Three different approaches are used to explain the generation of home-based trips in terms of demographic characteristics. The traditional multiple regression methods express zonal trip totals as a function of zonal charcteristics. The more recent household classification methods provide a catalogue of characteristic mean trip rates for particular types of household, and household regression applies regression analysis to disaggregated household data. All three methods were applied to both sets of data. The results indicate that a considerable amount of stability existed between 1962 and 1971, giving a forecasting accuracy of a few per cent of the mean value. This is well within the 10 to 30 per cent accuracy of representation obtained for the main trip purposes and modes of travel for the base year. Zonal regression is not recommended for explaining trip generation below zone level. Household classification methods make better use of the essential variability of the dissaggregated household survey data and household regression combines this advantage with increased ease of application.