Chevy versus Ford NASCAR Race Effect Size A Meta-Analysis .


33 views
Uploaded on:
Description
Information Description. Every one of the 256 NASCAR Races for 1993-2000 SeasonRace Finishes Among all Ford and Chevy Drivers (Ranks)Ford: 5208 Drivers (20.3 for each race)Chevrolet: 3642 Drivers (14.2 for each race)For every race, Compute Wilcoxon Rank-Sum Statistic (Large-example Normal Approximation)Effect Size = Z/SQRT(NFord NChevy).
Transcripts
Slide 1

Chevy versus Ford NASCAR Race Effect Size – A Meta-Analysis

Slide 2

Data Description All 256 NASCAR Races for 1993-2000 Season Race Finishes Among all Ford and Chevy Drivers (Ranks) Ford: 5208 Drivers (20.3 for every race) Chevrolet: 3642 Drivers (14.2 for each race) For every race, Compute Wilcoxon Rank-Sum Statistic (Large-example Normal Approximation) Effect Size = Z/SQRT(N Ford + N Chevy )

Slide 3

Wilcoxon Rank-Sum Test (Large-Sample)

Slide 4

Evidence that Chevrolet has a tendency to show improvement over Ford

Slide 5

Effect Sizes Appear to be roughly Normal

Slide 6

Combining Effect Sizes Across Races Weighted Average of Race-Specific Effect Sizes Weight Factor  1/V(d i ) = 1/N i = 1/(N Ford,i +N Chevy,i )

Slide 7

Test for Homogeneity of Effect Sizes

Slide 9

Testing for Year Effects

Slide 10

Testing for Year Effects

Slide 13

Testing for Year and Race/Track Effects Regression Model Relating Effect Size to: Season (8 Dummy Variables (No Intercept)) Track Length Number of Laps Race Length (Track Length x # of Laps) Weighted Least Squares with weight i = N i

Slide 14

Regression Coefficients/t-tests Controlling for every other indicator, none seem huge

Slide 15

C 2 – Tests for Sub-Models and Overall

Slide 16

Sources Hedges, L.V. what\'s more, I. Olkin (1985). Measurable Methods for Meta-Analysis , Academic Press, Orlando, FL. Champ, L. (2006). "NASCAR Winston Cup Race Results for 1975-2003," Journal of Statistical Education, Volume 14, #3 www.amstat.org/distributions/jse/v14n3/datasets.winner.html

Recommended
View more...