In a case–control study of liver disease and alcohol consumption, all patients in a community who had been newly diagnosed with liver disease were recruited as cases and people without liver disease...

In a case–control study of liver disease and alcohol consumption, all patients in a community who had been newly diagnosed with liver disease were recruited as cases and people without liver disease were selected at random from the community to act as controls. All of the cases and controls were then asked about their alcohol intake. Only 25% of the controls selected from the community agreed to take part in the study

(a) Do you think that people with a high alcohol intake would be more or less likely to agree to take part in the study than average? (b) Is alcohol consumption in the controls likely to be higher, the same as, or lower than in the whole community? (c) What effect would this have on the estimate of the association between alcohol and liver disease?