Article Text

Download PDFPDF

OP50 Push and/or pull: A systematic review and meta-analysis of studies evaluating the effectiveness of ‘carrot’, ‘stick’, and combined interventions on modifying travel behaviour
  1. Christina Xiao,
  2. Esther van Sluijs,
  3. Richard Patterson,
  4. David Ogilvie,
  5. Jenna Panter
  1. MRC Epidemiology Unit, University of Cambridge, Cambridge, UK


Background While active travel policies may positively affect health and the environment, evidence suggests small or inconsistent effects in these policies in changing travel behaviour. To identify which types of interventions are more effective, this systematic review and meta-analysis aims to (1) compare the effectiveness of positive (‘carrot’) strategies, negative (‘stick’) strategies, or a combination of the two on modifying travel behaviour and (2) assess which functions have greater impacts on travel outcomes.

Methods Nine databases were searched for controlled before-and-after studies of population-level interventions and travel behaviour (e.g. driving, active travel, public transit, walking, and cycling) from adults in the general population. Interventions were categorized into functions (how the intervention seeks to affect behaviour). Depending on whether gains or losses of functions could occur, interventions were classified as carrots (e.g. new bikeshare programs), combined carrot and stick (e.g. traffic calming), or stick interventions (e.g. congestion charging). Harvest plots were used to visually summarize the findings weighted by study quality. Where possible, outcomes were converted into standardized mean differences (SMD) and random-effects meta-analyses were conducted.

Results We extracted data from 83 publications reporting 98 interventions. From these, we identified 20 intervention types and eight function categories. The majority of interventions were carrots (n=64), followed by carrot and stick (n=17) and stick (n=17). Harvest plots demonstrated that most evaluations, particularly those classified as higher quality, found changes in favour of the intervention. Results for carrot interventions, however, were more less consistent than for stick or combined interventions. This was consistent with findings from the meta-analysis, which were statistically nonsignificant but had point-estimates of greater magnitude for driving outcomes for sticks (SMD -0.21; 95%CI -0.43, 0.01) and combined carrot and stick interventions (-0.17; -0.65, 0.31) compared to carrots (-0.09; -0.21, 0.03). Likewise, for active travel outcomes, combined carrot and stick interventions had a higher SMD (0.39; -0.01, 0.78) compared to carrot interventions (0.10; -0.06, 0.25). Financial functions were found to be the most effective for driving outcomes, whereas access, convenience, safety, and space were more effective for active travel outcomes.

Discussion This is the first review to compare whether positive, negative, or combined strategies and their functions differ in terms of effectiveness on travel behaviour, which can aid policymakers in designing sustainable transportation policies. Further research is needed for interventions with a stick component, which suggest greater effectiveness yet remain less well-studied, possibly because they are less conducive to experimental manipulation.

  • Active travel
  • Interventions
  • Systematic Review

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.