# PSU STAT 497 - Interpreting Results (2 pages)

Previewing page 1 of 2 page document
View Full Document

## Interpreting Results

Previewing page 1 of actual document.

View Full Document
View Full Document

## Interpreting Results

86 views

Lecture Notes

Pages:
2
School:
Pennsylvania State University
Course:
Stat 497 - Special Topics
##### Special Topics Documents
• 18 pages

• 6 pages

• 8 pages

Unformatted text preview:

Ordinal Logistic Regression OrdinalYears versus Adj Forty Link Function Logit Response Information Variable OrdinalYears Value 0 1 2 Total Count 9 10 11 30 Logistic Regression Table Predictor Const 1 Const 2 Adj Forty Coef 3 66970 2 25094 0 0292074 SE Coef 3 35366 3 30836 0 0342799 Z 1 09 0 68 0 85 P 0 274 0 496 0 394 Odds Ratio 1 03 95 CI Lower Upper 0 96 1 10 Log Likelihood 32 529 Test that all slopes are zero G 0 659 DF 1 P Value 0 417 Goodness of Fit Tests Method Pearson Deviance Chi Square 60 2893 65 0573 DF 57 57 P 0 358 0 217 Measures of Association Between the Response Variable and Predicted Probabilities Pairs Concordant Discordant Ties Total Number 170 125 4 299 Percent 56 9 41 8 1 3 100 0 Summary Measures Somers D Goodman Kruskal Gamma Kendall s Tau a 0 15 0 15 0 10 Interpreting the results Response Information displays the number of observations that fall into each of the response categories and the number of missing observations The ordered response values from lowest to highest are shown Here we use the default coding scheme which orders the values from lowest to highest 0 is zero years in NFL 1 more than zero but not more than 3 years and 2 more than three years Logistic Regression Table shows the estimated coefficients standard error of the coefficients zvalues and p values When you use the logit link function you see the calculated odds ratio and a 95 confidence interval for the odds ratio The values labeled Const 1 and Const 2 are estimated intercepts for the logits of the cumulative probabilities of survival for 0 years and for 0 to 3 years respectively Because the cumulative probability for the last response value is 1 there is not need to estimate an intercept for more than 3 years 1 There is one estimated coefficient for each covariate which gives parallel lines for the factor levels Here the estimated coefficient for the single covariate Adj Forty is 0 0292 with a pvalue of 0 394 The p value indicates that for most levels there is insufficient

View Full Document

Unlocking...