The Sample .

The Sample was planned as 3000 interviews with housewives, 500 in A Class, 500 in B Class, 1200 in C Class and 800 in D Class. These numbers were considered large enough to give reliable class breakdowns. For the total figures they were weighted in the proportions of 1:4:3:4.

This sample was distributed over the Ministry of Food Regions, (except North Wales), and the Class proportions were kept constant throughout.

The following tables show the distribution as planned, in practice some small deviations occurred, due in some measure to households without wireless. The final returns were, however, over 3000. (Tables 1 and 2).

Households Without Wireless .

It was decided that where the housewife had no wireless the interview was to be discounted, but where the wireless set was temporarily out of order owing to the shortage of batteries or valves etc., the interview was included.