Stats final stuff
- 1. Conformity at Hope College Samuel Fishman Gabriela Olaguibel
- 2. Summary • Recreation of Asch’s Conformity Experiment at Hope College • Unanimous confederates state incorrect answers • Subject is either conformist or nonconformist to their point of views • Conformity Questionnaire • Levels based on amounts of people pressuring to conform
- 4. Purpose • Test the effects of social influences on a person, and the respective person’s gender on conformity. • Conformity is modification in behavior that happens as result of real or perceived pressure from a group. • Social influence is the impact that others have on a person.
- 5. Literature Review Asch, (1955) Opinions and Social Pressure : Psychology lab of famous study on conformity. Our study is loosely based off of Asch’s experiment. Latane, (1981) The Psychology of Social Impact : Outlines Social Impact Theory. Uses equation I=F(SIN) do describe social impact forces. As numbers increase, impact forces exponentially increase. Myers, (2009) Social Psychology : Conformity increases with number but there are decreasing returns after five people. Informational influence happens when people go along with a group in order to be correct. Normative influence occurs when people go along with group in order to fit in. Cacioppo, Petty (1980) Sex Differences in Influenceability: Towards Specifying the Underlying Processes : Women are more susceptible to social influence than men are.
- 6. Research Question #1 • How much does a person think he or she will conform depending on amount of confederates exerting pressure? • Hypothesis: A person thinks that a larger amount of confederates present exert a greater amount of influence (in terms of conformity). • ANOVA
- 7. Research Question #2 • Are people more likely to conform in small or large groups? • Hypothesis: People are more likely to conform in large groups. • Chi-Square
- 8. Research Question #3 • Does conformity vary with between gender? • Hypothesis: There is a difference in levels of conformity between gender. • Chi-Square
- 10. Participants • Wyckoff, Scott, and Phelps Halls • Convenience Sample • Confederates are asked previously • Exclude Psychology majors and minors • All subjects are subjected to control and manipulated variable
- 11. Setting • Wyckoff, Scott, and Phelps Halls • Dorm room • Chairs are set up in semicircle • 10 pm start
- 12. Materials • Agreement survey: determines gender, agreement, and name- these were thrown out in order to ensure anonymity • Pictures of circles: 1A, 1B, 2A, 2B (circles are of obvious size difference)
- 13. Variables • Gender: dichotomous and qualitative • Confederates: qualitative and continuous (also divided into control and manipulated levels) • Incorrect answers: qualitative and continuous
- 14. Procedure • Recruit and instruct confederates and prepare material and setting • Recruit subjects • Instructions for subjects • Round 1: Control (no conformity affects) • Round 2: Manipulated (test conformity affects)
- 15. Chi-Square • Experiment • Population: Wyckoff, Scott, Phelps Hall • Sample: Residents • N = 46 • Ho: There is no relationship between where a the number of confederates and incorrect answers. • Ha: There is a relationship between the number of confederates and incorrect answers.
- 16. How does the number of social influences affect conformity? IV: Number of Confederates (3 or 7) DV: Incorrect Answers (Correct(0) or Incorrect(1)) Real Conclusion: The Null is plausible (p-value=.475). The number of social influences does not have an Influence on incorrect answers.
- 17. Fathom • Failed Assumption: less than 5 in 2 cells Test of Collection 1 Goodness of Fit Attribute: (categorical): Incorrect Function Plot Test of Collection 1 Count 0 0.35 0 30 0.30 Incorrect 1 10 0.25 Column Summary 40 0.20 Ho: Categories of Incorrect are equally likely Number of categories: 2 0.15 Chi-square: 9.478 0.10 DF: 1 P-value: 0.0021 0 0.05 0.00 0 2 4 6 8 10 12 14 16 chi-square y = chiSquareDensity ( ,df ) x
- 18. Does the Number of Confederates Affect Conformity? NO
- 19. Chi-Square • Experiment • Population: Phelps and Scott Halls • Sample: Residents • N = 31 H0: There is no relationship between gender and incorrect answers. HA: There is relationship between gender and incorrect answers.
- 20. How does the number of social influences affect conformity? IV: Gender (Male or Female) DV: Incorrect Answers (Correct(0) or Incorrect(1)) Real Conclusion: The Null is plausible (p-value=.458). Gender does not have an influence on incorrect answers.
- 21. Fathom • Failed assumption: less than 5 in 1 cell Test of Collection 2 Goodness of Fit Test of Collection 2 Function Plot Attribute: (categorical): Incorrect 0 0.35 Count 0.30 C 23 0.25 Incorrect I 8 0.20 Column Summary 31 0.15 Ho: Categories of Incorrect are equally likely Number of categories: 2 0.10 Chi-square: 7.258 0.05 DF: 1 P-value: 0.0071 0 0.00 0 2 4 6 8 10 12 14 16 chi-square y = chiSquareDensity ( , ) x df
- 22. Are male or females more likely to conform? No
- 23. Survey Information ANOVA
- 24. Participants • Wyckoff, Scott, and Phelps Halls • Random Sample • People in the hallways or with open rooms are asked if they would take the survey
- 25. Materials • Conformity Survey • Four Levels: 1, 3, 7, or 15 social influences • Questions are based off of conformity scenario with friends from college • Questions are on scale of how likely a person is to conform from 1 (very unlikely) through 5 (very likely)
- 26. Variables • Social Influences: qualitative • Likelihood of conformity: quantitative
- 27. Procedure • Ask students if they would fill out the survey • Read introduction script • Allow them to fill out survey • When they complete the survey, read them the debriefing script
- 28. ANOVA • Survey • Population: Wyckoff, Scott, Phelps Halls • Sample: Residents • N=18 • Ho: There is no significant mean difference in conformity among those with numbers of social influences. µ1=µ3=µ7=µ15 • Ha: µ1 ≠ µ3; µ1 ≠ µ7; µ1 ≠ µ15; µ3 ≠ µ7; µ3 ≠ µ 15; µ7 ≠ µ15
- 29. Does a Greater Number of Influences Make One More Likely to Conform? IV: Influences (7 group, 3 group, 7 control group, 3 control group) DV: Incorrect Answers P O o Conclusion: The r null is rejected in favor of the alternative (p-value=.000). p There is a difference in the likelihood of o conformity when there are different numbers of influences.
- 30. Assumptions + Normality: X Homogeneity of variance: + Outliers: X SD: .51131(2) < 1.45621 +Independence: + Limitation: Sample Size Failed!
- 31. ANOVA Fathom Mean Difference= Measures from Scrambled Collection 1 Histogram 140 4.83333 120 100 P-value < .001 80 60 40 20 0 1 2 3 4 5 6 Meandifferencepart1 PASW is confirmed!
- 32. Bias Reduction for Experiment • Shield subjects from true nature of experiment • Confidentiality for three days • Room obscurity/similarity • Confederates sign permission slips as well • Psych majors and minors are excluded • Use of scripts • Confederates confidentiality
- 33. Lurking Variables in Experiment • Lack of random sampling • Many people knew of Asch’s experiment before our study • Sample is only made up of three dorms • Gender of confederates probably played a role in conformity between gender • ELIMINATED: cases with errors (i.e. answered before confederates, etc).
- 34. Bias Reduction in Survey • People are not informed of topic before completing • Scenarios are similar • Use of scripts
- 35. Lurking Variables in Survey • Lack of random sampling • There may be underlying preference for certain scenarios even without number of social influences • Sample is only made up of three dorms
- 36. Conclusions for Hope College There is no significant difference in social conformity between gender among Hope students. There is no significant difference in conformity regarding number of social influences in real life situations. Hope students perceive that the presence of 7 social influences is the optimal number (of the four choices) of social influences for a high likelihood of conformity. All of this can only be attributed to Hope College.
- 37. Conformity in Real Life
- 38. What is True About Conformity? • It exists. • The number of influences does determine the amount of conformity as found in other experiments (our experiment did not back this up however). • Gender of social influences may play a role on subjects depending subjects’ gender (females influence females more and visa versa). • There are various types of conformity.
- 39. Is Conformity All Bad or Is it Sometimes Good? • Conformity is necessary for society to run smoothly. • Conformity can be negative or positive depending on the result and intention.
- 40. References Latane, B. (1981). The Psychology of Social Impact. American Psychologist, 36, 343-356. Asch, S. E. (1955). Opinions and Social Pressure. Scientific American, 31-35. Myers, D. G. (2009). Social Psychology (10th ed.). New York, NY: McGraw-Hill, 211, 215-216. Cacioppo, J. T., & Petty, R. E. (1980). Sex Differences in Influenceability: Toward Specifying the Underlying Processes. Personality and Social Psychology Bulletin, 6(4), 651-656.