IELTS Academic · Reading
Passage · Suggested time:
When an average hides the journey
A bus service can become faster on average without becoming easier to use. For a passenger who must arrive before a fixed appointment, an occasional long delay may matter more than a small reduction in the usual journey time. Published averages can conceal that distinction. They combine uneventful journeys and seriously disrupted ones into a single figure, although passengers may plan around the latter. This does not make average travel time worthless. It means that a claim about improvement should identify which part of the experience has improved and whether that change reaches people whose journeys involve tight deadlines or several connecting services.
In the fictional city of Carrow, planners compared two bus corridors with similar average waiting times. On the first, buses arrived at reasonably regular intervals. On the second, several often arrived together after a long gap. Passengers arriving during the gap experienced a very different service from those reaching the stop just before the cluster. Increasing the number of buses alone would not necessarily remove this pattern. If vehicles continued to catch one another, the additional capacity could arrive in the same cluster. The relevant issue was the distribution of service over time, not just the number of vehicles assigned to the corridor.
The city's first dashboard focused on punctuality at the final stop. Drivers could improve this measure by waiting at selected points to restore the timetable. For passengers making a connection midway along the route, however, that recovery came too late. A journey could be classified as punctual at its destination while still causing a missed connection earlier. The planners consequently added observations at important interchanges. They did not abandon destination punctuality; they stopped treating it as a complete description of the trip. The revised approach recognised that the same vehicle could provide different levels of usefulness to passengers travelling along different sections.
Cancellations created a second problem in the statistics. The original punctuality calculation included only buses that completed their journeys. A cancelled trip could therefore disappear from the measure rather than count as a failure. In an extreme case, removing a delayed journey might improve the reported percentage even though passengers had fewer services available. Carrow began publishing the share of scheduled journeys completed alongside punctuality. This did not answer every question about reliability, but it made the denominator visible. Readers could distinguish a high proportion of punctual arrivals from a high proportion of the promised service actually being delivered.
Passengers also responded to uncertainty by changing their own behaviour. Some left home early enough to absorb a substantial delay. Others avoided a useful connection and chose a slower direct route. Those decisions could make their recorded journeys appear orderly while imposing an unrecorded cost. A survey of actual travel times alone would miss the extra time reserved before departure or the opportunities given up to avoid risk. The planners therefore asked about time allowances and rejected alternatives as well as completed journeys. Reliability was understood partly as the ability to make plans with confidence, not simply as the speed observed once a passenger was on board.
Real-time information helped, but it was not a substitute for dependable service. A display that accurately predicted a long wait reduced uncertainty without shortening the wait itself. Repeatedly changing predictions could even weaken confidence in the system. The city examined forecast accuracy separately from service delivery and checked whether signs were available at the stops where disruption mattered most. It also retained a timetable that could be understood without a phone. Information was treated as another service with its own coverage and performance, rather than as evidence that irregular operations had been solved. Passengers needed both useful predictions and buses that arrived.
Different improvements benefited different travellers. Regular spacing mattered especially to people who reached a stop without consulting a timetable, while a dependable scheduled departure could be more useful on an infrequent route. Better connections might greatly shorten a journey for a small group even if the citywide average changed little. These distinctions argue against choosing a single headline indicator as the sole funding criterion. A defensible comparison should connect each measure to a particular need, report who benefits and disclose what remains difficult. It should also avoid implying that every traveller values the same trade-off between speed, waiting time and certainty.
Carrow's experience suggests that reliability is a relationship between a service and the plans people build around it. No one statistic can represent that relationship for everyone. A practical dashboard can nevertheless remain concise: completed service, the spread of waiting times, performance at key connections and the accuracy of information each reveal a different part of the experience. The purpose is not to overwhelm passengers with measures. It is to prevent an improvement in one convenient statistic from concealing deterioration elsewhere. A service becomes more usable when people can predict its limits and when those limits allow the journeys they need to make.