1 Thinking Critically With Psychological Science

Transcription

1 CHAPTER 1 Thinking Critically With Psychological Thinking Critically Science With Psychological Science Preview The scientific attitde reflects an eagerness to skeptically scrtinize competing ideas with an openminded hmility before natre. This attitde, copled with scientific principles for sifting reality from illsion, prepares s to think criti cally. Three reliable phenomena hindsight bias, jdgmental overconfidence, and the tendency to perceive order in random events illstrate the limits of everyday intition and or need for scientific inqiry and critical thinking. Psychologists constrct theories that organize observations and imply testable hypotheses. In their research, they se case stdies, natralistic observation, and srveys to describe behavior; correlation to assess the relationship between variables; and experimentation to ncover case-effect relationships. Researchers se statistics to describe their data, to assess relationships between variables, and to determine whether differences are significant. This chapter concldes by briefly answering several qestions that stdents commonly ask abot psychology. These inclde concern over the simplification of reality in laboratory experiments, the generalizability of research in terms of cltre and gender, the prpose of animal stdies, the adeqacy of research ethics, and the potential misse of psychology s knowledge. NOTE: Several items in the Instrctor s Resorces (indicated here by a ) have alternative ses. See the Instrctor s Resorces for explanations of those ses. Introdctory Exercise: Fact or Falsehood? The correct answers to Handot 1 1 are as follows: 1. F 2. T 3. F 4. T 5. F 6. F 7. T 8. F 9. F 10. F 7

2 8 Chapter 1 Thinking Critically With Psychological Science HANDOUT 1 1 Fact or Falsehood? 1. Hman intition is remarkably accrate and free from error. 2. Most people seem to lack confidence in the accracy of their beliefs. 3. Most people sffer from nrealistically low self-esteem. 4. Research sggests that college stdents spend more than 25 percent of their waking hors in conversation. 5. The opinions of 1500 randomly selected people can provide a very accrate pictre of the opinions of an entire nation. 6. The scientific finding that depressed people tend to have low self-esteem proves that depression cases people to be down on themselves. 7. The prpose of the experiment is to re-create behaviors exactly as they occr in everyday life. 8. Under ethical and legal gidelines, researchers mst ensre the comfort, health, and hmane treatment of animals. 9. As a science, psychology is objective and vale-free. 10. The wealthiest 20 percent in the United States possess 58 percent of the wealth.

3 Chapter 1 Thinking Critically With Psychological Science 9 Gide Learning Objectives Every qestion in the Test Banks is keyed to one of these objectives. The Need for Psychological Science 1-1. Explain how or everyday thinking sometimes leads s to a wrong conclsion Explain how the three main components of the scientific attitde relate to critical thinking. Research Strategies: How Psychologists Ask and Answer Qestions 1-3. Describe how theories advance psychological science Describe how psychologists se case stdies, natralistic observations, and srveys to observe and describe behavior, and explain why random sampling is important Explain what it means when we say two things are correlated, and describe positive and negative correlations Define regression toward the mean Discss why correlations enable prediction bt not case-effect explanation Describe the characteristics of experimentation that make it possible to isolate case and effect Discss whether laboratory conditions can illminate everyday life Explain why psychologists stdy animals, and describe the ethical gidelines that safegard hman and animal research participants. Discss how hman vales inflence psychology. Statistical Reasoning in Everyday Life Explain how we describe data sing three measres of central tendency, and discss the relative seflness of the two measres of variation Explain how we know whether an observed difference can be generalized to other poplations. The Need for Psychological Science Why We Cannot Rely on Intition and Common Sense Lectres: Misremembering the Cases of Behavior; Extraordinary Events and Chance: Yor Birth Date in Pi?) Exercises: The Limits of Hman Intition; The Birthday Coincidence and Other Remarkable Facts; The Hindsight Bias and Predicting Research Otcomes; The Overconfidence Phenomenon; Stdent Overconfidence; Overconfidence and the Confirmation Bias; The Gambler s Fallacy Exercise/Project: The Propensity Effect 1-1. Explain how or everyday thinking sometimes leads s to a wrong conclsion. Intition is an effortless, immediate, atomatic feeling or thoght, as contrasted with explicit, conscios reasoning. The hindsight bias, also known as the I-knew-it-all-along phenomenon, is the tendency to believe, after learning an otcome, that one wold have foreseen it. Finding ot that something has happened makes it seem inevitable. Ths, after learning the reslts of a stdy in psychology, they may seem to be obvios common sense. However, experiments have fond that events seem far less obvios and predictable beforehand than in hindsight. Sometimes, psychological findings even jolt or common sense. Or everyday thinking is also limited by or tendency to think we know more than we do. Asked how sre we are of or answers to factal qestions, we tend to be more confident than correct. Experts predictions of world events made with 80 percent confidence were right less than 40 percent of the time. Another limitation in everyday thinking is the tendency to perceive order in random events becase random seqences often don t look random. In actal random seqences, patterns and streaks occr more often than people expect.

4 10 Chapter 1 Thinking Critically With Psychological Science The Scientific Attitde and Critical Thinking Lectres: Yor Teaching Strategies and Critical Thinking ; Critical Thinking Exercises: Exit Tickets (p. xx); Critical Inqiry and Psychology ; A Psychic Reading Project: Evalating Media Reports of Research 1-2. Explain how the three main components of the scientific attitde relate to critical thinking. The scientific attitde reflects a hard-headed criosity to explore and nderstand the world withot being fooled by it. The eagerness to skeptically scrtinize competing claims reqires hmility becase it means we may have to reject or own ideas. This attitde, copled with scientific principles for sifting reality from fantasy, helps s winnow sense from nonsense. It carries into everyday life as critical thinking in which we examine assmptions, appraise the sorce, discern hidden vales, evalate evidence, and assess conclsions. Research Strategies: How Psychologists Ask and Answer Qestions The Scientific Method Exercise: Astrology and the Scientific Method Projects/Exercises: Is Psychology a Science? Testing Proverbs PsychSim 6: Understanding Psychological Research LanchPad: Research Methods 1-3. Describe how theories advance psychological science. A sefl theory is an explanation sing an integrated set of principles that effectively organizes a wide range of observations and implies testable predictions, called hypotheses. By enabling s to test and reject or revise a particlar theory, sch predictions give direction to research. They specify in advance what reslts wold spport the theory and what reslts wold disconfirm it. As an additional check on their own biases, psychologists report their reslts precisely with clear operational definitions of concepts. Sch statements of the procedres sed to define research variables allow others to replicate, or repeat, their observations. Often, research leads to a revised theory that better organizes and predicts observable behaviors or events. Description Lectres: Case Stdies; The Power of Vivid Cases; Srveys, Evalation Apprehension, and Natralistic Observation; Predicting Elections Exercises: The Wording of Srvey Qestions; Condcting a National Srvey; An M&M s Sampling Demonstration Exercise/Lectre Break: Finding the Good and Bad in Case Stdies Project/Exercise: Natralistic Observation in the Dining Hall 1-4. Describe how psychologists se case stdies, natralistic observations, and srveys to observe and describe behavior, and explain why random sampling is important. The case stdy is the method by which psychologists analyze one or more individals in great depth in the hope of revealing things tre of s all. While individal cases can sggest fritfl ideas, any given individal may be atypical, making the case misleading. Natralistic observation consists of observing and recording behavior in natral environments. Like the case stdy and srvey methods, this research strategy describes behavior bt does not explain it. The srvey looks at many cases in less depth and asks people to report their behavior or opinions. Asking qestions is tricky becase even sbtle changes in the order or wording of qestions can dramatically affect responses. In everyday experience, we are tempted to generalize from a few vivid bt nrepresentative cases. The srvey ascertains the self-reported attitdes or behaviors of a poplation by qestioning a representative, random sample.

5 Chapter 1 Thinking Critically With Psychological Science 11 Correlation Lectres: Understanding Correlation ; Misinterpreting Correlations Exercises: Correlation and Predicting Exam Performance ; Correlating Test-Taking Time and Performance ; Positive and Negative Correlations ; Displaying Scatterplots; Illsory Correlation ; Illsory Correlation Demonstration ; The Power of Disconfirming Evidence: Do Dreams Predict the Ftre? Exercise/Project: Correlation Does Not Imply Casation PsychSim 6: Correlation Television Show: Home Simpson and Illstory Correlation LanchPad: Correlation and Casation 1-5. Explain what it means when we say two things are correlated, and describe positive and negative correlations. When srveys and natralistic observations reveal that one trait or behavior accompanies another, we say the two correlate. A correlation coefficient is a statistical measre of the relationship between two things (from 1.00 to +1.00). A positive correlation indicates a direct relationship, meaning that two things increase together or decrease together. A negative correlation indicates an inverse relationship: As one thing increases, the other decreases. Researchers depict scores on graphs called scatterplots; each point plots the vale of two variables. The correlation coefficient helps s to see the world more clearly by revealing the extent to which two things relate Define regression toward the mean. Noticing and recalling confirming instances of relationships reslts in illsory correlations. Illsory correlations feed an illsion of control that chance events are sbject to or personal control. The illsion that we can control ncontrollable events is fed by a statistical phenomenon called regression toward the mean the tendency for nsal events (inclding emotions) to retrn to their average state Discss why correlations enable prediction bt not case-effect explanation. Perhaps the most irresistible thinking error is to assme that correlation proves casation. Correlation reveals how closely two things vary together and ths how well one predicts the other. However, the fact that events are correlated does not mean that one cases the other. Ths, while correlation enables prediction, it does not provide explanation. Experimentation Exercise: Introdcing the Experiment; Random Assignment; Main Effects and Interactions or It All Depends Projects/Exercises: The Placebo Effect; Identifying Variables and Grops and Correlational Verss Experimental Stdies LanchPad: Does Self-Confidence Intimidate Others? 1-8. Desscribe the characteristics of experimentation that make it possible to isolate case and effect. The experiment is a research method in which the investigator maniplates one or more factors to observe their effect on some behavior or mental process, while controlling other relevant factors. If a behavior changes when we vary an experimental factor, then we know the factor is having a casal effect. In many experiments, control is achieved by randomly assigning people either to an experimental grop, which is exposed to the treatment, or a control grop, which is not exposed. Often, the research participants are blind (ninformed) abot what treatment, if any, they are receiving. One grop might receive the treatment, while the other grop receives an inert placebo (a psedotreatment). Often, both the participant and the research assistant who collects the data will not know which condition the participant is in (the doble-blind procedre). The placebo effect is well docmented. Jst thinking one is receiving treatment can lead to symptom relief.

6 12 Chapter 1 Thinking Critically With Psychological Science A variable is anything that varies. The independent variable is the experimental factor that is maniplated. It is the variable whose effect is being stdied. The dependent variable is the variable that may change in response to the maniplations of the independent variable. It is the otcome factor. Confonding variables are factors other than the independent variable that might prodce an effect. Lectres: Field and Laboratory Experiments 1-9. Discss whether laboratory conditions can illminate everyday life. The experimenter intends the laboratory experiment to be a simplified reality, one in which important featres can be simlated and controlled. The experiment s prpose is not to re-create the exact behaviors of everyday life bt to test theoretical principles. It is the reslting principles not the specific findings that help explain many behaviors. Exercise: Psychological Research Methods Psychology s Research Ethics Lectres: APA Gidelines for Ethical Condct in the Care and Use of Nonhman Animals in Research; Invasion of Privacy; Research Ethics Exercises: Animal Rights; Teaching Ethics in the Introdction to Psychology Corse: Research Methods LanchPad: Ethics in Animal Research: The Sad Case of Boee the Chimp; Ethics in Hman Research: Violating One s Privacy; Death of a Sbject: The Ethics of Mental Health Research Explain why psychologists stdy animals, and describe the ethical gidelines that safegard hman and animal research participants. Discss how hman vales inflence psychology. Some psychologists stdy animals ot of an interest in animal behaviors. Others do so becase knowledge of the physiological and psychological processes of animals enables them to better nderstand the similar processes that operate in hmans. The debate between animal protection organizations and researchers has raised two important isses: Is it right to place the well-being of hmans above that of animals? What safegards are in place to protect the well-being of animals in research? Many professional organizations and fnding agencies have developed extensive gidelines for the hmane se of animals. In their statements of ethical principles for the treatment of hman participants, the U.S. APA and Britain s BPS rge investigators to obtain informed consent, protect participants from harm and discomfort, treat information abot individals confidentially, and flly explain the research afterward (debrief them). Lectres: Psychology and Hman Vales ; The Instrctor s Perspectives and Vales Exercise: Observing Verss Interpreting Psychologists vales can inflence their choice of research topic, their theories and observations, their interpretations, and their professional advice. Knowledge is power that can be sed for good or evil. Applications of psychology s principles have so far been mostly for the good, and psychology addresses some of hmanity s greatest problems and deepest longings. Statistical Reasoning in Everyday Life Describing Data Lectre: The Case for Statistical Analysis Exercise: Teaching Statistical Concepts Exercise/Project: Describing Data Project: Organizing and Interpreting Data PsychSim 6: Descriptive Statistics

7 1-11. Explain how we describe data sing the three measres of central tendency, and discss the relative seflness of the two measres of variation. Researchers se descriptive statistics to organize their data meaningflly. Measres of central tendency neatly smmarize data. The mode is the most freqently occrring score in a distribtion. The mean is the arithmetic average of a distribtion, obtained by adding the scores and then dividing by the nmber of scores. If the distribtion is skewed by even a few extreme scores, the mean will be biased. The median is the middle score in a distribtion; half the scores are above it and half are below it. The range of scores the gap between the lowest and highest score provides only a rogh estimate of variation. The more standard measre of how scores deviate from one another is the standard deviation, which describes how mch scores vary arond the mean score. It better gages whether scores are packed together or dispersed becase it ses information from each score. Many types of scores are distribted along a bell-shaped distribtion, or a normal crve (normal distribtion). Roghly 68 percent of the cases fall within one standard deviation of the mean. Abot 95 percent fall within two standard deviations. Significant Differences Chapter 1 Thinking Critically With Psychological Science 13 Lectres: The Law of Large Nmbers and the Gambler s Rin ; Differences Between Grops Exercises: More Cases Are Better Than Fewer Explain how we know whether an observed difference can be generalized to other poplations. Important principles to remember in making generalizations inclde the following: a. Representative samples are better than biased samples. We are particlarly prone to overgeneralize from vivid cases at the extremes. b. Less-variable observations are better than those that are more variable. Averages are more reliable when derived from scores with low variability. c. More cases are better than fewer. Small samples provide less reliable estimates of the average than do large samples. Psychologists se tests of statistical significance to help them determine whether differences between two grops are reliable. When the averages of the samples drawn from the grops are reliable, and the difference between them is relatively large, we say the difference has statistical significance. This means that the difference very likely reflects a real difference and is not de to chance variation between the samples. Given large enogh or homogeneos enogh samples, a difference between them may be statistically significant yet have little practical significance.

8