Ap psy methods, stats
- 1. Chapter 1: Thinking Critically with Psychological Science Aaron Biggar AP Psychology
- 2. I. The Need for Psychological Science
- 3. 98% Certainty 1. The area of the US in square miles? 2. The population of Afghanistan in 2001? 3. American battle deaths in Spanish- American War? 4. GM advertising budget in 1999? 5. Female officers in US Army in 2000?
- 4. 98% Certainty 1. Area of US: 3.6 million sq. miles 2. Afghan population: 26.8 million 3. Battle deaths: 385 4. GM advertising: $2.9 billion 5. Female officers: 9.953
- 5. I. The Need for Psychological Science A. Limits of intuition and common sense 1. Hindsight bias 2. Overconfidence
- 6. I. The Need for Psychological Science A. Limits of intuition and common sense B. Scientific attitude C. Scientific method 1. Theory & hypothesis 2. Operational definitions 3. Replication 4. Goals:
- 8. II. Description A. Case studies B. Surveys (questionnaires & interviews) 1. Wording effects
- 9. 72% interested in “plants & trees” 39% interested in “botany” 62% whites agree “problems faced by blacks brought on by blacks themselves” (white interviewer) 46% whites agree with black interviewer A majority oppose an amendment “prohibiting abortions” A majority support “protecting the life of an unborn child”
- 10. 1. Is the Mississippi River longer or shorter than 500 miles? 2. How many miles long is it?
- 11. 1. Is the Mississippi River longer or shorter than 3000 miles? 2. How many miles long is it?
- 12. II. Description A. Case studies B. Surveys (questionnaires & interviews) 1. Wording effects 2. Samples & populations a. False consensus & other biases b. Random & representative techinques
- 14. II. Description A. Case studies B. Surveys (questionnaires & interviews) C. Naturalistic observation & observer effects
- 15. III. Correlation
- 16. III. Correlation A. Positive & negative B. Scatterplots
- 18. III. Correlation A. Positive & negative B. Scatterplots C. Correlation coefficient (r)
- 19. III. Correlation A. Positive & negative B. Scatterplots C. Correlation coefficient (r) D. Causation v. predictive value Correlation between TV watching & GPA
- 23. III. Correlation A. Positive & negative B. Scatterplots C. Correlation coefficient (r) D. Causation v. predictive value E. Illusory correlation F. Order in random events
- 25. IV. Experimentation Purpose of an expt: General hypothesis: food effects learning Specific (operationalized) hypothesis: students who eat an oatmeal raisin cookie before class each day will have higher average scores on the semester final than students who don’t eat a cookie.
- 26. Eating cookies before class each day will lead to higher average scores. Variables: Independent (IV) Controlled by experimenter The “cause” variable Dependent (DV) Predicted by experimenter The “effect” variable
- 27. Eating cookies before class each day will lead to higher average scores. What if kids get cookies and A’s? Groups (conditions): to establish different levels of the IV Experimental group Exposed to IV Get cookie Control group Not exposed to IV No cookie
- 28. Eating cookies before class each day will lead to higher average scores. Confounding IV DV Variables Expt. Gp. Cookie 95% Environmental Cntrl. Gp. No Cookie 82% Expectations Individual differences
- 29. Random Sampling To select participants from population Allows you generalize results Random Assignment To divide participants into groups Controls confounding variables
- 30. Eating cookies before class each day will lead to higher average scores. IV DV Expt. Gp. Cookie 95% Cntrl. Gp. No Cookie 82% 85% 93%
- 31. Statistical Significance p value likelihood a result is caused by chance can be no greater than 5% p ≤ .05
- 32. Replication Non-replicated results are preliminary. Linus Pauling (1970). Vitamin C prevents colds. IV DV Expt. Gp. Vit C Expt. Gp. 45% Cntrl. Gp. Placebo Fewer colds
- 33. Experimental Design Terms • Hypothesis • Operational definitions • Participant selection • IV & DV • Experimental & control groups • Confounding variables • Random assignment • Placebo control • Double blind procedure • Statistical significance (p value) • Replication
- 34. Quasi-experiments • Men are better drivers than women. • Cigarette smoking causes lung cancer.
- 36. V. Statistical Reasoning A. Describing data 1. Frequency distributions
- 37. Multiple Choice Composite Essay 13 A+ 40 13 A+ 12 A 39 4 41% A 12 23 52% 12 A 11 39% 38 11 A- 11 10 11 A- 14 11 A- 37 11 B+ 10 15 10 B+ 9 10 B+ 36 6 B 9 5 41% 9 B 12 45% 9 B 35 4 31% B- 8 6 8 B- 8 34 5 C+ 7 2 7 C+ 2 8 B- 33 5 C 6 1 5% 6 C+ 2 11% 7 C+ 32 4 C- 5 5 C- 3 6 C 31 3 19% D+ 4 2 4 D+ 30 3 D 3 3% 3 D 3 5% 5 C- 29 2 D- 2 2 D- 4 D+ 28 F 1 0% 1 F 0% 3 D 27 4 8% 0 26 1 Mean=9.3 2 D- 25 Mean=10.2 SD=2.3 1 F 24 2% SD=2.0 <24 1 Mean=34.3 SD=4.2
- 38. V. Statistical Reasoning A. Describing data 1. Frequency distributions 2. Histograms & frequency polygons
- 39. Histogram 50 Percentage of students 40 30 20 10 0 A B C D Grades
- 40. Frequency Polygon 50 Percentage of students 40 30 20 10 0 A B C D Grades
- 41. V. Statistical Reasoning A. Describing data B. Measures of central tendency Mode=most common=4 4 Mean=arithmetic average=20/5=4 3 5 Median=middle score=4 4 4
- 42. Central Tendency 1968 TOPPS Baseball Cards Nolan Ryan $1500 Elston Howard $2.25 Billy Williams $8 Jim Bouton $2 Luis Aparicio $5 Rocky Colavito $2 Harmon Killebrew $5 Boog Powell $2 Orlando Cepeda $3.50 Luis Tiant $2 Maury Wills $3.50 Tim McCarver $1.75 Jim Bunning $3 Tug McGraw $1.75 Tony Conigliaro $3 Joe Torre $1.5 Tony Oliva $3 Rusty Staub $1.25 Lou Pinella $3 Curt Flood $1 Mickey Lolich $2.50 With Ryan: Without Ryan: Median=$2.50 Median=$2.38 Mean=$74.14 Mean=$2.85
- 44. V. Statistical Reasoning A. Describing data B. Measures of central tendency C. Measures of variation 1. Range 2. Variance & standard deviation
- 45. Standard Deviation Punt Deviation Deviation Distance from Mean Squared 36 -4 16 std. dev.= 38 -2 4 variance= 41 +1 1 45 +5 25 11.5=3.4 yds Mean= 46 160/4= 40 yds 46/4=11.5= variance
- 46. V. Statistical Reasoning A. Describing data B. Measures of central tendency C. Measures of variation D. Characteristics of the normal curve
- 47. V. Statistical Reasoning A. Describing data B. Measures of central tendency C. Measures of variation D. Characteristics of the normal curve E. Inference 1. Does the sample represent the pop.? a. Non-biased sample-good b. Low variability-good c. Larger samples-good
- 48. V. Statistical Reasoning E. Inference 1. Does the sample represent the pop.? 2. Are differences between groups statistically significant? a. Big differences-good b. Low variability-good c. Big groups-good
- 49. VI. FAQs
- 50. VI. FAQs A. Can lab expts illuminate real life? B. Does behavior depend on one’s culture? C. Does behavior vary with gender? D. Why do psychologists study animals? E. Is it ethical to experiment on animals? F. Is it ethical to experiment on people? G. Is psychology free of value judgments? H. Is psychology potentially dangerous?
- 51. VI. FAQs F. Is it ethical to experiment on people? Yes, if the APA’s ethical guidelines are followed. 1. Informed consent 2. Protection from harm 3. Confidentiality 4. Debriefing
- 52. How should participants be chosen? • Good answers use and make clear your understanding of the terms “sample” and “population.” • Good answers explain how to draw a random sample. • Ex. “All the assembly line workers for a company could be the population. I would draw a random sample of 40 of them by picking names from a hat.”
- 53. Create operationalized hypothesis • Good answers make an educated guess (“Workers who are complimented will work harder”) • Good answers develop specific, measurable ways to define compliments and hard work. • Ex. “10 compliments will be given at 4 specific times.” “Supervisors will give compliments from a pre-developed list.” “Hard work will be measured by using the number of items produced each hour.”
- 54. Identify IV & DV • IV=compliments • DV=work effort • Do not propose a direction when stating the variable (“the DV is an increase in work.”) • People are never the variables (“the IV is the people who get complimented.”)
- 55. Describe the groups • Experimental group gets compliments. • Control group doesn’t get compliments. • Use the terms!
- 56. Confounding variables • Do not confuse random assignment with random sampling. • Be reasonable!!! • “The company policy will be to have people sleep in a hotel with no TV or radio.” • “I will provide a seminar on the importance of sleep.” • “All workers must be paid the same.”