The impact of real-time information on passengers’ value of bus waiting time
Open accessAbstract
In the past few years, technological changes have made it possible for bus passengers to learn about the expected arrival time when waiting for buses or even before arrival at the stop. In London, live bus arrival information is provided by Countdown signs at over 2500 bus stops. The bus arrival information is also available through Smartphone or tablet apps, the internet and SMS. Though these services are widely used, their impacts on passengers’ perceived waiting time and passengers’ behaviour remain unclear. The objective of this study is to assess how having real-time bus arrival information impacts on both passenger behaviour and how they value their waiting time at the bus stop. To address these research questions, this study is based on a survey containing a stated preference discrete choice experiment, which is supplemented with some qualitative research questions. 1690 completed interviews were achieved. Respondents were either recruited from bus users at bus stops who were invited to participate in an online or telephone survey, or from bus users supplied by TfL from their Oyster user database who were directed to the online survey. Discrete choice models were estimated and the resulting model coefficients quantify how different forms of real-time information impact on respondents’ value of bus waiting time and how the values varied by different sub-groups of population. This study provides up-to-date, London-specific waiting time relative values (multiplier to the in-vehicle time) and provides critical implications in current appraisal of schemes on public transport services investment.
AI snapshot
Written from the paper’s full text.
TL;DR
This study uses a stated preference discrete choice experiment with 1,690 London bus users to assess how real-time bus arrival information affects the value of bus waiting time relative to in-vehicle time. It finds an overall waiting-time multiplier of about 2.0, lower than the DfT WebTAG value of 2.5, with lower multipliers for passengers who check live bus information.
Key findings
- Overall, London bus travellers value waiting time changes about 2 times more than in-vehicle time changes; the paper recommends a bus wait time multiplier of 2.0 for appraisal, below the DfT WebTAG 2.5. Weighted multipliers by information checking were 2.2 for no access/haven’t checked, 1.7 for mobile, 1.0 for internet, and 0.8 for both mobile and internet.
- Checking live bus information before the journey reduced expected waiting time: on low-frequency routes from 6.7 minutes (no information) and 8.4 minutes (information but not checked) to 4.5 minutes (checked); on high-frequency routes from 5.9 and 5.4 minutes to 4.3 minutes. For those who checked information, the high/low frequency difference was very small at 0.2 minutes per trip.
- Passengers who checked live information were less sensitive to expected waiting time and more sensitive to in-vehicle time, lowering the waiting-time multiplier; internet users had a multiplier close to or smaller than 1.
- Other factors mattered: the in-vehicle time coefficient ratio for commuting/education/personal business versus other leisure purposes was 1.3; longer journeys had higher in-vehicle time valuations and lower waiting multipliers; ages 17–20 and low-frequency route users were less sensitive to waiting time.
Methods
- Mixed-mode survey: at-bus-stop CAPI recruitment at 21 London bus stops with follow-up online or telephone surveys, plus an online sample from TfL’s Oyster user database; 1,690 participants were analysed (269 pilot + 1,421 main survey).
- Stated preference discrete choice experiment: each participant saw 8 scenarios, each with two bus journey alternatives and a “not travel by bus” option; attributes were expected waiting time and in-vehicle time, tailored as percentage changes around current times using bands with thresholds of 5/15 minutes for waiting and 10/30 minutes for in-vehicle time.
- Analysis used multinomial logit discrete choice models with scale parameters by recruitment method and journey distance band, bootstrap resampling for repeated observations, and sample enumeration with weighting to the TfL bus user survey to calculate waiting-time multipliers.
Limitations
- The paper states that no monetary term or new policy/infrastructure was included in the choice set, so it judges hypothetical bias as relatively low; countdown information impact became non-significant after bootstrapping and was removed from the final combined model; SMS users were removed due to very small sample size; night bus journeys (n=49) and employer business trips (n=45) were removed.
- The paper states the sample under-sampled the 17–20 age group (7% vs 10% in the bus travel survey), and weights were applied to the value of waiting time calculation.
- Evident from the text: the study relies on stated choices and self-reported expected/actual wait times rather than observed behavioural data; the low-frequency “information but not checked” group had longer expected waiting time (8.4 minutes) than the no-information group (6.7 minutes), with significance only at the 90% confidence interval.
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
0 of 25 cited sources checked · 0 supported · 0 flagged
- Findingown data
Bus users who did not use live bus arrival time information had longer expected wait times than those who did: 6 minutes on average compared to 5.1 minutes.
“Bus users who did not use live bus arrival time information had longer expected wait times than those did: 6 minutes on average compared to 5.1 minutes.”
Quote in the paperSubstantial claim - Findingown data
Bus users on low frequency routes have a slightly longer expected waiting time compared to those on high frequency routes.
“Bus users on low frequencyfoot_1 routes have a slightly longer expected waiting time compared to those on the high frequency routes, as would be expected without taking account of information provision.”
Quote in the paperNot a substantial claim - Findingown data
For passengers who have no access to live information, expected waiting times are 5.9 and 6.7 minutes per trip for high frequency and low frequency services, respectively.
“For passengers who have no access to the live information, their expected waiting times are 5.9 and 6.7 minutes per trip, for high frequency and low frequency services, respectively.”
Quote in the paperNot a substantial claim - Findingown data
The expected wait time (mean wait of 5.5 minutes) was slightly shorter than the actual wait time (6.1 minutes).
“The expected wait time (mean wait of 5.5 minutes) was slightly shorter than the actual wait time (6.1 minutes).”
Quote in the paperNot a substantial claim - Findingreasoning
Compared to participants who made shorter journeys (less than 10 minutes), those who made longer journeys value in-vehicle time more negatively per minute.
“Compared to participants who made journeys with a shorter journey length (less than 10 minutes in this study), those who made journeys with a longer length are found to value in-vehicle time more negatively (per minute).”
- Also stated in: § 6.1. Discussion of the key findings · plain
“With regard to journey length, passengers on longer journeys attached a higher value to in-vehicle travel time savings than those making shorter journeys.”
Quote in the paperSubstantial claim - Also stated in: § 6.1. Discussion of the key findings · plain
- Findingown data
Compared to commuting, personal business, and education purposes, those travelling for other leisure purposes value in-vehicle journey time less negatively.
“Compared to participants who are commuting, or travelling for personal business and education 7 journey purposes, those who travelled for other leisure purposes are found to value in-vehicle journey time less negatively.”
Quote in the paperNot a substantial claim - Findingown data
Compared to online survey respondents, responses from those recruited at the bus stop and undertaking telephone interviews show higher level of noise (smaller scale parameter).
“In terms of scale parameter, compared to the responses of those completing the online survey, the responses from those recruited at the bus stop and undertaking telephone interviews show higher level of noise (smaller scale parameter).”
Quote in the paperNot a substantial claim - Findingown data
After controlling for socio-demographic factors and journey characteristics, live bus information has a significant impact on bus users' value of expected waiting time.
“Third, after controlling the influence of socio-demographic factors and journey characteristics, live bus information is found to have a significant impact on bus users' value of expected waiting time.”
Quote in the paperSubstantial claim - Findingown data
This study's overall waiting time multiplier is smaller than the currently recommended DfT WebTAG value of 2.5 for commuting and other purposes.
“This overall multiplier is smaller than the currently recommended value by DfT (WebTAG of 2.5 for commuting and other purposes).”
- Also stated in: § 6.2. Policy implications and recommendations · plain
“Second, this new survey shows that the London-specific waiting time multiplier is lower than the current DfT WebTAG recommended value.”
Quote in the paperSubstantial claim0 of 1 sources checked- cites WebTAG
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
- Also stated in: § 6.2. Policy implications and recommendations · plain
- Findingown data
The waiting time-in-vehicle time multiplier calculated from the choice models is generally in line with or in the same order of magnitude as DfT WebTAG guidance and other UK public transport value of time meta-analysis results (Wardman 2014).
“Moreover, the waiting timein-vehicle time multiplier calculated from the choice models are generally in line with or in the same order of magnitude as DfT WebTAG guidance and other UK public transport value of time metaanalysis results (Wardman 2014).”
Quote in the paperNot a substantial claim0 of 1 sources checked- cites Wardman 2014
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
- Findingown data
Bus frequency has an impact on participants' value of waiting time.
“Bus frequency was found to have an impact on the participants' value of waiting time.”
Quote in the paperSubstantial claim - Findingown data
Age has a significant impact on the waiting time attribute.
“We also found some significant impact of age on the waiting time attribute.”
- Also stated in: § 6.1. Discussion of the key findings · plain
“In addition, age was found to have an impact on the valuation of waiting time.”
Quote in the paperNot a substantial claim - Also stated in: § 6.1. Discussion of the key findings · plain
- Findingown data
Compared to participants who do not have access to live bus information and those who have access but did not check before their reference journey, those who checked their bus arrival information by mobile phone and internet have differences in their valuation of waiting time and journey time.
“Compared to the participants who do not have access to live bus information and those who have access but did not check before their reference journey, those who did check their bus arrival information by mobile phone and internet were found to have differences in their valuation of waiting time and journey time.”
Quote in the paperSubstantial claim - Findingown data
Passengers who checked live bus information before their journey reported spending less time waiting.
“First, as can be seen in Table 11, it was observed that those passengers that checked live bus information before their journey reported spending less time waiting.”
Quote in the paperSubstantial claim - Findingown data
Participants' socio-demographic features and their current journey characteristics have an impact on their value of the waiting time and in-vehicle time.
“Second, we found participants' socio-demographic features and their current journey characteristics have an impact on their value of the waiting time and in-vehicle time.”
Quote in the paperSubstantial claim - Findingown data
People aged between 17 and 20 have less negative values for waiting time per minute.
“People who are aged between 17 and 20 are found to have less negative values for waiting time (per minute).”
- Also stated in: § 6.1. Discussion of the key findings · plain
“Participants aged between 17 and 20 were found to be less sensitive to changes in their waiting time, after controlling the journey characteristics factors and differences in accessing bus information.”
Quote in the paperNot a substantial claim - Also stated in: § 6.1. Discussion of the key findings · plain
- Findingreasoning
Compared to participants who travelled on high frequency routes and those for whom bus frequency information is not available, those who travelled on low frequency routes have less negative valuations of waiting time.
“We found that compared to participants who travelled on the high frequency routes and those for whom bus frequency information is not available, those who travelled on the low frequency routes have less negative valuations of waiting time.”
- Also stated in: § 6.1. Discussion of the key findings · plain
“Passengers who travelled on the low frequency routes were found to be less sensitive to changes in their expected waiting time.”
Quote in the paperSubstantial claim - Also stated in: § 6.1. Discussion of the key findings · plain
- Findingown data
The ratio of waiting and journey time (multipliers) differ according to the means used of checking live bus information.
“Second, the modelling outputs indicate that the ratio of waiting and journey time (multipliers) differ according to the means used of checking live bus information.”
Quote in the paperSubstantial claim - Findingown data
On average, current London bus travellers value changes in their waiting time 2 times more than changes in their in-vehicle time.
“Overall, we find that on average current London bus travellers value changes in their waiting time 2 times more than changes in their in-vehicle time.”
Quote in the paperSubstantial claim - Findingown data
For passengers who have checked their bus information before their trip, there is no significant difference in the expected waiting time between passengers on high and low frequency routes.
“It is noticeable that for passengers who have checked their bus information before their trip, there is no significant difference in the expected waiting time between the passengers on the high and low frequency routes.”
Quote in the paperSubstantial claim - Findingown data
The magnitude of the impact of live bus information varies by different means of checking the information.
“The magnitude varies by different means of checking the information.”
Quote in the paperSubstantial claim - Findingown data
The multiplier for people who checked using Internet is lower than those who used Mobile phone.
“We observed that the multiplier for people who checked using Internet is lower than those who used Mobile phone.”
Quote in the paperNot a substantial claim - Findingreasoning
Participants who used both approaches were found to have the lowest multiplier.
“Participants who used both approaches were found to have the lowest multiplier.”
Quote in the paperNot a substantial claim - Findingreasoning
The multiplier for participants who used Internet to access live information is close to or smaller than 1.
“We notice that the multiplier for participants who used Internet to access the live information is close to or small than 1.”
Quote in the paperSubstantial claim - Premisecitation
LBAI has been found to reduce passengers' perception of the length of wait times at bus stops.
“Decreased expected waiting time: LBAI has been found to reduce passengers' perception of the length of wait times at bus stops.”
Quote in the paperSubstantial claim0 of 3 sources checked- cites Kronborg et al. 2002
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
- cites Schweiger, 2003
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
- cites Dziekan and Kottenhoff 2007
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
- Premisecitation
When passengers have access to LBAI, they are able to plan their departure to minimise their waiting time at bus stop.
“Moreover, when passengers have access to LBAI, they are able to plan their departure to minimise their waiting time at bus stop.”
Quote in the paperNot a substantial claim0 of 2 sources checked- cites Watkins et al., 2011 source
- cites Cats et al., 2011
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
- Premisecitation
In Seattle, Washington, actual waiting time for passengers with mobile LBAI were almost two minutes less than those of non-users.
“In Seattle, Washington, a recent study found that the actual waiting time for passengers with mobile LBAI were almost two minutes less than those of non-users”
Quote in the paperNot a substantial claim0 of 1 sources checked- cites Watkins et al., 2011 source
- Study design and conductown data
89% of the participants thought that the SP choices were easy or moderately easy to make.
“89% of the participants thought that the SP choices were easy or moderately easy to make.”
Quote in the paperNot a substantial claim - Study design and conductown data
The preferred model structure contained two levels of scales: firstly by recruiting method and second by different journey distance band.
“After several iterations, the preferred model structure contained two levels of scales: firstly by recruiting method and second by different journey distance band.”
Quote in the paperNot a substantial claim - Study design and conductown data
Over 92% of the participants stated that they could understand the SP choices.
“Over 92% of the participants stated that they could understand the SP choices.”
Quote in the paperNot a substantial claim - Study design and conductcitation
The application of the bootstrap procedure ensures that the standard errors and t-ratios produced by the models are a realistic statement of the true errors of the model parameters.
“The application of the 'bootstrap' procedure ensures that the standard errors and t-ratios produced by the models are a realistic statement of the true errors of the model parameters.”
Quote in the paperSubstantial claim0 of 1 sources checked- cites Efron and Tibshirani, 1993 source
- Rests on an assumption: Naive models that do not account for correlated responses from the same individual underestimate standard errors and inflate statistical significance.
- Interpretationcitation
The finding that longer journey length is associated with more negative in-vehicle time valuation is consistent with previous meta-analysis (Wardman, 2001).
“This finding is consistent with previous meta-analysis of public transport value of in-vehicle time evidence (Wardman, 2001), which reflect the increasing discomfort of long distance journeys and perhaps higher opportunity cost of time spent travelling for longer periods at a time.”
Quote in the paperNot a substantial claim0 of 1 sources checked- cites Wardman, 2001 source
- Premisecitation
In Seattle, Washington, perceived wait times of mobile LBAI users were approximately 30 per cent less than those of non-users.
“and the perceived wait times of users were approximately 30 per cent less than those of the non-users (Watkins et al., 2011).”
Quote in the paperNot a substantial claim0 of 1 sources checked- cites Watkins et al., 2011 source
- Interpretationreasoning
Longer journey lengths result in a lower waiting time multiplier.
“This results in a lower waiting time multiplier for journeys of longer lengths.”
Quote in the paperSubstantial claim - Interpretationown data
Checking live bus information helps bus users to reduce their expected waiting time, particularly for low frequency services.
“This implies that checking live bus information helps bus users to reduce their expected waiting time, particularly for low frequency services.”
Quote in the paperSubstantial claim - Interpretationown data
Participants are less likely to select the 'Not travel by bus' option compared to the options of making bus journeys.
“This indicates that participants are less likely to select this option compared to the options of making bus journeys.”
Quote in the paperNot a substantial claim - Premisecitation
Ferris et al. (2010) and Gooze et al. (2013) found that approximately one third of passengers reported an increase in the number of non-work/school trips per week made on transit because of the introduction of the LBAI.
“Ferris et al. (2010) and Gooze et al. (2013) also found that approximately one third of passengers reported an increase in the number of non-work/school trips per week made on transit because of the introduction of the LBAI.”
Quote in the paperNot a substantial claim0 of 2 sources checked- cites Ferris et al. (2010)
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
- cites Gooze et al. (2013) source
- Study design and conductown data
The quality of the choice data appears to be high.
“First, the quality of the choice data appears to be high.”
Quote in the paperSubstantial claim - Interpretationreasoning
Higher time pressures on the journey tend to result in higher values of time.
“demonstrating that higher time pressures on the journey tend to result in higher values of time.”
Quote in the paperSubstantial claim - Interpretationreasoning
Using live bus information helps passengers manage their expected waiting time.
“This implies that using the live bus information helps passengers manage their expected waiting time.”
Quote in the paperSubstantial claim - Interpretationreasoning
These results all resulted in a lower ratio of waiting time to journey time (multiplier) for commuting and longer journeys.
“These all resulted in a lower ratio of waiting time to journey time (multiplier) for commuting and longer journeys.”
Quote in the paperSubstantial claim - Study design and conductcitation
The likelihood of having hypothetical bias in this study is relatively low.
“Therefore we judge the likelihood of having hypothetic bias as relatively low in this study.”
Quote in the paperNot a substantial claim0 of 1 sources checked- cites Wardman 2001 source
- Interpretationreasoning
Over time as the use of live bus information technology increases, the average multiplier is expected to be reduced further.
“In addition, live bus information acts to reduce the multiplier and therefore over time as the use of this technology increases we would expect the average multiplier to be reduced further.”
Quote in the paperNot a substantial claim - Interpretationreasoning
Bus users who checked live bus information prior to making their journey had a lower multiplier.
“They therefore had a lower multiplier.”
Quote in the paperSubstantial claim - Interpretationreasoning
For a well pre-planned bus journey, passengers valued their waiting time the same or even less than their journey in-vehicle time.
“This implies for a well pre-planned bus journey, passengers valued their waiting time the same or even less than their journey in-vehicle time.”
Quote in the paperSubstantial claim
- May combine two claims. 25
- The quote only partly states this claim. 1
- Classified as probably not a substantial claim. 21
May combine two claims.
Findingown dataThose who checked live bus arrival time information before getting to the bus stop reported both shorter expected and actual wait times than those who did not.
“Those who checked live bus arrival time information before getting to the bus stop reported both shorter expected and actual wait times than those who did not.”
Quote in the paperSubstantial claimMay combine two claims.
Findingown dataCompared to short and medium journeys, long journey group responses show a higher level of noise as the scale parameter for the long journey group is significantly lower than that of the reference group.
“Compared to the responses from those making short and medium journeys, the responses from the long journeys group show a higher level of noise as the scale parameter for the long journey group is significantly lower than that of the reference group.”
Quote in the paperNot a substantial claimMay combine two claims.
Findingown dataFor participants who checked bus information before the bus trip, expected waiting time is less compared to the first two groups and the difference in waiting time between high and low frequency routes is very small at 0.2 minutes per trip.
“It is noticeable that for participants who have checked the bus information before the bus trip, their expected waiting time is less compared to the first two groups and the difference in the waiting time between the high and low frequency routes is very small at 0.2 minutes per trip.”
Quote in the paperSubstantial claimMay combine two claims.
Findingown dataFor low frequency services, the average expected waiting time for passengers without information (6.7 minutes) is shorter than that of passengers who have information but did not check prior to their journey (8.4 minutes), although the impact is only significant at the 90% confidence interval.
“Interestingly we found that for low frequency services, the average expected waiting time for passengers without information (6.7 minutes) is shorter than that of the passengers who have information but did not check prior to their journey (8.4 minutes), although the impact is only significant at the 90% confidence interval.”
Quote in the paperNot a substantial claimMay combine two claims.
Findingown dataSlightly over half (53%) of bus users actually waited about as long as they expected to wait, 27% waited longer and 20% waited shorter than expected.
“A matrix of expected and actual wait times (by time ranges) shows that slightly over half (53%) actually waited about as long as they expected to wait, 27% waited longer and 20% waited shorter than expected.”
Quote in the paperNot a substantial claimMay combine two claims.
Findingown dataJourney length has an impact on the in-vehicle time coefficient and was significant in the current study.
“For instance, journey length was found to have impact on the in-vehicle time coefficient from previous evidence and was also found to be significant in the current study; therefore it was retained in the combined model.”
Quote in the paperSubstantial claimMay combine two claims.
Findingown dataJourney purpose is correlated with time of travel and travel frequency.
“we found that journey purpose was correlated with time of travel and travel frequency (i.e.travellers for commuting purposes normally travel during the AM peak /PM peak and are quite frequent travellers compared to other purposes).”
Quote in the paperNot a substantial claimMay combine two claims. The quote only partly states this claim.
Findingown dataLive traveller information acts to lower the average waiting time values and reduce the multiplier.
“However, it is up to date, London-specific, and takes account of the emerging impacts of live traveller information which can be observed to act to lower the average values.”
- Also stated in: § 6.2. Policy implications and recommendations · plain
“In addition, live bus information acts to reduce the multiplier and therefore over time as the use of this technology increases we would expect the average multiplier to be reduced further.”
Quote in the paperSubstantial claim- Also stated in: § 6.2. Policy implications and recommendations · plain
May combine two claims.
Findingown dataAge and journey purposes were found to have a significant influence within the choice models.
“The weights were applied in two dimensions: age and journey purposes as these two factors were found to have a significant influence within the choice models.”
Quote in the paperSubstantial claimMay combine two claims.
Findingown dataLive bus information impacts both on participants' values of expected waiting time and in-vehicle time.
“We found that live bus information 8 impacts both on the participants' values of expected waiting time and invehicle time.”
Quote in the paperSubstantial claimMay combine two claims.
Findingown dataThe average expected waiting time is 6.7 minutes with no information, 8.4 minutes with information but not checked, and 4.5 minutes per trip for passengers on low frequency routes; and 5.9, 5.4, and 4.3 minutes per trip for those on high frequency routes.
“On average, the reduction of expected waiting time is from a range of 6.7 minutes (with no information) to 8.4 minutes (with information but not checked) down to 4.5 minutes per trip for passengers who travelled on the low frequency routes, and the range of 5.9 to 5.4 down to 4.3 minutes per trip for those on the high frequency routes.”
Quote in the paperSubstantial claimMay combine two claims.
Findingown dataThese values vary by different journey purposes and journey length.
“More specifically, these values vary by different journey purposes and journey length.”
Quote in the paperSubstantial claimMay combine two claims.
Findingown dataNearly all participants have a good understanding and stated that it was easy or moderately easy to make the choices in the SP exercises.
“The responses to diagnostic questions included within the survey suggest that nearly all the participants have a good understanding and stated that it was easy or moderately easy to make the choices in the SP exercises.”
Quote in the paperSubstantial claimMay combine two claims.
FindingreasoningThe monetary value of waiting time calculated using VoTT 2015 values is £20.50/hr for commuting and £12.11/hr for others.
“Monetary value of waiting time (using VoTT 2015) (£/hr) 20.50 12.11”
Quote in the paperSubstantial claim0 of 1 sources checked- cites Value of Travel Time study (2015)
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
May combine two claims.
FindingreasoningBus users who checked live bus information prior to making their journey were less sensitive to changes in expected waiting time and more sensitive to changes in in-vehicle time.
“Bus users that stated they had checked live bus information prior to making their journey both were less sensitive to changes in the expected waiting time and more sensitive to changes in their in-vehicle time.”
Quote in the paperSubstantial claimProbably not a substantial claim: may not pass S3 (load-bearing).
Study design and conductreasoningThe analysis included data from 1,690 participants (269 from pilot survey and 1,421 from the main survey).
“As a result, the analysis included data from 1,690 participants (269 from pilot survey and 1,421 from the main survey).”
Quote in the paperNot a substantial claimProbably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.
Study design and conductnone givenThe levels for the attributes were balanced and distributed approximately equally between values better and worse than the existing values.
“The levels for the attributes were balanced and distributed approximately equally between values better and worse than the existing values.”
Quote in the paperNot a substantial claimProbably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.
Study design and conductown dataThe SP choice design and format was tested through a pilot survey, which showed that the SP experiment worked as intended, so the design was not amended for the main survey and the pilot data were included in the main analysis.
“The SP choice design and format was tested through a pilot survey.This showed that the SP experiment worked as intended and so the design was not amended for the main survey and the data were included in the main analysis.”
Quote in the paperNot a substantial claimProbably not a substantial claim: may not pass S3 (load-bearing).
PremisecitationLBAI facilitates route choice changes that would minimise travel time.
“LBAI also facilitates route choice changes that would minimise travel time (Cats et al., 2011).”
Quote in the paperNot a substantial claim0 of 1 sources checked- cites Cats et al., 2011
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
Probably not a substantial claim: may not pass S3 (load-bearing).
Study design and conductnone givenThe SP design was able to cover a wide range of the waiting time to in-vehicle time multiplier.
“The design was able to cover a wide range of the waiting time to in-vehicle time multiplier.”
Quote in the paperNot a substantial claimProbably not a substantial claim: may not pass S3 (load-bearing).
Study design and conductreasoningThere were no dominant alternatives in the SP design (no cases where one alternative was preferable for both waiting and journey time attributes).
“There were no dominant alternatives (i.e.we presented no cases where one alternative was preferable for both waiting and journey time attributes).”
Quote in the paperNot a substantial claimClassified as probably not a substantial claim.
Study design and conductreasoningThe model coefficients reflect the results after bootstrapping to take account of repeated observations being collected from a single individual.
“The model coefficients reflect the results after bootstrapping to take account of repeated observations being collected from a single individual.”
Quote in the paperSubstantial claim- Rests on an assumption: Naive models that do not account for correlated responses from the same individual underestimate standard errors and inflate statistical significance.
May combine two claims.
Interpretationown dataAll else equal, participants prefer options with lower waiting and lower journey times.
“Both time coefficients are strongly estimated with a correct sign (negative), indicating that all the other conditions being equal, participants prefer options with lower waiting and lower journey times.”
Quote in the paperNot a substantial claimProbably not a substantial claim: may not pass S3 (load-bearing).
InterpretationcitationThe finding for 'Other' purposes is in line with the meta-analysis by Wardman (2014).
“Again, this finding is in line with the meta-analysis by Wardman (2014).”
Quote in the paperNot a substantial claim0 of 1 sources checked- cites Wardman (2014)
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
Classified as probably not a substantial claim. May combine two claims.
Study design and conductown dataA fractional factorial orthogonal plan was used to specify the combinations of attributes and levels, giving 16 choice scenarios, with orthogonal blocking to ensure each set of scenarios was balanced across levels.
“A fractional factorial orthogonal plan was used to specify the combinations of attributes and levels to be presented in the choice experiment scenarios.This gave a total of 16 choice scenarios in total.Each participant was presented with 8 scenarios, and these were assigned using orthogonal blocking to ensure that each set of scenarios was balanced across levels.”
Quote in the paperSubstantial claimProbably not a substantial claim: may not pass S3 (load-bearing).
Study design and conductown dataCompared to the sample composition of TfL bus user survey, our study under-sampled the 17-20 years subgroup: 10% of the bus travel survey sample was aged 17-20 whilst in our study there were 7%.
“Compared to the sample composition of TfL bus user survey, our study under-sampled the 17 -20 years subgroup.In the bus travel survey, 10% of the sample was aged 17 -20 whilst in our study, there were 7%.”
Quote in the paperNot a substantial claimProbably not a substantial claim: may not pass S3 (load-bearing).
Interpretationown dataThe scale parameter finding is in line with the perceived difficulty of choice making from the diagnostic questions.
“This finding is in line with what we found about the perceived difficulty of choice making from the diagnostic questions.”
Quote in the paperNot a substantial claimProbably not a substantial claim: may not pass S3 (load-bearing).
InterpretationcitationThe pattern of results is in line with Wardman (2014, Table 4).
“The pattern of the results is in line with the study by Wardman (2014, Table 4).”
Quote in the paperNot a substantial claim0 of 1 sources checked- cites Wardman (2014, Table 4)
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
May combine two claims.
Study design and conductown dataThe majority of participants were able to understand the SP experiment and make the choices, with only a small percentage finding the choice experiments very difficult.
“However, the majority of participants were able to understand the SP experiment and make the choices, with only a small percentage finding the choice experiments”
Quote in the paperNot a substantial claimProbably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.
PremisecitationThe provision of LBAI can act to increase the level of passengers' perceived security and improve their overall satisfaction with bus service.
“Increased satisfaction with bus service: the provision of LBAI can act to increase the level of passengers' perceived security and improve their overall satisfaction with bus service.”
Quote in the paperNot a substantial claim0 of 3 sources checked- cites Wardman (2003)
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
- cites Schweiger (2003)
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
- cites Zhang et al. (2008) source
Probably not a substantial claim: may not pass S3 (load-bearing).
PremisecitationTang and Thakuriah (2012) found a modest increase in overall route-level ridership attributable to LBAI in Chicago: 126 rides per route per day, which is 1.8-2.2% of average route-level weekday bus ridership.
“Tang and Thakuriah (2012), through an empirical evaluation of Chicago bus ridership, found a”
Quote in the paperNot a substantial claim0 of 1 sources checked- cites Tang and Thakuriah (2012)
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
Classified as probably not a substantial claim.
Study design and conductreasoningScale parameters were incorporated into the choice models to take account of potential differences in error between different sources of data and experimental designs.
“To pool the data from different sources, scale parameters were incorporated into the choice models to take account of the potential differences in error between the different sources of data and experimental designs.”
Quote in the paperSubstantial claimProbably not a substantial claim: may not pass S3 (load-bearing).
Study design and conductnone givenWeights were applied to the calculation of the value of waiting time.
“Weights were applied to the calculation of the value of waiting time.”
Quote in the paperNot a substantial claimProbably not a substantial claim: may not pass S3 (load-bearing).
PremisecitationWardman (2003) reported that 46 percent of respondents felt safer at their bus stop if they knew when the bus was due to arrive.
“Wardman (2003) examined anxiety of waiting at a bus stop at night and reported that 46 percent of respondents felt safer at their bus stop if they knew when the bus was due to arrive.”
Quote in the paperNot a substantial claim0 of 1 sources checked- cites Wardman (2003)
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.
Study design and conductnone givenBus users were approached using a random 1 in n approach at bus stops, and sample quotas were set on main journey purpose, access mode to bus stop, age and gender.
“Bus users were approached using a random 1 in n approach at bus stops.We set sample quotas on main journey purpose, access mode to bus stop, age and gender.”
Quote in the paperNot a substantial claimMay combine two claims.
Interpretationown dataParticipants treated the SP experiment seriously and answered the choice questions in a rational way.
“In addition, examination of the responses to the background questions and the outputs from the separate models suggested that participants treated the SP experiment seriously and answered the choice questions in a rational way.”
Quote in the paperSubstantial claimProbably not a substantial claim: may not pass S3 (load-bearing).
InterpretationcitationThe finding that longer journeys have higher in-vehicle time value is in line with other evidence (Wardman 2014).
“This is in line with the other evidence (Wardman 2014).”
Quote in the paperNot a substantial claim0 of 1 sources checked- cites Wardman 2014
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.
InterpretationreasoningThese users will feel that they have good control over their waiting time, and possibly more so than their control over the in-vehicle time once they have boarded.
“This is to be anticipated as these users will feel that they have good control over their waiting time, and possibly more so than their control over the in-vehicle time once they have boarded.”
Quote in the paperNot a substantial claimClassified as probably not a substantial claim.
Study design and conductcitationThe recommended bus wait time multiplier of 2.0 is the weighted values using the sample composition from TfL bus user survey to better reflect the bus user population profile.
“This multiplier is the weighted values using the sample composition from TfL bus user survey to better reflect the bus user population profile.”
Quote in the paperSubstantial claim0 of 1 sources checked- cites TfL bus user survey
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
May combine two claims.
Interpretationown dataLive bus information has the ability to improve bus users' experience by providing information to them before they reach the bus stop and changing how long they have to wait, particularly for those travelling on low service frequency routes.
“First, live bus information has the ability to improve bus users' experience by providing information to them before they reach the bus stop and changing how long they have to wait, particularly for those travelling on low service frequency routes.”
Quote in the paperSubstantial claim
The paper relies on this without support.
There is a significant evidence gap in London on the impacts of LBAI on passengers' perceived waiting time and behaviour.
“There is a growing literature on exploring the impacts of LBAI on passengers' perceived waiting time and passengers' behaviour change, however there were significant evidence gaps in London.”
The authors state this as an assumption.
Participants chose the alternatives that provide them with the highest utility (the highest overall value to themselves).
“The estimation procedure assumes that participants chose the alternatives that provide them with the highest utility (the highest overall value to themselves).”
The authors state this as an assumption.
An error term in each utility function reflects unobservable factors in the individual's utility.
“An error term is included in each utility function to reflect unobservable factors in the individual's utility.”
The paper relies on this without support.
Naive models that do not account for correlated responses from the same individual underestimate standard errors and inflate statistical significance.
“It is known that naive models that do not take account of the fact that individuals provide a number of potentially correlated responses will underestimate the standard errors of the coefficient estimates, leading to inflated levels of statistical significance.”
The authors state this as an assumption.
The influence of various factors are included in the model using incremental effects (additive terms), so coefficients can be added to get a combined effect.
“It should be noted that the influence of the various factors are included in the model using incremental effects (additive terms).Therefore, the coefficients can be added to get a combined effect.”
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
Each contradiction between two passages listed here was confirmed by an independent check.
Conflicting percentages for respondents aged 17-20
MajorThe text says 7% of this study's sample was aged 17-20, while Table 5 reports 19%. This discrepancy matters because age was used to weight the waiting-time multipliers.
“In the bus travel survey, 10% of the sample was aged 17 -20 whilst in our study, there were 7%.”
“| | 17 -20 | 19% |”
Telephone-group constants are described as significant despite low t-ratios
MinorThe paper says the not-travel constants are significantly negative for every recruitment method, but the three reported telephone-recruitment t-ratios are −0.3, −0.1 and −0.3.
“The constants on the "Not travel by bus" terms are significantly negative for each journey length group by each recruitment method.”
“Not travel -short journey length telephone recruitment -9.795 -0.3Not travel -medium journey length telephone recruitment -16.854 -0.1Not travel -long journey length telephone recruitment -14.612 -0.3”
Cites
An Introduction to the Bootstrap
Efron, B., & Tibshirani, R. J. (1993) · doi:10.1007/978-1-4899-4541-9 · Cited by 21,230 · see citing works
Discrete Choice Methods with Simulation
Train, K. E. (2003) · Cambridge University Press eBooks · doi:10.1017/cbo9780511753930 · Cited by 3,950 · see citing works
Discrete choice analysis: Theory and application to travel demand
Prashker, J. N. (1988) · Transportation Research Part A General · doi:10.1016/0191-2607(88)90064-7 · Cited by 3,340 · see citing works
A Review of British Evidence on Time and Service Quality Valuations
Wardman, M. (2001) · Transportation Research Part E: Logistics and Transportation Review · doi:10.1016/s1366-5545(00)00012-0 · Cited by 254 · see citing works
Watkins, K. E., Ferris, B., Borning, A., Rutherford, G. S., & Layton, D. (2011) · Transportation Research Part A: Policy and Practice · doi:10.1016/j.tra.2011.06.010 · Cited by 190 · see citing works
Passenger travel time and path choice implications of real-time transit information
Hickman, M. D., & Wilson, N. H. M. (1995) · Transportation Research Part C Emerging Technologies · doi:10.1016/0968-090x(95)00007-6 · Cited by 141 · see citing works
Dynamic at-stop real-time information displays for public transport: effects on customers
Dziekan, K., & Kottenhoff, K. (2007) · Transportation Research Part A: Policy and Practice · doi:10.1016/j.tra.2006.11.006 · Cited by 123 · see citing works
Brakewood, C., Barbeau, S., & Watkins, K. E. (2014) · Transportation Research Part A Policy and Practice · doi:10.1016/j.tra.2014.09.003 · Cited by 102 · see citing works
Benefits of real-time transit information and impacts of data accuracy on rider experience
Gooze, A., Watkins, K. E., & Borning, A. (2013) · Transportation Research Record: Journal of the Transportation Research Board · doi:10.3141/2351-11 · Cited by 65 · see citing works
Politis, I., Papaioannou, P., Basbas, S., & Dimitriadis, N. (2010) · Research in Transportation Economics · doi:10.1016/j.retrec.2010.07.031 · Cited by 63 · see citing works
A Stated Preference Analysis of Real-Time Public Transit Stop Information
Caulfield, B. M., & O’Mahony, M. (2009) · Journal of Public Transportation · doi:10.5038/2375-0901.12.3.1 · Cited by 55 · see citing works
Examination of traveler responses to real-time information about bus arrivals using panel data
Zhang, F., Shen, Q., & Clifton, K. J. (2008) · Transportation Research Record: Journal of the Transportation Research Board · doi:10.3141/2082-13 · Cited by 55 · see citing works
Will Psychological Effects of Real-Time Transit Information Systems Lead to Ridership Gain?
Tang, L., & Thakuriah, P. V. (2011) · Transportation Research Record Journal of the Transportation Research Board · doi:10.3141/2216-08 · Cited by 40 · see citing works
Flagged first, then by citations
Research works printed but not linked: 2
- Bradley, M.A. and Daly, A.J. (1991) Estimation of Logit Choice Models using Mixed Stated Preference and Revealed Preference Information Presented to the 6th International Conference on Travel Behaviour, Québec; revised version presented to the Duke International Symposium on Choice Modelling and Behaviour, 1993. In Understanding Travel Behaviour in an Era of Change, edited by P. Stopher & M. Lee-Gosselin, 209–31. Oxford: Pergamon.
- Ferris, B., Watkins, K. and Borning, A. (2010) One Bus Away: Results from Providing Real-Time Arrival Information for Public Transit. Proceedings: CHI, 1807–1816
Cited by
Sorted by citations
Witowska, J., Schmidt, S., & Wittmann, M. (2020) · Acta Psychologica · doi:10.1016/j.actpsy.2020.103061 · Cited by 120 · see citing works
Sadrani, M., Tirachini, A., & Antoniou, C. (2022) · Transportation Research Part C Emerging Technologies · doi:10.1016/j.trc.2022.103793 · Cited by 46 · see citing works
de Castro, A. B. R., Sandoval, A. D. O., & Odamtten, G. (2022) · Journal of Transport Geography · doi:10.1016/j.jtrangeo.2022.103388 · Cited by 25 · see citing works
Romero, C., de Cáceres, A. M., Alonso, A., & Julio, R. (2021) · International Journal of Sustainable Transportation · doi:10.1080/15568318.2020.1871131 · Cited by 22 · see citing works
Comparison between different territorial policies to support intermodality of public transport.
Pazzini, M., Lantieri, C., Vignali, V., Simone, A., Dondi, G., Luppino, G., & Grasso, D. (2022) · Transportation research procedia · doi:10.1016/j.trpro.2021.12.010 · Cited by 20 · see citing works
Loría, L. E., Watson, V., Kiso, T., & Phimister, E. (2019) · Journal of Choice Modelling · doi:10.1016/j.jocm.2019.05.001 · Cited by 15 · see citing works
Huo, Y., Zhao, J., Li, X., & Guo, C. (2021) · Journal of Public Transportation · doi:10.5038/2375-0901.23.2.3 · Cited by 14 · see citing works
The impact of a passenger-safety-driven acceleration limit on the operation of a bus service
Karekla, X., Gkiotsalitis, K., & Tyler, N. (2020) · Accident Analysis & Prevention · doi:10.1016/j.aap.2020.105790 · Cited by 13 · see citing works
Bo, K., Teng, J., Zhang, C., & Han, D. (2021) · Journal of Advanced Transportation · doi:10.1155/2021/6622461 · Cited by 12 · see citing works
Pazzini, M., Lantieri, C., Vignali, V., Simone, A., Dondi, G., Luppino, G., & Grasso, D. (2020) · Advances in intelligent systems and computing · doi:10.1007/978-3-030-61075-3_7 · Cited by 11 · see citing works
Impact of Real-Time Information for Travellers: A Systematic Review
Akhla, A. A. N., Thong, C. L., Shibghatullah, A. S., Chit, S. M., Cherukuri, A. K., Chaw, L. Y., & Lee, C. Y. (2022) · Journal of Information & Knowledge Management · doi:10.1142/s0219649222500654 · Cited by 9 · see citing works
Designing limited-stop bus services for minimizing operator and user costs under crowding conditions
Sadrani, M., Jafarian-Moghaddam, A. R., Esfahani, M. A., & Rahimi, A. M. (2022) · Public Transport · doi:10.1007/s12469-022-00307-2 · Cited by 9 · see citing works
da Silva, C. F. A., de Andrade, M. O., Maia, M. L. A., dos Santos, A. M., & Portis, G. T. (2022) · GeoJournal · doi:10.1007/s10708-022-10595-7 · Cited by 8 · see citing works
Bogumil, V., & Vlasov, V. M. (2018) · Transportation research procedia · doi:10.1016/j.trpro.2018.12.044 · Cited by 8 · see citing works
Data-Driven Real-Time Denied Boarding Prediction in Urban Railway Systems
Tuncel, K. S., Koutsopoulos, H. N., & Ma, Z. (2023) · Transportation Research Record Journal of the Transportation Research Board · doi:10.1177/03611981231184237 · Cited by 7 · see citing works
Henríquez-Jara, B. I., Arriagada, J., Montenegro, K., Tirachini, A., & Munizaga, M. A. (2025) · Travel Behaviour and Society · doi:10.1016/j.tbs.2025.101024 · Cited by 7 · see citing works
Safiullin, R., & Arias, Z. P. (2024) · The Open Transportation Journal · doi:10.2174/0126671212320514240611100437 · Cited by 7 · see citing works
Would You Wait? Bus Choice Behavior Analysis Considering Various Incentives
Pan, L., Waygood, E. O. D., & Patterson, Z. (2022) · Transportation Research Record Journal of the Transportation Research Board · doi:10.1177/03611981221076843 · Cited by 7 · see citing works
Cui, H., Ren, Z., Ma, X., & Zhu, M. (2025) · Transportation Research Record Journal of the Transportation Research Board · doi:10.1177/03611981251318340 · Cited by 5 · see citing works
Wang, G., Singh, V. K., & Zhang, D. (2025) · Proceedings of the ACM on Human-Computer Interaction · doi:10.1145/3710921 · Cited by 4 · see citing works
Hu, Y., Cao, J., & Liu, J. (2021) · Journal of Transport Geography · doi:10.1016/j.jtrangeo.2021.103206 · Cited by 3 · see citing works
Roadspace allocation between autos, buses, and bicycles with heterogeneous demand
Gao, Y., Fielbaum, A., & Levinson, D. M. (2026) · Transportation Research Part A Policy and Practice · doi:10.1016/j.tra.2026.104884 · Cited by 2 · see citing works
Azege, I. S., Das, D. K., & Awuzie, B. O. (2019) · Springer eBooks · doi:10.1007/978-3-030-26528-1_25 · Cited by 2 · see citing works
Development of bus transit system control measures with open transit data
TARAR, A., Mukherjee, D. K., & Rao, K. R. (2021) · Scientific Journal of Silesian University of Technology Series Transport · doi:10.20858/sjsutst.2021.111.15 · Cited by 2 · see citing works
The Value of Security, Access Time, Waiting Time, and Transfers in Public Transport
Espino, R., Ortúzar, J. D. D., & Rizzi, L. I. (2021) · Elsevier eBooks · doi:10.1016/b978-0-08-102671-7.10020-x · Cited by 2 · see citing works
Souassi, M. A., & Hnaka, Z. (2025) · Lecture notes in networks and systems · doi:10.1007/978-3-031-90921-4_97 · Cited by 1 · see citing works
Rider Wellbeing as a Planning Metric for Dubai’s Bus System: A GSCA Model
Rahman, B. A., & Rashid, H. S. J. (2026) · Future Transportation · doi:10.3390/futuretransp6020062 · Cited by 1 · see citing works
Shen, W., Cao, H., & Zhao, J. (2025) · Sustainability · doi:10.3390/su17062645 · Cited by 1 · see citing works
Research and Simple Calculations on Public Transport Card Data
Yang, Y. (2021) · Journal of Physics Conference Series · doi:10.1088/1742-6596/1865/4/042055 · Cited by 1 · see citing works
Humić, R., & Luketić, H. (2017)
Data driven models for congested urban mobility networks
Tuncel, K. S. (2021) · doi:10.17760/d20416806
Study of Investment Feasibility for Smart Bus Stations at Silpakorn University
Yuenyong, P., Plubjang, N., Thongkonghan, T., Pullteap, S., & Kheovichai, K. (2023) · International STEM Education Conference (iSTEM-Ed) · doi:10.1109/istem-ed59413.2023.10305755
Development of passenger’s waiting time model at Bus Public Transit Terminal
Ibrahim, M. N., & Ede, J. K. (2024) · Nigerian Journal of Technology · doi:10.4314/njt.v42i4.3
Bus user itinerary choice: Can crowding information help shift riders?
Pan, L., Waygood, E. O. D., & Patterson, Z. (2025) · Case Studies on Transport Policy · doi:10.1016/j.cstp.2025.101375
- Authors
- Lu, H., Burge, P., Heywood, C., Sheldon, R., Lee, P., Barber, K., & Phillips, A.
- Affiliations
- RAND Europe (GB)
- Transport for London (GB)
- Corresponding author
- Lu, Hui
- ORCID
- Venue
- Transportation Research Procedia
- Publisher
- Elsevier BV
- Published
- 2018
- Volume
- 31
- Pages
- 18-34
- ISSN
- 2352-1457 (ISSN-L)
- 2352-1465
- OpenAlex
- W2892460332
- Cited by
- 34
- References
- 12
- Citations by year
- 2026: 2 · 2025: 6 · 2024: 2 · 2023: 2 · 2022: 7 · 2021: 8 · 2020: 3 · 2019: 2 · 2018: 1 · 2017: 1
- FWCI
- 4.12
- Type
- journal article
- Language
- en
- Open access
- diamond
- doi.org
- Licence
- Locations
- Transportation research procedia (publishedVersion, cc-by-nc-nd)
- Topics
- Social Sciences › Transportation › Transportation Planning and Optimization
- Physical Sciences › Engineering › Automotive Engineering › Transportation and Mobility Innovations
- Social Sciences › Economics, Econometrics and Finance › Economics and Econometrics › Economic and Environmental Valuation
- Keywords
- bus waiting time, passenger behavior, public transport information, mobile application, stated preference, discrete choice experiment
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Claims read by deepseek-v4.1-flash under definition version 3 (extract-v3) on 2026-11-06.
Classified by jev-1.13-20260917 (gold-guideline-v0.2) on 2026-11-06.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
