Findings from the Travel and Places Survey (Wave 1)

Johara Meyer

Research Associate

University of Westminster

J.Meyer@westminster.ac.uk

Dr Jamie Furlong

Research Fellow

University of Westminster

J.Furlong@westminster.ac.uk

The Travel and Places Survey (Wave 1)

Key information (1)

  • June-July 2022
  • Second wave of longitudinal study with 4,912 responses (baseline: 12,449)
  • Survey completed online using Qualtrics software
  • Questions on: past-week travel behaviour, local area perceptions, attitudes to TfL/LA funding
  • Demographic and household characteristics (e.g. age, gender, ethnicity, car ownership)
  • Postcode data allows identification of spatial variables e.g. connectedness to LTNs

Ensuring complete and representative data

Key information (2)

  • Generally, non-response in key variables was under 10%. We dealt with missing data in 2 stages:

 

 

 

  • Unrepresentativeness in the wave 1 sample was dealt by creating weights based on demography and LTN awareness

  1. Matched baseline to wave1 and wave 1 to baseline

  2. Imputation based on individual and area-level sociodemographic characteristics

Before matching missing with completed baseline responses After matching missing with completed baseline responses
Age 9.3% 0.2%
Household Income 26.3% 14.1%
Long Term Illness 9.2% 0%

Key research questions

  • Have there been any notable changes in past-week walking, cycling, and car use between baseline and wave 1? Are there any differences amongst demographic groups?

  • Can we find any notable changes in local area perceptions between baseline and wave 1?

  • To what extent do local environment perceptions and attitudes to TfL support for travel modes differ from the baseline?

Section 1: Comparing baseline (2021) with wave 1 (2022)

  • To what extent is there an association between living in a more/less connected area to an LTN and travel behaviour at baseline and at wave 1?

  • Can we find evidence that being more connected to an LTN at baseline is associated with a change in travel behaviour between baseline and wave 1?

  • To what extent is there evidence of new LTNs impacting upon past-week travel behaviour?

  • To what extent does moving into a more/less LTN-connected area lead to any travel behaviour change?

Section 2: The impacts of LTNs on travel behaviour

Methodology: LTN Connectedness

24.1% Baseline LTNs

6.2% Historic LTNs

0% New LTNs

Methodology: Connectedness classification

Baseline LTNs (March 2020-June 2021) New LTNs
​​​​​​​(July 2021-June 2022)		 Wave 1 LTNs (Baseline plus New LTNs) Historic LTNs
​​​​​​​(pre-March 2020)
Cut-off N= 4322 N= 4322 N= 4322 N= 4322
<5% of roads within LTN 3288 4292 3285 2353
5-15% of roads within LTN 393 30 379 1391
15-30% of roads within LTN 369 0 385 478
30-45% of roads within LTN 178 0 178 91
45-60% of roads within LTN 70 0 71 9
>60% of roads within LTN 24 0 24 0
Connectedness to LTN classification Cut-off Respondent count (wave 1 LTNs) Respondent count (baseline LTNs)
Low <15% of roads in LTNs 3664 3681
Medium 15-30% of roads in LTNs 385 369
High >30% of roads in LTNs 273 272
Connectedness to LTN classification Cut-off Respondent count (new LTNs)
Low 0% of roads within new LTN 4224
High >0% of roads within new LTN 96

The classification used in analysis of Wave 1 and Baseline LTNs

The classification used in analysis of New LTNs

The wave 1 sample by percentage of roads within LTN by LTN type

Comparing Baseline with Wave 1:

Past week cycling, walking, car use

  • most of the wave 1 and baseline sample does little cycling 
  • cycling behaviour largely remained largely unchanged
  • only increased among Black or Black British and Mixed respondents

RQ1: Is there a detectable change in past-week travel behaviour between baseline and wave 1?

 

RQ2: Is there a detectable change in demographic predictors of past-week travel behaviour between baseline and wave 1?

  • we can see a small increase in past-week walking
  • 5 percentage point decrease in respondents who reported to have done no walking 
  • we can see a small decrease in past-week car use
  • increase from 37% to 41% in respondents who did not use a car and those using a car over 240 minutes reduced by 7%
No
evidence		 Limited
evidence		 Some evidence Clear evidence
Change in minutes
cycled		 x
Change in minutes walked increase
Change in minutes of
car use		 decrease
 

RQ1: Is there a detectable change in past-week travel behaviour between baseline and wave 1?

RQ2: Is there a detectable change in demographic predictors of past-week travel behaviour between baseline and wave 1?

Change in cycling among Black or Black British & Mixed respondents increase
Change in other predictors x

Comparing Baseline with Wave 1:

Perceptions of local environment

RQ1: Is there a detectable change in perceptions of the local environment between baseline and wave 1.

No
evidence		 Limited
evidence		 Some evidence Clear evidence
Change in perceptions overall x
Change in local area is pleasant for walking increase
Change in local area has enough place to rest and shelter increase
 

RQ1: Is there a detectable change in perceptions of the local environment between baseline and wave 1.

5.1 point increase in respondents agreeing

Comparing Baseline with Wave 1:

Attitudes towards TfL & Local Authority funding

RQ1: Is there a detectable change in attitudes towards TfL and local authority funding for different transport modes between baseline and wave 1.

No
evidence		 Limited
evidence		 Some evidence Clear evidence
Change in desire for funding for cycling increase
Change in desire for funding for buses increase
Change in desire for funding for cars & vans decrease
 
Change in desire for funding for walking & running, Tube & DLR, Train, Tram, Overground increase

RQ1: Is there a detectable change in attitudes towards TfL and local authority funding between baseline and wave 1.

The impacts of LTNs on travel behaviour:

Connectedness to Wave 1 and Baseline LTNs

RQ1: To what extent is there an association between living in a more/less connected area to LTNs and travel behaviour at baseline and at wave 1?

No
evidence		 Limited
evidence		 Some evidence Clear evidence
Association with number of minutes cycled higher
Association with likelihood of ANY cycling higher
Association with number of minutes walked x
 
Association with likelihood of ANY walking x
Association with number of minutes of car use lower
Association with likelihood of ANY car use lower

RQ1: To what extent is there an association between living in an an area connected to LTNs and travel behaviour at baseline and at wave 1?

  • for every 10% increase in the proportion of roads within 15 minutes that fall within an LTN, there is a significant mean 7.8 minutes per week increase in cycling
  • living in an area of High LTN connectedness is associated with a mean 33.85 minutes per week increase in cycling compared to being in an area with zero LTN connectedness
  • direction of all relationships is as we would expect but car use and walking are not statistically significant

RQ2: To what extent can we find evidence that being in an LTN at baseline is associated with a change in travel behaviour between baseline and wave 1?

The impacts of LTNs on travel behaviour:

Connectedness to Wave 1 and Baseline LTNs

  • there is no evidence that being more connected to an LTN area at baseline is associated with any change in time spent walking, cycling, or using a car between baseline and wave 1
No
evidence		 Limited
evidence		 Clear
evidence
Change in minutes
cycled		 x
Change in minutes walked x
Change in minutes of
car use		 x

RQ2: To what extent can we find evidence that being in an LTN at baseline is associated with a change in travel behaviour between baseline and wave 1?

The impacts of LTNs on travel behaviour:

Connectedness to New LTNs

Cycling:

There is no evidence that being more connected to a new LTN is associated with more cycling or any cycling

Car use

Compared to those with no connectedness, a lower proportion of respondents with High connectedness reported car use over 120 minutes and higher proportion reported not driving at all in the past week

Walking:

A lower proportion of respondents living with High connectedness to new LTNs reported no past week walking.

RQ1: To what extent is there an association between living in an an area connected to new LTNs and travel behaviour at wave 1?

No
evidence		 Limited
evidence		 Clear evidence
Association with number of minutes cycled x
Association with likelihood of ANY cycling x
Association with number of minutes walked x
Association with likelihood of ANY walking higher
Association with number of minutes of car use lower
Association with likelihood of ANY car use lower

RQ1: To what extent is there an association between living in an an area connected to new LTNs and travel behaviour at wave 1?

The impacts of LTNs on travel behaviour:

Connectedness to New LTNs

RQ1: To what extent can an we find evidence that becoming connected to a new LTN is associated with a change in travel behaviour between baseline and wave 1?

  • conducted both a difference-in-differences and fixed effects analysis
  • in both models, there are no discernible statistically significant effects associated with the new LTNs yet
  • as the number of new LTNs increases, we anticipate greater statistical power to determine possible effects on behaviour change
No
evidence		 Limited
evidence		 Clear
evidence
Change in minutes
cycled		 x
Change in minutes walked x
Change in minutes of
car use		 x

RQ2: To what extent can we find evidence that becoming connected to a new LTN is associated with a change in travel behaviour between baseline and wave 1?

The impacts of LTNs on travel behaviour:

Moving into areas with more/less connectedness

Walking

  • Those who moved to an area with a higher proportion of roads within an LTN, there is a small increase in mean past week walking (from 241.4 to 271.2 minutes).
  • Those who moved to an area with a lower proportion of roads within an LTN, there is a small decrease in the mean and median past week walking

RQ1: To what extent is there an association between moving into a more/less LTN-connected area and travel behaviour change?

Weighted Mean Standard deviation Weighted Median Weighted Mean Standard deviation Weighted Median
No change 245.4 343.1 145 195.1 225.5 132.5
Higher proportion of LTN roads 241.4 272.2 162.5 271.2 297.1 165.5
Lower proportion of LTN roads 236.1 221.4 175 218.4 206.4 168.3

Baseline
(pre move)		 Wave 1
(post move)
No
evidence		 Limited
evidence		 Clear evidence
Association with number of minutes cycled x
Association with number of minutes walked
 		 higher/lower
Association with number of minutes of car use x

RQ1: To what extent is there an association between moving into a more/less LTN-connected area and travel behaviour change?

The impacts of LTNs on travel behaviour:

Moving into areas with more/less connectedness

RQ1: To what extent can we find evidence that becoming more connected to LTNs after moving is associated with a change in travel behaviour between baseline and wave 1?

  • we cannot yet identify any statistically significant past-week travel effects for movers of being in an area more or less connected to existing LTNs
  • as more respondents move and more LTNs are implemented, we may have more statistical power to assess this in future waves
No
evidence		 Limited
evidence		 Clear evidence
Association with number of minutes cycled x
Association with number of minutes walked x
 
Association with number of minutes of car use x

RQ2: To what extent can we find evidence that becoming more connected to LTNs after moving is associated with a change in travel behaviour between baseline and wave 1?

Key conclusions

  1. Travel behaviour and attitudes have changed more than we anticipated
  2. People that are more connected to LTNs continue to cycle more and use a car less…

  3. …But we cannot show any clear effects of LTNs on behaviour change

  4. No evidence does not necessarily mean no effects

The case for and against future waves of Travel and Places (1)

The case for and against future waves of Travel and Places (2)