STAT Factor Analysis in SAS

Similar documents
NEUROBLASTOMA DATA -- TWO GROUPS -- QUANTITATIVE MEASURES 38 15:37 Saturday, January 25, 2003

Subescala D CULTURA ORGANIZACIONAL. Factor Analysis

Subescala B Compromisso com a organização escolar. Factor Analysis

1. Objective: analyzing CD4 counts data using GEE marginal model and random effects model. Demonstrate the analysis using SAS and STATA.

HZAU MULTIVARIATE HOMEWORK #2 MULTIPLE AND STEPWISE LINEAR REGRESSION

Effect of Source and Level of Protein on Weight Gain of Rats

APÊNDICE 6. Análise fatorial e análise de consistência interna

Chapter 6 Measures of Bivariate Association 1

Notes for laboratory session 2

Here are the various choices. All of them are found in the Analyze menu in SPSS, under the sub-menu for Descriptive Statistics :

isc ove ring i Statistics sing SPSS

Statistics as a Tool. A set of tools for collecting, organizing, presenting and analyzing numerical facts or observations.

Hungry Mice. NP: Mice in this group ate as much as they pleased of a non-purified, standard diet for laboratory mice.

Understandable Statistics

HS Exam 1 -- March 9, 2006

General Example: Gas Mileage (Stat 5044 Schabenberger & J.P.Morgen)

Use the above variables and any you might need to construct to specify the MODEL A/C comparisons you would use to ask the following questions.

bivariate analysis: The statistical analysis of the relationship between two variables.

What you should know before you collect data. BAE 815 (Fall 2017) Dr. Zifei Liu

List of Figures. List of Tables. Preface to the Second Edition. Preface to the First Edition

Survey research (Lecture 1) Summary & Conclusion. Lecture 10 Survey Research & Design in Psychology James Neill, 2015 Creative Commons Attribution 4.

Survey research (Lecture 1)

ANOVA in SPSS (Practical)

Psych 5741/5751: Data Analysis University of Boulder Gary McClelland & Charles Judd. Exam #2, Spring 1992

2. Scientific question: Determine whether there is a difference between boys and girls with respect to the distance and its change over time.

Chapter 9. Factorial ANOVA with Two Between-Group Factors 10/22/ Factorial ANOVA with Two Between-Group Factors

Explore. sexcntry Sex according to country. [DataSet1] D:\NORA\NORA Main File.sav

Chapter 5. Describing numerical data

NORTH SOUTH UNIVERSITY TUTORIAL 2

Small Group Presentations

Applications. DSC 410/510 Multivariate Statistical Methods. Discriminating Two Groups. What is Discriminant Analysis

Stat Wk 8: Continuous Data

Business Research Methods. Introduction to Data Analysis

2.4.1 STA-O Assessment 2

Assessing the Validity and Reliability of the Teacher Keys Effectiveness. System (TKES) and the Leader Keys Effectiveness System (LKES)

Summary & Conclusion. Lecture 10 Survey Research & Design in Psychology James Neill, 2016 Creative Commons Attribution 4.0

SPSS Portfolio. Brittany Murray BUSA MWF 1:00pm-1:50pm

Sociology 63993, Exam1 February 12, 2015 Richard Williams, University of Notre Dame,

RESULTS. Chapter INTRODUCTION

Quantitative Methods in Computing Education Research (A brief overview tips and techniques)

Factor Analysis. MERMAID Series 12/11. Galen E. Switzer, PhD Rachel Hess, MD, MS

Chapter 1: Explaining Behavior

Unit 1 Exploring and Understanding Data

Stepwise method Modern Model Selection Methods Quantile-Quantile plot and tests for normality

Answer to exercise: Growth of guinea pigs

PRINTABLE VERSION. Quiz 1. True or False: The amount of rainfall in your state last month is an example of continuous data.

Data Analysis Using SPSS. By: Akmal Aini Othman

Table of Contents. Plots. Essential Statistics for Nursing Research 1/12/2017

Introduction to Factor Analysis. Hsueh-Sheng Wu CFDR Workshop Series June 18, 2018

Population. Sample. AP Statistics Notes for Chapter 1 Section 1.0 Making Sense of Data. Statistics: Data Analysis:

Age (continuous) Gender (0=Male, 1=Female) SES (1=Low, 2=Medium, 3=High) Prior Victimization (0= Not Victimized, 1=Victimized)

Chapter 1: Exploring Data

The Association Design and a Continuous Phenotype

Ecological Statistics

SAS Data Setup: SPSS Data Setup: STATA Data Setup: Hoffman ICPSR Example 5 page 1

Lev Sverdlov, Ph.D.; John F. Noble, Ph.D.; Gabriela Nicolau, Ph.D. Innapharma, Inc., Upper Saddle River, NJ

Daniel Boduszek University of Huddersfield

Types of Statistics. Censored data. Files for today (June 27) Lecture and Homework INTRODUCTION TO BIOSTATISTICS. Today s Outline

Using SAS to Conduct Pilot Studies: An Instructors Guide

Application of Local Control Strategy in analyses of the effects of Radon on Lung Cancer Mortality for 2,881 US Counties

Principal Components Factor Analysis in the Literature. Stage 1: Define the Research Problem

ANOVA. Thomas Elliott. January 29, 2013

Lab 7 (100 pts.): One-Way ANOVA Objectives: Analyze data via the One-Way ANOVA

Linear Regression in SAS

Deanna Schreiber-Gregory Henry M Jackson Foundation for the Advancement of Military Medicine. PharmaSUG 2016 Paper #SP07

11/18/2013. Correlational Research. Correlational Designs. Why Use a Correlational Design? CORRELATIONAL RESEARCH STUDIES

Department of Statistics TEXAS A&M UNIVERSITY STAT 211. Instructor: Keith Hatfield

Midterm Exam ANSWERS Categorical Data Analysis, CHL5407H

Regression Output: Table 5 (Random Effects OLS) Random-effects GLS regression Number of obs = 1806 Group variable (i): subject Number of groups = 70

AP Statistics. Semester One Review Part 1 Chapters 1-5

Chapter 17: Exploratory factor analysis

1. Below is the output of a 2 (gender) x 3(music type) completely between subjects factorial ANOVA on stress ratings

Introduction to Quantitative Methods (SR8511) Project Report

International Conference on Humanities and Social Science (HSS 2016)

On the purpose of testing:

The FASTCLUS Procedure as an Effective Way to Analyze Clinical Data

Regression. Page 1. Variables Entered/Removed b Variables. Variables Removed. Enter. Method. Psycho_Dum

Multiple Linear Regression Analysis

4.3 Measures of Variation

Chapter 4 Data Analysis & Results

Lab #7: Confidence Intervals-Hypothesis Testing (2)-T Test

Measurement Error 2: Scale Construction (Very Brief Overview) Page 1

Unit outcomes. Summary & Conclusion. Lecture 10 Survey Research & Design in Psychology James Neill, 2018 Creative Commons Attribution 4.0.

Unit outcomes. Summary & Conclusion. Lecture 10 Survey Research & Design in Psychology James Neill, 2018 Creative Commons Attribution 4.0.

What Causes Stress in Malaysian Students and it Effect on Academic Performance: A case Revisited

Analysis and Interpretation of Data Part 1

STATISTICAL METHODS FOR DIAGNOSTIC TESTING: AN ILLUSTRATION USING A NEW METHOD FOR CANCER DETECTION XIN SUN. PhD, Kansas State University, 2012

Professor Rose-Helleknat's PCR Data for Breast Cancer Study

Multiple Bivariate Gaussian Plotting and Checking

Two-Way Independent Samples ANOVA with SPSS

SPECIFIC FEATURES OF THE DECATHLON

Lecture Outline. Biost 517 Applied Biostatistics I. Purpose of Descriptive Statistics. Purpose of Descriptive Statistics

A CONSTRUCT VALIDITY ANALYSIS OF THE WORK PERCEPTIONS PROFILE DATA DECEMBER 4, 2014

Modern Regression Methods

Standard Scores. Richard S. Balkin, Ph.D., LPC-S, NCC

Identifying or Verifying the Number of Factors to Extract using Very Simple Structure.

Regression Including the Interaction Between Quantitative Variables

LAB ASSIGNMENT 4 INFERENCES FOR NUMERICAL DATA. Comparison of Cancer Survival*

Introduction to Statistical Data Analysis I

Normal Q Q. Residuals vs Fitted. Standardized residuals. Theoretical Quantiles. Fitted values. Scale Location 26. Residuals vs Leverage

Transcription:

STAT 5600 Factor Analysis in SAS The data for this example come from the decathlon results in the 1988 Olympics. The decathlon is a two-day competition, with the 100 m race, long jump, shot put, high jump, and 400 m on day 1, and 110 m hurdles, discus, pole vault, javelin, and 1500 m on day 2. Participants and their scores are shown in the table below: Athlete 100 m Long Jump Shot High Jump 400 m 110 m Hurdles Discus Pole Vault Javelin 1500 m TOTAL Schenk 11.25 7.43 15.48 2.27 48.9 15.13 49.28 4.7 61.32 268.95 8488 Voss 10.87 7.45 14.97 1.97 47.71 14.46 44.36 5.1 61.76 273.02 8399 Steen 11.18 7.44 14.2 1.97 48.29 14.81 43.66 5.2 64.16 263.2 8328 Thompson 10.62 7.38 15.02 2.03 49.06 14.72 44.8 4.9 64.04 285.11 8306 Blondel 11.02 7.43 12.92 1.97 47.44 14.4 41.2 5.2 57.46 256.64 8286 Plaziat 10.83 7.72 13.58 2.12 48.34 14.18 43.06 4.9 52.18 274.07 8272 Bright 11.18 7.05 14.12 2.06 49.34 14.39 41.68 5.7 61.6 291.2 8216 De wti 11.05 6.95 15.34 2.00 48.21 14.36 41.32 4.8 63 265.86 8189 Johnson 11.15 7.12 14.52 2.03 49.15 14.66 42.36 4.9 66.46 269.62 8180 Tarnovetsky 11.23 7.28 15.25 1.97 48.6 14.76 48.02 5.2 59.48 292.24 8167 Keskitalo 10.94 7.45 15.34 1.97 49.94 14.25 41.86 4.8 66.64 295.89 8143 Gaehwiler 11.18 7.34 14.48 1.94 49.02 15.11 42.76 4.7 65.84 256.74 8114 Szabo 11.02 7.29 12.92 2.06 48.23 14.94 39.54 5 56.8 257.85 8093 Smith 10.99 7.37 13.61 1.97 47.83 14.7 43.88 4.3 66.54 268.97 8083 Shirley 11.03 7.45 14.2 1.97 48.94 15.44 41.66 4.7 64 267.48 8036 Poelman 11.09 7.08 14.51 2.03 49.89 14.78 43.2 4.9 57.18 268.54 8021 Olander 11.46 6.75 16.07 2.00 51.28 16.06 50.66 4.8 72.6 302.42 7869 Freimuth 11.57 7 16.6 1.94 49.84 15 46.66 4.9 60.2 286.04 7860 Warming 11.07 7.04 13.41 1.94 47.97 14.96 40.38 4.5 51.5 262.41 7859 Hraban 10.89 7.07 15.84 1.79 49.68 15.38 45.32 4.9 60.48 277.84 7781 Werthner 11.52 7.36 13.93 1.94 49.99 15.64 38.82 4.6 67.04 266.42 7753 Gugler 11.49 7.02 13.8 2.03 50.6 15.22 39.08 4.7 60.92 262.93 7745 Penalver 11.38 7.08 14.31 2.00 50.24 14.97 46.34 4.4 55.68 272.68 7743 Kruger 11.3 6.97 13.23 2.15 49.98 15.38 38.72 4.6 54.34 277.84 7623 Lee Fu-An 11 7.23 13.15 2.03 49.73 14.96 38.06 4.5 52.82 285.57 7579 Mellado 11.33 6.83 11.63 2.06 48.37 15.39 37.52 4.6 55.42 270.07 7517 moser 11.1 6.98 12.69 1.82 48.63 15.13 38.04 4.7 49.52 261.9 7505 Valenta 11.51 7.01 14.17 1.94 51.16 15.18 45.84 4.6 56.28 303.17 7422 O'Connell 11.26 6.9 12.41 1.88 48.24 15.61 38.02 4.4 52.68 272.06 7310 Richards 11.5 7.09 12.94 1.82 49.27 15.56 42.32 4.5 53.5 293.85 7237 Gong 11.43 6.22 13.98 1.91 51.25 15.88 46.18 4.6 57.84 294.99 7231 Miller 11.47 6.43 12.33 1.94 50.3 15 38.72 4 57.26 293.72 7016 Kwang-Ik 11.57 7.19 10.27 1.91 50.71 16.2 34.36 4.1 54.94 269.98 6907 Kunwar 12.12 5.83 9.71 1.7 52.32 17.05 27.1 2.6 39.1 281.24 5339

With the SAS commands below, we want to carry out some crude exploratory analyses, and then apply a factor analysis to these data along with an orthogonal and oblique rotation. Some fundamental questions we would like to answer: Are there any outliers in the dataset? What patterns do we observe in examining correlations between the events (variables)? Based on a factor analysis, how many latent factors appear to reasonably represent the 10 variables? How do the results compare between different factor rotations? Applying a factor rotation, can we identify any substantive patterns among the determined factors? SAS CODE: options ls=79 nodate; data decathlon; infile 'c:\decathlon88.txt' expandtabs; input run100 Ljump shot Hjump run400 hurdle discus polevlt javelin run1500 score; ***** EXPLORATORY ANALYSIS -- POTENTIAL OUTLIERS?; proc univariate data=decathlon plots; var score; run; ***** TRANSFORM RACING SCORES SO THAT DIRECTION IS CONSISTENT WITH OTHER EVENTS; data decathlon; set decathlon; if score > 6000; run100=run100*-1; run400=run400*-1; hurdle=hurdle*-1; run1500=run1500*-1; run; ***** COMPUTE CORRELATION MATRIX; proc corr data=decathlon; var run100--run1500; run; title "FACTOR ANALYSIS WITH NO ROTATION"; proc factor data=decathlon method=ml mineigen=1 rotate=none; var run100--run1500; run; title "FACTOR ANALYSIS WITH OBLIQUE ROTATION"; proc factor data=decathlon method=ml mineigen=1 rotate=obvarimax; var run100--run1500; run; title "FACTOR ANALYSES WITH ORTHOGONAL ROTATIONS"; proc factor data=decathlon method=ml mineigen=1 rotate=orthomax; var run100--run1500; run; proc factor data=decathlon method=ml mineigen=1 rotate=factorparsimax; var run100--run1500; run;

SAS OUTPUT: The UNIVARIATE Procedure Variable: score Moments N 34 Sum Weights 34 Mean 7782.85294 Sum Observations 264617 Std Deviation 594.582723 Variance 353528.614 Skewness -2.2488675 Kurtosis 7.67309194 Uncorrected SS 2071141641 Corrected SS 11666444.3 Coeff Variation 7.63964997 Std Error Mean 101.970096 Basic Statistical Measures Location Variability Mean 7782.853 Std Deviation 594.58272 Median 7864.500 Variance 353529 Mode. Range 3149 Interquartile Range 663.00000 Tests for Location: Mu0=0 Test -Statistic- -----p Value------ Student's t t 76.32486 Pr > t <.0001 Sign M 17 Pr >= M <.0001 Signed Rank S 297.5 Pr >= S <.0001 Quantiles (Definition 5) Quantile Estimate 100% Max 8488.0 99% 8488.0 95% 8399.0 90% 8306.0 75% Q3 8180.0 50% Median 7864.5 25% Q1 7517.0 10% 7231.0 5% 6907.0 1% 5339.0 0% Min 5339.0

The UNIVARIATE Procedure Variable: score Extreme Observations ----Lowest---- ----Highest--- Value Obs Value Obs 5339 34 8286 5 6907 33 8306 4 7016 32 8328 3 7231 31 8399 2 7237 30 8488 1 Stem Leaf # Boxplot 84 09 2 82 27913 5 80 248914789 9 +-----+ 78 667 3 *-----* 76 24458 5 + 74 2028 4 +-----+ 72 341 3 70 2 1 68 1 1 66 64 62 60 58 56 54 52 4 1 * ----+----+----+----+ Multiply Stem.Leaf by 10**+2 Normal Probability Plot 8500+ +++ * +**** * * ******** **+++ ****++ ****++ ***+++ *++++ 6900+ *+++ +++ ++++ ++ 5300+ * +----+----+----+----+----+----+----+----+----+----+ -2-1 0 +1 +2

The CORR Procedure 10 Variables: run100 Ljump shot Hjump run400 hurdle discus polevlt javelin run1500 Simple Statistics Variable N Mean Std Dev Sum Minimum Maximum run100 33-11.19636 0.24332-369.48000-11.57000-10.62000 Ljump 33 7.13333 0.30434 235.40000 6.22000 7.72000 shot 33 13.97636 1.33199 461.22000 10.27000 16.60000 Hjump 33 1.98273 0.09398 65.43000 1.79000 2.27000 run400 33-49.27667 1.06966-1626 -51.28000-47.44000 hurdle 33-15.04879 0.50677-496.61000-16.20000-14.18000 discus 33 42.35394 3.71913 1398 34.36000 50.66000 polevlt 33 4.73939 0.33442 156.40000 4.00000 5.70000 javelin 33 59.43879 5.49600 1961 49.52000 72.60000 run1500 33-276.03848 13.65710-9109 -303.17000-256.64000 Pearson Correlation Coefficients, N = 33 Prob > r under H0: Rho=0 run100 Ljump shot Hjump run400 run100 1.00000 0.53957 0.20797 0.14591 0.60590 0.0012 0.2455 0.4178 0.0002 Ljump 0.53957 1.00000 0.14190 0.27314 0.51533 0.0012 0.4309 0.1241 0.0021 shot 0.20797 0.14190 1.00000 0.12208-0.09458 0.2455 0.4309 0.4986 0.6006 hurdle discus polevlt javelin run1500 run100 0.63836 0.04722 0.38914 0.06471 0.26103 <.0001 0.7941 0.0252 0.7205 0.1423 Ljump 0.47800 0.04192 0.34993 0.18167 0.39559 0.0049 0.8168 0.0459 0.3116 0.0227 shot 0.29572 0.80635 0.47998 0.59767-0.26883 0.0947 <.0001 0.0047 0.0002 0.1303 run100 Ljump shot Hjump run400 Hjump 0.14591 0.27314 0.12208 1.00000 0.08750 0.4178 0.1241 0.4986 0.6282 run400 0.60590 0.51533-0.09458 0.08750 1.00000 0.0002 0.0021 0.6006 0.6282 hurdle 0.63836 0.47800 0.29572 0.30674 0.54603 <.0001 0.0049 0.0947 0.0825 0.0010 discus 0.04722 0.04192 0.80635 0.14742-0.14219 0.7941 0.8168 <.0001 0.4130 0.4299 polevlt 0.38914 0.34993 0.47998 0.21323 0.31866 0.0252 0.0459 0.0047 0.2335 0.0707 javelin 0.06471 0.18167 0.59767 0.11594-0.12035 0.7205 0.3116 0.0002 0.5206 0.5047 run1500 0.26103 0.39559-0.26883 0.11409 0.58728 0.1423 0.0227 0.1303 0.5273 0.0003

hurdle discus polevlt javelin run1500 Hjump 0.30674 0.14742 0.21323 0.11594 0.11409 0.0825 0.4130 0.2335 0.5206 0.5273 run400 0.54603-0.14219 0.31866-0.12035 0.58728 0.0010 0.4299 0.0707 0.5047 0.0003 hurdle 1.00000 0.11050 0.52155 0.06282 0.14330 0.5404 0.0019 0.7284 0.4263 discus 0.11050 1.00000 0.34397 0.44291-0.40232 0.5404 0.0500 0.0098 0.0203 polevlt 0.52155 0.34397 1.00000 0.27424 0.03150 0.0019 0.0500 0.1225 0.8619 javelin 0.06282 0.44291 0.27424 1.00000-0.09638 0.7284 0.0098 0.1225 0.5937 hurdle discus polevlt javelin run1500 run1500 0.14330-0.40232 0.03150-0.09638 1.00000 0.4263 0.0203 0.8619 0.5937

FACTOR ANALYSIS WITH NO ROTATION Initial Factor Method: Maximum Likelihood Prior Communality Estimates: SMC run100 Ljump shot Hjump run400 0.55457740 0.45256490 0.80445733 0.21341936 0.68782612 hurdle discus polevlt javelin run1500 0.62561769 0.73943287 0.42243571 0.41879238 0.55199588 Preliminary Eigenvalues: Total = 15.8533372 Average = 1.58533372 Eigenvalue Difference Proportion Cumulative 1 8.89838207 2.21951018 0.5613 0.5613 2 6.67887188 5.79654533 0.4213 0.9826 3 0.88232656 0.43776469 0.0557 1.0382 4 0.44456186 0.19762584 0.0280 1.0663 5 0.24693603 0.18365534 0.0156 1.0819 6 0.06328069 0.18838261 0.0040 1.0859 7 -.12510193 0.14994679-0.0079 1.0780 8 -.27504872 0.04051625-0.0173 1.0606 9 -.31556497 0.32974126-0.0199 1.0407 10 -.64530624-0.0407 1.0000 2 factors will be retained by the MINEIGEN criterion. Iteration Criterion Ridge Change Communalities 1 0.7285725 0.0000 0.1488 0.59384 0.46599 0.94737 0.08387 0.72801 0.56093 0.68895 0.44242 0.32332 0.40324 2 0.7221374 0.0000 0.0335 0.59884 0.45475 0.96928 0.06850 0.71726 0.57614 0.68533 0.42789 0.35677 0.38623 3 0.7216597 0.0000 0.0120 0.60212 0.45394 0.97536 0.06753 0.71508 0.57833 0.68328 0.42613 0.36047 0.37424 4 0.7216141 0.0000 0.0031 0.60328 0.45358 0.97709 0.06716 0.71355 0.57973 0.68252 0.42554 0.36149 0.37114 5 0.7216093 0.0000 0.0012 0.60368 0.45352 0.97750 0.06713 0.71308 0.58021 0.68238 0.42541 0.36180 0.36991 6 0.7216088 0.0000 0.0004 0.60382 0.45350 0.97766 0.06714 0.71288 0.58038 0.68231 0.42539 0.36186 0.36956 Convergence criterion satisfied. Significance Tests Based on 33 Observations Pr > Test DF Chi-Square ChiSq H0: No common factors 45 137.1386 <.0001 HA: At least one common factor H0: 2 Factors are sufficient 26 19.1226 0.8312 HA: More factors are needed

Chi-Square without Bartlett's Correction 23.091482 Akaike's Information Criterion -28.908518 Schwarz's Bayesian Criterion -67.817715 Tucker and Lewis's Reliability Coefficient 1.129187 Squared Canonical Correlations 0.97932390 0.87049551 Eigenvalues of the Weighted Reduced Correlation Matrix: Total = 54.0867551 Average = 5.40867551 Eigenvalue Difference Proportion Cumulative 1 47.3650155 40.6432756 0.8757 0.8757 2 6.7217399 6.0853763 0.1243 1.0000 3 0.6363636 0.2644105 0.0118 1.0118 4 0.3719530 0.1504848 0.0069 1.0186 5 0.2214682 0.1425255 0.0041 1.0227 6 0.0789427 0.1557214 0.0015 1.0242 7-0.0767787 0.1917958-0.0014 1.0228 8-0.2685744 0.0537115-0.0050 1.0178 9-0.3222859 0.3188028-0.0060 1.0119 10-0.6410887-0.0119 1.0000 Initial Factor Method: Maximum Likelihood Factor Pattern run100 0.21972 0.74536 Ljump 0.15527 0.65528 shot 0.98867-0.01359 Hjump 0.13490 0.22123 run400-0.08346 0.84018 hurdle 0.30854 0.69658 discus 0.81265-0.14808 polevlt 0.49326 0.42671 javelin 0.59918-0.05364 run1500-0.27318 0.54297 Variance Explained by Each Factor Factor Weighted Unweighted 47.3650155 2.50756391 6.7217399 2.72691426 Final Communality Estimates and Variable Weights Total Communality: Weighted = 54.086755 Unweighted = 5.234478 Variable Communality Weight run100 0.60384600 2.5241161 Ljump 0.45349671 1.8298317 shot 0.97765555 44.7534978 Hjump 0.06714079 1.0719691 run400 0.71286418 3.4829107 hurdle 0.58041971 2.3831226 discus 0.68232597 3.1477626 polevlt 0.42538373 1.7402965 javelin 0.36189692 1.5670528 run1500 0.36944861 1.5861952

FACTOR ANALYSIS WITH OBLIQUE ROTATION Initial Factor Method: Maximum Likelihood Prior Communality Estimates: SMC run100 Ljump shot Hjump run400 0.55457740 0.45256490 0.80445733 0.21341936 0.68782612 hurdle discus polevlt javelin run1500 0.62561769 0.73943287 0.42243571 0.41879238 0.55199588 Preliminary Eigenvalues: Total = 15.8533372 Average = 1.58533372 Eigenvalue Difference Proportion Cumulative 1 8.89838207 2.21951018 0.5613 0.5613 2 6.67887188 5.79654533 0.4213 0.9826 3 0.88232656 0.43776469 0.0557 1.0382 4 0.44456186 0.19762584 0.0280 1.0663 5 0.24693603 0.18365534 0.0156 1.0819 6 0.06328069 0.18838261 0.0040 1.0859 7 -.12510193 0.14994679-0.0079 1.0780 8 -.27504872 0.04051625-0.0173 1.0606 9 -.31556497 0.32974126-0.0199 1.0407 10 -.64530624-0.0407 1.0000 2 factors will be retained by the MINEIGEN criterion. Pr > Test DF Chi-Square ChiSq H0: No common factors 45 137.1386 <.0001 HA: At least one common factor H0: 2 Factors are sufficient 26 19.1226 0.8312 HA: More factors are needed Chi-Square without Bartlett's Correction 23.091482 Akaike's Information Criterion -28.908518 Schwarz's Bayesian Criterion -67.817715 Tucker and Lewis's Reliability Coefficient 1.129187 Squared Canonical Correlations 0.97932390 0.87049551 Eigenvalues of the Weighted Reduced Correlation Matrix: Total = 54.0867551 Average = 5.40867551 Eigenvalue Difference Proportion Cumulative 1 47.3650155 40.6432756 0.8757 0.8757 2 6.7217399 6.0853763 0.1243 1.0000 3 0.6363636 0.2644105 0.0118 1.0118 4 0.3719530 0.1504848 0.0069 1.0186 5 0.2214682 0.1425255 0.0041 1.0227 6 0.0789427 0.1557214 0.0015 1.0242 7-0.0767787 0.1917958-0.0014 1.0228 8-0.2685744 0.0537115-0.0050 1.0178 9-0.3222859 0.3188028-0.0060 1.0119 10-0.6410887-0.0119 1.0000

Initial Factor Method: Maximum Likelihood Factor Pattern run100 0.21972 0.74536 Ljump 0.15527 0.65528 shot 0.98867-0.01359 Hjump 0.13490 0.22123 run400-0.08346 0.84018 hurdle 0.30854 0.69658 discus 0.81265-0.14808 polevlt 0.49326 0.42671 javelin 0.59918-0.05364 run1500-0.27318 0.54297 Variance Explained by Each Factor Factor Weighted Unweighted 47.3650155 2.50756391 6.7217399 2.72691426 Final Communality Estimates and Variable Weights Total Communality: Weighted = 54.086755 Unweighted = 5.234478 Variable Communality Weight run100 0.60384600 2.5241161 Ljump 0.45349671 1.8298317 shot 0.97765555 44.7534978 Hjump 0.06714079 1.0719691 run400 0.71286418 3.4829107 hurdle 0.58041971 2.3831226 discus 0.68232597 3.1477626 polevlt 0.42538373 1.7402965 javelin 0.36189692 1.5670528 run1500 0.36944861 1.5861952 Rotation Method: Oblique Varimax Oblique Transformation Matrix 1 2 1 0.08424332 0.98862745 2 1.0008292-0.1771208 Inter-Factor Correlations 1.00000 0.09317 0.09317 1.00000 Rotated Factor Pattern (Standardized Regression Coefficients) run100 0.76449 0.08520 Ljump 0.66890 0.03744 shot 0.06969 0.97983 Hjump 0.23278 0.09418 run400 0.83384-0.23133 hurdle 0.72315 0.18165 discus -0.07975 0.82964 polevlt 0.46862 0.41207 javelin -0.00321 0.60187 run1500 0.52041-0.36625

Reference Axis Correlations 1.00000-0.09317-0.09317 1.00000 Reference Structure (Semipartial Correlations) run100 0.76117 0.08483 Ljump 0.66599 0.03728 shot 0.06938 0.97557 Hjump 0.23177 0.09377 run400 0.83022-0.23032 hurdle 0.72000 0.18086 discus -0.07940 0.82603 polevlt 0.46658 0.41028 javelin -0.00320 0.59925 run1500 0.51815-0.36465 Variance Explained by Each Factor Eliminating Other Factors Factor Weighted Unweighted 7.0076789 2.78160269 46.1010308 2.39760046 Factor Structure (Correlations) run100 0.77243 0.15643 Ljump 0.67239 0.09976 shot 0.16097 0.98633 Hjump 0.24155 0.11586 run400 0.81229-0.15364 hurdle 0.74007 0.24902 discus -0.00245 0.82221 polevlt 0.50701 0.45573 javelin 0.05286 0.60157 run1500 0.48629-0.31776 Variance Explained by Each Factor Ignoring Other Factors Factor Weighted Unweighted 7.9857246 2.83687771 47.0790765 2.45287548 Final Communality Estimates and Variable Weights Total Communality: Weighted = 54.086755 Unweighted = 5.234478 Variable Communality Weight run100 0.60384600 2.5241161 Ljump 0.45349671 1.8298317 shot 0.97765555 44.7534978 Hjump 0.06714079 1.0719691 run400 0.71286418 3.4829107 hurdle 0.58041971 2.3831226 discus 0.68232597 3.1477626 polevlt 0.42538373 1.7402965 javelin 0.36189692 1.5670528 run1500 0.36944861 1.5861952

FACTOR ANALYSES WITH ORTHOGONAL ROTATIONS Initial Factor Method: Maximum Likelihood Prior Communality Estimates: SMC run100 Ljump shot Hjump run400 0.55457740 0.45256490 0.80445733 0.21341936 0.68782612 hurdle discus polevlt javelin run1500 0.62561769 0.73943287 0.42243571 0.41879238 0.55199588 Preliminary Eigenvalues: Total = 15.8533372 Average = 1.58533372 Eigenvalue Difference Proportion Cumulative 1 8.89838207 2.21951018 0.5613 0.5613 2 6.67887188 5.79654533 0.4213 0.9826 3 0.88232656 0.43776469 0.0557 1.0382 4 0.44456186 0.19762584 0.0280 1.0663 5 0.24693603 0.18365534 0.0156 1.0819 6 0.06328069 0.18838261 0.0040 1.0859 7 -.12510193 0.14994679-0.0079 1.0780 8 -.27504872 0.04051625-0.0173 1.0606 9 -.31556497 0.32974126-0.0199 1.0407 10 -.64530624-0.0407 1.0000 2 factors will be retained by the MINEIGEN criterion. Pr > Test DF Chi-Square ChiSq H0: No common factors 45 137.1386 <.0001 HA: At least one common factor H0: 2 Factors are sufficient 26 19.1226 0.8312 HA: More factors are needed Chi-Square without Bartlett's Correction 23.091482 Akaike's Information Criterion -28.908518 Schwarz's Bayesian Criterion -67.817715 Tucker and Lewis's Reliability Coefficient 1.129187 Squared Canonical Correlations 0.97932390 0.87049551 Eigenvalues of the Weighted Reduced Correlation Matrix: Total = 54.0867551 Average = 5.40867551 Eigenvalue Difference Proportion Cumulative 1 47.3650155 40.6432756 0.8757 0.8757 2 6.7217399 6.0853763 0.1243 1.0000 3 0.6363636 0.2644105 0.0118 1.0118 4 0.3719530 0.1504848 0.0069 1.0186 5 0.2214682 0.1425255 0.0041 1.0227 6 0.0789427 0.1557214 0.0015 1.0242 7-0.0767787 0.1917958-0.0014 1.0228 8-0.2685744 0.0537115-0.0050 1.0178 9-0.3222859 0.3188028-0.0060 1.0119 10-0.6410887-0.0119 1.0000

Initial Factor Method: Maximum Likelihood Factor Pattern run100 0.21972 0.74536 Ljump 0.15527 0.65528 shot 0.98867-0.01359 Hjump 0.13490 0.22123 run400-0.08346 0.84018 hurdle 0.30854 0.69658 discus 0.81265-0.14808 polevlt 0.49326 0.42671 javelin 0.59918-0.05364 run1500-0.27318 0.54297 Variance Explained by Each Factor Factor Weighted Unweighted 47.3650155 2.50756391 6.7217399 2.72691426 Final Communality Estimates and Variable Weights Total Communality: Weighted = 54.086755 Unweighted = 5.234478 Variable Communality Weight run100 0.60384600 2.5241161 Ljump 0.45349671 1.8298317 shot 0.97765555 44.7534978 Hjump 0.06714079 1.0719691 run400 0.71286418 3.4829107 hurdle 0.58041971 2.3831226 discus 0.68232597 3.1477626 polevlt 0.42538373 1.7402965 javelin 0.36189692 1.5670528 run1500 0.36944861 1.5861952 Orthogonal Transformation Matrix 1 2 1 0.17070 0.98532 2 0.98532-0.17070 Rotated Factor Pattern run100 0.77193 0.08927 Ljump 0.67216 0.04114 shot 0.15537 0.97648 Hjump 0.24101 0.09515 run400 0.81360-0.22565 hurdle 0.73902 0.18511 discus -0.00719 0.82600 polevlt 0.50464 0.41318 javelin 0.04942 0.59955 run1500 0.48837-0.36186 Variance Explained by Each Factor Factor Weighted Unweighted 7.9059780 2.83367841 46.1807773 2.40079975

Final Communality Estimates and Variable Weights Total Communality: Weighted = 54.086755 Unweighted = 5.234478 Variable Communality Weight run100 0.60384600 2.5241161 Ljump 0.45349671 1.8298317 shot 0.97765555 44.7534978 Hjump 0.06714079 1.0719691 run400 0.71286418 3.4829107 hurdle 0.58041971 2.3831226 discus 0.68232597 3.1477626 polevlt 0.42538373 1.7402965 javelin 0.36189692 1.5670528 run1500 0.36944861 1.5861952

FACTOR ANALYSES WITH ORTHOGONAL ROTATIONS Initial Factor Method: Maximum Likelihood Prior Communality Estimates: SMC run100 Ljump shot Hjump run400 0.55457740 0.45256490 0.80445733 0.21341936 0.68782612 hurdle discus polevlt javelin run1500 0.62561769 0.73943287 0.42243571 0.41879238 0.55199588 Preliminary Eigenvalues: Total = 15.8533372 Average = 1.58533372 Eigenvalue Difference Proportion Cumulative 1 8.89838207 2.21951018 0.5613 0.5613 2 6.67887188 5.79654533 0.4213 0.9826 3 0.88232656 0.43776469 0.0557 1.0382 4 0.44456186 0.19762584 0.0280 1.0663 5 0.24693603 0.18365534 0.0156 1.0819 6 0.06328069 0.18838261 0.0040 1.0859 7 -.12510193 0.14994679-0.0079 1.0780 8 -.27504872 0.04051625-0.0173 1.0606 9 -.31556497 0.32974126-0.0199 1.0407 10 -.64530624-0.0407 1.0000 2 factors will be retained by the MINEIGEN criterion. Convergence criterion satisfied. Significance Tests Based on 33 Observations Pr > Test DF Chi-Square ChiSq H0: No common factors 45 137.1386 <.0001 HA: At least one common factor H0: 2 Factors are sufficient 26 19.1226 0.8312 HA: More factors are needed Chi-Square without Bartlett's Correction 23.091482 Akaike's Information Criterion -28.908518 Schwarz's Bayesian Criterion -67.817715 Tucker and Lewis's Reliability Coefficient 1.129187 Squared Canonical Correlations 0.97932390 0.87049551 Eigenvalues of the Weighted Reduced Correlation Matrix: Total = 54.0867551 Average = 5.40867551 Eigenvalue Difference Proportion Cumulative 1 47.3650155 40.6432756 0.8757 0.8757 2 6.7217399 6.0853763 0.1243 1.0000 3 0.6363636 0.2644105 0.0118 1.0118 4 0.3719530 0.1504848 0.0069 1.0186 5 0.2214682 0.1425255 0.0041 1.0227 6 0.0789427 0.1557214 0.0015 1.0242 7-0.0767787 0.1917958-0.0014 1.0228 8-0.2685744 0.0537115-0.0050 1.0178 9-0.3222859 0.3188028-0.0060 1.0119 10-0.6410887-0.0119 1.0000

Initial Factor Method: Maximum Likelihood Factor Pattern run100 0.21972 0.74536 Ljump 0.15527 0.65528 shot 0.98867-0.01359 Hjump 0.13490 0.22123 run400-0.08346 0.84018 hurdle 0.30854 0.69658 discus 0.81265-0.14808 polevlt 0.49326 0.42671 javelin 0.59918-0.05364 run1500-0.27318 0.54297 Variance Explained by Each Factor Factor Weighted Unweighted 47.3650155 2.50756391 6.7217399 2.72691426 Final Communality Estimates and Variable Weights Total Communality: Weighted = 54.086755 Unweighted = 5.234478 Variable Communality Weight run100 0.60384600 2.5241161 Ljump 0.45349671 1.8298317 shot 0.97765555 44.7534978 Hjump 0.06714079 1.0719691 run400 0.71286418 3.4829107 hurdle 0.58041971 2.3831226 discus 0.68232597 3.1477626 polevlt 0.42538373 1.7402965 javelin 0.36189692 1.5670528 run1500 0.36944861 1.5861952 Rotation Method: Factor Parsimax Orthogonal Transformation Matrix 1 2 1-0.12296 0.99241 2 0.99241 0.12296 Rotated Factor Pattern run100 0.71269 0.30971 Ljump 0.63121 0.23467 shot -0.13506 0.97950 Hjump 0.20297 0.16108 run400 0.84406 0.02048 hurdle 0.65335 0.39185 discus -0.24689 0.78827 polevlt 0.36282 0.54198 javelin -0.12691 0.58804 run1500 0.57245-0.20434

Variance Explained by Each Factor Factor Weighted Unweighted 7.3362764 2.64149774 46.7504789 2.59298042 Final Communality Estimates and Variable Weights Total Communality: Weighted = 54.086755 Unweighted = 5.234478 Variable Communality Weight run100 0.60384600 2.5241161 Ljump 0.45349671 1.8298317 shot 0.97765555 44.7534978 Hjump 0.06714079 1.0719691 run400 0.71286418 3.4829107 hurdle 0.58041971 2.3831226 discus 0.68232597 3.1477626 polevlt 0.42538373 1.7402965 javelin 0.36189692 1.5670528 run1500 0.36944861 1.5861952