Counterfactual analysis
Introduction
As outlined within Scottish Transport Appraisal Guidance (STAG), when evaluating the success of a scheme it is important to determine the extent any change in the outcomes or outputs monitored are a result of the intervention, as opposed to other external factors. Within this evaluation this has been determined by comparing actual outcomes with those of a counterfactual location which represents the transport situation had the project not gone ahead.
At the baseline stage, a counterfactual survey was undertaken at Ancrum and surrounding local communities. This was selected as a suitable location as it is an area with a similar population and public transport connections to the Reston study area (prior to scheme delivery). Ancrum acts as a proxy for what would have happened in the Reston area if the new station had not been built.
Ancrum is a small village within the Scottish Borders that is situated between Jedburgh and St Boswells. Similar to Reston before the station opened, there is no rail access in the village, with the closest station located 12 miles away at Tweedbank.
Within the baseline reporting, the counterfactual survey was designed to meet the following objective:
Collect baseline data to establish a comparator / counterfactual using primary research.
At this stage of the evaluation, this survey has been replicated to help understand what would have happened if the scheme had not been delivered in Reston and help attribute benefits to the scheme.
The results of counterfactual surveys along with other available data (socio-economic data) for Ancrum and Reston have been used to set out trends affecting both areas and not related to the new station. This helped to provide a credible estimate of the additional change attributable to the station itself, isolating impacts from other external factors.
Methodology
Target Area
Figures 89 and 90 show the that area adopted for the Ancrum counterfactual survey mirrors that of the baseline reporting to ensure a similar sample undertake the survey. The catchment adopted for the counterfactual survey, similar to the resident survey, is expanded to approximately 10km to capture an appropriate sample size.
Survey Material
The counterfactual survey was developed collaboratively between IBP, Jacobs and Transport Scotland. This survey was designed to align with the baseline survey as much as possible to allow comparisons to be made whilst also introducing questions to understand whether travel patterns and behaviours have changed since the last survey (with these also being compared to relevant questions in the current Residents Survey). The final survey is presented within Appendix A.
Survey Delivery
The baseline survey was conducted by means of telephone interview. However, a face-to-face approach was chosen for the Stage? 1 evaluation survey, both for practical reasons in relation to securing a suitable sample size, and in ensuring that a reasonable proportion of interviews could be secured from Ancrum itself (this breakdown being unclear from the baseline reporting).
Given the large extent of the catchment considered for counterfactual surveys there was a significant risk that respondents would be located some distance from the centre of Ancrum. For this reason, a target of 50 responses from Ancrum and 50 responses from the wider settlements of Jedburgh and St. Boswell were set. A minimum target size of 100 responses was therefore set to match the baseline sample size of 101 and this sample size was achieved.
Comparison to Baseline Survey
The baseline counterfactual survey also applied age weighting relative to National Records of Scotland population estimates for relevant Local Authority areas. Again, the relationship between the survey catchment areas and these wider population geographies was not a direct one and weighting was not applied in relation to other demographic or contextual variables. Sensitivity analysis suggests that age weighting had limited impact on overall findings and broad trends.
Accordingly, unweighted data has been used for primary comparative analysis in order to maintain consistency between the baseline and evaluation datasets. Sub-group analysis has been undertaken where appropriate, and any relevant considerations or limitations relating to comparability are noted throughout the report.
Research Limitations
The counterfactual survey is intended to provide an indicative comparison with individuals who do not currently use Reston Rail Station or who may use alternative locations or travel modes. As with all counterfactual exercises of this nature, the findings should be interpreted with awareness that responses are based on stated perceptions, intentions and hypothetical behaviours rather than directly observed outcomes. Respondents’ views on how they might behave under different circumstances may not necessarily reflect actual future behaviour. In addition, the comparison between survey groups is influenced by differences in respondent characteristics, travel needs and geographic context that cannot be fully controlled within a practical fieldwork exercise. The counterfactual findings should therefore be regarded as providing contextual insight and indicative comparison rather than a definitive measure of causation and should be read alongside the wider survey and evaluation evidence.
Data & Analysis
Respondent Profile
Similar to the Reston resident survey, respondents were asked various questions regarding their personal characteristics (such as gender, age, ethnicity etc.). This demographic information allows for a comparison with the Reston resident survey to determine whether Ancrum and surrounding areas are similar in make up to the Reston resident survey sample.
Figure 91 presents the gender profile of Ancrum residents (and surrounding areas) who partook in the survey. Similar to the Reston resident survey, a greater proportion of female respondents were captured as part of the survey.
Respondents were asked to identify their age category. Again, similar to the Reston resident survey, Figure 92 shows that the Ancrum (and surround area) population shows a similar profile with spikes for the 65-74 years of age and 45-54 years of age brackets. This demonstrates that the population of Ancrum captured as part of this survey is broadly similar to Reston in terms of gender and age.
Respondents also provided their employment status to ensure a similar market is available to that at Reston. Table 52 shows that at this time, the number of residents in full time employment (29%) and retired (40%) are the most prevalent. These trends are similar for residents in Reston.
| Employment Status | Baseline | 2026 |
|---|---|---|
| Full-time employment | 17% | 29% |
| Part-time employment | 16% | 11% |
| Full or part time further education | 0% | 1% |
| Full or part time higher education | 0% | 1% |
| Not in employment due to illness or disability | 0% | 4% |
| Not in employment / looking after home or family | 0% | 9% |
| Unemployed | 5% | 5% |
| Retired | 62% | 40% |
| Other | 0% | 0% |
| Prefer not to say | 0% | 0% |
| Base | 101 | 100 |
In terms of annual household income there is some disparity between the baseline and stage 1 counterfactual. Table 53 shows that at baseline stage, 32% of respondents earnt less than £20,000 whereas this was 7% at Stage 1. When compared to the Reston resident survey a similar trend is observed with the most common income bracket being £20,000 to £39,999. Again, this helps to validate that this is a suitable counterfactual location as people taking part in the survey have similar incomes and can therefore afford to make similar transport choices.
| Baseline Income Band | Baseline % | 2026 Income Band | 2026 % |
|---|---|---|---|
| Less than £10,000 | 11% | Under £20,000 | 7% |
| £10,001 to £20,000 | 21% | £20,000 to £39,999 | 18% |
| £20,001 to £30,000 | 13% | £40,000 to £59,999 | 5% |
| £30,001 to £40,000 | 8% | £60,000 to £79,999 | 3% |
| £40,001 - £50,000 | 1% | £80,000 to £99,999 | 1% |
| £50,001 - £60,000 | 0% | More than £100,000 | 1% |
| Over £60,000 | 4% | Don't know | 2% |
| Don't know | 7% | Prefer not to say | 63% |
| Prefer not to say | 36% | Base | 100 |
| Base | 101 | No data | No data |
Respondents were also asked to identify whether they have a physical and /or mental health condition and whether this impacts their mobility. Figure 93 shows that around a quarter of respondents identified as having a physical and/or mental health condition within and around Ancrum, which is marginally less than that observed in the Reston resident survey.
Similar to the Reston resident survey a significant proportion of those who identified as having a physical and/or mental health condition class themselves as being mobility impaired. Figure 94 shows that 75% (from a base of 24 respondents) identified that their disability impairs their ability to travel.
Respondents were also asked to identify whether they are National Entitlement Card (NEC) owners. Figure 95 shows that the proportion of card holders is very similar to that witnessed within the resident survey for Reston with just under half holding an NEC (48% in Reston).
Respondents were also asked to identify whether they have access to a car. Figure 96 shows that many respondents have full access to a car (49%) however this is noticeably lower than the proportion of Reston residents with full access to a car (73%). However, when looking at all respondents who have access to a car to some extent, within Ancrum and surrounding areas this represents 88% of respondents and in Reston and surrounding areas this represents 89% of respondents demonstrating broad alignment between the study area and this counterfactual location.
Finally, respondents were asked to identify their ethnicity. Table 54 shows that similar to Reston, a significant proportion of respondents identified as being either White – Scottish or White – Other British, totalling 96% in and around Ancrum and 94% in and around Reston.
| Ethnic Origin | Baseline | 2026 |
|---|---|---|
| White - Scottish | 67% | 86% |
| White - Other British | 24% | 10% |
| White - Irish | 0% | 1% |
| White - Gypsy / Traveller | 0% | 0% |
| White - Polish | 0% | 1% |
| White - Any other White ethnic group | 3% | 1% |
| Mixed or Multiple Ethnic Group (please specify below) | 0% | 0% |
| Asian - Pakistani, Pakistani Scottish or Pakistani British | 0% | 0% |
| Asian - Indian, Indian Scottish or Indian British | 1% | 0% |
| Asian - Bangladeshi, Bangladeshi Scottish or Bangladeshi British | 0% | 0% |
| Asian - Chinese, Chinese Scottish or Chinese British | 0% | 0% |
| Asian - Any other Asian ethnic group | 0% | 0% |
| African - African, African Scottish or African British | 0% | 0% |
| African - Any other African ethnic group | 0% | 0% |
| Caribbean or Black - Caribbean, Caribbean Scottish or Caribbean British | 0% | 0% |
| Caribbean or Black - Black, Black Scottish or Black British | 0% | 0% |
| Caribbean or Black - Any other Caribbean or Black ethnic group | 0% | 0% |
| Other - Arab, Arab Scottish or Arab British | 0% | 0% |
| Other - Any other ethnic group | 0% | 0% |
| Don't know | 1% | 0% |
| Prefer not to say | 4% | 1% |
| Base | 101 | 100 |
Travel to Work and Education
Similar to the Reston resident survey, respondents were asked questions regarding their travel choices to help understand whether changes in Reston and surrounding areas can be attributed to the introduction of Reston Railway Station or are as a result of wider change. Figure 97 shows that within Ancrum, a similar trend in terms of working patterns is apparent with the majority travelling to a fixed place of work / education (62%), this aligns closely with Reston where 64% travel to a fixed place of work/education.
Ancrum also demonstrates similarities in terms of the frequency of commuting journeys. Figure 98 shows that the significant proportion of journeys to work/education are undertaken five days per week (45%) which aligns closely with travel patterns in Reston (38% travelling five days per week).
Respondents were then asked to identify their primary mode of transport used to travel to work/education. As shown within Table 55, similar to Reston, Ancrum and its surrounding areas is dominated by car travel, with 77% travelling as either a car driver or passenger. This is mostly due to the rural geographies around Ancrum and a perceived convenience attributed to car travel. However, it is noticeable that 0% of respondents use rail as a commuting mode in and around Ancrum; this differs the Reston where 4% of respondents identify as using rail for access to work/education. This could be attributed as being as a result of the scheme opening, as it makes rail a viable commuting alternative.
| Mode of Transport | Baseline | 2026 |
|---|---|---|
| Walk / Wheel | 21% | 13% |
| Cycle | 0% | 0% |
| Bus | 14% | 6% |
| Rail | 0% | 0% |
| Bus & Rail | 0% | 0% |
| Park & Ride (Bus) | 0% | 0% |
| Park & Ride (Rail) | 0% | 3% |
| Car / Van (passenger) | 7% | 6% |
| Car / Van (drive myself) | 59% | 71% |
| Other | 0% | 0% |
| Base | 29 | 31 |
Respondents were also asked to consider their commuting journey times to their place of work. The length of journey time is shown in Table 56. Respondents generally work in close proximity to Ancrum with many commuting to nearby Jedburgh (55% of respondents), hence why journey times are relatively low. There is evidence, however, of a small number of residents commuting further afield to Edinburgh and Eyemouth. For those commuting to Edinburgh, journey times are identified as being in excess of one hour. Without a direct rail connection to Edinburgh, journey times in excess of one hour are unfavourable hence why limited commuter journeys currently take place.
| Duration | Baseline | 2026 |
|---|---|---|
| Length of journey commuting to work / education | 25 minutes | 18 minutes |
| Base | Unknown | 31 |
Figure 99 shows that of those who currently drive to their place of work/education, 86% are able to park at their workplace. This is similar to that observed with Reston residents. This availability of parking and its convenience can make public transport less favourable and therefore a modal shift away from private cars more difficult to achieve.
Leisure and Personal Travel
Respondents were asked to identify locations they have travelled to over the course of the last 30 days for personal or leisure purposes. Table 57 shows that the majority of respondents identified regular travel to Jedburgh (62%), Galashiels (39%) and Edinburgh (35%), Kelso (30%) and Tweedbank (30%). The majority of these destinations are short distance journeys within 2km of Ancrum. Within the Reston resident survey, a greater percentage of residents identified travel to destinations further afield for leisure and personal purposes with greater proportions visiting Newcastle (14% compared to 6% in Ancrum). This may be facilitated by longer distance rail connections that are provided at Reston; however, this is broadly similar to the results of the baseline resident survey making it difficult to attribute to the scheme directly.
| Location | Baseline (Personal Business) | Baseline (Leisure Purposes) | 2026 |
|---|---|---|---|
| Berwick Upon Tweed | 7% | 11% | 13% |
| Carlisle | 3% | 9% | 3% |
| Edinburgh | 37% | 32% | 35% |
| Galashiels | 68% | 62% | 39% |
| Hawick | 56% | 38% | 17% |
| Jedburgh | 61% | 46% | 62% |
| Kelso | 58% | 36% | 30% |
| Melrose | 39% | 26% | 15% |
| Newcastle | 2% | 9% | 6% |
| Peebles | 6% | 8% | 1% |
| St Boswells | 37% | 12% | 24% |
| Tweedbank | 25% | 14% | 30% |
| Other | 3% | 6% | 5% |
| I have not travelled for personal or leisure purposes in the past 30 days | 3% | 10% | 10% |
| Base | Unknown | Unknown | 100 |
Table 58 shows the top two modes used to travel to various destinations for leisure and personal based journeys over the last 30 days. Use of car/van appears to be the most favourable mode of travel to various destinations. However, rail does appear in the stage 1 counterfactual survey mostly for longer distance journeys to Berwick-Upon-Tweed, Edinburgh and Newcastle. This is consistent with the finding of the Reston resident survey where residents opted for rail for longer distance travel.
| Location | Top two modes | Baseline (personal business) | Baseline (leisure purposes) | 2026 |
|---|---|---|---|---|
| Berwick Upon Tweed | Car/Van Driver | 50% | 53% | 31% |
| Berwick Upon Tweed | Car/Van Passenger | 33% | 35% | - |
| Berwick Upon Tweed | Rail | - | - | 38% |
| Carlisle | Car/Van Driver | 57% | 29% | 67% |
| Carlisle | Car/Van Passenger | 29% | 43% | 33% |
| Edinburgh | Car/Van Driver | 49% | 39% | 20% |
| Edinburgh | Rail | 14% | 19% | 60% |
| Galashiels | Car/Van Driver | 63% | 58% | 44% |
| Galashiels | Car/Van Passenger | 24% | 29% | - |
| Galashiels | Bus | - | - | 33% |
| Hawick | Car/Van Driver | 79% | 65% | 65% |
| Hawick | Car/Van Passenger | 13% | 26% | - |
| Hawick | Bus | - | - | 24% |
| Jedburgh | Car/Van Driver | 61% | 53% | 38% |
| Jedburgh | Car/Van Passenger | 18% | - | - |
| Jedburgh | Walk/Wheel | - | 22% | 25% |
| Kelso | Car/Van Driver | 73% | 64% | 73% |
| Kelso | Car/Van Passenger | 20% | 25% | 20% |
| Melrose | Car/Van Driver | 65% | 61% | 67% |
| Melrose | Car/Van Passenger | 28% | 32% | 20% |
| Newcastle | Car/Van Driver | 67% | 100% | 33% |
| Newcastle | Rail | - | - | 67% |
| Newcastle | Other | 33% | - | - |
| Peebles | Car/Van Driver | 65% | 62% | 100% |
| Peebles | Car/Van Passenger | 27% | 38% | - |
| St Boswells | Car/Van Driver | 59% | 52% | 46% |
| St Boswells | Car/Van Passenger | 25% | 19% | 29% |
| Tweedbank | Car/Van Driver | 58% | 58% | 34% |
| Tweedbank | Car/Van Passenger | 27% | 33% | 38% |
Table 59 presents the average journey times from Ancrum to various destinations. Compared to the baseline survey. There have been notable reduction in journey times, in particular to Edinburgh and Carlise where journey times reductions of 10 minutes or more have been realised. Similar journey time reductions were also observed in the Reston resident survey, suggesting this cannot be directly attributed to the opening of Reston Railway Station as it has also been witnessed in and around Ancrum.
| Location | Baseline (personal business) Average minutes | Baseline (personal business) Base | Baseline (leisure purposes) Average minutes | Baseline (leisure purposes) Base | 2026 Average minutes | 2026 Base |
|---|---|---|---|---|---|---|
| Berwick Upon Tweed | 55.1 | 12 | 57.4 | 17 | 46.3 | 12 |
| Carlisle | 73 | 7 | 88.3 | 6 | 60 | 3 |
| Edinburgh | 64.3 | 35 | 68.1 | 31 | 53.6 | 35 |
| Galashiels | 26.2 | 67 | 26.4 | 45 | 20.9 | 38 |
| Hawick | 21.6 | 47 | 18.9 | 34 | 25.3 | 17 |
| Jedburgh | 11.7 | 56 | 12.2 | 36 | 11.6 | 60 |
| Kelso | 21.2 | 60 | 19.1 | 44 | 20.8 | 29 |
| Melrose | 17.5 | 46 | 17.5 | 31 | 18.5 | 15 |
| Newcastle | 62 | 3 | 78 | 5 | 57.5 | 6 |
| Peebles | 45.5 | 12 | 48.5 | 13 | 22.5 | 1 |
| St Boswells | 15.4 | 44 | 10.7 | 21 | 12.3 | 24 |
| Tweedbank | 16.9 | 26 | 16.7 | 12 | 23.4 | 28 |
Use of the Rail Network
Residents of Ancrum and surrounding areas were asked a series of questions regarding their current use of the rail network. Figure 100 shows that despite not having a direct connection to the rail network in close proximity, the majority of respondents (55%) do travel by rail in some capacity.
Of these rail users, the majority travel to Tweedbank station to access the rail network (95%) with others travelling to Galashiels. Respondents outlined that rail is mostly used less often than once per week (78%) or once per week (15%).
Figure 101 shows that Ancrum residents outlined that when accessing the rail network, car was the most used access mode, with 87% either travelling as a car driver or passenger. Owing to the distance to nearby stations and the rural nature of Ancrum, very few respondents travelled to existing stations by public transport or active modes. This is consistent with responses at the baseline stage for Reston residents prior to scheme delivery.
Figure 102 shows that the majority of rail users do so for leisure-based trips (93%), with 11% using rail for access to employment. This is consistent with the Reston resident survey which demonstrates that Reston Railway Station is currently mostly used for infrequent leisure-based trips.
Non-rail users were asked to identify the major reasons for this decision. Table 60 shows that similar to Reston Railway Station non-users, the most common reason for not travelling by train was the preferred convenience of travelling by car, that public transport options for station access are not convenient, bus travel is free using their NEC card, and that rail is considered too expensive and more time consuming.
| Reasons | Major Reason | Minor Reason | Not a Reason | Base |
|---|---|---|---|---|
| My National Entitlement Card makes bus travel free | 30% | 25% | 45% | 20 |
| I like the convenience of travelling by car / van | 56% | 9% | 35% | 43 |
| I need my car / van for business purposes | 14% | - | 86% | 43 |
| There is not enough parking at my local station to allow me to park and ride | 9% | 7% | 84% | 43 |
| Public transport options between my house and the station aren't convenient / don't exist | 53% | 9% | 37% | 43 |
| Public transport options between the end station and my desired destinations aren't convenient / don't exist | 42% | 9% | 49% | 43 |
| Travel by rail would take longer | 23% | 9% | 67% | 43 |
| Rail travel is too expensive | 26% | 21% | 53% | 43 |
| The timetable does not suit my needs | 12% | 2% | 86% | 43 |
| The Sunday rail timetable frequency does not suit my needs | 9% | 2% | 88% | 43 |
| Rail services at peak times are very busy and there is no guarantee of a seat | 12% | 5% | 84% | 43 |
| Accessing the train is challenging due to physical limitations / disability | 12% | - | 88% | 43 |
| Other | 100% | - | - | 2 |
Views on Public Transport
Respondents in and around Ancrum were asked to outline their level of satisfaction with the current public transport services that are available to them. Figure 103 shows that this generated a broadly positive response, with 49% of respondents outlining that they are satisfied with the current services available to them. This demonstrates an improvement on the baseline position. This, however, does not guarantee that public transport is working to its full potential as shown by earlier results where rail travel in particular is mostly used for infrequent leisure-based journeys.
Figure 104 shows that when asked about the factors that influence this decision, the majority of residents highlighted the availability of public transport and overall journey times as the reason for their satisfaction/dissatisfaction.
Respondents were then asked to consider how public transport services have changed over the course of the last three years considering whether various factors have improved, worsened or remained unchanged. Table 61 shows that for all identified factors the majority of respondents (in many cases 60% or more) considered no change has taken place over the last three years. However, 49% of respondents consider that fares have worsened either a little or a lot, and 27% state that the length of the operating day has worsened. Furthermore, 20% of respondents outlined that the current public transport overall has worsened with respect to their household travel needs.
| Factors | Improved a lot | Improved a little | No change | A Little Worse | A Lot Worse | Don't know / Not relevant | Base |
|---|---|---|---|---|---|---|---|
| Availability of public transport services | - | 17% | 60% | 13% | 2% | 8% | 99 |
| Frequency | 1% | 8% | 64% | 14% | 7% | 6% | 99 |
| Fares | - | 6% | 33% | 31% | 18% | 12% | 97 |
| Length of operating day | - | 5% | 55% | 20% | 7% | 13% | 99 |
| Overall journey time | - | 6% | 76% | 6% | 6% | 6% | 99 |
| Reliability | - | 14% | 61% | 5% | 12% | 8% | 99 |
| Current public transport overall with respect to your household's travel needs | - | 10% | 64% | 15% | 5% | 6% | 99 |
Summary of Findings
The counterfactual analysis enables a comparison of trends observed in Reston against those in a similar settlement (Ancrum) without a new rail intervention. This enables scheme‑specific effects and wider background changes in travel behaviour, accessibility and socio‑economic conditions to be identified.
Overall, many observed trends (such as reductions in perceived journey times, car dominance, and limited changes in overall trip frequency) are evident in both Reston and Ancrum areas. This suggests that a proportion of the changes identified in the primary and secondary data reflect wider background changes, including post pandemic travel patterns, economic conditions and evolving working behaviours, rather than being solely attributed to the opening of Reston Railway Station.
Despite this, a number of differences between the two areas point to scheme‑related impacts. Reston residents demonstrate higher levels of rail usage, particularly for longer-distance leisure trips to destinations such as Edinburgh and Newcastle, while rail use in the comparator settlement remains limited. This suggests that the introduction of the station has improved access to the rail network and enabled a modest shift toward rail for specific journey types. However, this has not translated into significant changes in commuting behaviour or rail modal share.
Travel patterns in both locations remain car dependent for both commuting and leisure-based travel. In Ancrum, rail plays no role in commuting trips, whereas in Reston it has achieved a marginal increase. This reinforces the finding that, whilst the station has expanded travel options, it has not fundamentally altered core travel behaviours at this point in time. The analysis also highlights that improvements in journey times have been experienced in both areas, indicating that some accessibility benefits arise from wider network or external factors rather than the station alone.
Nevertheless, Reston shows evidence of increased connectivity to key regional centres via rail, supporting longer-distance travel patterns not observed to the same extent in Ancrum.
From a socio‑economic perspective, trends in indicators such as employment, income and benefit claims are broadly similar between the two areas, with no clear evidence of economic uplift uniquely associated with the station.
Key barriers to behavioural change are consistent across both areas, including the continued convenience of car travel, cost perceptions, and limitations in public transport provision. In Reston, additional scheme‑specific constraints like service frequency, timetable suitability and weak integration with local bus services, limit the extent to which the station can influence modal shift.
The counterfactual analysis therefore indicates that the reopening of Reston Railway Station is beginning to generate modest transport-related impacts, particularly in terms of enabling rail use for longer-distance travel and improving perceived accessibility. The greatest impact can be seen in close proximity to the station (within Reston itself) where residents are now using rail who did not previously and some degree of substitution for car-based journeys is observed. However, most broader trends are mirrored in the comparator settlement, suggesting that the overall scale of change remains limited at this stage. The station contributes positively to local connectivity, but its wider behavioural and economic effects are still emerging and are currently constrained by service and integration factors, consistent with the findings from the Reston area resident and business surveys.