Chapter 10 For Explaining Psychological Statistics, 4th ed. by B. Cohen 1 A perfect correlation implies the ability to predict one score from another perfectly.

Slides:



Advertisements
Similar presentations
Lesson 10: Linear Regression and Correlation
Advertisements

Chapter 12 Simple Linear Regression
Review ? ? ? I am examining differences in the mean between groups
Linear Regression. PSYC 6130, PROF. J. ELDER 2 Correlation vs Regression: What’s the Difference? Correlation measures how strongly related 2 variables.
1 Simple Linear Regression and Correlation The Model Estimating the Coefficients EXAMPLE 1: USED CAR SALES Assessing the model –T-tests –R-square.
Describing Relationships Using Correlation and Regression
Correlation & Regression Chapter 15. Correlation statistical technique that is used to measure and describe a relationship between two variables (X and.
Chapter 15 (Ch. 13 in 2nd Can.) Association Between Variables Measured at the Interval-Ratio Level: Bivariate Correlation and Regression.
LINEAR REGRESSION: Evaluating Regression Models Overview Assumptions for Linear Regression Evaluating a Regression Model.
LINEAR REGRESSION: Evaluating Regression Models. Overview Assumptions for Linear Regression Evaluating a Regression Model.
1-1 Regression Models  Population Deterministic Regression Model Y i =  0 +  1 X i u Y i only depends on the value of X i and no other factor can affect.
Chapter 10 Simple Regression.
9. SIMPLE LINEAR REGESSION AND CORRELATION
Lecture 11 PY 427 Statistics 1 Fall 2006 Kin Ching Kong, Ph.D
REGRESSION What is Regression? What is the Regression Equation? What is the Least-Squares Solution? How is Regression Based on Correlation? What are the.
Chapter Topics Types of Regression Models
Regression Chapter 10 Understandable Statistics Ninth Edition By Brase and Brase Prepared by Yixun Shi Bloomsburg University of Pennsylvania.
REGRESSION Predict future scores on Y based on measured scores on X Predictions are based on a correlation from a sample where both X and Y were measured.
C82MCP Diploma Statistics School of Psychology University of Nottingham 1 Linear Regression and Linear Prediction Predicting the score on one variable.
PSY 307 – Statistics for the Behavioral Sciences Chapter 7 – Regression.
Correlation and Regression Analysis
Simple Linear Regression Analysis
Chapter 9 For Explaining Psychological Statistics, 4th ed. by B. Cohen 1 What is a Perfect Positive Linear Correlation? –It occurs when everyone has the.
EDUC 200C Section 4 – Review Melissa Kemmerle October 19, 2012.
Relationships Among Variables
Correlation and Linear Regression
Correlation and Linear Regression
McGraw-Hill/Irwin Copyright © 2010 by The McGraw-Hill Companies, Inc. All rights reserved. Chapter 13 Linear Regression and Correlation.
Chapter 8: Bivariate Regression and Correlation
Lecture 16 Correlation and Coefficient of Correlation
Introduction to Linear Regression and Correlation Analysis
Linear Regression and Correlation
MAT 254 – Probability and Statistics Sections 1,2 & Spring.
CORRELATION & REGRESSION
Introduction to Regression Analysis. Two Purposes Explanation –Explain (or account for) the variance in a variable (e.g., explain why children’s test.
Chapter 15 Correlation and Regression
Chapter 6 & 7 Linear Regression & Correlation
Chapter 12 Examining Relationships in Quantitative Research Copyright © 2013 by The McGraw-Hill Companies, Inc. All rights reserved. McGraw-Hill/Irwin.
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.
© 2014 by Pearson Higher Education, Inc Upper Saddle River, New Jersey All Rights Reserved HLTH 300 Biostatistics for Public Health Practice, Raul.
Introduction to Linear Regression
Examining Relationships in Quantitative Research
Sullivan – Fundamentals of Statistics – 2 nd Edition – Chapter 4 Section 2 – Slide 1 of 20 Chapter 4 Section 2 Least-Squares Regression.
Inference for Regression Simple Linear Regression IPS Chapter 10.1 © 2009 W.H. Freeman and Company.
© Copyright McGraw-Hill Correlation and Regression CHAPTER 10.
Chapter 16 Data Analysis: Testing for Associations.
STA291 Statistical Methods Lecture LINEar Association o r measures “closeness” of data to the “best” line. What line is that? And best in what terms.
Chapter Twelve The Two-Sample t-Test. Copyright © Houghton Mifflin Company. All rights reserved.Chapter is the mean of the first sample is the.
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.
Chapter 14 Correlation and Regression
Correlation & Regression Analysis
Correlation They go together like salt and pepper… like oil and vinegar… like bread and butter… etc.
LESSON 6: REGRESSION 2/21/12 EDUC 502: Introduction to Statistics.
Advanced Statistical Methods: Continuous Variables REVIEW Dr. Irina Tomescu-Dubrow.
Copyright © 2011 by The McGraw-Hill Companies, Inc. All rights reserved. McGraw-Hill/Irwin Simple Linear Regression Analysis Chapter 13.
1 Simple Linear Regression and Correlation Least Squares Method The Model Estimating the Coefficients EXAMPLE 1: USED CAR SALES.
© 2011 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part, except for use as permitted in a license.
CORRELATION ANALYSIS.
Intro to Statistics for the Behavioral Sciences PSYC 1900 Lecture 7: Regression.
©The McGraw-Hill Companies, Inc. 2008McGraw-Hill/Irwin Linear Regression and Correlation Chapter 13.
Chapter 13 Linear Regression and Correlation. Our Objectives  Draw a scatter diagram.  Understand and interpret the terms dependent and independent.
Correlation and Simple Linear Regression
Correlation and Simple Linear Regression
Correlation and Regression
Correlation and Simple Linear Regression
Simple Linear Regression and Correlation
Review I am examining differences in the mean between groups How many independent variables? OneMore than one How many groups? Two More than two ?? ?
Correlation and Simple Linear Regression
Correlation and Simple Linear Regression
Presentation transcript:

Chapter 10 For Explaining Psychological Statistics, 4th ed. by B. Cohen 1 A perfect correlation implies the ability to predict one score from another perfectly. Perfect predictions: –When dealing with z scores, the z score you predict for the Y variable is exactly the same as the z-score for the X variable –That is, when r = +1.0: z Y’ = z X –And, when r = -1.0: z Y’ = -z X When r is less than perfect, this rule must be modified, according to the strength of the correlation. The modified rule is the standardized regression equation, as shown on the next slide. Chapter 10: Linear Regression

Chapter 10 For Explaining Psychological Statistics, 4th ed. by B. Cohen 2 Predicting with z scores –Standardized Regression Equation: z Y’ = r z X If r = –1 or +1, the magnitude of the predicted z score is the same as the z score from which we are predicting. If r = 0, the z score prediction is always zero (i.e., the mean), which implies that, given no other information, our best prediction for a variable is its own mean. –As the magnitude of r becomes smaller, there is less of a tendency to expect an extreme score on one variable to be associated with an equally extreme score on the other. This is consistent with Galton’s concept of “regression toward mediocrity” (i.e., regression toward the mean).

Chapter 10 For Explaining Psychological Statistics, 4th ed. by B. Cohen 3 Raw score graph z score graph

Chapter 10 For Explaining Psychological Statistics, 4th ed. by B. Cohen. 4 Regression Formulas When Dealing With a Population –A basic formula for linear regression in terms of population means and standard deviations is as follows: –This formula can be simplified to the basic equation for a straight line: where and

Chapter 10For Explaining Psychological Statistics, 4th ed. by B. Cohen 5 Regression Formulas for Making Predictions From Samples –The same raw-score regression equa- tion is used when working with samples: except that the slope of the line is now found from the unbiased SDs: and the Y-intercept is now expressed in terms of the sample means:

Chapter 10 For Explaining Psychological Statistics, 4th ed. by B. Cohen 6 Quantifying the Errors Around the Regression Line –Residual: The difference between the actual Y value and the predicted Y value (Y – Y’). Each residual can be thought of as an error of prediction. –The positive and negative residuals will balance out so that the sum of the residuals will always be zero. –The linear regression equation gives us the straight line that minimizes the sum of the squared residuals (i.e., the sum of squared errors). Therefore, it is called the least-squares regression line. –The regression line functions like a running average of Y, in that it passes through the mean of the Y values (approximately) for each value of X.

Chapter 10 For Explaining Psychological Statistics, 4th ed. by B. Cohen 7 The Variance of the Estimate in a Population –Quantifies the average amount of squared error in the predictions : –The variance of the estimate (or residual variance) is the variance of the data points around the regression line. –As long as r is not zero, σ 2 est Y will be less than σ 2 Y (the ordinary variance of the Y values); the amount by which it is less represents the advantage of performing regression. –Larger rs (in absolute value) will lead to less error in prediction (i.e., points closer to the regression line), and therefore a smaller value for σ 2 est Y. –This relation between σ 2 est Y and Pearson’s r is shown in the following formula:

Chapter 10 For Explaining Psychological Statistics, 4th ed. by B. Cohen 8 Coefficient of Determination –The proportion of variance in the predicted variable that is not accounted for by the predicting variable is found by rearranging the formula for the variance of the estimate in the previous slide. 1 – r 2 = unexplained variance = σ 2 estY total variance σ 2 Y –The ratio of the variance of the estimate to the ordinary variance of Y is called the coefficient of nondetermination, and it is sometimes symbolized as k 2. –Larger absolute values of r are associated with smaller values for k 2. –The proportion of the total variance that is explained by the predictor variable is called the coefficient of determination, and it is simply equal to r 2 : r 2 = explained variance = 1 – k 2 total variance

Chapter 10For Explaining Psychological Statistics, 4th ed. by B. Cohen 9 Example from Lockhart, Robert S. (1998). Introduction to statistics and data analysis. New York: W. H. Freeman & Company. Here Is a Concrete Example of Linear Regression …

Chapter 10 For Explaining Psychological Statistics, 4th ed. by B. Cohen 10 Estimating the Variance of the Estimate From a Sample –When using a sample to estimate the variance of the estimate, we need to correct for bias, even though we are basing our formula on the unbiased estimate if the ordinary variance: Standard Error of the Estimate The standard error of the estimate is just the square root of the variance of the estimate. When estimating from a sample, the formula is:

Chapter 10 For Explaining Psychological Statistics, 4th ed. by B. Cohen 11 Assumptions Underlying Linear Regression –Independent random sampling –Bivariate normal distribution –Linearity of the relationship between the two variables –Homoscedasticity (i.e., the variance around the regression line is the same for every X value) Uses for Linear Regression –Prediction –Statistical control (i.e., removing the linear effect of one variable on another) –Quantifying the relationship between a DV and a manipulated IV with quantita- tive levels

Chapter 10For Explaining Psychological Statistics, 4th ed. by B. Cohen 12 The Point-Biserial Correlation Coefficient –An ordinary Pearson’s r calculated for one continuous multivalued variable and one dichotomous (i.e., grouping) variable. The sign of r pb is arbitrary and therefore usually ignored. –A r pb can be tested for significance with a one-sample t test as follows: –By solving for r pb, we obtain a simple formula for converting a two-sample pooled-variance t value into a correl- ational measure of effect size:

Chapter 10For Explaining Psychological Statistics, 4th ed. by B. Cohen 13 The Proportion of Variance Accounted for in a Two-Sample Comparison –Squaring r pb gives the proportion of vari- ance in your DV accounted for by your two-level IV (i.e., group membership). –Even when you obtain a large t value it is possible that little variance is accounted for; therefore r pb is a useful supplement to the two-sample t value. –r pb is an alternative to g for expressing the effect size found in your samples. The two measures have a fairly simple relationship: where N is the total number of cases across both groups, and df = N – 2

Chapter 10For Explaining Psychological Statistics, 4th ed. by B. Cohen 14 Estimating the Proportion of Variance Accounted for in the Population –r 2 pb from a sample tends to over- estimate the proportion of variance accounted for in the population. This bias can be corrected with the following formula: –ω 2 and d 2 are two different measures of the effect size in the population. They have a very simple relationship, as shown by the following formula: