This preview shows page 1. Sign up to view the full content.
Unformatted text preview: Chapter 2 Chapter Studying Behavior Studying Scientifically Scientifically How do We Get Knowledge about Behavior? Behavior?
Through: Common sense, family, reason, intuition, Common personal opinion, religion,….. personal OR OR Psychological scientific inquiry based on: Curiosity Skepticism Open-mindedness Steps in The Scientific Process
(Gathering Evidence) (Gathering 1. 1. 2. 3. 4. 5. 6. Observation (curiosity) (curiosity) Hypothesis: tentative explanation; (If-Then) tentative Research (Test hypothesis) Research (Test Analyze data (draw tentative conclusions) (draw Report findings (further research) Theory building (broader than hypotheses) (broader New hypotheses (derived from theory) (derived Example: Kitty Genovese incident (Darley & Latane) (Darley Example: Two Approaches to Studying Behavior Behavior Hindsight (after-the-fact understanding): most common method; might seem reasonable; can provide most insight, but multiple explanations are possible insight, Prediction, Control, and Theory Building Prediction, Theory A good theory:
• Organizes existing info clearly & meaningfully with existing k/l • Is testable • Its predictions are supported by new research • Conforms to the law of parsimony Conforms law Defining & Measuring Variables Defining Variable: any characteristic or factor that can Variable: vary (e.g. gender, IQ, age, religion,….) vary (e.g. Operational definition: defines variable in Operational terms of the specific procedures used to produce or measure it; relies on: produce Self-report measures & Reports by others (social(socialdesirability) Observations of Behavior (direct, unobtrusive, unobtrusive
archival) archival Psychological Tests (IQ, personality, neuropsychological) Physiological Measures (heart rate, blood pressure) (heart Ethical Principles to Research Research Five Ethical Principles Five
1. 1. 1. 1. 1. Beneficence Responsibility Integrity Justice: (enhancing access) (enhancing Respect Ethics in Human Research Ethics Informed Consent: before people agree to participate in research, they should be informed about: about:
Study’s purpose, procedures Study’s potential benefits Study’s potential risks Right to decline participation & withdraw at any time Whether responses will be confidential & if not, how privacy Whether will be safeguarded will When and how do we use deception? When deception No other feasible alternative available Benefits outweigh costs Debriefing will take place Research Methods Research Descriptive Research… Descriptive
(Recording Events) Case Study:
in-depth analysis of individual/group/event
• • • • Rare phenomenon studied closely Challenge validity of theory or widely held scientific belief Source of new ideas to examine thru other research tools Good for evaluating treatments or training programs BUT: 1. Cause-effect relationship poorly determined (low chance to Cause-effect rule out alternative explanations) rule 2. Generalizability problems Generalizability problems 3. Possible lack of objectivity (in data collection & interpretation) …Descriptive Research… Naturalistic Observation:
observe behavior as it occurs in natural setting without trying to change it (no intrusion) trying Non-participant vs. Participant
1. 2. Clues of possible origins of human behavior Observe children BUT: BUT: No clear causal conclusions Many influencing variables can’t be disentangled Possible observer bias Mere presence may disrupt behavior (habituation) …Descriptive Research Survey Research:
Information gathered thru questionnaires or interviews to many participants (large data, inexpensive) participants Population Sample
• Random sampling BUT:
Cause & Effect conclusions not possible Social desirability bias, interviewer bias, or inaccurate Social perceptions of own behavior, poorly phrased items perceptions Unrepresentative sample faulty generalizations Occasionally, & by chance, sample may not be Occasionally, representative (<5%) representative Correlational Research Correlational 1. 1. 2. 3. Measuring associations between naturally Measuring occurring events occurring
Measure one variable X (hrs of TV watching) Measure second variable Y (grades) Statistically determine whether X and Y are related Correlation coefficient (direction & strength): -1/ +1 (direction
BUT Bidirectionality (two-way causality) interpretative Bidirectionality problem problem Third-variable problem (Z) Cannot draw causal conclusions, but can predict Experimental Methods… Experimental Optimal for cause & effect relations Optimal cause
1. 2. 3. Manipulate one/more variables (groups) Measure manipulation influence on other variables Attempt to control extraneous factors What is the LOGIC behind that?? …Experimental Methods…
• • I. II. Independent Variable (cause, manipulated) (cause, Dependent Variable (effect, depends on IV) (effect, Experimental Group: receives treatment or active level of IV active Control Group: standard for comparison Two designs: − Random assignment − Counterbalancing Experimental Methods - Example Experimental Question: Question: Does TV violence cause aggression? Does 1. Observation 1. • Problem ... Correlational Experimental Methods - Example 2. Experiment 2.
Between - Groups Design OR OR Within – Group Design Experimental Methods - Example Identify variables Identify Independent Dependent Manipulated Measured Experimental Methods Group 1 Watch violent TV show
Measure Aggression Experimental Methods Group 2 Watch nonviolent TV show
Measure Aggression Experimental Methods Group 3 Watch No TV nonviolent (Control) TV show
Measure Aggression Experimental Methods- Example Was that a Between-Groups or Within-Groups Design? • A Within - Groups Design means: All participants exposed to all
conditions Experimental Methods - Example In a within-groups experiment:
All watch violent shows All Measure aggression Measure All watch non-violent shows Measure aggression Problem? Experimental Methods - Example Potential order effect order Counterbalance Experimental Methods - Example How to Measure Aggression? How (Operational Definition) • Self-report • Verbal attack • Physical attack Threats To Validity Threats Internal Validity: clear causal conclusions supported confidence that IV caused variation in DV Confounding Variables: what influenced DV (music effect on mood) effect Demand Characteristics: cues about expected behavior behavior Placebo Effects: expectations influence behavior expectations Placebo: substance that has no pharmacological effect substance Experimenter Expectancy effects: subtle & unintentional unintentional Double-blind procedure (blind to experimental condition) Double-blind Generalizing the Findings Generalizing
How do we know? External validity: the degree to which results can be generalized to other populations, settings, & conditions Replication: can original findings be duplicated (Cross-cultural replication is duplicated increasingly popular) increasingly Importance of Critical Thinking Importance Be smart & critical when reading statistics Be (evaluate evidence) (evaluate Critical thinking does not imply finding fault; it implies evaluation or judgment. implies Why Do We Need Statistics? Why Descriptive Statistics: to summarize & describe the characteristics of a set (or distribution) of data distribution) Describing Data thru Statistics Describing Measures of Central Tendency (typical score) (typical
- Mode (most frequently occurring score; may not be Mode
representative) representative) - Median (midpoint; not affected by extreme scores) (midpoint; from every score most commonly used measure) - Mean (average; affected by extreme scores, but captures info Measures of Variability (whether scores cluster
together or vary widely) together - Range (simplest but least informative) Range (simplest - Standard Deviation (how much each score differs from the Standard
mean) Making Inferences Thru Statistics Making Inferential statistics: Drawing conclusions about Drawing results being due to a mere chance occurrence Statistical significance @ p<.05: Only < 5% of the times may the results have occurred by chance Meta-Analysis Meta-Analysis Statistical procedure for combining results of Statistical different studies that examine the same topic different Each study treated as a “single participant” Considered by many as most objective way to Considered integrate findings of multiple studies & reach overall conclusions about behavior overall Question: A psychologist is testing the effectiveness of a weight loss program. One group receives a diet and counseling program. A second group receives the diet, but not the counseling program. Subjects are weighed weekly for 6 months. What is the independent variable? Choose one answer:
a. b. c. d. e. The six month time frame. Average weekly weight. The diet program. The counseling program. The subjects. Correct answer is:
a. b. c. d. e. The six month time frame. Average weekly weight. The diet program. The counseling program. The subjects. Question: Dr. Jung argues that increases in temperature cause increases in aggression. To support his claim, he recorded the temperature each day and noted the number of reported assaults. More assaults were reported on hotter days. His claim is inaccurate because: Choose an answer:
a. b. c. d. e. The data are based on self-report. The data are correlational. This is a case study. The variables are controlled. Assaults are negatively related to aggression. Correct answer is:
a. b. c. d. e. The data are based on self-report. The data are correlational. This is a case study. The variables are controlled. Assaults are negatively related to aggression. Which statement is a FACT & which is an OPINION? OPINION Women just can’t seem to be happy Women unless they are in a relationship with a man. man. Females scored significantly higher than Females males on every measure administered in the study of gender and affiliation. the Which statement is a FACT & FACT which is an OPINION? OPINION Women are naturally more empathic than Women men—it’s just part of their nature. men—it’s Women who put up with battering from Women men can’t have a very high opinion of themselves. themselves. Which statement is a FACT & which is an OPINION? OPINION Jealousy causes more problems in human Jealousy relations than just about anything else you can think of. can Jealousy and overall anger were Jealousy significantly positively correlated in males but unrelated in females. but Experimental, Descriptive, or Correlational Design? Correlational Students will learn better in a cool room than in a Students hot one Students do better on exams with spaced Students reviews than with last minute cramming reviews Individuals who were abused as children are Individuals more likely to become abusive in dating relationships relationships Drinking a cup of coffee before an exam will Drinking improve performance improve It is easier to remember words that rhyme than It words that are completely unrelated words Chapter 2 Studying Behavior Scientifically Research Methods ...
View Full Document
This note was uploaded on 02/07/2011 for the course PSYC 202 taught by Professor Dialalawand during the Fall '07 term at American University of Beirut.
- Fall '07