Street epistemology


The auto-rickshaws of Bangalore are the best of India. They stop and pick you up; you announce your destination; and they have this automatic movement by which they turn around leftwards to re-set the meter. Recently there was an increase of fare, so all of them are carrying a small chart with the mileage, the old price (as appearing on the meter) and the new price (that the customer is supposed to pay). However, most of them know the correspondence by heart now and whatever they would announce is usually true (In my work, we call that monitoring the baseline situation with regular surveillance).

In that context, I tried to get a rick in Malleshwaram to go to work on Monday morning. The first I stopped asked for four times the regular amount and refused the meter. That seemed unusual. The second went into a similar game. I started to wonder (That we call a signal). But after six unsuccessful attempts and a frustrating settlement over an excessive price, I knew there was a problem (The signal had become an alert).

Arrived at the office, I asked around whether there was something wrong with the rick drivers in Bangalore. Saddened by my perplexity, my senior colleague mentioned that there were roads being laid in Malleshwaram and that the rick drivers where in disarray with the one-way streets and other headaches. As a result, they would refuse selected destinations and overcharge for others (A good environmental investigation). Overall, the cluster of the recent weeks affected the rick drivers of Malleshwaram only (That is a Time-Place and Person analysis for you) and our hypothesis was that the cause of the problem was the recent roadwork: “If you cross the road that is in front the office, you enter the part of the city that is not affected” said the colleague. The office is more or less on the edge of Malleshwaram.

By that time, I felt reassured that the problem was framed because I hated to accept the idea that something had happened to the best rick drivers of India. But all of that was only speculation, right? (What people call hypothesis generation.) So I had to find out more.

In the evening, after a good day of work, I decided to go downtown for some errands. So I crossed the road opposite to the office. As I was doing so, I realized that I was going back into the part of Bangalore that was supposed to be unaffected. So as I was trying to stop a rick, I thought: “If what we elaborated is true, then the rick I will stop should be OK”. I stopped the rick. He picked me up. I announced my destination. He had this automatic movement by which he turned around leftwards to re-set the meter (This is hypothesis testing).

One of the points of the training during the day with the students was: Never test a hypothesis using the data that were used to generate it!

Kommentarer

Anonym sa…
You concluded that "the auto-rickshaws of Bangalore are the best of India" based on assessment of a limited sample (Chennai, Delhi etc.). I will soon conduct a case control study to prove the hypothesis that Mumbai auto-rickshaws are even better than Bangalore auto rickshaws.