Is wine healthier than beer?
There have been numerous discussions about the live extending benefits of moderate wine consumption - opposed to beer. Several recent studies have shown that the general living conditions of the consumers seem to be responsible for this effect rather than the beverages themselves. Most of these studies have been done by surveys among test-persons.
Today I read in the NZZ am Sonntag that Danish Scientists haven done a market basket analysis for the food items correlated with the purchase of wine or beer. Obviously this removes the bias that might be contained in the answers from test persons. I just found that the study is available online at British Medical Journal. Actually I would have presumed they use association rules for the market basket analysis, however, it turns out that they use correspondence analysis and logistic regression to analyse the data. I wonder why, i.e. what the benefits of this method are?? It would also be interesting to see if one got the same results with association rules.