smoking: Survival and Smoking

In SMPracticals: Practicals for Use with Davison (2003) Statistical Models

Description

Twenty-year survival and smoking status for 1314 women from Whickham, near Newcastle-upon-Tyne.

Usage

data(smoking)

Format

A data frame with 14 observations on the following 4 variables.

age

Age group (factor)

smoker

Smoking status (1=smoker, 0=non-smoker)

alive

Number alive after 20 years

dead

Number dead after 20 years

Source

Appleton, D. R., French, J. M. and Vanderpump, M. P. J. (1996) Ignoring a covariate: An example of Simpson's paradox. The American Statistician, 50, 340–341.

References

Davison, A. C. (2003) Statistical Models. Cambridge University Press. Page 258.

Examples

data(smoking)
summary(glm(cbind(dead,alive)~smoker,data=smoking,binomial))
# note sign change for smoker covariate, due to Simpson's paradox
summary(glm(cbind(dead,alive)~age+smoker,data=smoking,binomial))

Example output

Loading required package: ellipse

Call:
glm(formula = cbind(dead, alive) ~ smoker, family = binomial, 
    data = smoking)

Deviance Residuals: 
   Min      1Q  Median      3Q     Max  
-9.052  -5.674  -1.869   5.776  12.173  

Coefficients:
            Estimate Std. Error z value Pr(>|z|)    
(Intercept) -0.78052    0.07962  -9.803  < 2e-16 ***
smoker      -0.37858    0.12566  -3.013  0.00259 ** 
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

(Dispersion parameter for binomial family taken to be 1)

    Null deviance: 641.5  on 13  degrees of freedom
Residual deviance: 632.3  on 12  degrees of freedom
AIC: 683.29

Number of Fisher Scoring iterations: 4


Call:
glm(formula = cbind(dead, alive) ~ age + smoker, family = binomial, 
    data = smoking)

Deviance Residuals: 
     Min        1Q    Median        3Q       Max  
-0.72545  -0.22836   0.00005   0.19146   0.68162  

Coefficients:
              Estimate Std. Error z value Pr(>|z|)    
(Intercept)    -3.8601     0.5939  -6.500 8.05e-11 ***
age25-34        0.1201     0.6865   0.175 0.861178    
age35-44        1.3411     0.6286   2.134 0.032874 *  
age45-54        2.1134     0.6121   3.453 0.000555 ***
age55-64        3.1808     0.6006   5.296 1.18e-07 ***
age65-74        5.0880     0.6195   8.213  < 2e-16 ***
age75+         27.8073 11293.1431   0.002 0.998035    
smoker          0.4274     0.1770   2.414 0.015762 *  
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

(Dispersion parameter for binomial family taken to be 1)

    Null deviance: 641.4963  on 13  degrees of freedom
Residual deviance:   2.3809  on  6  degrees of freedom
AIC: 65.377

Number of Fisher Scoring iterations: 20

SMPracticals documentation built on May 2, 2019, 11:12 a.m.