The Effectiveness of Behavioural Interventions on Residential Location Choices and Commute Behaviours: Experimental Evidence from China
Round 1
Reviewer 1 Report
Comments and Suggestions for AuthorsRigour: The research employs a sound methodological approach by using randomised controlled trials (RCTs) to test the effectiveness of three behavioural interventions on residential location choices and commuting behaviours. Conducting the experiment online via Credamo allowed for a broad reach within the targeted demographic of renters in Xi'an who needed to commute. The use of a control group and multiple treatment groups facilitates the comparison of intervention effectiveness. However, it's important to note that the online setting might introduce limitations regarding the depth of engagement and the validity of the simulated decision-making process.
Validity and Reliability: The study's internal validity is potentially impacted by the observed heterogeneity in the sample groups, despite the use of regression analysis to control for these differences. The external validity of the findings is limited by the specific context of Xi'an, and the effectiveness of these interventions might differ in other urban contexts with varying transportation infrastructure and cultural norms. The study acknowledges this limitation and calls for further exploration in broader contexts.
Significance: The research addresses a highly important issue in urban planning: promoting sustainable transportation and integrating these goals into residential location decisions. With the transportation sector being a significant contributor to greenhouse gas emissions and urban challenges like congestion and pollution, exploring cost-effective demand-side interventions is crucial.
Impact of the Findings: The findings indicating that behavioural interventions can significantly increase respondents' preference for green commute modes and reduce their tendency to use private cars in a Chinese city are potentially impactful. This suggests that such interventions could be a valuable tool for urban transport management in countries facing rapid urbanization and increasing car ownership.
Relevance: The research is highly relevant to the field of urban planning, transportation demand management, and behavioural economics. It directly investigates the application of behavioural science principles to address real-world urban challenges.
Originality: The research contributes to the field by providing empirical evidence on the effectiveness of these behavioural interventions in a non-Western urban setting. The finding that the social norm intervention had the largest impact in this context offers valuable insights for tailoring interventions in similar urban environments.
Unique Perspective: The study offers a unique perspective by focusing on a major urban center in a rapidly developing country with a well-established public transportation system. The exploration of how behavioural interventions can influence both commuting preferences and, potentially, long-term residential location decisions provides a holistic view relevant to sustainable urban development.
Author Response
Thank you very much for your thoughtful review and positive feedback. We have revised the manuscript in response to the comments from all reviewers. Please feel free to let us know if you have any further suggestions.
Reviewer 2 Report
Comments and Suggestions for AuthorsThis study explores the application of behavioral interventions in Chinese cities through randomized controlled trials, demonstrating practical significance and methodological innovation. It is recommended to further refine the sample, operational details, and the depth of statistical model interpretation to enhance the study's scientific validity and persuasiveness.
- Regarding the section on recruiting experimental participants, the article only mentions an online panel data platform in China (www.credamo.co). The website's pages and related links are in English, yet all experimental participants are young Chinese individuals working and living in Xi’an. Could this have impacted recruitment? Please provide the specific recruitment process for experimental participants.
- Please replace the Chinese text in Figure 3 with English to facilitate reader comprehension, and clarify in the article that the images shown to experimental participants were in Chinese.
- The penultimate paragraph of Section 3.3 states that the focalism bias intervention is in Appendix 2, but I was unable to locate where Appendix 2 is positioned.
- Given the significant success of China’s new energy vehicle policies, a high proportion of electric vehicles are now used in many major Chinese cities. When analyzing the data, should a certain proportion of private cars be classified as "green commuting"? Additionally, categorizing all "taxis" as "green commuting" may be controversial, as many taxis are still internal combustion engine vehicles with high carbon emissions, differing in nature from low-carbon modes such as walking and public transportation.
- The regression model does not provide multicollinearity tests (such as VIF values), and control variables like "housing area" and "rent" may be highly correlated, potentially affecting the accuracy of coefficient estimates.
Author Response
Thank you very much for your thoughtful review and feedback. We appreciate your time and below please find a list of your comments and our replies. We have also updated the manuscript so please feel free to let us know if you have any further suggestions.
Comment 1: Regarding the section on recruiting experimental participants, the article only mentions an online panel data platform in China (www.credamo.co). The website's pages and related links are in English, yet all experimental participants are young Chinese individuals working and living in Xi’an. Could this have impacted recruitment? Please provide the specific recruitment process for experimental participants.
Reply 1: Thank you for this valuable comment. Credamo offers both English and Simplified Chinese versions of its platform and automatically switches to the Chinese version when accessed from an IP address within China. Our questionnaire was also in Chinese, so participants interacted with the platform entirely in Chinese throughout the process. We have added a footnote (highlighted in yellow) to clarify this point. (Page 11)
Comment 2: Please replace the Chinese text in Figure 3 with English to facilitate reader comprehension, and clarify in the article that the images shown to experimental participants were in Chinese.
Reply: Thanks for the comment and we have changed the figure accordingly. (Page 10)
Comment 3: The penultimate paragraph of Section 3.3 states that the focalism bias intervention is in Appendix 2, but I was unable to locate where Appendix 2 is positioned.
Reply: Thanks for the comment and we have added the appendixes between “Conclusion” and “References.” (Page 18)
Comment 4: Given the significant success of China’s new energy vehicle policies, a high proportion of electric vehicles are now used in many major Chinese cities. When analyzing the data, should a certain proportion of private cars be classified as "green commuting"? Additionally, categorizing all "taxis" as "green commuting" may be controversial, as many taxis are still internal combustion engine vehicles with high carbon emissions, differing in nature from low-carbon modes such as walking and public transportation.
Reply:
Thank you for your comment regarding the categorization issue. A certain proportion of private cars are indeed electric and could be considered "green commuting." However, our questionnaire did not collect information on vehicle types, so we are unable to distinguish which respondents were driving electric vehicles. The impact of this limitation should be relatively small. As of early 2022, municipal government data showed that only 3.8% of all vehicles in Xi'an were electric. Although electric vehicles accounted for 23% of new car sales that year, the majority of private cars remained internal combustion engine vehicles.
Regarding taxis, the situation is different. As of 2022, Xi'an had 15,457 licensed taxis and 24,721 private hire vehicles. The government began promoting electric taxis in 2015 and added 6,000 more electric taxis in 2019. Since the end of 2021, all newly licensed taxis have been electric. Private hire vehicles are also largely electric due to their lower operating costs. The government aims to fully electrify public transportation, including private hire vehicles, by 2025. Based on this context, we categorized "taxis" as a "green commuting" option.
Nonetheless, we acknowledge that more precise wording and clarification of local contexts are important for a better understanding. Necessary explanation of local context has been added in the last four paragraphs of “Discussion” section. (Page 16 &17)
Related report:
Government report on vehicle numbers in Mandarin Chinese: https://mp.weixin.qq.com/s/LQ4Cxeo9acpqaejcc9upGQ
https://mp.weixin.qq.com/s/OodMLuG9lf5kjh_iqZpzfQ
Comment 5: The regression model does not provide multicollinearity tests (such as VIF values), and control variables like "housing area" and "rent" may be highly correlated, potentially affecting the accuracy of coefficient estimates.
Reply: Thank you for the comment. We have added the multicollinearity test results in the final paragraph of Section 4 (Results). The changes are highlighted in yellow (Page 13). We have also included two additional variables—rent and distance to city center—in the OLS regression, and Table 6 has been updated accordingly. We selected distance to city center to reflect the housing area, as Xi’an is a monocentric city with multiple ring roads surrounding its center. (Table 6)
Reviewer 3 Report
Comments and Suggestions for AuthorsA very interesting paper with sound methodology and results.
The conclusion part should be improved.
As stated in the introduction part: The paper "seeks to determine whether these interventions are effective in the unique urban context of China and which intervention is superior. The study’s objective is to not only identify effective tools but to discern the most efficient among them. Previous research often examined one intervention at a time, making direct comparisons difficult. This study addresses this by implementing three interventions concurrently for direct comparison.
Notwithstanding, in the conclusion part this comparison between the three interventions is absent.
Some minor comments:
The introduction section could be enriched with more references and numbers.
Definition of randomised controlled trials (RCTs) and quasi-experiments could be given in the review part of the paper.
Explanation of anchoring and herding effects in psychology
"For example, in studies of the effect of the personal travel planning intervention, relatively straightforward interventions were found ineffective in shifting commuter behaviour" (page 4). What is a straightforward intervention in personal travel planning?
"The focalism intervention reminded respondents to notice the important but intangible features of housing units" (page 5) More details could be provided to understand the "feature of housing unit" and the connection with travel choices.
"Although the population density of Xi’an is much higher than that of London, Xi’an residents enjoy better accessibility to sustainable transportation options. " Denser cities have better accessibility options so it is not correct to use the term "Although".
"The city’s temperature ranges from -5 to 30 C" In my view it is better to use a more precise measure as "daily high temperature" or "daily mean temperature".
Figure 3: The walk and cycle map are in wrong position.
Appendixes 1 and 2 are missing.
Author Response
Thank you very much for your thoughtful review and feedback. We appreciate your time and below please find a list of your comments and our replies. We have also updated the manuscript so please feel free to let us know if you have any further suggestions.
Comment 1: In the conclusion part this comparison between the three interventions is absent.
Reply: Thank you for the comment. We have added a new Discussion section, and the first paragraph compares the effectiveness of the three interventions. (Page 16)
Comment 2: The introduction section could be enriched with more references and numbers.
Reply: Thank you for your comments. We have added the necessary references in the Introduction section (Pages 2–3).
Comment 3: Definition of randomised controlled trials (RCTs) and quasi-experiments could be given in the review part of the paper.
Reply: We have added these definitions at the beginning of the Literature Review section. The changes are highlighted in green. (Page 3)
Comment 4: Explanation of anchoring and herding effects in psychology.
Reply: We have added explanations of the two effects and their applications in transport demand management. The changes are highlighted in green. (Page 4)
Comment 5: "For example, in studies of the effect of the personal travel planning intervention, relatively straightforward interventions were found ineffective in shifting commuter behaviour" (page 4). What is a straightforward intervention in personal travel planning?
Reply: Thank you for bringing this to our attention. The term "straightforward intervention" was ambiguous in our submission. Since the main purpose of this section is to illustrate that studies using the same type of intervention can report different outcomes, we have revised the terminology and emphasized the distinction between generic and personalized interventions. Specifically, we have revised the wording of this part as follows: For example, in studies on the effects of personal travel planning interventions, generic approaches such as letters and emails were found to be ineffective in shifting commuter behavior, while more participant-specific interventions produced significantly positive results. A footnote explaining the details of the two types of interventions has also been added and highlighted in green. (Page 5)
Comment 6: "The focalism intervention reminded respondents to notice the important but intangible features of housing units" (page 5) More details could be provided to understand the "feature of housing unit" and the connection with travel choices.
Reply: Thank you for the comment. We have updated the phrasing of this section and added more explanations of the two interventions used in previous literature. The changes are highlighted in green. (Page 6)
Comment 7: "Although the population density of Xi’an is much higher than that of London, Xi’an residents enjoy better accessibility to sustainable transportation options. " Denser cities have better accessibility options so it is not correct to use the term "Although".
Reply: Thank you for the comment. We have deleted the first half of this sentence. It now states that Xi’an residents enjoy good accessibility to sustainable transportation options. The following sentences briefly introduce the public transportation system in Xi’an. The changes are highlighted in green. (Page 6)
Comment 8: "The city’s temperature ranges from -5 to 30 C" In my view it is better to use a more precise measure as "daily high temperature" or "daily mean temperature"
Reply 8: Thank you for the comment. We have revised the expression accordingly. The updated part is highlighted in green. (Page 7)
Comment 9: Figure 3: The walk and cycle map are in wrong position.
Reply: Thank you for the comment. We have switched the positions of the two maps and translated the text into English for better understanding. (Page 10)
Comment 10: Appendixes 1 and 2 are missing.
Reply: Thanks for the comment and we have added the appendixes after the conclusion. (Page 18)
Reviewer 4 Report
Comments and Suggestions for AuthorsThe article presents an interesting pilot case study of a methodologically robust intervention, addressing a highly relevant topic that is deeply connected to the sustainability of cities—not only environmental, but more importantly, social. Given the substantial work carried out by the authors and the generally encouraging, albeit not definitive, results, I would like to offer some broader observations that concern the overall framing of the study.
First, mobility is presented as a largely mechanistic issue, which is not entirely accurate—as the authors themselves acknowledge in the conclusion, referring to private car use as a matter of social status. In this sense, the introduction could be enriched by trimming certain potentially redundant sections and broadening the perspective to include, for example, the role of advertising in shaping perceptions of mobility. This is particularly relevant in the case of cars: in Europe, and even more so in the United States, it has been the automotive industry that has fundamentally transformed cities into car-dependent spaces. Even today, there is strong resistance to change, particularly when it comes to reclaiming public space from private vehicles and returning it to people. In this regard, the ongoing efforts in China to strengthen public transport are not only commendable, but should be highlighted and promoted as a cornerstone of harmonious and sustainable urban development.
The article also touches only marginally on the socio-economic factors that underlie phenomena such as the rent gap and the filtering effect, whereby better-serviced areas (including by public transport) tend to have higher housing prices. If price control policies exist in the Chinese context, they should be discussed, as in many Western countries the decision about where to live is rarely truly "free", but instead overwhelmingly determined by income and affordability. This has major implications for the equitable enjoyment of the city and for the realization of the right to the city.
It would also be interesting to further explore the effectiveness of the interventions across subgroups, as I imagine that one of the three behavioural treatments may have been more effective for specific categories (e.g., younger individuals, those with higher levels of education, people with children, etc.). Such insights would be extremely helpful for designing targeted local policy measures, especially in urban contexts where there is substantial room for improving public transport use.
Finally, I would briefly note that, in urban contexts, mobility planning must go hand in hand not only with the planning of urban functions, but also with overall urban planning. This is essential to avoid spatial mismatches between where people live, where they work, and where they access urban services and amenities. Again—and based on what I can tell—Chinese cities appear to be significantly more advanced in this respect compared to many of their Western counterparts.
In my view, these conceptual refinements would greatly enhance the value of the article and could make it not only publishable, but also a highly valuable contribution to the broader debate on sustainable urban mobility.
Author Response
Thank you very much for your thoughtful review and feedback. We appreciate your time and below please find a list of your comments and our replies. We have also updated the manuscript so please feel free to let us know if you have any further suggestions.
Comment 1: First, mobility is presented as a largely mechanistic issue, which is not entirely accurate—as the authors themselves acknowledge in the conclusion, referring to private car use as a matter of social status. In this sense, the introduction could be enriched by trimming certain potentially redundant sections and broadening the perspective to include, for example, the role of advertising in shaping perceptions of mobility. This is particularly relevant in the case of cars: in Europe, and even more so in the United States, it has been the automotive industry that has fundamentally transformed cities into car-dependent spaces. Even today, there is strong resistance to change, particularly when it comes to reclaiming public space from private vehicles and returning it to people. In this regard, the ongoing efforts in China to strengthen public transport are not only commendable, but should be highlighted and promoted as a cornerstone of harmonious and sustainable urban development.
Reply: Thanks very much for this constructive suggestion. We have added the following paragraph to the Introduction section to address the socio-cultural dimensions of transport choices.
“While offering promising tools for transport demand management, behavioural approaches are also critical in recognising that transport mode choice extends beyond mere utility maximisation to encompass complex socio-cultural dimensions. Private car ownership in particular represents not simply a mobility solution but a powerful status symbol, shaped significantly by commercial advertising and industry influence. In Western contexts, automotive manufacturers have systematically transformed urban spaces into car-dependent environments through decades of marketing, lobbying and infrastructure development that prioritised private vehicles over public alternatives. This transformation created entrenched resistance to reclaiming public space from cars, even when environmental concerns demand such changes. The Chinese urban context presents a notably different trajectory, with substantial governmental investment in public transport infrastructure occurring alongside rapid motorisation. This parallel development creates a distinct opportunity to examine behavioural interventions that might counteract the status-driven appeal of car ownership. By acknowledging these socio-cultural forces, this study recognises that effective transport policy must address both individual decision-making processes and the broader structural factors that shape mobility perceptions.”
Comment 2: The article also touches only marginally on the socio-economic factors that underlie phenomena such as the rent gap and the filtering effect, whereby better-serviced areas (including by public transport) tend to have higher housing prices. If price control policies exist in the Chinese context, they should be discussed, as in many Western countries the decision about where to live is rarely truly "free", but instead overwhelmingly determined by income and affordability. This has major implications for the equitable enjoyment of the city and for the realization of the right to the city.
Reply: Thank you for your comment; it addresses an important issue in Xi’an. We have expanded the Discussion section to provide more context about the socio-economic factors that influence housing choices in our study area. The sixth paragraph (page 16) now includes information about Xi'an's mixed residential development patterns, where diverse housing options exist throughout the city, even in areas with premium public transport access. We have explained how urban villages and older gated communities provide affordable housing options alongside luxury developments, creating conditions where renters with average incomes maintain reasonable choice flexibility. We have also briefly addressed price control policies (page 17), noting that while state-owned properties are subject to government regulation, they constitute a separate segment from the private rental market examined in our study. These additions provide important context for understanding how the socio-economic landscape of Xi'an differs from many Western cities and how this might influence the effectiveness of behavioural interventions in our study. We have added these discussions at the end of the Discussion section to address contextual differences between Xi’an and other cities.
Comment 3: It would also be interesting to further explore the effectiveness of the interventions across subgroups, as I imagine that one of the three behavioural treatments may have been more effective for specific categories (e.g., younger individuals, those with higher levels of education, people with children, etc.). Such insights would be extremely helpful for designing targeted local policy measures, especially in urban contexts where there is substantial room for improving public transport use.
Reply: We agree that interventions may have different effects on different subgroups. However, our study did not conduct such an analysis due to the relatively small sample size (around 90 participants in each group). Further subdividing the dataset would result in even smaller sample sizes, making the statistical results less reliable.
Comment 4: Finally, I would briefly note that, in urban contexts, mobility planning must go hand in hand not only with the planning of urban functions, but also with overall urban planning. This is essential to avoid spatial mismatches between where people live, where they work, and where they access urban services and amenities. Again—and based on what I can tell—Chinese cities appear to be significantly more advanced in this respect compared to many of their Western counterparts.
Reply: Thank you for this insightful observation about the importance of integrated urban and mobility planning. We agree that effective transportation demand management requires coordination with broader urban planning to minimize spatial mismatches between residential areas, workplaces, and urban amenities. Your observation that Chinese cities may be more advanced in this integration aligns with our experience in Xi'an, where compact, mixed-use development and coordinated infrastructure investment have created relatively good conditions for sustainable mobility choices. We have incorporated this context in our expanded Discussion section (Page 16-17), where we note Xi'an's mixed residential patterns and density advantages. This urban form indeed creates favorable conditions for the behavioral interventions we studied, as it helps minimize the structural barriers that might otherwise limit the effectiveness of such interventions. This integrated approach to urban development in Chinese cities represents an important contextual factor when interpreting our findings and considering their applicability to other urban settings.
Reviewer 5 Report
Comments and Suggestions for AuthorsThis manuscript investigates the effectiveness of focalism, social norm, and visualisation behavioural interventions on residential location and commuting preferences in Xi'an, China, using randomised controlled trials with renters. The study addresses sustainable urban development and transportation challenges.It examines behavioural interventions in China, contributing to generalizability.The direct comparison of three interventions is a novel aspect.The use of RCTs and online panel data (Credamo) is well-described.
The following improvements are suggested:
1)Revisit the focalism intervention's simplification and its potential impact.
2)Further explore the unexpected shift in housing preferences (private kitchens/toilets).
3）Further examine the potential decrease in satisfaction levels due to interventions.
4）Acknowledge and consider local context (built environment, policies, culture) more thoroughly in future research.
A separate discussion section is needed to address the above issue. Meanwhile, it is recommended to use "the study", "the paper", "the authors" instead of "we"/ "our" in journal articles.
Author Response
Thank you very much for your thoughtful review and feedback. We appreciate your time and below please find a list of your comments and our replies. We have also updated the manuscript so please feel free to let us know if you have any further suggestions.
Comment 1: Revisit the focalism intervention's simplification and its potential impact.
REPLY: Thank you for the comment. We have added a Discussion section, and the first paragraph of that section compares the three interventions and discusses the influence of the simplification made to the focalism intervention. (Page 16)
Comment 2: Further explore the unexpected shift in housing preferences (private kitchens/toilets).
REPLY: In the newly added Discussion section, the fourth paragraph explores the unexpected shift in preferences for private kitchens and toilets. (Page 16)
Comment 3: Further examine the potential decrease in satisfaction levels due to interventions.
REPLY: In the newly added discussion section, the second and third paragraphs explore the influence of interventions on satisfaction levels. (Page 16)
Comment 4: Acknowledge and consider local context (built environment, policies, culture) more thoroughly in future research.
REPLY: In the newly added discussion section, the fifth to eighth paragraphs mention some local context influencing the experiment results. (Page 16-17)
Comment 5: A separate discussion section is needed to address the above issue. Meanwhile, it is recommended to use "the study", "the paper", "the authors" instead of "we"/ "our" in journal articles.
REPLY: Thank you for the comment. We have revised the wording throughout the article.
Round 2
Reviewer 4 Report
Comments and Suggestions for AuthorsThe authors should provide appropriate references for the sections added in the discussion and conclusion. Once this is done, the article can be published without reservations. The manuscript has been improved, and the responses provided in the first round of revision are fully appropriate. Excellent work!
Author Response
Dear Reviewer,
Thanks for your comment and we have added the relative references to the discussion and conclusion sections in the newly updated manuscript.
Best wishes
