SPSP 2010 1 Estimating Mediated Effects of Personality and Social Psychological Processes Patrick E. Shrout, Ph.D. NYU Niall Bolger, Ph.D. Columbia U.

Slides:



Advertisements
Similar presentations
Agency for Healthcare Research and Quality (AHRQ)
Advertisements

Can a Variable be both a Mediator and a Moderator? Dominique Muller (University of Grenoble/University Institute of France) Charles M. Judd (University.
Mediation: Multiple Variables David A. Kenny. 2 Mediation Webinars Four Steps Indirect Effect Causal Assumptions.
Cross Cultural Research
Seven Deadly Sins of Dyadic Data Analysis David A. Kenny February 14, 2013.
 Graduate symposium deadline Friday  CSBS student conference April 25 (deadline April 10)  Thesis defenses  Outline due on Friday.
Cross Sectional Designs
Copyright EM LYON Par accord du CFC Cession et reproduction interdites Research in Entrepreneurship- The problem of unobserved heterogeneity Frédéric Delmar.
Mediation and Moderation It’s a multivariate world out there… Todd D. Little, Director Institute for Measurement, Methodology, Analysis and Policy.
FUNDAMENTAL RESEARCH ISSUES © 2012 The McGraw-Hill Companies, Inc.
Estimation and Reporting of Heterogeneity of Treatment Effects in Observational Comparative Effectiveness Research Prepared for: Agency for Healthcare.
Culture and psychological knowledge: A Recap
 Confounders are usually controlled with the “standard” response regression model.  The standard model includes confounders as covariates in the response.
Non-Experimental designs: Developmental designs & Small-N designs
Chapter 4 Multiple Regression.
Non-Experimental designs: Developmental designs & Small-N designs
“Ghost Chasing”: Demystifying Latent Variables and SEM
Structural Equation Modeling
American Psychological Association’s Task Force on Statistical Inference TFSI 1999.
SPSP Mediation and Moderation as Tools for Theory Development in Social and Personality Psychology A Symposium SPSP 2010.
Estimating and Testing Mediation
G Lect 31 G Lecture 3 SEM Model notation Review of mediation Estimating SEM models Moderation.
Latent variables in psychology and social sciences: theoretical positions, assumptions, and methodological conundrums Alina Zlati.
Part 3 of 3 By: Danielle Davidov, PhD & Steve Davis, MSW, MPA INTRODUCTION TO RESEARCH: SAMPLING & DESIGN.
Mediation: Sensitivity Analysis David A. Kenny. 2 You Should Know Assumptions Detailed Example Solutions to Assumption Violation.
G Lecture 61 G SEM Lecture 6 An Example Measures of Fit Complex nonrecursive models How can we tell if a model is identified? Direct and.
Chapter 4 Research Methods
Chapter 5 Research Methods in the Study of Abnormal Behavior Ch 5.
Correlation and Correlational Research Slides Prepared by Alison L. O’Malley Passer Chapter 5.
Chapter 1: Introduction to Statistics
Reading Scientific Papers Shimae Soheilipour
Copyright 2003, D. P. MacKinnon 1 Mediator and Moderator Methods David P. MacKinnon Arizona State University AABT November 22, 2003 PAlternative Methods.
G Lecture 41 Panel Designs Panel Path Analyses Examples Lord’s Paradox Extensions.
1 Advanced Topics in Regression Quantile Regression Analysis of Causality Mediation Analysis Hierarchical Linear Modeling Compiled by Nick Evangelopoulos,
1. Transformationa l Leadership M 1 :Affective Commitment OCB M 2 :Calculative Commitment.
G Lect 21 G Lecture 2 Regression as paths and covariance structure Alternative “saturated” path models Using matrix notation to write linear.
Group Quantitative Designs First, let us consider how one chooses a design. There is no easy formula for choice of design. The choice of a design should.
Causal inferences During the last two lectures we have been discussing ways to make inferences about the causal relationships between variables. One of.
Chapter 1 Introduction to Statistics. Statistical Methods Were developed to serve a purpose Were developed to serve a purpose The purpose for each statistical.
SS440 Seminar: Unit 4 Research in Psychopathology Dr. Angie Whalen Kaplan University 1.
1 G Lect 7M Statistical power for regression Statistical interaction G Multiple Regression Week 7 (Monday)
Testing Theories: The Problem of Sampling Error. The problem of sampling error It is often the case—especially when making point predictions—that what.
By: Amani Albraikan.  Pearson r  Spearman rho  Linearity  Range restrictions  Outliers  Beware of spurious correlations….take care in interpretation.
Chapter 4 Linear Regression 1. Introduction Managerial decisions are often based on the relationship between two or more variables. For example, after.
STUDYING BEHAVIOR © 2009 The McGraw-Hill Companies, Inc.
Mediation: Solutions to Assumption Violation
Mediation: The Four Steps David A. Kenny davidakenny.net.
1 G Lect 14M Review of topics covered in course Mediation/Moderation Statistical power for interactions What topics were not covered? G Multiple.
1 Statistics in Research & Things to Consider for Your Proposal May 2, 2007.
Controlling for Baseline
 Descriptive Methods ◦ Observation ◦ Survey Research  Experimental Methods ◦ Independent Groups Designs ◦ Repeated Measures Designs ◦ Complex Designs.
Mediation Example David A. Kenny.
The State of Mediation Analysis in Psychological Science Issues and Solutions Rick Hoyle Duke University.
Non-Experimental designs Psych 231: Research Methods in Psychology.
Chapter 10 Mediation Class 6 Spring 2015 Ang &Huan (2006)
Mediation: The Causal Inference Approach David A. Kenny.
Past and Present Methods for testing Mediation
Mediation: Assumptions David A. Kenny davidakenny.net.
Mediation. 1.Definition 2. Testing mediation using multiple regression Baron, R. M., & Kenny, D. A. (1986). The moderator-mediator variable distinction.
Research Methods in Business Fall 2011 Hypothesis Development: Moderation & Mediation Dr. Stefan Wuyts Associate Professor Marketing Koç University
G Lecture 2 Regression as paths and covariance structure
Complex Experimental Designs Chp 10
Chapter 4: Studying Behavior
Experimental Design.
Class 8 Spring 2017 Ang &Huan (2006)
Mediation MODERATION THREE WAYS OF DOING ANALYSIS Kun
Special Topic: Longitudinal Mediation Analysis
SRM II Review of key concepts
Multicollinearity What does it mean? A high degree of correlation amongst the explanatory variables What are its consequences? It may be difficult to separate.
Presentation transcript:

SPSP Estimating Mediated Effects of Personality and Social Psychological Processes Patrick E. Shrout, Ph.D. NYU Niall Bolger, Ph.D. Columbia U

SPSP An Example: Feeling Excluded Bernstein, Sacco, Brown, Young & Claypool (JESP, 2009)  Randomly assigned Ss to write about exclusion experience or another experience  Measured self esteem, belonging, control, & meaningful existence  Measured preference to 20 faces Duchenne smiles involving two muscle groups Non-Duchenne smiles involving one voluntary muscle group Summarized results as difference score

SPSP Exclusion affected face perception Those writing about exclusion were more likely to prefer “genuine” smiles to possibly staged smiles. BUT WHY?  Self esteem seemed to mediate the Exclusion effect  “The fact that self-esteem alone fully mediated the effect warrants further discussion. Self-esteem is the mechanism by which Sociometer Theory operates (Leary et al.,1995). In this model, self-esteem acts as a gauge of belongingness, and when a threat occurs, individuals take actions to ameliorate that threat.” Bernstein et al, (2009) Theory was tested using Baron & Kenny mediation model

SPSP B&K(1986) Step 1: Find an effect to explain Y X M e Bernstein et al (2009) showed that Exclusion led to increased preference for natural smiles.  c=0.26 c

SPSP B&K(1986) Step 2: Show X is related to mediator Y X M eYeY Bernstein et al showed that Exclusion was related to M: Self-esteem was lower in the exclusion condition.  a = -.88 a eMeM

SPSP B&K(1986) Step 3: Show M is Related to Y Y X M eYeY ADJUSTING for X, M must be related to Y Bernstein et al reported b=  Increased self esteem decreased interest in natural smile, adjusting for Exclusion b eMeM

SPSP B&K(1986) Step 4: Test the indirect effect Y X M eYeY Indirect effect is quantified by the product a*b  Formal test by Sobel test, joint-significance test, bootstrap confidence interval Bernstein et al found indirect path was significant using Sobel test a b eMeM

SPSP B&K (1986) Step 5: Distinguish Full from partial mediation Test direct effect, c’, while adjusting for M. The adjusted (direct) effect in Bernstein example was c’=0.18, which was not significantly different from zero  Authors interpreted result as Full Mediation Y X M eYeY C’ eMeM

SPSP Mediation and Theory Construction When mediation is complete, researcher has “explained the effect”  Other explanations apparently not needed  Often those other explanations not tested Bernstein et al. (2009) did test theory- driven alternate mediators based on Williams (2007)  Self esteem vs. belonging, efficacy needs

SPSP Some Vexing Problems Claiming complete mediation is too easy  If the total effect is just significant, not much reduction is needed to make adjusted direct effect non- significant  Multiple mediators are often of theoretical interest but not usually tested Estimates of indirect effects are often biased  If based on mediators that are measured with error  If based on wrong model

SPSP Model Specification Baron and Kenny (1986) assume model is correct What does this entail?  Causal paths are interpretable  Variables are measured without error  Residual (error) values uncorrelated Implies that important causes are represented

SPSP Causal Pathways and Time Causal Assumptions in Mediation  X is prior to M and Y  Change in X is associated with change in Y  Change in X is associated with change in M  Change in M is associated with change in Y  Measurements taken at times that reflect causal action Y X M eYeY C’ eMeM a b

SPSP Causal Pathways and Time Causal Assumptions in Mediation  X is prior to M and Y  Change in X is associated with change in Y  Change in X is associated with change in M  Change in M is associated with change in Y  Measurements taken at times that reflect causal action Y X M eYeY C’ eMeM a b

SPSP Causal Pathways and Time Causal Assumptions in Mediation  X is prior to M and Y  Change in X is associated with change in Y  Change in X is associated with change in M  Change in M is associated with change in Y  Measurements taken at times that reflect causal action Y X M eYeY C’ eMeM a b

SPSP Causal Pathways and Time Causal Assumptions in Mediation  X is prior to M and Y  Change in X is associated with change in Y  Change in X is associated with change in M  Change in M is associated with change in Y  Measurements taken at times that reflect causal action Y X M eYeY C’ eMeM a b

SPSP Causal Pathways and Time Causal Assumptions in Mediation  X is prior to M and Y  Change in X is associated with change in Y  Change in X is associated with change in M  Change in M is associated with change in Y  Measurements taken at times that reflect causal action Y X M eYeY C’ eMeM a b

SPSP Inferring Within-Person Change from Between-Person Data Systematic consideration of time draws us to psychological process  Within person changes  Effects of manipulations on persons Traditional designs substitute between person differences for within person change  Justified in experiments  Harder to justify in surveys In nature, between person associations are rarely the same as within person associations

SPSP Revisiting Bernstein et al. (2009) Randomized design makes temporal order clear  X->Y: Exclusion experience (randomized) was related to face preference  X->M: Exclusion was also related to Self esteem (apparent mediator) Efficacy needs (not found as mediator) M->Y: Temporal relation of self-esteem and face preference not clear  What might contribute the correlation between M & Y?

SPSP Possible Between Person Confounding of M->Y Y3Y3 X1X1 M2M2 G

SPSP If “third variable” is ignored, error terms are correlated Y3Y3 X1X1 M2M2 However, the correlation can not be estimated in traditional Baron & Kenny Mediation model.

SPSP Baseline Measures Can Reduce Confounding Y3Y3 X1X1 M2M2 M1M1 Y1Y1 a b c' g1g1 g2g2 r my Design adds within person information so that change can be estimated.

SPSP But most ignore baseline What are implications?  Total effect (c) is not biased.  Effect on M (a) is not biased.  BUT Effect of M on Y may be biased The more stable the processes (g 1, g 2 ), the more the bias for nonzero correlations of M and Y. The more the correlation of baseline M and Y the more the bias for stable processes. Y3Y3 X1X1 M2M2 M1M1 Y1Y1 a c' g1g1 g2g2 r my b

SPSP Quantifying Bias: A Numerical Example Y3Y3 X1X1 M2M2 M1M1 Y1Y g1g1 g2g2 r my.40 Direct effect.28 Indirect effect.28

SPSP g=.2 If we ignore baseline, what do we estimate as indirect effect? g=.0 g=.4 g=.6 g=.8 Y3Y3 X1X1 M2M2 M1M1 Y1Y1 a c' g1g1 g2g2 r my b

SPSP Quantifying Bias for Direct Effects g=.0 g=.2 g=.4 g=.6 g=.8

SPSP Extensions Will correlations of M and Y error terms also cause problems in cross-sectional studies?  You betcha!  The M->Y path needs to approximate within person change.  Additional covariates will be needed  But see Cole and Maxwell (2005) about plausibility of cross sectional models

SPSP Objections What if taking baseline measures in experiments would prime processes that are left un-primed?  Often possible to estimate Corr(M,Y) and the stability of M and Y in separate samples  Combining the data from the two samples will require structural equation methods.

SPSP Conclusions Social psychology theory is ready for next generation mediation analysis  Will aid in communication with other scientists  Will refine thinking about process Combination of new heuristic steps and systematic thinking about process will serve us well

SPSP Time for a Ten Step Program? 1)*Argue that X can be a causal agent of Y 2)Show that X is related to Y. 3)Show that X is related to M, the mediator 4)*Show that M is measured with little error. 5)*Identify plausible competing mediators and include them in the model 6)Show that M is related to Y adjusting for X 7)*Adjust for correlation between M and Y that is prior to causal process 8)Show that indirect path (X->M->Y) is present 9)Estimate/test direct effect of X->Y after adjusting for M. 10)*Report ratio of mediated effect. If it is nearly 1 then claim full mediation.

SPSP Time for a Ten Step Program? 1)*Argue that X can be a causal agent of Y 2)Show that X is related to Y. 3)Show that X is related to M, the mediator 4)*Show that M is measured with little error. 5)*Identify plausible competing mediators and include them in the model 6)Show that M is related to Y adjusting for X 7)*Adjust for correlation between M and Y that is prior to causal process 8)Show that indirect path (X->M->Y) is present 9)Estimate/test direct effect of X->Y after adjusting for M. 10)*Report ratio of mediated effect. If it is nearly 1 then claim full mediation.

SPSP Time for a Ten Step Program? 1)*Argue that X can be a causal agent of Y 2)Show that X is related to Y. 3)Show that X is related to M, the mediator 4)*Show that M is measured with little error. 5)*Identify plausible competing mediators and include them in the model 6)Show that M is related to Y adjusting for X 7)*Adjust for correlation between M and Y that is prior to causal process 8)Show that indirect path (X->M->Y) is present 9)Estimate/test direct effect of X->Y after adjusting for M. 10)*Report ratio of mediated effect. If it is nearly 1 then claim full mediation.

SPSP Help from our friends Margarita Krochik Turu Stadler Couples lab members at NYU and Columbia Grant R01-AA from NIAAA

2010 SPSP 33 References Baron, R. M., & Kenny, D. A. (1986). The moderator-mediator variable distinction in social psychological research: Conceptual, strategic and statistical considerations. Journal of Personality and Social Psychology, 51, Bernstein, M.J. et al. (2009). A preference for genuine smiles following social exclusion. Journal of Experimental Social Psychology, doi: /j.jesp Cole DA, Maxwell SE. (2003). Testing mediational models with longitudinal data: questions and tips in the use of structural equation modeling. J. Abnormal Psychology, 112:558–577. Gollob, H.F. & Reichardt, C.S. (1987). Taking account of time lags in causal models. Child Development, 58(1), Kraemer, H., Kiernan, M., Essex, M., & Kupfer, D. J. (2008). How and why criteria defining moderators and mediators differ between the Baron & Kenny and MacArthur approaches. Health Psychology, 27(2, Suppl), S101-S108 MacKinnon DP (2008). Introduction to statistical mediation analysis. New York: LEA Maxwell, S. E., & Cole, D. A. (2007). Bias in cross-sectional analyses of longitudinal mediation. Psychological Methods, 12(1), Shrout, P.E. (in press). Integrating causal analysis into psychopathology research. In Causality and Psychopathology: Finding the Determinants of Disorders and their Cures. P.E. Shrout, K. Keyes, K. Ornstein (Eds). New York: Oxford U. Press. Spencer, S.J., Zanna, M.P. & Fong, G.T. (2005). Establishing a causal chain: Why experiments are often more effective than mediational analyses in examining psychological processes. Journal of Personality and Social Psychology, 89(6),