Mandatory Prac 4 Automaticity of Social Behavior Direct

Mandatory Prac 4 Automaticity of Social Behavior: Direct Effects of Trait Construct and Stereotype Activation on Action Bargh, Chen, & Burrows, 1996 Gerry Tehan CRICOS QLD 00244 B | NSW 02225 M TEQSA: PRV 12081

The Scientific Method and the research process 1. What is the research question? • What are theoretical variables / constructs. • What are the operational variables 13. Has the hypothesis been confirmed 14. What is the answer to the research question. theory about how something works testing: comparing the observations with theory 10. What statistics do we use? 11. What is the value of the statistic and p-value 12. Is the p-value less than (<). 05 2. What are the hypotheses? 3. What do you expect to find if theory is correct generating predictions Scientific Method (what would theory lead you to observe? ) systematic empirical observations 6. Find Means, Standard Deviation 7. What type of graph should we use 8. Make appropriate Graph 9. Does the data conform with expectations 4. What type of investigation is being used 5. What are IVs and/or DV’s

The Scientific Method and the research process theory about how something works testing: comparing the observations with theory generating predictions Scientific Method systematic empirical observations (what would theory lead you to observe? )

Theories Conceptual structure that is supported by large and varied set of data. ” (Stanovich, 2004, p 21. ) Conceptual Structure = 2 concepts and their connection theory about how something works generating predictions testing: comparing the observations with theory systematic empirical observations (what would theory lead you to observe? ) Physical activity promotes psychological wellness. Economic prosperity is related to form of Government Climate change has an effect on global warming • Historians of science have argued that “good theories” tend to have the following qualities: (1) They are generative (2) They make precise (i. e. , risky) predictions (3) They can be unambiguously tested (falsifiable) (4) They are simple (parsimonious) (5) They have Good Track Records (previous predictions have been tested and supported by systematic observation)

Concepts to Variables Concepts: Variables: Operationalised Come from theory Used in the Experiment Physical activity promotes psychological wellness. Physical Activity Operationalised Psychological Wellness 10, 000 Steps per day 30 min Gym 3 time week Daily walk the dog Psychological Wellbeing Scale Beck Depression Inventory Enjoyment in walking the dog Walking 10, 000 steps a day will lead to higher scores on the Psychological Wellbeing scale. 30 minutes in the gym three times a week will lead to more enjoyment in walking the dog Daily walking the dog will lead to lower scores on the Beck Depression Inventory

Summary of Definitions Theory: Conceptual structure that is supported by large and varied set of data. Hypotheses are proposed relations between concepts (theoretical hypotheses) or variables (operational hypotheses) Variable: Any phenomenon that can vary along some dimension Continuous variable: varies continuously (body weight, height) Categorical variable: can take on fixed values (gender, political affiliation)

Types of Investigation There are two broad ways of investigating a research question: 1) By observing what naturally happens 2) By manipulating some aspect of the environment and observing what effect it has on another variable of interest. Correlation designs involve observing what naturally goes on in the world without directly interfering with it Experimental designs involves direct manipulation of one or more variables Type of investigation is indicated by the way in which the research question is framed

Research Question Relationships (positive, negative) Research Paradigm: Experimental IVs and DV Research Paradigm Correlational DVs Data Presentation A B C 5. 952381 3. 095238 8. 142857 4. 095238 3. 047619 7. 142857 Exp CTL 10 8 6 4 2 0 Differences (better than, less than) Exp CTL A B C 0. 809524 5. 761905 0. 428571 1. 857143 4. 571429 0. 904762 8 CTL Exp 6 CTL C 0 R 2 = 0, 2127 2 2 B 6 4 4 A r =. 41 0 A B C 0 Statistics 5 10 Statistics Parametric Non Parametric Independent Groups t-test Wilcoxon Signed Ranks Pearson Bivariate Spearman’s Rho Paired Sample Paired-sample ttest Mann-Whitney U Test

The Bargh et al. study

What is the Research Question For many years, social psychologists have studied the effects of priming on the individual's subsequent impressions of others. Priming refers to the incidental activation of knowledge structures, such as trait concepts and stereotypes, by the current situational context. Many studies have shown that the recent use of a trait construct or stereotype, carries over for a time to exert an unintended, passive influence on the interpretation of behavior (see Bargh, 1994; Higgins, 1989; Wyer&Srull, 1989, for reviews) We argue here that such passive, automatic effects of stereotype priming can be observed not only in the perception of behaviours but also in the enactment of active social behaviours. (not their words - my summary of their argument) Priming social stereotypes has an automatic effect on related stereotypical actions

Concepts to Variables Concepts: Come from theory Operationalised Variables: Used in the Experiment Priming social stereotypes has an automatic effect on related stereotypical actions Priming Social Stereotypes Scrambled Sentences Test Elderly Vs Neutral Operationalised Action Walking time There will be differences in walking time between those who were given “elderly” words on the Scrambled Sentences Test and those given neutral words on the Scrambled Sentences Test Walking times will be longer in elderly condition than neutral condition (Elderly > Neutral)

Research Question Differences Research Paradigm: Bargh Study: Experimental IVs and DV Data Presentation A B C 5. 952381 3. 095238 8. 142857 4. 095238 3. 047619 7. 142857 Exp CTL 10 8 6 4 2 0 Exp CTL A B C 0. 809524 5. 761905 0. 428571 1. 857143 4. 571429 0. 904762 8 CTL Exp 6 CTL 4 2 A B C 0 A B C Statistics Parametric Non Parametric Independent Groups t-test Wilcoxon Signed Ranks Paired Sample Mann-Whitney U Test Paired-sample t -test Expose people to either elderly stereotype of neutral stimuli Test Walking time.

Elements in an Experimental Design Experiments ask whether systematic variation in one variable produces variation in another variable Independent variable (IV) Manipulated by experimenter Has multiple levels Is categorical Dependent variable (DV): What is measured Participants response Is continuous Experiments investigate the effect of the IV on the DV. Are there differences on the DV between the levels of the IV

This experiment investigates the effect of priming stereotype on action Priming Stereotype: Elderly Vs Neutral Action: Multiple Choice, Short Answer Prime Action Does it Vary Yes Is it being Manipulated Yes No Does it have Multiple Levels 2 - Elderly & Neutral No Is it categorical Yes No Is it being Measured No Yes Is it the Participant Response No Yes Is it continuous No Yes IV, DV or neither IV DV

Research Question Differences Research Paradigm: Experimental Ivs and DV Data Presentation A B C 5. 952381 3. 095238 8. 142857 4. 095238 3. 047619 7. 142857 Exp CTL 10 8 6 4 2 0 Exp A B C 0. 809524 5. 761905 0. 428571 1. 857143 4. 571429 0. 904762 Exp CTL 8 CTL Exp 6 CTL 4 2 A B C 0 A B C Statistics Parametric Non Parametric Independent Groups t-test Wilcoxon Signed Ranks Paired Sample Mann-Whitney U Test Paired-sample t -test

Priming social stereotypes has an automatic effect on related stereotypical act Elderly Prime N=15 Neutral Prime N=15 30 – five word lists in scrambled order Set A: 30 elderly prime words Set B: 30 neutral words Walking Time (secs) Bargh et al. , 1998 9 8, 5 8 7, 5 7 6, 5 6 5, 5 5 4, 5 4 3, 5 3 Elderly Neutral Walk time: 9. 75 m Priming Stereotype: Elderly vs Neutral Action: Walking time 1. Do scrambled sentences 2. Debrief 3. Walk to elevator t (28) = 2. 86, p <. 05 Walking time slower in elderly prime condition.

Outcomes of Bargh study Walking times were longer in elderly condition than neutral condition (Elderly > Neutral) There were differences in walking time between those who were given “elderly” words on the Scrambled Sentences Test and those given neutral words on the Scrambled Sentences Test Priming social stereotypes has an automatic effect on related stereotypical actions

Mandatory practical: Use a correlational research design to investigate the relationship between stereotypes and behaviour by replicating the 1996 investigation by John Bargh, Mark Chen and Lara Burrows (Experiment 2). How do we replicate? Conceptual replication: Same theoretical variables, same theoretical hypothesis What aspects of study do we replicate: keep, change or modify some or all of the operational variables Keep all: direct replication Change some or all: extension

Mandatory practical: Use a correlational research design to investigate the relationship between stereotypes and behaviour by replicating the 1996 investigation by John Bargh, Mark Chen and Lara Burrows (Experiment 2). What are conceptual variables Stereotypes Behaviour (Actions) What are the operational variables Elderly related words Walking Time

Mandatory practical: Use a correlational research design to investigate the relationship between stereotypes and behaviour by replicating the 1996 investigation by John Bargh, Mark Chen and Lara Burrows (Experiment 2). Problem: Bargh et al. used an experimental design – we need to do a conceptual replication, but adapt to use a correlational design.

How to get from experimental paradigm to correlational paradigm Stereotype priming is unlikely to be uniform across participants. We can adopt an individual differences approach If stereotype priming has an effect on performance then those who show strong stereotype priming effects should also show similarly strong action effects That is, there is a positive relationship between stereotype priming effects and action effects

Research Question Relationships (positive, negative) Research Paradigm Correlational DVs Data Presentation r =. 41 6 R 2 = 0, 2127 4 2 0 0 5 10 Statistics Parametric Non Parametric Pearson Bivariate Spearman’s Rho

• Define the research question, state theoretical hypotheses, and determine the concepts to testing: comparing the observations with theory Priming social stereotypes has an automatic effect on actions Is stereotype priming related to action generating predictions (what would theory lead you to observe? ) systematic empirical observations Research Question Is there a relationship between stereotype priming effects and action effects Theoretical Hypothesis What about Stereotype priming What action Identify Concepts

• determine the variables and state the operational hypotheses testing: comparing the observations with theory Priming social stereotypes has an automatic effect on actions systematic empirical observations Generating risky predictions (what would theory lead you to observe? ) Is stereotype priming related to action Is there a relationship between stereotype priming effects and action effects What about Stereotype Priming What about the action Research Question Theoretical Hypothesis Identify Variables Turning Concepts into Variables Operationalisation Stereotype Priming: 24 -item Scrambled Sentences Action: Walking time over fixed distance Is there a relationship between scores on the scrambled sentences test and walking time. Those who have higher scores on the scrambled sentences task will have longer walking times. Opereational Hypothesis

Interlude #2 Design & Methodology Type of Investigation Experiments ask whether systematic variation in one variable produces variation in another variable. Does one variable have an effect on another Does exposure to different stereotype priming tasks produce differences on an action task. Correlational designs investigate whethere is a relationship between two or more variables. Is performance on a stereotype priming task related to performance on an action task

Interlude #3 Elements of an Experiments ask whether systematic variation in one variable produces variation in another variable Independent variable (IV) Manipulated by experimenter Has multiple levels Is categorical Dependent variable (DV): What is measured Participants response Is continuous Experiments investigate the effect of the IV on the DV

Correlational designs investigate whethere is a relationship between two or more variables. To what extent is the order of one set of scores (measured, participant response, continuous) maintained in a second set of scores (measured, participant response, continuous) from the same person

This research explores the relationship between stereotype priming and action Stereotype Priming: 24 -item Scrambled Sentences Action: Walking time over fixed distance Scrambled Sentences Walking Does it Vary Yes Is it being Manipulated No No Does it have Multiple Levels No No Is it categorical No No Is it being Measured Yes Is it the Participant Response Yes Is it continuous Yes IV, DV or neither DV DV

testing: comparing the observations with theory Priming social stereotypes has an automatic effect on actions systematic empirical observations As scores on Scrambled Sentence test increase walking time will increase As the priming scores on the scrambled sentences task increases (DV – Continuous), walking time will increase (DV – Continuous)

Gathering systematic empirical observations testing: comparing the observations with theory Priming social stereotypes has an automatic effect on actions systematic empirical observations As scores on Scrambled Sentence test increase walking time will increase Pre Experiment 1. Get Ethics approval 2. Have a rationale for people walking 3. Set up room with two tables 4. Make sure video camera can clearly record walking 5. 6. Find two rooms with corridor linking them Find an assistant to run the memory component of the study. Run the Experiment 1. Test people individually 2. Get informed consent to participate 3. Turn video to record session. 4. Present the cover story and then present the scrambled Sentences test 5. Place paper on one table 6. Walk to the second table to pick up response sheet 7. Present the memory test 8. Debrief and Score data 9. 10. 11. 12. Send participant to Room 2 Have confederate record walking time Confederate to present the Memory test. (optional: do square walking and stair climbing test)

Step 1: Summarise the Data Mean SD SEM Scrambled Sentence 4 7 4 8 2 6 7 9 6 11 9 5 7 8 6 7 5 7 9 5 10 6 6 Walking Time 4. 3 5. 3 4. 1 5. 5 4. 4 4. 5 5. 1 7. 2 5. 4 5. 0 4. 2 7. 4 4. 6 5. 0 4. 6 5. 1 4. 2 5. 0 5. 1 4. 1 5. 6 4. 4 6. 79 2. 03 0. 41 5. 10 0. 95 0. 19 Any set of numbers can be described by: 1. A measure of central tendency of which the mean is one 2. A measure of variability (dispersion) of which standard deviation and standard error of the mean are two.

Scrambled Sentence 4 7 4 8 2 6 7 9 6 11 9 5 7 8 6 7 5 7 9 5 10 6 6 Walking Time 4. 3 5. 3 4. 1 5. 5 4. 4 4. 5 5. 1 7. 2 5. 4 5. 0 4. 2 7. 4 4. 6 5. 0 4. 6 5. 1 4. 2 5. 0 5. 1 4. 1 5. 6 4. 4 Some hypothetical data Correlational designs are commonly more interested in the pairs of scores than in means and standard deviations Scrambled Sentence: Number of “old” words selected in the sentence (Max = 12) Walking Time: Time to transition between two points on the corridor

Step 2: Present Summary 8 Mean & SD Choice of “old” word Walking Time 6. 79 (2. 03) 1. 43 5. 10 (0. 95). 43 1 Can now assess the hypotheses of the study Walking Time (secs) 7 6 5 4 3 R 2 = 0, 1856 2 1 0 0 2 4 6 8 Choice of "old" word 10 12 As the priming scores on the scrambled sentences task increases walking time will increase Yes

Interlude #3 Hypothesis Testing & Statistics Those who have higher scores on the scrambled sentences task will have longer walking times Alternative Hypothesis What statistics are designed to actually test Scores on the scrambled sentences task has no bearing on walking times Null Hypothesis Can we reject the Null Hypothesis? If so we can accept the alternative hypothese. S

Interlude #1 Measurement Classification Type of Scale Objective Categorical Nominal Sort stimuli into discrete categories Ordinal Rank-order stimuli on a single dimension Interval Specify the distance between stimuli on a given dimension Ratio Specify the distance between stimuli on a given dimension and express ratios of scale values Continuous “The topic of scales of measurement is one that some writers think is crucial and others think irrelevant. Although this book will tend to side with the latter group it is important that you have some familiarity with the general issue” – Howell, 1987

Interlude #4 What Stats to use when Hypothesis Variable 2 DV Variable 1 # IV’s Levels of IV Between / Within Test Group Differences DV – Continuous Interval IV Categorical Nominal 1 2 Between Independent Groups ttest Group Differences DV – Continuous Interval IV Categorical Nominal 1 2 Within Paired Sample ttest Group Differences DV – Continuous Ordinal IV Categorical Nominal 1 2 Between Mann. Whitney U Test Group Differences DV – Continuous Ordinal IV Categorical Nominal 1 2 Within Wilcoxon Signed Ranks Relationships DV Continuous Interval - - Within Pearson Correlation Relationships DV Continuous Ordinal - - Within Spearman Correlation

Step 2: Present Summary 8 Mean & SD Choice of “old” word Walking time Choice of “old” word Walking Time 6. 79 (2. 03) 1. 43 5. 10 (0. 95). 43 1 Walking Time (secs) 7 6 5 4 3 R 2 = 0, 1856 2 1 0 0 2 4 6 8 Choice of "old" word 10 12 Can now statistically evaluate the hypotheses of the study Can we reject the Null Hypothesis? Is p <. 05 As the priming scores on the scrambled sentences task increases walking time will increase R=. 43, p =. 032. Reject the null hypothesis

• Draw Conclusions Observations consistent with theory Priming social stereotypes has an automatic effect on actions As scores on Scrambled Sentence test increase walking time will increase 10 0 R 2 = 0, 1856 0 Empirical data 20 supports Operational Hypothesis supports Theoretical Hypothesis As the priming scores on the scrambled sentences task increased walking time increased There was a positive relationship between stereotypes and walking time There is a relationship between stereotype priming effects and action effects

Final Interlude What about……. Empirical data supports Operational Hypothesis supports Theoretical Hypothesis & multiple other theoretical hypotheses

If this was useful you might like to check out Graduate Certificate in Science (Psychology in Schools) Find out more: tehan@usq. edu. au https: //www. usq. edu. au/handbook/current/science s/GCSC. html CRICOS QLD 00244 B | NSW 02225 M TEQSA: PRV 12081