The Scottish Government

« Previous | Contents | Next »

Listen

RURAL ACCESSIBILITY

CHAPTER EIGHT ANALYSIS OF TRAVEL PATTERNS

8.1 The analysis sought to identify consistent ways of understanding and predicting travel behaviour by rural residents. Previous research (Scottish Executive 2000) had shown that accessibility measures could describe transport choices from the viewpoint of travellers. It was therefore decided to calculate accessibility measures for a range of trip purposes in each of the study areas to investigate whether observed travel perceptions and behaviour from the survey work were described by the accessibility measures.

8.2 This Chapter therefore describes accessibility measures for each of the five areas by trip purpose and examines the extent to which these measures explain the surveyed travel choices.

Data Collection

8.3 Data from each Council was collected for all the public transport and community transport services. Separation (deterrence) values were then estimated for car available and non car available travel for each of the study areas using the local zone structure defined in Chapter 2.

8.4 The assumptions used in calculating the average generalised time of travel between each of these zones are set out in Appendix D.

8.5 Data on the opportunities available in each zone were obtained as follows:

• Population - Mid year population figures for 1997 at a district level from GROS.
• Employment - Figures were obtained from the national employment census for the number in employment in December 1999 and the number of unemployed people for March 2001.
• Health - The location of all GP surgeries in Scotland was derived from statistics maintained by ISD and supplied by SEGIS.
• Supermarkets - Supplied by SEGIS from their database.
• Banks and Building Societies - Supplied by SEGIS from their database.
• Chemists - Derived from commercial data provided by Retail Locations and supplied by SEGIS.
• Petrol Stations - Derived from a survey by ERM in 1998 and supplied by SEGIS
• Post Offices - Data from Post Office Counters supplied by SEGIS
• Schools - Data from the Scottish Executive education statistics division supplied by SEGIS.

8.6 Opportunities were represented in the accessibility calculation as described in Appendix D.

8.7 Accessibility indices were then calculated for each zone to zone opportunity by factoring the opportunities in each zone by a deterrence function ³. For the initial analysis the calibration factor was set to the values shown in Table 17. The aim was to start with a consistent look at the five study areas using the same calibration factors.

Table 17 - Calibration factors

8.8 The indices calculated are shown in Appendix E. The results are summarised for each area as follows.

East Ayrshire

8.9 For car available households in East Ayrshire, accessibility is generally good. Zones closest to the major towns of Ayr and Kilmarnock rate particularly highly for access to employment and supermarkets. Overall Mauchline, Cumnock and Auchinleck have good access to a range of services. The lowest levels of accessibility are for the most sparsely populated zones of Dalblair and Muirkirk.

8.10 The indices also allow the travel choices faced by residents of each zone to be examined by trip purpose. Figure 25 shows the attractiveness of different zones for each travel purpose.

Figure 25 - Car Travel Choices for New Cumnock Residents

8.11 New Cumnock residents can be expected to undertake most of their personal business in New Cumnock or Cumnock but to travel to larger towns for food shopping and employment. Given the relatively low number of jobs locally and the strong attraction of Scotland's largest employment centre Glasgow only 40 miles away it is interesting to note that this analysis predicts that 70 per cent of the car available work trips will be to Glasgow. The observed data shows that very few people actually travel to Glasgow and that about 80 per cent of residents work in the Ayr and Kilmarnock areas. This emphasises that employment markets need to be broken down by population and employment sector for these data to be calibrated meaningfully. Nevertheless the accessibility analysis does demonstrate approximately the correct proportion of people who need to travel to regional centres for work.

8.12 Although non car accessibility is typically a factor of five to ten lower than for car accessibility, non car access is also good for a rural area from most zones. However for non car trips the difference between the larger and smaller towns is very clear. Whereas for car travel accessibility in New Cumnock is just over half that for that for the more central town of Auchinleck, for non car travel New Cumnock only has about a quarter of the level of accessibility of Auchinleck. The accessibility analysis suggests that more non car trips will be made locally in the New Cumnock and Cumnock area.

8.13 Figure 26 shows the equivalent anticipated distribution of trips for non-car travellers.

Figure 26 - Non Car Travel Choices for New Cumnock Residents

8.14 Commuting trips to Glasgow are difficult to achieve given the time to reach rail at Kilmarnock so this option has been excluded from the analysis. Kilmarnock has fewer jobs than Ayr/Prestwick but would be expected to attract more work trips from new Cumnock because of the direct bus service in the morning peak. For non car travellers the employment opportunities are also likely to be more attractive.

8.15 Observed data on non car trips is too limited to allow callibration by trip purpose but overall 80 to 100 per cent of trips have destinations in Cumnock or New Cumnock. The accessibility indices represent this for banks and GPs but suggest that more people would be expected to travel elsewhere for supermarket shopping.

8.16 There is also no evidence from the data of non car work trips to Ayr and Kilmarnock even though the analysis would suggest that most work trips would have destinations in these zones. It appears that people in this area are travelling less far than the accessibility analysis would predict.

West Aberdeenshire

8.17 There is a very wide range of accessibility across this study area. Strathdon and Braemar are approaching levels of accessibility more common in remote areas but close to Banchory and Alford there is very good accessibility particularly for car available trips.

8.18 Figure 27 contrasts car and non car available accessibility for access to supermarkets for residents of Strathdon and Aboyne.

Figure 27 - Comparison of Supermarket Shopping Destinations for Strathdon and Aboyne

8.19 The accessibility index is expressed as square feet of shopping floorspace. It can be seen that for non car available households in Strathdon, supermarket shopping is not a practical option. For car available households the scale of the shopping opportunities in Aberdeen suggests that people would choose to travel there if they had the time available. However, it can be seen that there are at least some shopping opportunities in each zone which may prove attractive for some trips.

8.20 The Tables in Appendix E demonstrate that Aberdeen is also the most likely destination for employment opportunities but that local rural destinations should be attractive for other trip purposes. However there were no chemists outside Aberdeen in the SEGIS database though this is known to be incorrect. Some updating of this database is therefore required which would substantially improve accessibility.

8.21 For all trip purposes, non car accessibility is very much worse away from the core bus routes. The combination of limited provision of services and limited public transport gives these residents very poor accessibility. For example Strathdon has levels of non car accessibility about one hundredth of the car available accessibility. This reinforces the points made by people in this area that car travel is a necessity either by taking lifts or by owning cars.

Wigtownshire

8.22 Of all the case study areas Wigtownshire is the most homogeneous in terms of accessibility characteristics. For a deeply rural area, the bus services are good, and there are a reasonable range of services in the towns of Newton Stewart, Gatehouse of Fleet, Kirkcudbright and Castle Douglas in addition to a wider range of services in Stranraer and Dumfries.

8.23 It would therefore be expected that observed trip patterns would show a higher than average use of local centres and the trip distribution spread relatively evenly across the zones. Although the trip numbers from each zone are small, over 80 per cent of trip making is elsewhere in the origin zone or to a neighbouring zone. The main surprise from the observed data is that Dumfries does not attract more trips.

8.24 Car and non car accessibility to supermarkets are compared for Newton Stewart in Figure 28. It can be seen that non car trips are most likely within Newton Stewart but that a significant number of trips to Stranraer are also likely. This is also the general trip patterns observed for other trip purposes amongst this group.

Figure 28 - Comparison of Supermarket Shopping Destinations for Newton Stewart

8.25 For car trips there is a wider range of destinations to choose, but Stranraer, Dumfries and Castle Douglas are likely to be the most attractive based on square footage of floorspace of supermarkets. In practice car available trips are less scattered than this. About half the observed trips are to Stranraer which is broadly in line with the expectations from the accessibility indicators but there are no observed trips to Dumfries. It may be that the supermarkets in Newton Stewart and Stranraer provide a range and choice which is acceptable to the majority of the local population. Alternatively it may be that the higher than average age of the local population reduces the propensity to travel.

Caithness and Sutherland

8.26 This area includes some very remote locations and the analysis zone sizes are very large. To adopt smaller zones would rarely be practical since in most cases the cost of the analysis would become disproportionate to the transport or service changes being considered. In considering the results of the accessibility analysis it should therefore be borne in mind that the measures are intended to give only a general indication of levels of accessibility.

8.27 The first point to note from the Tables in Appendix E is that for many trip purposes much of the area has negligible non car accessibility to fixed services such as Chemists or GP practices. Even for car available accessibility, practical levels of accessibility are only achieved in the more populated areas. In the very remote areas it is often necessary and much more efficient to make services mobile, to provide the community with a basic level of accessibility. Many residents of this area highlighted this in the telephone interviews.

8.28 The indicators suggest that only between Golspie, Dornoch and Tain and Halkirk, Lybster, Wick and Thurso are there likely to be non car available people working outwith the zone where they live. However for car available employment trips from the south of the area the indicators suggest that Inverness may be as attractive an employment location working within the zone where people live. Journey times to work of an hour or more are not uncommon near to major cities but the observed data suggests that travel to work areas for these remote locations are likely to be much shorter for the majority of people. Calibration of the accessibility measure to observed behaviour would therefore suggest much greater sensitivity of trip making to travel time.

8.29 The characteristics for supermarket trips show similar trends to the employment patterns. Almost every zone has some supermarket floorspace and this provides typically 25 to 50 per cent of the overall accessibility to supermarkets. For most zones, Inverness, Thurso and Wick provide more than half of the accessibility opportunity. Although the observed data shows very much fewer trips to these strategic locations than this it is likely that if spend were studied at each location the accessibility measure would be a reasonable representation.

East Lothian

8.30 For nearly all the zones in the East Lothian area, the employment opportunities in Edinburgh provide the greatest contribution to accessibility for both car and non car travel. Only for non car travel from the remoter zones, does local accessibility become most important.

8.31 The analysis also suggests that Edinburgh will be an important supermarket shopping destination for up to a third of trips from the area. However the observed travel data indicates very few supermarket shopping trips to Edinburgh. People in East Lothian mainly choose to shop in local supermarkets or in more major stores in local centres such as Haddington.

8.32 Other trip purposes also show similar trends with most personal business trips being captured locally although for banking the analysis suggest that Edinburgh will attract more trips.

Calibration of the accessibility analysis

8.33 The above discussion highlights that the calibration factors in Table 17 can give a good indication of the choices facing rural residents and that in many cases observed travel behaviour matches the opportunities identified by the accessibility indices.

8.34 There are a number of lessons from the above analysis as follows:

• For assessing travel to work behaviour it is important to consider each sector of the employment market separately. Highly skilled jobs will not always be suitable for socially excluded people seeking to re-enter work, so the catchments for each employment market should be analysed separately. Unfortunately analysis at this level of detail is beyond the scope of this project.
• Analysis to help define accessibility problems and solutions in rural areas should not be undertaken in isolation. There are many factors such as attitudes and perceptions in the local community which are also strong influences on travel behaviour and can only be understood by working with local people when developing improvements.
• Opportunities such as work, shopping, healthcare etc are best defined in terms of the level of activity at each location and the quality of the service provided. For example floorspace was used as a proxy for retail activity but if turnover of each store could be used then that would be much more accurate.
• The analysis highlights the accessibility benefits of providing facilities as close as possible to where people live for many types of trip, such as access to local health centre. However to analyse what transport measures are necessary locally to help people travel to these local facilities would require very small zones. Accessibility analysis with large zones is quick and easy to undertake and provides a valuable overview of areas of poor accessibility. Analysing schemes at a very local level would be much more time consuming and in many cases uneconomic.

8.35 Given the absence of detailed employment data and the local nature of many of the other trip purposes, the only trip purpose that merits more detailed calibration at this stage is for access to food shopping. For each of the areas, people were observed travelling very different distances for this purpose and it is useful to examine how much of this difference can be explained simply by the availability of shopping opportunities.

8.36 The accessibility analysis was therefore run for the supermarket shopping trip purpose to obtain the calibration factor with the best fit to the observed data for each of the trip purposes. The results of this analysis are shown in Table 18.

Table 18 - Calibration factors for Access to Supermarkets

8.37 It can be seen that, for car trips, three of the areas calibrate with factors close to 0.09. This is significantly higher than the 0.05 used in the analysis above. If 0.09 had been used in the analysis then the value of local supermarkets would have been found to be even greater. Given the large differences in the characteristics of these three case study areas, the consistency in calibration may suggest that accessibility measures can account for different degrees of rurality.

8.38 Caithness and Sutherland indicates a lower propensity to travel further. As indicated above this is likely to be a result of the way that supermarket activity was represented. Through the trip frequency information and the telephone surveys it is clear that people from the remoter areas make less frequent trips to larger supermarkets but buy more on each trip. Therefore if retail turnover was considered it is likely that the Caithness and Sutherland factor would be closer to 0.09.

8.39 It was noted from the telephone and postal surveys that East Ayrshire has some very different characteristics from the other rural areas. It also has the highest density of: socially excluded people, people in public rented accommodation and people on lower incomes. The lower propensity to travel is consistent with all these factors. Not only do lower disposable incomes lead to pressure to reduce transport costs but many people do not look beyond the boundaries of their own community for better opportunities elsewhere.

8.40 For non car trips, the sample sizes were smaller but again there seems to be some consistency for a calibration factor of just over 0.02. Travellers without cars generally spend longer travelling so it is not surprising that the sensitivity to travel time is lower than for car travel. For all the areas, more of these trips are very local so, despite its remoteness, Caithness and Sutherland shows a similar value to less remote areas.

8.41 For East Lothian, car ownership was higher and only seven non car supermarket shopping trips were identified from the data. Therefore this is not a reliable figure.

8.42 East Ayrshire shows a greater propensity to travel further which is the opposite from the car trips. This is likely to be because concessionary fare pricing was not allowed for in the analysis. Many people from towns in the East Ayrshire area can travel to Ayr and Kilmarnock for 40 pence which is the same as it would cost them to travel to Cumnock. Part of the reason for this rural fare subsidy by SPT is to enhance the quality of opportunities available to socially excluded residents. It is therefore likely that the sample is affected by, probably elderly people, travelling cheaply to Ayr or Kilmarnock when their counterparts in the other study areas would travel more locally.

« Previous | Contents | Next »