CORRELATIONS ~ PARTIAL REGRESSION COEFFICIENTS (GROWTH STUDY PAPER #29) and. Charles E. Werts

Size: px
Start display at page:

Download "CORRELATIONS ~ PARTIAL REGRESSION COEFFICIENTS (GROWTH STUDY PAPER #29) and. Charles E. Werts"

Transcription

1 RB-69-6 ASSUMPTIONS IN MAKING CAUSAL INFERENCES FROM PART CORRELATIONS ~ PARTIAL CORRELATIONS AND PARTIAL REGRESSION COEFFICIENTS (GROWTH STUDY PAPER #29) Robert L. Linn and Charles E. Werts This Bulletin is a draft for interoffice circulation. Corrections and suggestions for revision are solicited. The Bulletin should not be cited as a reference without the specific permission of the authors. It is automatically superseded upon fonnal publication of the material. Educational Testing Service Princeton~ New Jersey January 1969

2 ASSUMPrIONS IN MAKING CAUSAL INFERENCES FROM PART CORRELATIONS} PARTIAL CORRELATIONS AND PARTIAL REGRESSION COEFFICIENTS Robert L. Linn and Charles E. Werts Educational Testing Service Abstract Given a linear model and that X is antecedent to Y,a third variable, W which is antecedent to both X and Y,is o~ten used as a control variable to remove any spurious association between X and Y Four different methods that have been used as measures of the "influence" of X on Yare described. The implicit assumptions that are made in using each of these methods to make causal inferences are stated and compared.

3 ASSUMPTIONS IN MAKING CAUSAL INFERENCES FROM PART CORRELATIONS, PARTIAL CORRELATIONS AND PARTIAL REGRESSION COEFFICIENTS Robert L. Linn and Charles E. Werts Educational Testing Service It is frequently the case in quasi-experimental and naturalistic studies that the investigator wishes to make "causal" inferences about the "influence" of variable X on variable Y when it is clear from the logic or design of the analysis that X is antecedent to Y To rule out alternative explanations for the observed association between X and Y,variables believed to be antecedent to both X and Yare controlled to remove "spurious" association. For example, the input-output model employed in studies of schools employs this logic since inferences are made about the effect of the school environment on an output variable with input variables controlled to eliminate the alternate hypothesis that the observed association between school environment and output may be the spurious resultant of both being influenced by input. In these studies it is reasonable to suppose that the school atmosphere may be influenced by the kinds of students who enter that school and that the output will be influenced by the input. In the above situation four rather different measures of "influence" of X on Y have been employed, often interchangeably: Method 1: The partial correlation between X and Y with prior variables controlled, e.g., in the case of one common antecedent variable, W, r XY W Method 2; The pg.rt correlation of X and Y with antecedent variables removed from X,e.g., r(x.w)y This correlation squared is I:arlington's (1968) "usefulness," Le., the proportion of

4 -2- variance that X adds to the prediction of Y in a stepwise regression after the common antecedent variables are entered. Method 3: The part correlation of X and Y with antecedent variables removed from Y, e.g., rx(y.w) Tnis procedure is especially common in school effects studies, in which a "residual" output is obtained by partialing input out of output and correlating the school variables with residual output. Most users of Method 3 have unwisely ignored McNemar's (1962) statement that this part correlation will be most useful when it can be argued that the common antecedent variables do not influence X, e.g., that the kind of student a school receives does not influence the school environment. Method 4: The standardized partial regression (i.e., beta) weight of Y on X,e.g., bh.w,obtained from the regression equation in which the antecedent variables and X are independent variables. Even though these four procedures may give Widely divergent estimates of the magnitude of the presumed influence (Werts & Linn, 1968) of X on Y (antecedents controlled), the size of the respective coefficients is interpreted as indicating a "large" or a "small" effect. Although it is clear that the different methods ask somewhat different questions and make somewhat different assumptions about the nature of the underlying phenomena, few investigators are aware of these differences or justify which method is relevant to the question under study. The assumptions (Duncan, Featherman, & Duncan, 1968) implied by the use of these methods for causal interpretation will be detailed in this paper, so thatinvest1gators may better decide which statistic 1s most appropriate.

5 -3- In discussing the interpretation of regression weights in a causal sense, Larlington (1968) notes the assumption that "all variables which might affect the dependent variable are included in the regression equation or are uncorrelated with the variables which are included." It follows that when a variable is not included as an independent variable the implicit causal assumption has been made that there are no other causes of the dependent variables which are correla.ted with the independent variables (linearity of all relationships is assumed throughout this paper). This principle derives from the necessity, when making causal interpretations, of assuming a closed theoretical system, i.e., no variables omitted. Since it is always the case that some variables are omitted the researcher must make the assumption that the variables omitted (Le., the "implicit" variables) do not affect the associations among observed variables. Thus if X 2 = b Xl l + e it is necessary to assume that e is 2 2 uncorrelated with the independent variable Xl vlliile many authors call e an error term, in a causal schema it represents the implicit (i.e., unmeasured) factors that act on the explicitly measured dependent variable (Blalock & Blalock, 1968, p. 200). X 3 = b l Xl + b 2 X 2 + e 3 If the causal system includes other variables, e.g., then it is necessary to assume that the implicit factors represented by e 2 and e 3 are uncorrelated in order for the causal model to be equivalent to a closed system. In actual research, if causal interpretations are to be made, at some point the simplifying assumption must be made that all variables which affect the dependent variables under study are either included in the system or are uncorrelated with the relevant independent variables and the implicit factors (Blalock, 1967). Because of the necessary assumptions about linearity, measurement error, and the effect of unmeasured variables, it will always be the case that causal inferences will be speculative.

6 ~- Since in a causal schema, factors influencing a particular dependent variable must be either included as independent variables or uncorrelated with the independent variables that are included, the implicit causal models for the four methods mentioned above can be specified since each procedure can be studied in terms of the sequence of regression equations used to obtain the respective coefficients. Thus the procedure for a partial correlation is to: (a) Use a regression equation with X as the dependent variable and the common antecedent variables as the independent variables to obtain the residual of X with antecedents controlled, i.e., herein labelled X'. (b) Use a regression equation with Y as the dependent variable and the common antecedent variables as independent variables to obtain the residuals of Y with antecedents controlled, which is herein labelled y t (c) In the regression equation with Y' as the dependent variable and XI X' the independent variable, the standardized regression weight for is the partial correlation coefficient which represents the influence of X' on y t It follows that when the partial correlation is interpreted in a causal sense, it is assumed that: (1) there are no other causes of X correlated with the common antecedent variables; (2) there are no other causes of Y correlated with the common antecedent variables, in particular X cannot be a cause of Y unless it 1s uncorrelated with the common antecedent variables; and (3) there are no other causes of y t which are correlated with X' The resulting implicit causal model for the partial correlation is shown in Figure 1 for the case of a single common antecedent variable, W It can be seen from Figure 1 that the

7 -5- zero order correlation between X and Y is considered to be totally spurious due to the common antecedent variables Wand X' (X' influences Y through the intervening variable Y'), the correlation r is the influence of W on WX X,the correlation r wy is the influence of W on Y,the correlation r X 'X( =,f I - rex).; I - rey) is the is the influence of X' on X the correlation ry,y( influence of y' on Y, the correlation rx'y' ( = r XY W ) is the influence of X' on y', the correlation r XY ' ( = r XY W v" I - rex) is completely spurious because of the common antecedent variable XI and the correlation results from X' influencing y' which in turn influences Y Insert Figure I about here In the case of method 2, the steps are: (a) Use a regression equation with X as the dependent variable and the common antecedent variables as independent variables to obtain the X residuals. (b) In the regression equation with X' as the independent variable and Y as the dependent variable, the standardized regression weight for Xl is the part correlation which measures the influence of XI on Y. It follows that method 2, when used in a causal sense, implies that all other influences on X are uncorrelated with the antecedent variables and that it is the hypothetical variable X' which influences Y as depicted in Figure 2. In Figure 2, r XY is the spurious resultant of the common antecedent variables Wand X' Note that both X' and W can be influences on Y since they are uncorrelated.

8 -6- Insert Figure 2 about here Method 3, the ~rt correlation of X with Y, common antecedent variables removed only from Y, proceeds as follows: (a) Use a regression equation with Y as the dependent variable and common antecedent variables as independent variables to obtain the Y residuals. (b) In the regression equation with X as the independent variable and y 1 as the dependent variable obtain the measure of the influence of X on Y' It follows from (a) that this ~rt correlation (e.g., rx(y.w» assumes that there are no other influences on Y correlated with the antecedent variables. However, if X is influenced by the antecedent variable(s) and X influences Y', then the antecedent variable(s) would be correlated with y r, which would. create a logical contradiction in the model since Y' is by definition uncorrelated with the antecedent variables. l Since we wish to measure the influence of X on y t we must assume that the antecedent variables do not influence X (1.e., r = 0) to avoid having contradictory assumptions in the model. WX Only when X is uncorrelated with (i.e., is not influenced by) the common antecedent variables, the standardized regression coefficient of Y' on X will equal the part correlation (e.g., rx(y.w» The causal model with this assumption is shown in Figure 3, for the case of a single common antecedent variable; the correlation r XY ( = rx(y.w) '/1 - r~ ) results from the influence of X on Y' which in turn influences Y The use of method 3 in the study of the influence of school environments on some output variable is not generally applicable since it 1s unusual for school variables to be uncorrelated with input variables.

9 -7- Insert Figure 3 about here Method 4, in contrast to the other methods, involves only the regression equation with X and the antecedent variables (e.g., W ) as independent variables and Y as the dependent variable. Although X and W may be correlated it is assumed that all other causes of Yare uncorrelated with both X and the common antecedent variables. As depicted in Figure 4, for the case of a single common antecedent variable (w), the correlation r XY has a nonspurious component bfx w representing the influence of X on Y and a spurious component (r XY - bh w) representing association due to the common antecedent variable W Algebraically r XY which can be recognized as a "normal" equation. The correlation r represents the total influence of W on Y,but this correlatio:::l is divided lnto WY two parts, bfw X the direct influence of W on Y and (ryw - bfw.x) which represents the indirect influence of W on Y via X Algebraically r yw b* which can be recognized as the other "normal" equation - b~.x = r vlx YXAI used to solve for the regression coefficients. An interesting property of the partial regression coefficients (Blalock & Blalock, 1968, p ) in these circumstances is that regardless of whether X is influenced by or influences the other independent variables in the equation (e.g., W ) the same regression coefficient for X will be obtained. It is the interpretation of (r yw - b~.x) which changes depending on the relationships of X to the other independent variables. Insert Figure 4 about here

10 -8- Algebraic Correspondences It will be noticed that the coefficient obtained in methods 1, 2, and 4 and in method 3 when X is uncorrelated with the antecedent variables (e.g., r wx = 0) is a standardized regression coefficient for a specified pair of independent and dependent variables. It may be shown further that the W'lstandardized regression coefficient between each of these pairs of variables will be the same for all four methods, i.e., each of the four procedures (for method 3, r wx = 0 ) is merely the W'lstandardized partial regression coefficient of Y on X ( W controlled) standardized by a different set of variances depending on what the independent and what the dependent variables are considered to be. The unstandardized coefficient for the case of a single common variable W is ) where 0y = standard deviation of y Ox = standard deviation of X Multiplying B yx W by the standard deviation of the independent variable and dividing by the standard deviation of the dependent variable we obtain: (1) r xy w = B yx W ox' 0y' 1/1 - r~, Ox r yx - r yw r XW = Byx W = 0y -V 1- r~ (2) r(x.w)y = Byx W ax, r WX r yx - ryyfj r - XW = B yx = 0y W 0y.yl- r 2 wx Ox 11-2

11 (3) rx(y.w) = Byx -W -9- Ox Ox r yx = Byx when 0, 0y' W ~l _ r 2 {I - r 2 r WX 0y WY WY (4) bh.w = Byx w Since all four methods when properly used are in fact simply variations of i~he unstandardized partial regression coefficient, why not simply study unstandardized regression coefficients in the manner of the econometricians? The standardized methods are highly subject to change due to changes in the variances of the variables from one sample to another. The effects of group heterogeneity on correlation coefficients are well known (see Lord & Novick, 1968, Chapter 6, for an excellent treatment of this topic). The standardized regression coefficients are also subject to this same influence. For this reason, the use of any of the four standardized methods must limit generalizations to very specifically defined populations since anything that restricted the range of the sample on one or more of the variables could lead to serious bias in the estimates and therefore the conclusions_ Unstandardized regression weights have the distinct advantage of not being affected by changes in the group heterogeneity. For this reason, Blalock (1967) argues that the unstandardized coefficients are more useful for purposes of stating general laws. A similar argument for the use of unstandardized weights rather than standardized weights is made by Tukey (1954). Tukey also argues that if the scales of the variables have any meaning, then the interpretation of the unstandardized weights is more straightforward. Due to the arbitrary nature of the scale units for most variables in the social sciences, however, the standardized coefficients will undoubtedly continue to be more generally used. Standardized measures

12 -10- lend themselves to the familiar language of "percentage of variance accounted forb which 1s often used to suggest that a given effect is "large ll or "small. tt Such language ascribes a meaning to an arbitrary scale, ignoring the crucial question of how this scale in any sense is a measure of the underlying phenomena. The language of prediction equations cannot be arbitrarily assumed to have meaning when used in the interpretation of causal systems. Summary Examination of Figures 1 through 4 shows that each of the four methods when used for causal interpretation accounts for the observed pattern of correlations with a different explanation or "theory" about the nature of the phenomena under study. In the case of a single common antecedent variable (w): Method 1 the partial correlation r XY W asserts that it is only Xl (the implicit factor(s) inf'luencing X ) which influences y' (the implicit factor(s) influencing Y ). Method 2 the part correlation r(x w)y asserts that it is only XI (the implicit factor(s) influencing X ) which ini'luences Y Method 3 the part correlation rx(y.w) asserts (when r WX = 0) that X inf'luences only Y' (the implicit factor(s) inf'luencing Y ). Method 4 the beta weight b~.w asserts that X influences Y ~.W is the (standardized) rate of change in Y when X varies, W having a fixed value (i.e., the inf'luence of X on Y, W con- trolled) Method. 4 differs from the others in that no hypothetical unmeasured variables, i.e., X' or Y',are invoked to explain the observed associations. The challenge to investigators who invoke implicit, unmeasured factors to explain observed data is obviously to measure these factors and then to

13 -11- demonstrate that they in fact account for the observed associations. Investigators who do not invoke unmeasured variables should be conscientious in searching for and incorporating new variables which would affect the variables under study.

14 -12- References Blalock, H. M. Causal inference, closed populations, and measures of association. American Political Science Review, 1967, 61, Blalock, H. M., & Blalock, A. B. Methodology in social research. New York: McGraw-Hill, Darlington, R. B. Multiple regression in research and practice. Psychological Bulletin, 1968, 69, Duncan, O. D., Featherman, D. L., & Duncan, B. Partials, partitions and paths. Socioeconomic background and occupational achievement. Final Report, Project No (EO-191), Contract No. OE , U. S. Office of Education. Ann Arbor: University of Michigan, Lord, F. M., & Novick, M. R. Statistical theories of mental test scores. New York: Addison-Wesley, McNemar, Q. Psychological statistics. (3rd ed.) New York: Wiley, P Tukey, J. W. Causation, regression, and path analyses. In O. Kempthorne, T. A. Bancroft, J. W. Gowen, & J. L. Lush (Eds.), Statistics and mathematics in biology. Ames, Iowa: Iowa State College Press, Werts, C. E., & Linn, R. L. Analyzing school effects: How to use the same data to support different hypotheses. Research Bulletin Princeton, N. J.: Educational Testing Service, 1968.

15 -13- Footnote lin Figure 3, the dilemma could be resolved by assuming that W influences Y' negatively, but this would contradict the intent of partialing W out of y to obtain a measure uninfluenced by W,e.g., in school studies the intent of partialing input out of output is to obtain a residual output uninfluenced by input. Furthermore if VI influenced YI then both Wand X would have to be included in the regression equation with Y' as the dependent variable and the part correlation rx(y.w) would not be the standardized weight for X in that equation.

16 -14- Figure Captions Fig. 1. The causal diagram corresponding to the causal interpretation of a partial correlation. Fig. 2. The causal diagram corresponding to method 2, the part correlation (r(x.w)y)' Fig. 3. The causal diagram corresponding to method 3, the part correlation rx(y'w). Fig. 4. The causal diagram corresponding to method 4, the standardized partial regression weight b~x.w'

17 e

18 e'

19 e" / " " " r wx ;: 0 ",,-,, "" /

20 e* el

Do not copy, post, or distribute

Do not copy, post, or distribute 14 CORRELATION ANALYSIS AND LINEAR REGRESSION Assessing the Covariability of Two Quantitative Properties 14.0 LEARNING OBJECTIVES In this chapter, we discuss two related techniques for assessing a possible

More information

VARIABILITY OF KUDER-RICHARDSON FOm~A 20 RELIABILITY ESTIMATES. T. Anne Cleary University of Wisconsin and Robert L. Linn Educational Testing Service

VARIABILITY OF KUDER-RICHARDSON FOm~A 20 RELIABILITY ESTIMATES. T. Anne Cleary University of Wisconsin and Robert L. Linn Educational Testing Service ~ E S [ B A U ~ L t L H E TI VARIABILITY OF KUDER-RICHARDSON FOm~A 20 RELIABILITY ESTIMATES RB-68-7 N T. Anne Cleary University of Wisconsin and Robert L. Linn Educational Testing Service This Bulletin

More information

The Use of Structural Equation Models in Interpreting Regression Equations Including Suppressor and Enhancer Variables

The Use of Structural Equation Models in Interpreting Regression Equations Including Suppressor and Enhancer Variables The Use of Structural Equation Models in Interpreting Regression Equations Including Suppressor and Enhancer Variables Robert M. McFatter University of Denver It is shown that the usual interpretation

More information

Key Algebraic Results in Linear Regression

Key Algebraic Results in Linear Regression Key Algebraic Results in Linear Regression James H. Steiger Department of Psychology and Human Development Vanderbilt University James H. Steiger (Vanderbilt University) 1 / 30 Key Algebraic Results in

More information

CHAPTER 4 THE COMMON FACTOR MODEL IN THE SAMPLE. From Exploratory Factor Analysis Ledyard R Tucker and Robert C. MacCallum

CHAPTER 4 THE COMMON FACTOR MODEL IN THE SAMPLE. From Exploratory Factor Analysis Ledyard R Tucker and Robert C. MacCallum CHAPTER 4 THE COMMON FACTOR MODEL IN THE SAMPLE From Exploratory Factor Analysis Ledyard R Tucker and Robert C. MacCallum 1997 65 CHAPTER 4 THE COMMON FACTOR MODEL IN THE SAMPLE 4.0. Introduction In Chapter

More information

RR R E E A H R E P DENOTING THE BASE FREE MEASURE OF CHANGE. Samuel Messick. Educational Testing Service Princeton, New Jersey December 1980

RR R E E A H R E P DENOTING THE BASE FREE MEASURE OF CHANGE. Samuel Messick. Educational Testing Service Princeton, New Jersey December 1980 RR 80 28 R E 5 E A RC H R E P o R T DENOTING THE BASE FREE MEASURE OF CHANGE Samuel Messick Educational Testing Service Princeton, New Jersey December 1980 DENOTING THE BASE-FREE MEASURE OF CHANGE Samuel

More information

Elementary Linear Algebra, Second Edition, by Spence, Insel, and Friedberg. ISBN Pearson Education, Inc., Upper Saddle River, NJ.

Elementary Linear Algebra, Second Edition, by Spence, Insel, and Friedberg. ISBN Pearson Education, Inc., Upper Saddle River, NJ. 2008 Pearson Education, Inc., Upper Saddle River, NJ. All rights reserved. APPENDIX: Mathematical Proof There are many mathematical statements whose truth is not obvious. For example, the French mathematician

More information

A TEST OF SIGNIFICANCE OF DIFFERENCE BETWEEN CORRELATED PROPORTIONS. John A. Keats

A TEST OF SIGNIFICANCE OF DIFFERENCE BETWEEN CORRELATED PROPORTIONS. John A. Keats ~ E S E B A U ~ L t L I-i E TI RB-55-20 A TEST OF SIGNIFICANCE OF DIFFERENCE BETWEEN CORRELATED PROPORTIONS John A. Keats N This Bulletin is a draft for interoffice circulation. Corrections and suggestions

More information

Chapter 5. Introduction to Path Analysis. Overview. Correlation and causation. Specification of path models. Types of path models

Chapter 5. Introduction to Path Analysis. Overview. Correlation and causation. Specification of path models. Types of path models Chapter 5 Introduction to Path Analysis Put simply, the basic dilemma in all sciences is that of how much to oversimplify reality. Overview H. M. Blalock Correlation and causation Specification of path

More information

DIFFERENT APPROACHES TO STATISTICAL INFERENCE: HYPOTHESIS TESTING VERSUS BAYESIAN ANALYSIS

DIFFERENT APPROACHES TO STATISTICAL INFERENCE: HYPOTHESIS TESTING VERSUS BAYESIAN ANALYSIS DIFFERENT APPROACHES TO STATISTICAL INFERENCE: HYPOTHESIS TESTING VERSUS BAYESIAN ANALYSIS THUY ANH NGO 1. Introduction Statistics are easily come across in our daily life. Statements such as the average

More information

An Introduction to Path Analysis

An Introduction to Path Analysis An Introduction to Path Analysis Developed by Sewall Wright, path analysis is a method employed to determine whether or not a multivariate set of nonexperimental data fits well with a particular (a priori)

More information

Variance Partitioning

Variance Partitioning Lecture 12 March 8, 2005 Applied Regression Analysis Lecture #12-3/8/2005 Slide 1 of 33 Today s Lecture Muddying the waters of regression. What not to do when considering the relative importance of variables

More information

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at Regression Analysis when there is Prior Information about Supplementary Variables Author(s): D. R. Cox Source: Journal of the Royal Statistical Society. Series B (Methodological), Vol. 22, No. 1 (1960),

More information

SAMPLE IS USED IN A NEW SAMPLE

SAMPLE IS USED IN A NEW SAMPLE EB 50-40 ~ E S E B A U R L t L Ii E TI EFFICIENCY OF PBEDICTION WHEN A REGRESSION EQUATION FROM ONE SAMPLE IS USED IN A NEW SAMPLE Frederic M. Lord (Prepublication draft) N ~----'-- This Bulletin is a

More information

Prentice Hall CME Project Algebra

Prentice Hall CME Project Algebra Prentice Hall CME Project Algebra 1 2009 Algebra 1 C O R R E L A T E D T O from March 2009 Algebra 1 A1.1 Relations and Functions A1.1.1 Determine whether a relation represented by a table, graph, words

More information

PIRLS 2016 Achievement Scaling Methodology 1

PIRLS 2016 Achievement Scaling Methodology 1 CHAPTER 11 PIRLS 2016 Achievement Scaling Methodology 1 The PIRLS approach to scaling the achievement data, based on item response theory (IRT) scaling with marginal estimation, was developed originally

More information

Introduction to Linear Regression

Introduction to Linear Regression Introduction to Linear Regression James H. Steiger Department of Psychology and Human Development Vanderbilt University James H. Steiger (Vanderbilt University) Introduction to Linear Regression 1 / 46

More information

Causal Inference Lecture Notes: Causal Inference with Repeated Measures in Observational Studies

Causal Inference Lecture Notes: Causal Inference with Repeated Measures in Observational Studies Causal Inference Lecture Notes: Causal Inference with Repeated Measures in Observational Studies Kosuke Imai Department of Politics Princeton University November 13, 2013 So far, we have essentially assumed

More information

AN INDEX OF THE DISCRIMINATING POWER OF A TEST. Richard Levine and Frederic M. Lord

AN INDEX OF THE DISCRIMINATING POWER OF A TEST. Richard Levine and Frederic M. Lord RB-58-13 ~ [ s [ B A U ~ L t L I-t [ T I N AN INDEX OF THE DISCRIMINATING POWER OF A TEST AT DIFFERENT PARTS OF THE SCORE RANGE Richard Levine and Frederic M. Lord This Bulletin is a draft for interoffice

More information

B. Weaver (24-Mar-2005) Multiple Regression Chapter 5: Multiple Regression Y ) (5.1) Deviation score = (Y i

B. Weaver (24-Mar-2005) Multiple Regression Chapter 5: Multiple Regression Y ) (5.1) Deviation score = (Y i B. Weaver (24-Mar-2005) Multiple Regression... 1 Chapter 5: Multiple Regression 5.1 Partial and semi-partial correlation Before starting on multiple regression per se, we need to consider the concepts

More information

Variance Partitioning

Variance Partitioning Chapter 9 October 22, 2008 ERSH 8320 Lecture #8-10/22/2008 Slide 1 of 33 Today s Lecture Test review and discussion. Today s Lecture Chapter 9: Muddying the waters of regression. What not to do when considering

More information

Math 300 Introduction to Mathematical Reasoning Autumn 2017 Proof Templates 1

Math 300 Introduction to Mathematical Reasoning Autumn 2017 Proof Templates 1 Math 300 Introduction to Mathematical Reasoning Autumn 2017 Proof Templates 1 In its most basic form, a mathematical proof is just a sequence of mathematical statements, connected to each other by strict

More information

1 Correlation and Inference from Regression

1 Correlation and Inference from Regression 1 Correlation and Inference from Regression Reading: Kennedy (1998) A Guide to Econometrics, Chapters 4 and 6 Maddala, G.S. (1992) Introduction to Econometrics p. 170-177 Moore and McCabe, chapter 12 is

More information

Propositional Logic Review

Propositional Logic Review Propositional Logic Review UC Berkeley, Philosophy 142, Spring 2016 John MacFarlane The task of describing a logical system comes in three parts: Grammar Describing what counts as a formula Semantics Defining

More information

Correlational Analysis and Causal Inferences

Correlational Analysis and Causal Inferences Correlational Analysis and Causal Inferences H. M. BLALOCK, JR. University of Michigan NE of the most difficult methodological and theoretical problems con- 0 fronting social scientists is that of making

More information

Instrumental Variables

Instrumental Variables James H. Steiger Department of Psychology and Human Development Vanderbilt University Regression Modeling, 2009 1 Introduction 2 3 4 Instrumental variables allow us to get a better estimate of a causal

More information

Introduction to Structural Equation Modeling

Introduction to Structural Equation Modeling Introduction to Structural Equation Modeling Notes Prepared by: Lisa Lix, PhD Manitoba Centre for Health Policy Topics Section I: Introduction Section II: Review of Statistical Concepts and Regression

More information

, (1) e i = ˆσ 1 h ii. c 2016, Jeffrey S. Simonoff 1

, (1) e i = ˆσ 1 h ii. c 2016, Jeffrey S. Simonoff 1 Regression diagnostics As is true of all statistical methodologies, linear regression analysis can be a very effective way to model data, as along as the assumptions being made are true. For the regression

More information

Ch. 16: Correlation and Regression

Ch. 16: Correlation and Regression Ch. 1: Correlation and Regression With the shift to correlational analyses, we change the very nature of the question we are asking of our data. Heretofore, we were asking if a difference was likely to

More information

Interaction effects for continuous predictors in regression modeling

Interaction effects for continuous predictors in regression modeling Interaction effects for continuous predictors in regression modeling Testing for interactions The linear regression model is undoubtedly the most commonly-used statistical model, and has the advantage

More information

The Dual of the Maximum Likelihood Method

The Dual of the Maximum Likelihood Method Open Journal of Statistics, 06, 6, 86-93 Published Online February 06 in SciRes. http://www.scirp.org/journal/ojs http://dx.doi.org/0.436/ojs.06.606 The Dual of the Maximum Likelihood Method Quirino Paris

More information

3/10/03 Gregory Carey Cholesky Problems - 1. Cholesky Problems

3/10/03 Gregory Carey Cholesky Problems - 1. Cholesky Problems 3/10/03 Gregory Carey Cholesky Problems - 1 Cholesky Problems Gregory Carey Department of Psychology and Institute for Behavioral Genetics University of Colorado Boulder CO 80309-0345 Email: gregory.carey@colorado.edu

More information

PROBABILITIES OF MISCLASSIFICATION IN DISCRIMINATORY ANALYSIS. M. Clemens Johnson

PROBABILITIES OF MISCLASSIFICATION IN DISCRIMINATORY ANALYSIS. M. Clemens Johnson RB-55-22 ~ [ s [ B A U R L t L Ii [ T I N PROBABILITIES OF MISCLASSIFICATION IN DISCRIMINATORY ANALYSIS M. Clemens Johnson This Bulletin is a draft for interoffice circulation. Corrections and suggestions

More information

The Matrix Algebra of Sample Statistics

The Matrix Algebra of Sample Statistics The Matrix Algebra of Sample Statistics James H. Steiger Department of Psychology and Human Development Vanderbilt University James H. Steiger (Vanderbilt University) The Matrix Algebra of Sample Statistics

More information

UNIT 4 RANK CORRELATION (Rho AND KENDALL RANK CORRELATION

UNIT 4 RANK CORRELATION (Rho AND KENDALL RANK CORRELATION UNIT 4 RANK CORRELATION (Rho AND KENDALL RANK CORRELATION Structure 4.0 Introduction 4.1 Objectives 4. Rank-Order s 4..1 Rank-order data 4.. Assumptions Underlying Pearson s r are Not Satisfied 4.3 Spearman

More information

Instrumental Variables

Instrumental Variables Instrumental Variables James H. Steiger Department of Psychology and Human Development Vanderbilt University James H. Steiger (Vanderbilt University) Instrumental Variables 1 / 10 Instrumental Variables

More information

Lewis 2 Definitions of T-terms

Lewis 2 Definitions of T-terms Lewis 2 Definitions of T-terms (pp. 431 438, 445 446) Patrick Maher Philosophy 471 Fall 2006 Review The Ramsey sentence of T says T is realized: x 1... x n [x 1... x n ] The Carnap sentence says if T is

More information

SC705: Advanced Statistics Instructor: Natasha Sarkisian Class notes: Introduction to Structural Equation Modeling (SEM)

SC705: Advanced Statistics Instructor: Natasha Sarkisian Class notes: Introduction to Structural Equation Modeling (SEM) SC705: Advanced Statistics Instructor: Natasha Sarkisian Class notes: Introduction to Structural Equation Modeling (SEM) SEM is a family of statistical techniques which builds upon multiple regression,

More information

PROGRAM STATISTICS RESEARCH

PROGRAM STATISTICS RESEARCH An Alternate Definition of the ETS Delta Scale of Item Difficulty Paul W. Holland and Dorothy T. Thayer @) PROGRAM STATISTICS RESEARCH TECHNICAL REPORT NO. 85..64 EDUCATIONAL TESTING SERVICE PRINCETON,

More information

NBER WORKING PAPER SERIES IS THE SPURIOUS REGRESSION PROBLEM SPURIOUS? Bennett T. McCallum. Working Paper

NBER WORKING PAPER SERIES IS THE SPURIOUS REGRESSION PROBLEM SPURIOUS? Bennett T. McCallum. Working Paper NBER WORKING PAPER SERIES IS THE SPURIOUS REGRESSION PROBLEM SPURIOUS? Bennett T. McCallum Working Paper 15690 http://www.nber.org/papers/w15690 NATIONAL BUREAU OF ECONOMIC RESEARCH 1050 Massachusetts

More information

Regression Analysis. Table Relationship between muscle contractile force (mj) and stimulus intensity (mv).

Regression Analysis. Table Relationship between muscle contractile force (mj) and stimulus intensity (mv). Regression Analysis Two variables may be related in such a way that the magnitude of one, the dependent variable, is assumed to be a function of the magnitude of the second, the independent variable; however,

More information

Section 1.2: Propositional Logic

Section 1.2: Propositional Logic Section 1.2: Propositional Logic January 17, 2017 Abstract Now we re going to use the tools of formal logic to reach logical conclusions ( prove theorems ) based on wffs formed by some given statements.

More information

Manual of Logical Style

Manual of Logical Style Manual of Logical Style Dr. Holmes January 9, 2015 Contents 1 Introduction 2 2 Conjunction 3 2.1 Proving a conjunction...................... 3 2.2 Using a conjunction........................ 3 3 Implication

More information

Feature selection and classifier performance in computer-aided diagnosis: The effect of finite sample size

Feature selection and classifier performance in computer-aided diagnosis: The effect of finite sample size Feature selection and classifier performance in computer-aided diagnosis: The effect of finite sample size Berkman Sahiner, a) Heang-Ping Chan, Nicholas Petrick, Robert F. Wagner, b) and Lubomir Hadjiiski

More information

UNIVERSITY OF ILLINOIS LIBRARY AT URBANA-CHAMPAIGN BOOKSTACKS

UNIVERSITY OF ILLINOIS LIBRARY AT URBANA-CHAMPAIGN BOOKSTACKS UNIVERSITY OF ILLINOIS LIBRARY AT URBANA-CHAMPAIGN BOOKSTACKS i CENTRAL CIRCULATION BOOKSTACKS The person charging this material is responsible for its renewal or its return to the library from which it

More information

Supplementary Logic Notes CSE 321 Winter 2009

Supplementary Logic Notes CSE 321 Winter 2009 1 Propositional Logic Supplementary Logic Notes CSE 321 Winter 2009 1.1 More efficient truth table methods The method of using truth tables to prove facts about propositional formulas can be a very tedious

More information

7.1 Significance of question: are there laws in S.S.? (Why care?) Possible answers:

7.1 Significance of question: are there laws in S.S.? (Why care?) Possible answers: I. Roberts: There are no laws of the social sciences Social sciences = sciences involving human behaviour (Economics, Psychology, Sociology, Political Science) 7.1 Significance of question: are there laws

More information

chapter 12 MORE MATRIX ALGEBRA 12.1 Systems of Linear Equations GOALS

chapter 12 MORE MATRIX ALGEBRA 12.1 Systems of Linear Equations GOALS chapter MORE MATRIX ALGEBRA GOALS In Chapter we studied matrix operations and the algebra of sets and logic. We also made note of the strong resemblance of matrix algebra to elementary algebra. The reader

More information

Research Design: Causal inference and counterfactuals

Research Design: Causal inference and counterfactuals Research Design: Causal inference and counterfactuals University College Dublin 8 March 2013 1 2 3 4 Outline 1 2 3 4 Inference In regression analysis we look at the relationship between (a set of) independent

More information

Correlation and regression

Correlation and regression NST 1B Experimental Psychology Statistics practical 1 Correlation and regression Rudolf Cardinal & Mike Aitken 11 / 12 November 2003 Department of Experimental Psychology University of Cambridge Handouts:

More information

y response variable x 1, x 2,, x k -- a set of explanatory variables

y response variable x 1, x 2,, x k -- a set of explanatory variables 11. Multiple Regression and Correlation y response variable x 1, x 2,, x k -- a set of explanatory variables In this chapter, all variables are assumed to be quantitative. Chapters 12-14 show how to incorporate

More information

A Guide to Proof-Writing

A Guide to Proof-Writing A Guide to Proof-Writing 437 A Guide to Proof-Writing by Ron Morash, University of Michigan Dearborn Toward the end of Section 1.5, the text states that there is no algorithm for proving theorems.... Such

More information

Applied Regression Analysis

Applied Regression Analysis Applied Regression Analysis Lecture 2 January 27, 2005 Lecture #2-1/27/2005 Slide 1 of 46 Today s Lecture Simple linear regression. Partitioning the sum of squares. Tests of significance.. Regression diagnostics

More information

Hypothesis Tests and Confidence Intervals in Multiple Regression

Hypothesis Tests and Confidence Intervals in Multiple Regression Hypothesis Tests and Confidence Intervals in Multiple Regression (SW Chapter 7) Outline 1. Hypothesis tests and confidence intervals for one coefficient. Joint hypothesis tests on multiple coefficients

More information

Multiple Regression. More Hypothesis Testing. More Hypothesis Testing The big question: What we really want to know: What we actually know: We know:

Multiple Regression. More Hypothesis Testing. More Hypothesis Testing The big question: What we really want to know: What we actually know: We know: Multiple Regression Ψ320 Ainsworth More Hypothesis Testing What we really want to know: Is the relationship in the population we have selected between X & Y strong enough that we can use the relationship

More information

A consideration of the chi-square test of Hardy-Weinberg equilibrium in a non-multinomial situation

A consideration of the chi-square test of Hardy-Weinberg equilibrium in a non-multinomial situation Ann. Hum. Genet., Lond. (1975), 39, 141 Printed in Great Britain 141 A consideration of the chi-square test of Hardy-Weinberg equilibrium in a non-multinomial situation BY CHARLES F. SING AND EDWARD D.

More information

Milton Friedman Essays in Positive Economics Part I - The Methodology of Positive Economics University of Chicago Press (1953), 1970, pp.

Milton Friedman Essays in Positive Economics Part I - The Methodology of Positive Economics University of Chicago Press (1953), 1970, pp. Milton Friedman Essays in Positive Economics Part I - The Methodology of Positive Economics University of Chicago Press (1953), 1970, pp. 3-43 CAN A HYPOTHESIS BE TESTED BY THE REALISM OF ITS ASSUMPTIONS?

More information

Propositions and Proofs

Propositions and Proofs Chapter 2 Propositions and Proofs The goal of this chapter is to develop the two principal notions of logic, namely propositions and proofs There is no universal agreement about the proper foundations

More information

STOCKHOLM UNIVERSITY Department of Economics Course name: Empirical Methods Course code: EC40 Examiner: Lena Nekby Number of credits: 7,5 credits Date of exam: Friday, June 5, 009 Examination time: 3 hours

More information

An Introduction to Mplus and Path Analysis

An Introduction to Mplus and Path Analysis An Introduction to Mplus and Path Analysis PSYC 943: Fundamentals of Multivariate Modeling Lecture 10: October 30, 2013 PSYC 943: Lecture 10 Today s Lecture Path analysis starting with multivariate regression

More information

Dr. StrangeLOVE, or. How I Learned to Stop Worrying and Love Omitted Variables. Adam W. Meade, Tara S. Behrend and Charles E.

Dr. StrangeLOVE, or. How I Learned to Stop Worrying and Love Omitted Variables. Adam W. Meade, Tara S. Behrend and Charles E. AU: Check that your name is presented correctly and consistently here against the TOC 4 Dr. StrangeLOVE, or How I Learned to Stop Worrying and Love Omitted Variables Adam W. Meade, Tara S. Behrend and

More information

International Journal of Statistics: Advances in Theory and Applications

International Journal of Statistics: Advances in Theory and Applications International Journal of Statistics: Advances in Theory and Applications Vol. 1, Issue 1, 2017, Pages 1-19 Published Online on April 7, 2017 2017 Jyoti Academic Press http://jyotiacademicpress.org COMPARING

More information

On the learning of algebra

On the learning of algebra On the learning of algebra H. Wu Department of Mathematics #3840 University of California, Berkeley Berkeley, CA 94720-3840 USA http://www.math.berkeley.edu/ wu/ wu@math.berkeley.edu One of the main goals

More information

Mgmt 469. Causality and Identification

Mgmt 469. Causality and Identification Mgmt 469 Causality and Identification As you have learned by now, a key issue in empirical research is identifying the direction of causality in the relationship between two variables. This problem often

More information

Draft Proof - Do not copy, post, or distribute. Chapter Learning Objectives REGRESSION AND CORRELATION THE SCATTER DIAGRAM

Draft Proof - Do not copy, post, or distribute. Chapter Learning Objectives REGRESSION AND CORRELATION THE SCATTER DIAGRAM 1 REGRESSION AND CORRELATION As we learned in Chapter 9 ( Bivariate Tables ), the differential access to the Internet is real and persistent. Celeste Campos-Castillo s (015) research confirmed the impact

More information

References. Regression standard errors in clustered samples. diagonal matrix with elements

References. Regression standard errors in clustered samples. diagonal matrix with elements Stata Technical Bulletin 19 diagonal matrix with elements ( q=ferrors (0) if r > 0 W ii = (1 q)=f errors (0) if r < 0 0 otherwise and R 2 is the design matrix X 0 X. This is derived from formula 3.11 in

More information

Quantitative Understanding in Biology Module II: Model Parameter Estimation Lecture I: Linear Correlation and Regression

Quantitative Understanding in Biology Module II: Model Parameter Estimation Lecture I: Linear Correlation and Regression Quantitative Understanding in Biology Module II: Model Parameter Estimation Lecture I: Linear Correlation and Regression Correlation Linear correlation and linear regression are often confused, mostly

More information

Ignoring the matching variables in cohort studies - when is it valid, and why?

Ignoring the matching variables in cohort studies - when is it valid, and why? Ignoring the matching variables in cohort studies - when is it valid, and why? Arvid Sjölander Abstract In observational studies of the effect of an exposure on an outcome, the exposure-outcome association

More information

An Introduction to Path Analysis

An Introduction to Path Analysis An Introduction to Path Analysis PRE 905: Multivariate Analysis Lecture 10: April 15, 2014 PRE 905: Lecture 10 Path Analysis Today s Lecture Path analysis starting with multivariate regression then arriving

More information

Class Notes Spring 2014

Class Notes Spring 2014 Psychology 513 Quantitative Models in Psychology Class Notes Spring 2014 Robert M. McFatter University of Louisiana Lafayette 5.5 5 4.5 Positive Emotional Intensity 4 3.5 3 2.5 2.5 1.25 2-2.5-2 -1.5-1

More information

CHAPTER 4 & 5 Linear Regression with One Regressor. Kazu Matsuda IBEC PHBU 430 Econometrics

CHAPTER 4 & 5 Linear Regression with One Regressor. Kazu Matsuda IBEC PHBU 430 Econometrics CHAPTER 4 & 5 Linear Regression with One Regressor Kazu Matsuda IBEC PHBU 430 Econometrics Introduction Simple linear regression model = Linear model with one independent variable. y = dependent variable

More information

Dr. Junchao Xia Center of Biophysics and Computational Biology. Fall /1/2016 1/46

Dr. Junchao Xia Center of Biophysics and Computational Biology. Fall /1/2016 1/46 BIO5312 Biostatistics Lecture 10:Regression and Correlation Methods Dr. Junchao Xia Center of Biophysics and Computational Biology Fall 2016 11/1/2016 1/46 Outline In this lecture, we will discuss topics

More information

CSE 355 Test 2, Fall 2016

CSE 355 Test 2, Fall 2016 CSE 355 Test 2, Fall 2016 28 October 2016, 8:35-9:25 a.m., LSA 191 Last Name SAMPLE ASU ID 1357924680 First Name(s) Ima Regrading of Midterms If you believe that your grade has not been added up correctly,

More information

Causality and research design: experiments versus observation. Harry Ganzeboom Research Skills #3 November

Causality and research design: experiments versus observation. Harry Ganzeboom Research Skills #3 November Causality and research design: experiments versus observation Harry Ganzeboom Research Skills #3 November 10 2008 Experiments All scientific questions are implicitly explanatory questions and ask for causal

More information

review session gov 2000 gov 2000 () review session 1 / 38

review session gov 2000 gov 2000 () review session 1 / 38 review session gov 2000 gov 2000 () review session 1 / 38 Overview Random Variables and Probability Univariate Statistics Bivariate Statistics Multivariate Statistics Causal Inference gov 2000 () review

More information

Alternatives to Difference Scores: Polynomial Regression and Response Surface Methodology. Jeffrey R. Edwards University of North Carolina

Alternatives to Difference Scores: Polynomial Regression and Response Surface Methodology. Jeffrey R. Edwards University of North Carolina Alternatives to Difference Scores: Polynomial Regression and Response Surface Methodology Jeffrey R. Edwards University of North Carolina 1 Outline I. Types of Difference Scores II. Questions Difference

More information

1.1 Simple functions and equations Polynomials and polynomial equations

1.1 Simple functions and equations Polynomials and polynomial equations 1 Preliminary algebra This opening chapter reviews the basic algebra of which a working knowledge is presumed in the rest of the book. Many students will be familiar with much, if not all, of it, but recent

More information

Linear Modelling in Stata Session 6: Further Topics in Linear Modelling

Linear Modelling in Stata Session 6: Further Topics in Linear Modelling Linear Modelling in Stata Session 6: Further Topics in Linear Modelling Mark Lunt Arthritis Research UK Epidemiology Unit University of Manchester 14/11/2017 This Week Categorical Variables Categorical

More information

Spatial Layout and the Promotion of Innovation in Organizations

Spatial Layout and the Promotion of Innovation in Organizations Spatial Layout and the Promotion of Innovation in Organizations Jean Wineman, Felichism Kabo, Jason Owen-Smith, Gerald Davis University of Michigan, Ann Arbor, Michigan ABSTRACT: Research on the enabling

More information

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at A Note on the Efficiency of Least-Squares Estimates Author(s): D. R. Cox and D. V. Hinkley Source: Journal of the Royal Statistical Society. Series B (Methodological), Vol. 30, No. 2 (1968), pp. 284-289

More information

REVIEW 8/2/2017 陈芳华东师大英语系

REVIEW 8/2/2017 陈芳华东师大英语系 REVIEW Hypothesis testing starts with a null hypothesis and a null distribution. We compare what we have to the null distribution, if the result is too extreme to belong to the null distribution (p

More information

Research Design - - Topic 19 Multiple regression: Applications 2009 R.C. Gardner, Ph.D.

Research Design - - Topic 19 Multiple regression: Applications 2009 R.C. Gardner, Ph.D. Research Design - - Topic 19 Multiple regression: Applications 2009 R.C. Gardner, Ph.D. Curve Fitting Mediation analysis Moderation Analysis 1 Curve Fitting The investigation of non-linear functions using

More information

Outline

Outline 2559 Outline cvonck@111zeelandnet.nl 1. Review of analysis of variance (ANOVA), simple regression analysis (SRA), and path analysis (PA) 1.1 Similarities and differences between MRA with dummy variables

More information

Two-Variable Regression Model: The Problem of Estimation

Two-Variable Regression Model: The Problem of Estimation Two-Variable Regression Model: The Problem of Estimation Introducing the Ordinary Least Squares Estimator Jamie Monogan University of Georgia Intermediate Political Methodology Jamie Monogan (UGA) Two-Variable

More information

4:3 LEC - PLANNED COMPARISONS AND REGRESSION ANALYSES

4:3 LEC - PLANNED COMPARISONS AND REGRESSION ANALYSES 4:3 LEC - PLANNED COMPARISONS AND REGRESSION ANALYSES FOR SINGLE FACTOR BETWEEN-S DESIGNS Planned or A Priori Comparisons We previously showed various ways to test all possible pairwise comparisons for

More information

Chapter 1 Elementary Logic

Chapter 1 Elementary Logic 2017-2018 Chapter 1 Elementary Logic The study of logic is the study of the principles and methods used in distinguishing valid arguments from those that are not valid. The aim of this chapter is to help

More information

Use of e-rater in Scoring of the TOEFL ibt Writing Test

Use of e-rater in Scoring of the TOEFL ibt Writing Test Research Report ETS RR 11-25 Use of e-rater in Scoring of the TOEFL ibt Writing Test Shelby J. Haberman June 2011 Use of e-rater in Scoring of the TOEFL ibt Writing Test Shelby J. Haberman ETS, Princeton,

More information

15: Regression. Introduction

15: Regression. Introduction 15: Regression Introduction Regression Model Inference About the Slope Introduction As with correlation, regression is used to analyze the relation between two continuous (scale) variables. However, regression

More information

1 Differentiability at a point

1 Differentiability at a point Notes by David Groisser, Copyright c 2012 What does mean? These notes are intended as a supplement (not a textbook-replacement) for a class at the level of Calculus 3, but can be used in a higher-level

More information

Tests about a population mean

Tests about a population mean October 2 nd, 2017 Overview Week 1 Week 2 Week 4 Week 7 Week 10 Week 12 Chapter 1: Descriptive statistics Chapter 6: Statistics and Sampling Distributions Chapter 7: Point Estimation Chapter 8: Confidence

More information

MBF1923 Econometrics Prepared by Dr Khairul Anuar

MBF1923 Econometrics Prepared by Dr Khairul Anuar MBF1923 Econometrics Prepared by Dr Khairul Anuar L4 Ordinary Least Squares www.notes638.wordpress.com Ordinary Least Squares The bread and butter of regression analysis is the estimation of the coefficient

More information

On the Impossibility of Certain Ranking Functions

On the Impossibility of Certain Ranking Functions On the Impossibility of Certain Ranking Functions Jin-Yi Cai Abstract Suppose all the individuals in a field are linearly ordered. Groups of individuals form teams. Is there a perfect ranking function

More information

Metainduction in Operational Set Theory

Metainduction in Operational Set Theory Metainduction in Operational Set Theory Luis E. Sanchis Department of Electrical Engineering and Computer Science Syracuse University Syracuse, NY 13244-4100 Sanchis@top.cis.syr.edu http://www.cis.syr.edu/

More information

Understanding Ding s Apparent Paradox

Understanding Ding s Apparent Paradox Submitted to Statistical Science Understanding Ding s Apparent Paradox Peter M. Aronow and Molly R. Offer-Westort Yale University 1. INTRODUCTION We are grateful for the opportunity to comment on A Paradox

More information

Mobolaji Williams Motifs in Physics April 26, Effective Theory

Mobolaji Williams Motifs in Physics April 26, Effective Theory Mobolaji Williams Motifs in Physics April 26, 2017 Effective Theory These notes 1 are part of a series concerning Motifs in Physics in which we highlight recurrent concepts, techniques, and ways of understanding

More information

Module 1 Linear Regression

Module 1 Linear Regression Regression Analysis Although many phenomena can be modeled with well-defined and simply stated mathematical functions, as illustrated by our study of linear, exponential and quadratic functions, the world

More information

Chapter 12 Summarizing Bivariate Data Linear Regression and Correlation

Chapter 12 Summarizing Bivariate Data Linear Regression and Correlation Chapter 1 Summarizing Bivariate Data Linear Regression and Correlation This chapter introduces an important method for making inferences about a linear correlation (or relationship) between two variables,

More information

Introduction to Uncertainty and Treatment of Data

Introduction to Uncertainty and Treatment of Data Introduction to Uncertainty and Treatment of Data Introduction The purpose of this experiment is to familiarize the student with some of the instruments used in making measurements in the physics laboratory,

More information

Applied Statistics and Econometrics

Applied Statistics and Econometrics Applied Statistics and Econometrics Lecture 6 Saul Lach September 2017 Saul Lach () Applied Statistics and Econometrics September 2017 1 / 53 Outline of Lecture 6 1 Omitted variable bias (SW 6.1) 2 Multiple

More information

EM375 STATISTICS AND MEASUREMENT UNCERTAINTY CORRELATION OF EXPERIMENTAL DATA

EM375 STATISTICS AND MEASUREMENT UNCERTAINTY CORRELATION OF EXPERIMENTAL DATA EM375 STATISTICS AND MEASUREMENT UNCERTAINTY CORRELATION OF EXPERIMENTAL DATA In this unit of the course we use statistical methods to look for trends in data. Often experiments are conducted by having

More information