This article attempts to analyse the origin and destination data provided by the 1981 and 1989 British Rail surveys about commuting. The data set comprises journeys only from stations of the County of Kent, and is part of the information given by the surveys of rail travel in London and the South-East, carried out every 10 years. The attributes of travel and the social class composition of the commuting population are combined in a descriptive and comparative analysis of the data. As explained in chapter 5, the station’s samples were grossed up by using the weight factors provided by British Rail. The resulting figures, which will be simply called ‘estimates’, are supposed to represent the population commuting by rail.


Despite the various differences between the 1981 and 1989 origin and destination surveys, the data analysis provided relevant information on accessibility to stations, inter-change of trains, location of commuters by distance to Central London, and socio-economic characteristic of travellers. Such issues are treated in this section, which initially presents the level of commuting to the main destination stations of Central London, followed by the flow changes according to ranges of travel distances. The growth in commuting by station and origin district are also presented. Some attributes of travel related to the period 1981-1989 are in the scope of the later article which shows the average travel time and cost of commuting to Central London, train inter-changes, the access time to local stations of Kent, the egress time to the final destination in Central London, and some modifications undertaken in the transport network. Section 7.3.4 deals with the influence of the Outer London stations on the commuting of the boundary area of Kent, and presents an analysis of the socio-economic characteristics of the commuters.


The figure 5.1 showed that some important rail lines in Kent were not covered by the 1989 survey, in particularthe section which comprises the main stations of Dartford and Gravesend. Therefore, the districts Dartford, Rochester and Gravesham were excluded from the analysis. During the period 1981-89, as seen in chapters 2 and 6, increases in the number of commuting trips from Kent to Central London were expected. For the set of stations surveyed in both years, the estimated number of rail commuters to Central London was 20756 in 1981 and 25186 in 1989, an increase of 21 % (see table 7.5). This increase was very slightly faster than the 20% average growth of the flows entering Central London by rail from all destinations (see table 2.1). Growth from Kent was also higher than growth in the flows on the South-East sub-sector of Network South-East as a whole. According to the figures shown by the Department of Transport (1990), the total flow from that sub-sector increased from 119,000 in 1981 to 139,000 in 1989, that is by 16.8%.


Overall, therefore, we conclude that the stations studied here experienced growth in central London commuting which was typical of that from all destinations, and higher than the rest of the South-East sub-sector. Flow changes experienced by Kent were higher than British Rail forecasts for Network South-East as a whole. According to the report on rail passengers in the South-East of England, by the Monopolies and Mergers Commission (1987), “the growth of traffic between 1984-85 and 1986-87 exceeds that previously forecast for the entire period of the 1985 Corporate Plan”. The report explains that “The 1985 Corporate Plan was based on a stable commuting market and increased leisure travel expected to result from marketing initiatives, producing a two percent Increase in passenger miles over the five years as a whole”.


In general, the stations of Kent presented growth in flows. However some few stations either kept about the same level of commuting or even reduced it. Significant increases were observed for the stations Sevenoaks, Gillingham, and Ashford. To a certain extent one may say that the largest growth of commuting to Central London occurred along the rail lines Tonbridge-Ashford and Faversharn-Rochester, whereas the decreases were scattered in various places of Kent. The destination Central London considered in this analysis comprises any of the following six major stations: Cannon St., Charing Cross, Holborn, London Bridge, Waterloo, and London Victoria.


A clear picture of increases in commuting distances is given by the analysis regarding the level of rail users of local stations. In 1981, about 39% of the total flow was originated in stations within 30 miles from Central London, whereas in 1989, the reduced proportion of approximately 36% was observed at the same range of distances. However, for distances above 40 miles, the figures given by the 1989 rail data are generally higher than the 1981 ones. This means that during the 1 980’s there was a tendency of rail users to live further out London. Such trends are corroborated by table 7.6, which depicts the proportion of rail commuting by distance, for data given by the 1981 and 1989 British Rail survey.


One should be aware of the problems involved in the identification of the rail users’ origin. The address codes given by both surveys were completely different. Additionally, the 1989 survey contained some cards with wrong and incomplete postcodes, as well as missing information about travellers addresses. These discrepancies made more difficult the task of an analysis of commuting by ward. However, at the district level a better view of the problem was given by the analysis of the raw data. Due to incomplete postcode, a total of 2269 (9%) addresses were not properly matched to wards.


However, 1997 of them could be associated to districts, and only 272 comprised missing information. Since the file provided by British Rail contained the data in a sequential order by stations surveyed, it was possible to guess the likely origin of the 272 missing postcodes by associating them with the other district or districts shown by that station. This adjustment in the 1989 estimated number of rail commuters to Central London allowed a compatible analysis of the 1981 and 1989 commuting changes to be undertaken. Given the small size of this adjustment, it is unlikely that significant errors were introduced in the flow distributions.


A probable reason for the discrepancies found in the 1989 flows from Tunbridge Wells is the fact that the main station in the district (Tunbridge station) was not covered by the British Rail survey, at the time the data analysed here was obtained. Another possible reason concerns commuters who initiate their trips in stations located in a district different from their origin addresses. As concerns the latter, the comparison of the flows shown by Maidstone and Sevenoaks in both tables suggest that a high number of non residents of these two areas initiate their commuting trips in stations of these districts. Apart from Gillingham, both tables show that most of the changes in the commuting flows occurred in Sevenoaks, Swale and Ashford.


Table 7.8 shows substantial increases in the flows from the districts Ashford (79%), Swale (50%), and Dover (38%). Shepway presented flow changes about the overall average commuting growth of 21%, whereas the remaining districts showed much lower figures. Given the relative accuracy of the information provided by the origin addresses compared with the coarse approach used in the computation of the district flows by association with station flows, only the 1989 distribution of rail commuting shown by table 7.8 will be considered in the remainder of this thesis.