample of people are asked questions about their attitudes or behavior. Surveys are a convenient way to measure people’s attitudes; for example, people can be telephoned and asked which candidate they will support in an upcoming election or how they feel about a variety of social issues. Psychologists often use surveys to help understand social behavior and attitudes—for example, by seeing whether the amount of pornography men say they read is correlated with their attitudes toward women. Surveys have a number of advantages, one of which is allowing researchers to judge the relationship between variables that are difficult to observe. Survey researchers go to great lengths to ensure that the people they test are typical. They select samples that are representative of the population on a number of characteristics important to a given research question (e.g., age, educational background, religion, gender, income level). They also make sure to use a random selection of people from the population at large, which is a way of ensuring that a sample of people is representative of a population by giving everyone in the population an equal chance of being selected for the sample. As long as the sample is selected randomly, we can assume that the responses are a reasonable match to those of the population as a whole. The great strength of correlational research is that it tends to occur in real-world settings where we can examine factors such as race, gender, and social status—factors that we cannot manipulate in the laboratory. Its great disadvantage lies in the ambiguity of the results. This point is so important that even if it fails to impress people the first 25 times they hear it, it is worth repeating a 26th time: Knowing that two variables change together (correlate) enables us to predict one when we know the other, but correlation does not specify cause and effect. A potential problem with survey data is the accuracy of the responses. Straightforward questions, regarding who people intend to vote for or what they typically do, are relatively easy to answer. But asking survey participants to predict how they might behave in some hypothetical situation or to explain why they behaved as they did in the past is an invitation to inaccuracy (Schuman & Kalton, 1985; Schwarz, Groves, & Schuman, 1998). Often people simply don’t know the answer—but they think they do. LIMITS OF THE CORRELATIONAL METHOD Correlation Does Not Equal Causation The major shortcoming of the correlational method is that it tells us only that two variables are related, whereas the goal of the social psychologist is to identify the causes of social behavior. We want to be able to say that A causes B, not just that A is correlated with B. If a researcher finds that there is a correlation between two variables, it means that there are three possible causal relationships between these variables. For example, researchers have found a correlation between the amount of violent television children watch and how aggressive they are. One explanation of this correlation is that watching TV violence causes kids to become more violent themselves. It is equally probable, however, that the reverse is true: that kids who are violent to begin with are more likely to watch violent TV. Or there might be no causal relationship between these two variables; instead, both TV watching and violent behavior could be caused by a third variable, such as having neglectful parents who do not pay much attention to their kids. When using the correlational method, it is wrong to jump to the conclusion that one variable is causing the other to occur. Correlation does not prove causation. Unfortunately, forgetting this adage is one of the most common methodological errors in the social sciences. Consider a study of birth control methods and sexually transmitted diseases (STDs) in women (Rosenberg, Davidson, Chen, Judson, & Douglas, 1992). The researchers examined the records of women who had visited a clinic, noting which method of birth control they used and whether they had an STD. Surprisingly, the researchers found that women who relied on condoms had significantly more STDs than women who used diaphragms or contraceptive sponges. This result was widely reported in the popular press, with the conclusion that the use of diaphragms and sponges caused a lower incidence of disease. Some reporters urged women whose partners used condoms to switch to other methods. Can you see the problem with this conclusion? The fact that the incidence of disease was correlated with the type of contraception women used is open to a number of causal interpretations. Perhaps the women who used sponges and diaphragms had sex with fewer partners. (In fact, condom users were more likely to have had sex with multiple partners in the previous month.) Perhaps the partners of women who relied on condoms were more likely to have STDs than were the partners of women who used sponges and diaphragms. There is simply no way of knowing. Thus, the conclusion that the birth control methods protected against STDs cannot be drawn from this correlational study.  4.3. Experimental Research Methods The only way to determine causal relationships is with the experimental method. Here, the researcher systematically orchestrates the event so that people experience it in one way (e.g., they witness an emergency along with other bystanders) or another way (e.g., they witness the same emergency but are the sole bystander). The experimental method is the method of choice in most social psychological research, because it allows the experimenter to make causal inferences. The difficulty of discerning cause and effect among naturally correlated events often prompts social psychologists to create laboratory simulations of everyday processes whenever this is feasible and ethical. INDEPENDENT AND DEPENDENT VARIABLES Social psychologists experiment by constructing social situations that simulate important features of our daily lives. By varying just one or two factors at a time— called independent variables—the experimenter pinpoints their influence. To illustrate the laboratory experiment, consider an experiment that offers a cause–effect explanation of the correlation between television viewing and children’s behavior. The more violent television children watch, the more aggressive they tend to be. So, are children learning and reenacting what they see on the screen? As we hope you now recognize, this is a correlational finding. Social psychologists have therefore brought television viewing into the laboratory, where they control the amount of violence the children see. By exposing children to violent and nonviolent programs, researchers can observe how the amount of violence affects behavior. Chris Boyatzis and colleagues (1995) showed some elementary school children, but not others, an episode of the most popular— and violent—children’s television program of the 1990s, Power Rangers. Immediately after viewing the episode, the viewers committed seven times as many aggressive acts per 2-minute interval as the non- viewers. The observed aggressive acts we call the dependent variable. Such experiments indicate that television can be one cause of children’s aggressive behavior. The variable a researcher measures to see if it is influenced by the independent variable the researcher hypothesises that the dependent variable will depend on the level of the independent variable Every social psychological experiment has two essential ingredients. We have just considered one—control. We manipulate one or more independent variables while trying to hold everything else constant. The other ingredient is random assignment. INTERNAL VALIDITY IN EXPERIMENTS How can we be sure that the differences in help across conditions in the Latané and Darley seizure study were due to the different numbers of bystanders who witnessed the emergency? Could something else have produced this effect? This is the beauty of the experimental method: We can be sure of the causal connection between the number of bystanders and helping, because Latané and Darley made sure that everything about the situation was the same in the different conditions except for the independent variable—the number of bystanders. Keeping everything but the independent variable the same in an experiment, making sure that nothing besides the independent variable can affect the dependent variable; this is accomplished by controlling all extraneous variables and by randomly assigning people to different experimental conditions is referred to as internal validity. There is a technique that allows experimenters to minimize differences among participants as the cause of the results: random assignment to condition. This is the process whereby all participants have an equal chance of taking part in any condition of an experiment; through random assignment, researchers can be relatively certain that differences in the participants’ personalities or backgrounds are distributed evenly across conditions. Even with random assignment, however, there is the (very small) possibility that different characteristics of people did not distribute themselves evenly across conditions. For example, if we randomly divide a group of 40 people into two groups, it is possible that those who know the most about epilepsy will by chance end up more in one group than in the other—just as it is possible to get more heads than tails when you flip a coin 40 times. This is a possibility we take seriously in experimental science. The analyses of our data come with a probability level (p-value), which is a number, calculated with statistical techniques, that tells researchers how likely it is that the results of their experiment occurred by chance and not because of the independent variable. The convention in science, including social psychology, is to consider results significant (trustworthy) if the probability level is less than 5 in 100 that the results might be due to chance factors rather than the independent variables studied. For example, if we flipped a coin 40 times and got 40 heads, we would probably assume that this was very unlikely to have occurred by chance and that there was something wrong with the coin (we might check the other side to make sure it wasn’t one of those trick coins with heads on both sides!). Similarly, if the results in two conditions of an experiment differ significantly from what we would expect by chance, we assume that the difference was caused by the independent variable (e.g., the number of bystanders present during the emergency). The p-value tells us how confident we can be that the difference was due to chance rather than the independent variable. To summarise, the key to a good experiment is to maintain high internal validity, which we can now define as making sure that the independent variable, and only the independent variable, influences the dependent variable. This is accomplished by controlling all extraneous variables and by randomly assigning people to different experimental conditions (Campbell & Stanley, 1967). When internal validity is high, the experimenter is in a position to judge whether the independent variable causes the dependent variable. This is the hallmark of the experimental method that sets it apart from the observational and correlational methods: Only the experimental method can answer causal questions, such as whether exposure to pornography causes men to commit violent acts. EXTERNAL VALIDITY IN EXPERIMENTS For all the advantages of the experimental method, there are some drawbacks. By virtue of gaining enough control over the situation so as to randomly assign people to conditions and rule out the effects of extraneous variables, the situation can become somewhat artificial and distant from real life. External validity is the extent to which the results of a study can be generalized to other situations and other people. Note that two kinds of generalizability are at issue: the extent to which we can generalize from the situation constructed by an experimenter to real-life situations, referred to as generalizability across situations, and the extent to which we can generalize from the people who participated in the experiment to people in general, referred to as generalizability across people. Social psychologists attempt to increase the generalizability of their results by making their studies as realistic as possible. But this is hard to do in a laboratory setting in which people are placed in situations they would rarely, if ever, encounter in everyday life. Instead, psychologists attempt to maximize the study’s psychological realism, which is the extent to which the psychological processes triggered in an experiment are similar to psychological processes that occur in everyday life (Aronson, Wilson, & Brewer, 1998). Even though Latané and Darley staged an emergency that in significant ways was unlike those encountered in everyday life, was it psychologically similar to real-life emergencies? Were the same psychological processes triggered? Did the participants have the same types of perceptions and thoughts, make the same types of decisions, and choose the same types of behaviors that they would in a real-life situation? If so, the study is high in psychological realism and we can generalize the results to everyday life. Psychological realism is heightened if people feel involved in a real event. To accomplish this, experimenters often tell participants a cover story—a disguised version of the study’s true purpose. Recall, for example, that Latané and Darley told people that they were studying the personal problems of college students and then staged an emergency. It would have been a lot easier to say to people, “Look, we are interested in how people react to emergencies, so at some point during this study we are going to stage an accident, and then we’ll see how you respond.” We think you’ll agree that such a procedure would be very low in psychological realism. In real life, we never know when emergencies are going to occur, and we do not have time to plan our responses to them. If participants knew that an emergency was about to happen, the kinds of psychological processes triggered would have been quite different from those of a real emergency, reducing the psychological realism of the study.  4.4. Field Experiements In a field experiment, researchers study behavior outside the laboratory, in its natural setting. As in a laboratory experiment, the researcher controls the occurrence of an independent variable (e.g., group size) to see what effect it has on a dependent variable (e.g., helping behavior) and randomly assigns people to the different conditions. Thus, a field experiment has the same design as a laboratory experiment, except that it is conducted in a real-life setting rather than in the relatively artificial setting of the laboratory. The participants in a field experiment are unaware that the events they experience are in fact an experiment. The external validity of such an experiment is high, because, after all, it is taking place in the real world, with real people who are more diverse than a typical college student sample. Many such field studies have been conducted in social psychology. For example, Latané and Darley (1970) tested their hypothesis about group size and bystander intervention in a convenience store outside New York City. Two “robbers” (with full knowledge and permission of the cashier and manager of the store) waited at the checkout counter until there were either one or two other customers approaching to get in line. They then asked the cashier to retrieve the most expensive beer the store carried. The cashier said he would have to check in the back to see how much of that brand was in stock. While the cashier was gone, the robbers picked up a case of beer in the front of the store, declared, “They’ll never miss this,” put the beer in their car, and drove off. Because the robbers were rather burly fellows, no one attempted to intervene directly to stop the theft. The question was, when the cashier returned, how many people would help by telling him that a theft had just occurred? As it turned out, the number of bystanders had the same inhibiting effect on helping behavior as in the laboratory seizure study: Significantly fewer people reported the theft when there was another customer-witness in the store than when they were alone. It might have occurred to you to ask why researchers conduct laboratory studies at all, given that external validity is so much better with field experiments. Indeed, it seems to us that the perfect experiment in social psychology would be one that was conducted in a field setting with a sample randomly selected from a population of interest and with extremely high internal validity (all extraneous variables controlled, people randomly assigned to the conditions). Sounds good, doesn’t it? The only problem is that it is very difficult to satisfy all these conditions in one study, making such studies virtually impossible to conduct. There is almost always a trade-off between internal and external validity—that is, between being able to randomly assign people to conditions and having enough control over the situation to ensure that no extraneous variables are influencing the results, and making sure that the results can be generalized to everyday life. We have the most control in a laboratory setting, but the laboratory may be unlike real life. Real life can best be captured by doing a field experiment, but it is very difficult to control all extraneous variables in such studies. For example, the astute reader will have noticed that Latané and Darley’s (1970) beer theft study differed from laboratory experiments in an important way: People could not be randomly assigned to the alone or in-pairs conditions. Were this the only study Latané and Darley had performed, we could not be sure whether the kinds of people who prefer to shop alone, as compared to the kinds of people who prefer to shop with a friend, differ in ways that might influence helping behavior. By randomly assigning people to conditions in their laboratory studies, Latané and Darley were able to rule out such alternative explanations. The trade-off between internal and external validity has been referred to as the basic dilemma of the social psychologist (Aronson & Carlsmith, 1968). The way to resolve this dilemma is not to try to do it all in a single experiment. Most social psychologists opt first for internal validity, conducting laboratory experiments in which people are randomly assigned to different conditions and all extraneous variables are controlled; here there is little ambiguity about what is causing what. Other social psychologists prefer to maximize external validity by conducting field studies. And many social psychologists do both. Taken together, both types of studies meet the requirements of our perfect experiment.  4.5. Replications and Meta-Analysis Replications are the ultimate test of an experiment’s external validity. Only by conducting studies in different settings, with different populations, can we determine how generalizable the results are. Often, though, when many studies on one problem are conducted, the results are somewhat variable. Several studies might find an effect of the number of bystanders on helping behavior, for example, while a few do not. How can we make sense of this? Does the number of bystanders make a difference or not? Fortunately, there is a statistical technique called meta-analysis that averages the results of two or more studies to see if the effect of an independent variable is reliable. Earlier we discussed p-values, which tell us the probability that the findings of one study are due to chance or to the independent variable. A meta-analysis essentially does the same thing, except that it averages the results of many different studies. If, say, an independent variable is found have an effect in only 1 of 20 studies, the meta-analysis will tell us that that one study was probably an exception and that, on average, the independent variable is not influencing the dependent variable. If an independent variable is having an effect in most of the studies, the meta-analysis is likely to tell us that, on average, it does influence the dependent variable. A handful of unreliable findings, some from researchers who committed fraud by faking data, have raised concerns about the reproducibility of medical and psychological research. Although “mere replications” of others’ research are unglamorous—they seldom make headline news—today’s science is placing greater value on replication studies. Researchers must precisely explain their stimuli and procedures so that others can match them. And we now expect them to file their methods and their detailed data in a public, online, “open science” archive (Brandt et al., 2014; Miguel et al., 2014). In recent years, efforts to reproduce studies—13 studies in one project, 100 in another—have produced both successful and failed replications (Anderson et al., 2016; Gilbert et al., 2016; Klein et al., 2014; Open Science, 2015). Amid the scientific debate, all agree that replication is important.  5. Lesson 4: Ethics of Experimentation Our television example illustrates why experiments can raise ethical issues. Social psychologists would not, over long periods, expose one group of children to brutal violence. Rather, they briefly alter people’s social experience and note the effects. Sometimes the experimental treatment is a harmless, perhaps even enjoyable, experience to which people give their knowing consent. Occasionally, however, researchers find themselves operating in a gray area between the harmless and the risky. Social psychologists often venture into that ethical gray area when they design experiments that engage intense thoughts and emotions. Experiments do not need to have mundane realism (Aronson et al., 1985). That is, laboratory behavior need not be like everyday behavior, which is typically mundane, or unimportant. But the experiment should have experimental realism—it should engage the participants. Experimenters do not want participants consciously play-acting or bored, they want to engage real psychological processes. An experiment on aggression. Forcing people to choose whether to give intense or mild electric shock to someone else can be a realistic measure of aggression. It functionally simulates real aggression. Achieving experimental realism sometimes requires deceiving people with a plausible cover story. If the person in the next room is actually not receiving the shocks, the experimenter does not want the participants to know that. That would destroy the experimental realism. Thus, approximately one-third of social psychological studies in past decades used deception (Korn & Nicks, 1993; Vitelli, 1988). Researchers often walk a tightrope in designing experiments that will be involving yet ethical. To believe that you are hurting someone, or to be subjected to strong social pressure, may be temporarily uncomfortable. Such experiments raise the age-old question of whether ends justify means. Do the risks exceed those we experience in everyday life (Fiske & Hauser, 2014)? The social psychologists’ deceptions are usually brief and mild compared with many misrepresentations in real life and in some of television’s reality shows. (One network reality TV series deceived women into competing for the hand of a handsome supposed millionaire, who turned out to be an ordinary laborer.) University ethics committees review social psychological research to ensure that it will treat people humanely and that the scientific merit justifies any temporary deception or distress. Ethical principles developed by the American Psychological Association (2010), the Canadian Psychological Association (2000), and the British Psychological Society (2009) mandate investigators to Tell potential participants enough about the experiment to enable their informed consent. Be truthful. Use deception only if essential and justified by a significant purpose and not “about aspects that would affect their willingness to participate.” Protect participants (and bystanders, if any) from harm and significant discomfort. Treat information about the individual participants confidentially. Debrief participants. Fully explain the experiment afterward, includingany deception. The only exception to this rule is when the feedback would be distressing, such as by making participants realize they have been stupid or cruel. The experimenter should be sufficiently informative and considerate that people leave feeling at least as good about themselves as when they came in. Better yet, the participants should be compensated by having learned something (Sharpe & Faye, 2009). When treated respectfully, few participants mind being deceived (Epley & Huff, 1998; Kimmel, 1998). Indeed, say social psychology’s advocates, professors provoke far greater anxiety and distress by giving and returning course exams than researchers provoke in their experiments.  6. Concepts to Remember social psychology The scientific study of how people think about, influence, and relate to one another. theory An integrated set of principles that explain and predict observed events. hypothesis A testable proposition that describes a relationship that may exist between events. field research Research done in natural, real-life settings outside the laboratory. correlational research The study of the naturally occurring relationships among variables. experimental research Studies that seek clues to cause–effect relationships by manipulating one or more factors (independent variables) while controlling others (holding them constant). independent variable The experimental factor that a researcher manipulates. dependent variable The variable being measured, so called because it may depend on manipulations of the independent variable. replication Repeating a research study, often with different participants in different settings, to determine whether a finding could be reproduced. random assignment The process of assigning participants to the conditions of an experiment such that all persons have the same chance of being in a given condition. (Note the distinction between random assignment in experiments and random sampling in surveys. Random assignment helps us infer cause and effect. Random sampling helps us generalize to a population.) mundane realism Degree to which an experiment is superficially similar to everyday situations. experimental realism Degree to which an experiment absorbs and involves its participants. informed consent An ethical principle requiring that research participants be told enough to enable them to choose whether they wish to participate. Module 2 References: Hewstone, Stroebe, & Jonas. An Introduction to Social Psychology (5th Edition) Aronson, Wilson, Akert, Sommers. Social Psychology (9the Edition)  7. Discussion Forum Click the link to access the discussion form https://college2425.neu.edu.ph/mod/forum/view.php?id=112763&group=3853