Day 168 — Listening: Planning a Bike-Share Study — Section 3（議論）

（色付きは解答後の確認用。青＝根拠、黄＝distractor）

Dr Hale: Come in, both of you. So, your project on the city bike-share scheme. Where have you got to?

Jonas: We've mostly been arguing about data, to be honest. Maya wanted to start with the trip logs from the operator.

Maya: Yes, every journey is recorded, with the start and end station. The operator will give us the logs free of charge, which helps. The trouble is that casual users, the people who pay at the kiosk, don't appear in them at all. In the suburbs, those might be the majority.

Jonas: I'd rather run an online questionnaire. It's quick to set up, and we could ask people why they ride. But I do worry about who actually answers. Realistically, it'd be people who already love cycling and are happy to talk about it.

Maya: And everyone else just ignores the link.

Dr Hale: What about interviewing people in the street?

Jonas: We looked at that. You get a much better mix of people. I thought we'd need a permit from the council, but it turns out that's just a form, and they confirmed it takes a week.

Maya: The real problem is time. Two of us can manage thirty interviews on a good afternoon, and we'd need to cover a dozen stations. It would eat the whole term.

Dr Hale: And the weather? I'd expect that to matter.

Jonas: Weather records are free and recorded hourly, which is ideal. But the nearest recording station is at the airport, twelve kilometres away, so it tells us nothing about the hills in the north, where the cold is worst.

Dr Hale: Fine. Anything else?

Maya: We were also sent a sample from the council's complaints database. It's up to date, to last month at least. But nearly all of it is about bikes left on pavements. It says nothing about whether the scheme actually works.

Dr Hale: Then the trip logs are your backbone, and you can ask the operator about casual users separately. Let's go through the plan. How much data will you request?

Maya: Six months, covering this spring and summer.

Dr Hale: I'd go for twelve. Otherwise any difference you find might simply be the season, not the location.

Jonas: That's fair. Twelve months, then.

Dr Hale: And cleaning the data?

Maya: We'll remove very short trips. Under one minute, probably, because people unlock a bike and then change their mind.

Jonas: The operator's own report treats anything under two minutes as a likely docking fault, so I'd use that instead.

Maya: Okay, two it is.

Dr Hale: Then you need to group the stations. By district?

Maya: District boundaries are political and very uneven, so no. It's better to group them by distance from the city centre: within two kilometres, two to five, and over five.

Dr Hale: Good. And how will you compare the groups?

Jonas: The average number of trips per day.

Dr Hale: Careful. One festival weekend will drag the mean up for the central stations. Use the median instead.

Jonas: Right, median trips per day.

Dr Hale: Last question. How will you present the results?

Maya: Bar charts, one for each zone?

Dr Hale: You could, but a heat map would show you where use is high at a glance. For a final presentation, I'd choose that.

Maya: Agreed. A heat map it is.