Question 3: How would you assess the evidence concerning alcohol consumption and sex as predictors of BMI?
The effect of alcohol consumption is not significant. If we control for the effect of sex on BMI, there is not enough evidence that large alcohol consumption is associated with higher BMI. There is only very weak evidence for this and the effect could be zero or even slightly negative.
The effect of sex is highly significant and we have strong evidence that men have a higher mean BMI than women who consume the same amount of alcohol. We estimate that the mean BMI for males is between 0.76 and 3.10 Kg/m2 greater than the mean BMI for females who consume the same amount of alcohol.
The coefficient for alcohol consumption is smaller when we include sex in the model (0.057) than it was when sex was not included (0.074). Although there may be a relationship between BMI and alcohol consumption, at least part of the relationship observed in the simple regression can be explained by the relationship of sex both to alcohol comsumption and to BMI.
Back to Exercise: BMI, alcohol consumption, and sex in the 2005 questionnaire.
To Applied Biostatistics index.
To Martin Bland's M.Sc. index.
This page maintained by Martin Bland.
Last updated: 8 December, 2006.