Issues in Experimental Design Reliability and ‘Error’

Slides:



Advertisements
Similar presentations
Properties of Least Squares Regression Coefficients
Advertisements

ANCOVA Workings of ANOVA & ANCOVA ANCOVA, Semi-Partial correlations, statistical control Using model plotting to think about ANCOVA & Statistical control.
RELIABILITY Reliability refers to the consistency of a test or measurement. Reliability studies Test-retest reliability Equipment and/or procedures Intra-
Reliability and Validity
Experiments Pre and Post condition.
3.2 OLS Fitted Values and Residuals -after obtaining OLS estimates, we can then obtain fitted or predicted values for y: -given our actual and predicted.
1 SSS II Lecture 1: Correlation and Regression Graduate School 2008/2009 Social Science Statistics II Gwilym Pryce
Prediction, Correlation, and Lack of Fit in Regression (§11. 4, 11
Smith/Davis (c) 2005 Prentice Hall Chapter Thirteen Inferential Tests of Significance II: Analyzing and Interpreting Experiments with More than Two Groups.
Correlation and Linear Regression
Understanding the General Linear Model
LINEAR REGRESSION: Evaluating Regression Models. Overview Standard Error of the Estimate Goodness of Fit Coefficient of Determination Regression Coefficients.
Longitudinal Experiments Larry V. Hedges Northwestern University Prepared for the IES Summer Research Training Institute July 28, 2010.
Chapter 9 - Lecture 2 Some more theory and alternative problem formats. (These are problem formats more likely to appear on exams. Most of your time in.
Basic Statistical Concepts Psych 231: Research Methods in Psychology.
ANCOVA Psy 420 Andrew Ainsworth. What is ANCOVA?
Lecture 4: Correlation and Regression Laura McAvinue School of Psychology Trinity College Dublin.
Lesson #32 Simple Linear Regression. Regression is used to model and/or predict a variable; called the dependent variable, Y; based on one or more independent.
Chapter 14 Conducting & Reading Research Baumgartner et al Chapter 14 Inferential Data Analysis.
One-Way Analysis of Covariance One-Way ANCOVA. ANCOVA Allows you to compare mean differences in 1 or more groups with 2+ levels (just like a regular ANOVA),
Chapter 10 - Part 1 Factorial Experiments.
Chapter 9 - Lecture 2 Computing the analysis of variance for simple experiments (single factor, unrelated groups experiments).
C82MCP Diploma Statistics School of Psychology University of Nottingham 1 Linear Regression and Linear Prediction Predicting the score on one variable.
DOCTORAL SEMINAR, SPRING SEMESTER 2007 Experimental Design & Analysis Further Within Designs; Mixed Designs; Response Latencies April 3, 2007.
Lorelei Howard and Nick Wright MfD 2008
Introduction to Regression Analysis, Chapter 13,
Relationships Among Variables
Basic Analysis of Variance and the General Linear Model Psy 420 Andrew Ainsworth.
Example of Simple and Multiple Regression
Introduction to Multilevel Modeling Using SPSS
Analysis of Variance. ANOVA Probably the most popular analysis in psychology Why? Ease of implementation Allows for analysis of several groups at once.
LEARNING PROGRAMME Hypothesis testing Intermediate Training in Quantitative Analysis Bangkok November 2007.
1 Experimental Statistics - week 10 Chapter 11: Linear Regression and Correlation Note: Homework Due Thursday.
Simple Linear Regression One reason for assessing correlation is to identify a variable that could be used to predict another variable If that is your.
Extension to Multiple Regression. Simple regression With simple regression, we have a single predictor and outcome, and in general things are straightforward.
Correlation and Regression Used when we are interested in the relationship between two variables. NOT the differences between means or medians of different.
Summarizing Bivariate Data
Examining Relationships in Quantitative Research
1 G Lect 7M Statistical power for regression Statistical interaction G Multiple Regression Week 7 (Monday)
MGS3100_04.ppt/Sep 29, 2015/Page 1 Georgia State University - Confidential MGS 3100 Business Analysis Regression Sep 29 and 30, 2015.
Regression Lesson 11. The General Linear Model n Relationship b/n predictor & outcome variables form straight line l Correlation, regression, t-tests,
Department of Cognitive Science Michael J. Kalsher Adv. Experimental Methods & Statistics PSYC 4310 / COGS 6310 Regression 1 PSYC 4310/6310 Advanced Experimental.
© Copyright McGraw-Hill Correlation and Regression CHAPTER 10.
1 G Lect 11a G Lecture 11a Example: Comparing variances ANOVA table ANOVA linear model ANOVA assumptions Data transformations Effect sizes.
Multiple Regression. Simple Regression in detail Y i = β o + β 1 x i + ε i Where Y => Dependent variable X => Independent variable β o => Model parameter.
ANCOVA. What is Analysis of Covariance? When you think of Ancova, you should think of sequential regression, because really that’s all it is Covariate(s)
General Linear Model.
Smith/Davis (c) 2005 Prentice Hall Chapter Fifteen Inferential Tests of Significance III: Analyzing and Interpreting Experiments with Multiple Independent.
Example x y We wish to check for a non zero correlation.
Experimental Statistics - week 9
Applied Quantitative Analysis and Practices LECTURE#28 By Dr. Osman Sadiq Paracha.
ANCOVA.
Chapter 8 Relationships Among Variables. Outline What correlational research investigates Understanding the nature of correlation What the coefficient.
Chapter 13 Understanding research results: statistical inference.
More repeated measures. More on sphericity With our previous between groups Anova we had the assumption of homogeneity of variance With repeated measures.
Regression Analysis: A statistical procedure used to find relations among a set of variables B. Klinkenberg G
Michael J. Kalsher PSYCHOMETRICS MGMT 6971 Regression 1 PSYC 4310 Advanced Experimental Methods and Statistics © 2014, Michael Kalsher.
Education 793 Class Notes ANCOVA Presentation 11.
Inferential Statistics Psych 231: Research Methods in Psychology.
Stats Methods at IC Lecture 3: Regression.
Multiple Regression.
Regression Analysis.
Reliability & Validity
12 Inferential Analysis.
Simple Linear Regression
Multiple Regression.
12 Inferential Analysis.
Inferential Statistics
Inference about the Slope and Intercept
MGS 3100 Business Analysis Regression Feb 18, 2016
Presentation transcript:

Issues in Experimental Design Reliability and ‘Error’

More things to think about in experimental design The relationship of reliability and power Treatment effect not the same for everyone –Some benefit more than others Sounds like no big deal (or even obvious), but all of these designs discussed assume equal effect of treatment for individuals

Reliability What is reliability? Often thought of as consistency, but this is more of a by- product of reliability –Not to mention that you could have perfectly consistent scores lacking variability (i.e. constants) for which one could not obtain measures of reliability Reliability may refer to a measure’s ability to capture an individual’s true score, to distinguish accurately one person from another on some measure It is the correlation of scores on some measure with their true scores regarding that construct

Classical True Score Theory Each subject’s score is true score + error of measurement Obs var = True var + Error var Reliability = True var / Obs var = 1 – Error var / Obs var

Reliability and power Reliability = True var / Obs var = 1 – Error var / Obs var If observed variance goes up, power will decrease However if observed variance goes up, we don’t know automatically what happens to reliability Obs var = True var + Error var If it is error variance that is causing the increase in observed variance, reliability will decrease 1 –Reliability goes down, Power goes down If it is true variance that is causing the increase in observed variance, reliability will increase –Reliability goes up, Power goes down The point is that psychometric properties of the variables play an important, and not altogether obvious role in how we will interpret results, and not having a reliable measure is a recipe for disaster

Error in ANOVA Typical breakdown in a between groups design –SS tot = SS b/t + SS e Variation due to treatment and random variation (error) The F statistic is a ratio of these variances F = MS b /MS e

Error in ANOVA Classical True Score Theory –Each subject’s score = true score + error of measurement MS e can thus be further partitioned –Variation due to true differences on scores between subjects and error of measurement (unreliability) MS e = MS er + MS es –MS er regards measurement error –MS es systematic differences between individuals MS es comes has two sources –Individual differences –Treatment differences »Subject by treatment interaction

Error in ANOVA The reliability of the measure will determine the extent to which the two sources of variability (MS er or MS es) contribute to the overall MS e If Reliability = 1.00, MS er = 0 –Error term is a reflection only of systematic individual differences If Reliability = 0.00, MS es = 0 –Error term is a reflection of measurement error only MS er = (1-Rel)MS e MS es = (Rel)MS e

Error in ANOVA We can actually test to see if systematic variation is significantly larger than variation due to error of measurement

Error in ANOVA With a reliable measure, the bulk of MS e will be attributable to systematic individual differences However with strong main effects/interactions, we might see sig F for this test even though the contribution to model is not very much Calculate an effect size (eta-squared) –SS es /SS total –Lyons and Howard suggest (based on Cohen’s rules of thumb) that <.33 would suggest that further investigation may not be necessary How much of the variability seen in our data is due to systematic variation outside of the main effects? –Subjects responding differently to the treatment

Gist Discerning the true nature of treatment effects, e.g. for clinical outcomes, is not easy, and not accomplished just because one has done an experiment and seen a statistically significant effect Small though significant effects with not so reliable measures would not be reason to go with any particular treatment as most of the variance is due poor measures and subjects that do not respond similarly to that treatment –One reason to perhaps suspect individual differences due to the treatment would be heterogeneity of variance –For example, lots of variability in treatment group, not so much in control Even with larger effects and reliable measures, a noticeable amount of the unaccounted for variance may be due to subjects responding differently to the treatment Methods for dealing with the problem are outlined in Bryk and Raudenbush (hierarchical linear modeling), but one strategy may be to single out suspected covariates and control for them (ANCOVA or Blocking)

Repeated Measure and Hierarchical Linear Modeling Another issue with ANOVA design again concerns the subject by treatment interaction, this time with regard to repeated measurements RM design can be seen as a special case of HLM where the RM (e.g. time) is nested within subjects The outcome is predicted by the repeated measure as before, but one can allow the intercept and slope(s) to vary over subjects, and that variance taken into account for the model In this manner the HLM approach is specifically examining the treatment by subject interaction, getting a sense of the correlation between starting point and subsequent change

Repeated Measures and Hierarchical Linear Modeling Briefly, HLM is a regression approach in which intercepts and/or coefficients are allowed to vary depending on other variables As an example, the basic linear model for RM is the same However, as an example, the intercept may be allowed to vary as a function of another variable (in this case Subject) Which gives a new regression equation (note how this compares to RM in the GLM notes)

Example with One-way From before, stress week before, the week of, or the week after their midterm exam Using lmer in R 1, allowing a random intercept for a linear model where time predicts stress level but the intercept is allowed to vary by subject reveals the same ANOVA –lmemod0=lmer(Score~Time+ (1|Subject),rmdata) –anova(lmemod0) SourcedfSSMSFp Subject time error Analysis of Variance Table Df Sum Sq Mean Sq F value Time

Example with One-way However, if I were allow the coefficients 1 to vary, I would also note that starting point matters, in that there is a negative relation with the intercept and the general effect of time If one starts out stressed, there is less of a jump during the midterm, and stronger decline by the end

Summary Even though ANOVA designs may seem straightforward on the surface, and even if one has control over the administration of the variable of interest, one can see that issues remain, and that the basic approach may be inadequate to resolving the true nature of effects

Resources Zimmerman & Williams (1986) Bryk & Raudenbush (1988) Lyons & Howard (1991)