Introduction to Path Analysis Reminders about some important regression stuff Boxes & Arrows – multivariate models beyond MR Ways to “think about” path.

Slides:



Advertisements
Similar presentations
Bivariate &/vs. Multivariate
Advertisements

Transformations & Data Cleaning
Canonical Correlation simple correlation -- y 1 = x 1 multiple correlation -- y 1 = x 1 x 2 x 3 canonical correlation -- y 1 y 2 y 3 = x 1 x 2 x 3 The.
Survey of Major Correlational Models/Questions Still the first & most important is picking the correct model… Simple correlation questions simple correlation.
Automated Regression Modeling Descriptive vs. Predictive Regression Models Four common automated modeling procedures Forward Modeling Backward Modeling.
Research Hypotheses and Multiple Regression: 2 Comparing model performance across populations Comparing model performance across criteria.
ANCOVA Workings of ANOVA & ANCOVA ANCOVA, Semi-Partial correlations, statistical control Using model plotting to think about ANCOVA & Statistical control.
Copyright © 2009 Pearson Education, Inc. Chapter 29 Multiple Regression.
Outline 1) Objectives 2) Model representation 3) Assumptions 4) Data type requirement 5) Steps for solving problem 6) A hypothetical example Path Analysis.
© 2005 The McGraw-Hill Companies, Inc., All Rights Reserved. Chapter 14 Using Multivariate Design and Analysis.
Research Hypotheses and Multiple Regression Kinds of multiple regression questions Ways of forming reduced models Comparing “nested” models Comparing “non-nested”
Statistical Control Regression & research designs Statistical Control is about –underlying causation –under-specification –measurement validity –“Correcting”
ANCOVA Workings of ANOVA & ANCOVA ANCOVA, Semi-Partial correlations, statistical control Using model plotting to think about ANCOVA & Statistical control.
Multiple Regression Models Advantages of multiple regression Important preliminary analyses Parts of a multiple regression model & interpretation Differences.
Multiple Regression Models: Some Details & Surprises Review of raw & standardized models Differences between r, b & β Bivariate & Multivariate patterns.
Multiple Regression Models
© 2005 The McGraw-Hill Companies, Inc., All Rights Reserved. Chapter 4 Choosing a Research Design.
Simple Regression correlation vs. prediction research prediction and relationship strength interpreting regression formulas –quantitative vs. binary predictor.
Introduction to Path Analysis & Mediation Analyses Review of Multivariate research & an Additional model Structure of Regression Models & Path Models Direct.
Multivariate Analyses & Programmatic Research Re-introduction to Programmatic research Factorial designs  “It Depends” Examples of Factorial Designs Selecting.
1 Psych 5500/6500 The t Test for a Single Group Mean (Part 5): Outliers Fall, 2008.
Multivariate Analyses & Programmatic Research Re-introduction to Multivariate research Re-introduction to Programmatic research Factorial designs  “It.
Multiple-group linear discriminant function maximum & contributing ldf dimensions concentrated & diffuse ldf structures follow-up analyses evaluating &
Introduction to Path Analysis Ways to “think about” path analysis Path coefficients A bit about direct and indirect effects What path analysis can and.
Multivariate Analyses & Programmatic Research Re-introduction to Multivariate research Re-introduction to Programmatic research Factorial designs  “It.
Bivariate & Multivariate Regression correlation vs. prediction research prediction and relationship strength interpreting regression formulas process of.
Correlational Designs
Multiple-Group Research Designs Limitations of 2-group designs “Kinds” of Treatment & Control conditions Kinds of Causal Hypotheses k-group ANOVA & Pairwise.
Week 14 Chapter 16 – Partial Correlation and Multiple Regression and Correlation.
Multiple Linear Regression A method for analyzing the effects of several predictor variables concurrently. - Simultaneously - Stepwise Minimizing the squared.
Correlation Nabaz N. Jabbar Near East University 25 Oct 2011.
McGraw-Hill © 2006 The McGraw-Hill Companies, Inc. All rights reserved. Correlational Research Chapter Fifteen.
Nonlinear Regression Functions
McGraw-Hill/Irwin Copyright © 2011 by The McGraw-Hill Companies, Inc. All rights reserved. Choosing a Research Design.
CHAPTER NINE Correlational Research Designs. Copyright © Houghton Mifflin Company. All rights reserved.Chapter 9 | 2 Study Questions What are correlational.
Understanding Statistics
Correlational Designs
Least-Squares Regression Section 3.3. Why Create a Model? There are two reasons to create a mathematical model for a set of bivariate data. To predict.
L 1 Chapter 12 Correlational Designs EDUC 640 Dr. William M. Bauer.
Extension to Multiple Regression. Simple regression With simple regression, we have a single predictor and outcome, and in general things are straightforward.
Observation & Analysis. Observation Field Research In the fields of social science, psychology and medicine, amongst others, observational study is an.
Chapter 9 Analyzing Data Multiple Variables. Basic Directions Review page 180 for basic directions on which way to proceed with your analysis Provides.
Correlational Research Chapter Fifteen Bring Schraw et al.
Statistical Control Statistical Control is about –underlying causation –under-specification –measurement validity –“Correcting” the bivariate relationship.
BIOL 582 Lecture Set 11 Bivariate Data Correlation Regression.
Coding Multiple Category Variables for Inclusion in Multiple Regression More kinds of predictors for our multiple regression models Some review of interpreting.
Testing Path Models, etc. What path models make you think about… Model Identification Model Testing –Theory Trimming –Testing Overidentified Models.
ANOVA and Linear Regression ScWk 242 – Week 13 Slides.
Copyright © 2008 Pearson Education, Inc. Publishing as Pearson Addison-Wesley Chapter 8 Linear Regression.
Time series Model assessment. Tourist arrivals to NZ Period is quarterly.
Measurement Models: Exploratory and Confirmatory Factor Analysis James G. Anderson, Ph.D. Purdue University.
Regression Models w/ 2 Quant Variables Sources of data for this model Variations of this model Main effects version of the model –Interpreting the regression.
Aron, Aron, & Coups, Statistics for the Behavioral and Social Sciences: A Brief Course (3e), © 2005 Prentice Hall Chapter 12 Making Sense of Advanced Statistical.
Algebra Simplifying and collecting like terms. Before we get started! Believe it or not algebra is a fairly easy concept to deal with – you just need.
© 2006 by The McGraw-Hill Companies, Inc. All rights reserved. 1 Chapter 12 Testing for Relationships Tests of linear relationships –Correlation 2 continuous.
Chapter 10 Finding Relationships Among Variables: Non-Experimental Research.
The Practice of Statistics, 5th Edition Starnes, Tabor, Yates, Moore Bedford Freeman Worth Publishers CHAPTER 4 Designing Studies 4.2Experiments.
Copyright © 2007 Pearson Education, Inc. Publishing as Pearson Addison-Wesley Slide 8- 1.
Single-Subject and Correlational Research Bring Schraw et al.
ANCOVA Workings of ANOVA & ANCOVA ANCOVA, partial correlations & multiple regression Using model plotting to think about ANCOVA & Statistical control Homogeneity.
Making Sense of Advanced Statistical Procedures in Research Articles
CHAPTER 4 Designing Studies
CHAPTER 4 Designing Studies
CHAPTER 4 Designing Studies
CHAPTER 4 Designing Studies
CHAPTER 4 Designing Studies
CHAPTER 4 Designing Studies
CHAPTER 4 Designing Studies
Correlation and Causality
CHAPTER 4 Designing Studies
Presentation transcript:

Introduction to Path Analysis Reminders about some important regression stuff Boxes & Arrows – multivariate models beyond MR Ways to “think about” path analysis A bit about direct and indirect effects A bit about mediation effects measured vs. manifested  the “when” of variables What path analysis can and can’t do for you… About non-recursive cause in path models Some ways to improve a path analysis model

The first multivariate model most of us learn is multiple regression y’ = b 1 x 1 + b 2 x 2 + b 3 x 3 + a & Z y ’ = β 1 Z x1 + β 2 Z x2 + β 3 Z x3 Things we study when learning this model … Colinearity -- the difference between the information and interpretation of bivariate r & b & or multivariate b how the amount and pattern of colinearity influences this difference between bivariate r & b & or multivariate b Suppressor effects – greater/differently signed multivariate contribution that expected based on bivariate r & b colinearity patterns at their most interesting & informative

Things we learn when learning this model (cont.) … Underspecification – our datasets/models don’t include all the relevant predictors (causal agents) combined with colinearity & suppressor effects, this means that the contribution of a predictor to “this model,” may not be the same as its contribution to the “complete model" Raw vs. Standardized weights -- the difference between b & β the greater comparability of β than b to compare the relative unique contribution of predictors to the model Causation -- the causal interpretability of the results of statistical analysis is dependent upon the design used to collect the data, not the statistical model used statistical control =/= experimental control “effect on” means “statistical relationship with” not “causal of”

One way to “think about” path analysis is as a way of “sorting out” the colinearity patterns amongst the predictors – asking yourself what may be the structure -- temporal and/or causal relationships -- among these predictors that produces the pattern of colinearity. “Structure” of a MR model – with hypotheses about which predictors will contribute A proposed structure for the colinearity among the predictors and how they relate to the criterion – with hypotheses about which paths will contribute Crit earlierlater final causelocal cause

Alternative ways to “think about” path analysis… to capture the “causal paths” among the predictors and to the criterion to capture the “temporal paths” among the predictors and to the criterion to distinguish “direct” and “indirect” paths of relationship to investigate “mediation effects”

… to distinguish “direct” and “indirect” paths of relationship… Crit 2 has a direct effect on Crit a “contributor” in both the regression and the path models the regression model Crit 5 does not have a direct effect on Crit – but does have multiple indirect effects not “contributing” in the regression model could mistakenly lead us to conclude it “5 doesn’t matter in understanding Crit”

…to distinguish “direct” and “indirect” paths of relationship…, cont Crit 3 also has an indirect effect on Crit there’s more to the 3  Crit relationship than was captured in the regression model Crit 3 has a direct effect on Crit

… to investigate “mediation effects”… Mediation effects and analyses highlight the difference between bivariate and multivariate relationships between a variable and a criterion (collinearity & suppressor effects). For example… For Teaching Quality & Exam Performance  r =.30, p =.01 for binary regression β = r, so we have the path model… TQEP β=.3 It occurs to one of the researchers that there just might be something else besides Teaching Quality related to (influencing, even) Exam Performance. The researcher decides that Study Time (ST) might be such a variable. Thinking temporally/causally, the researcher considers that Study Time “comes in between” Teaching and Testing. So the researcher builds a mediation model, getting the weights from a multiple regression with TQ and ST as predictors of EP

… to investigate “mediation effects”… The resulting model looks like … TQ EP β=.0 ST β=.4 β=.3 We might describe model as, “The effect of Teaching Quality on Exam Performance (r=.30) is mediated by Study Time.” We might describe the combination of the bivariate analysis and the multiple regression from which the path coefficients were obtained as, “While Teaching Quality has a bivariate relationship with Exam Performance (r=.30), it does not contribute to a multiple regression model (β=.0) that also includes Study Time (β=.40). Either analysis reminds us that the bivariate contribution of a given predictor might not “hold up” when we look at that relationship within a multivariate model! Notice that TQ is “still important” because it seems to have something to do with study time – an indirect effect upon Exam Performance.

What path analysis can and can’t accomplish… Cans -- for a given structural model you can… evaluate the contribution of any path or combination of paths to the overall fit of that structural model help identify sources of suppressor effects (indirect paths) Can’ts non-recursive (bi-directional) models help decide among alternative structural models provide tests of causality (unless experimental data) So… You have to convince yourself and your audience of the “reasonableness” of your structural model (the placing of the predictors), and then you can test hypotheses about which arrows amongst the variables have unique contributions.

The “when” of variables and their place in the model … When a variable is “measured”  when we collect the data: usually concurrent often postdictive (can be a problem – memory biases, etc.) sometimes predictive (hypothetical – can really be a problem) When a variable is “manifested”  when the value of the variable came into being when it “comes into being for that participant” may or may not be before the measure was taken E.g., State vs. Trait anxiety trait anxiety is intended to be “characterological,” “long term” and “context free”  earlier in model state anxiety is intended to be “short term” & “contextual”  depends when it was measured

About non-recursive (bi-directional) models Crit Sometimes we want to consider whether two things that “happen at the same time” might have “reciprocal causation” – so we want to put in a sideways arrow Neither of these can be “handled” by path analysis. However, this isn’t really a problem because both are a misrepresentation of the involved causal paths! The real way to represent both of these is … Sometimes we want to consider whether two things that “happen sequentially” might have “iterative causation” – so we want to put in a back-and-forth arrow Crit

The things to remember are that: 1. “cause takes time” or “cause is not immediate” even the fastest chemical reactions take time behavioral causes take an appreciable amount of time 2.Something must “be” to “cause something else to be” a variable has to be manifested as an effect of some cause before it can itself be the cause of another effect Cause comes before effect  not at the same time When you put these ideas together, then both “sideways” and “back-and-forth” arrows don’t make sense and are not an appropriate portrayal of the causations being represented. The causal path has to take these two ideas into account…

About non-recursive (bi-directional) models If “5” causes “4”, then “4” changes “5”, which changes “4” again, all before the criterion is caused, we need to represent that we have 2 “4s” and 2 “5s” in a hypothesized sequence Crit We also have to decide when1, 2 & 3 enter into the model, temporally &/or causally. Say … 4 5 Crit

About non-recursive (bi-directional) models, cont… Crit When applying these ideas to “sideways arrows” we need to remember that the cause comes before the effect Crit To do that, we have to decide (& defend) which comes first – often the hardest part) and then add in the second causation, etc.… As well as sort out where the other variables fall temporally &/or causally. Perhaps … 1

Some of the ways to improve a path analysis For a given model, consider these 4 things…. TQ EP ST 1.Antecedents to the current model Variables that “come before” or “cause” the variables in the model 2.Effects of the current model Variables that “come after” or “are caused by” the variables in the model 3.Intermediate causes Variables that “come in between” the current causes and effects. 4.Non-linear variations of the model Curvilinear & interaction effects of & among the variables