Sexual harassment on the streets costs women in many ways. A study has found that it compels women in Delhi, for one, to compromise on the quality of college they choose, makes travel more expensive and ultimately affects their economic mobility. Male students, of course, are far less driven by safety concerns.

For her 2017 paper, Safety First: Perceived Risk of Street Harassment and Educational Choices of Women, Brown University PhD scholar Girija Borker studied the choices made by 4,000 students attending Delhi University colleges. Her paper explores how women applicants weighed the quality of the college against the perceived safety of the route to that college and chose to trade quality for safety.

The national capital is widely considered to be an unsafe city for women. A survey conducted by the Thomson Reuters Foundation and released in October found that Delhi is considered the world’s worst megacity out of 19 megacities surveyed because of the sexual violence women face.

Constrained choices

Quantifying both quality of the institution and safety, Borker found that “women are willing to attend a college that is 13.04 percentage points lower in quality” than the institution they are eligible for if they feel the journey will be safer by a single unit. “This is equivalent to choosing a college that is 8.5 ranks lower,” wrote Borker. “Men on the other hand are willing to attend a college that is only 1.37 percentage points (or 0.9 ranks) lower in quality for an additional SD [standard deviation or, for the study, a single unit] of safety.”

Women are willing to travel longer and spend more too with regard to safety, according to the study. They are willing to spend Rs 20,000 more per year for one unit of safety while men will spend just Rs 1,200 more. “The difference of Rs 18,000 is…almost double the average annual tuition in [Delhi University],” Borker pointed out. And where women are willing to travel as much as 40 minutes more for a safer journey, men will increase their travel time by just four minutes for the same amount of safety. Women feel most insecure in buses.

“Street harassment imposes an external constraint on women’s behaviour that could potentially lead to sub-optimal choices,” writes Borker. “Choosing a worse ranked college is likely to have long-term consequences since college quality affects a student’s academic training…,network of peers…,access to labour opportunities, and lifetime earnings. In fact, such misallocation of students to colleges, where high achieving females sort to low quality colleges, may not only affect women’s long-term outcomes but could also have important aggregate productivity effects.”

Quantifying safety

Four thousand students from eight Delhi colleges participated in a detailed survey conducted by Borker in spring, 2016. The eight included two women’s colleges and one evening college. From these 4,000, Borker worked with a sample of 2,695 students (1,757 female) who live in Delhi with their parents and travel to college every day. Another 887 students from 32 colleges took shorter surveys and data from 669 of them was used for analysis.

For the study, both the quality of the college and safety were quantified. Borker used the minimum marks required for admission to a college, or cut-off as it is called, as a marker of its quality – higher the cut-off, better the quality – and ranked colleges accordingly. For every student included in the survey, their “choice set” – the set of colleges they had the marks for – was considered and ranked.

Quantifying the relative safety of travel routes was more complicated. To chart actual routes taken and potential routes and modes of transport – private car, public transport and walking – Borker used Google Maps. To assign a “safety score” to each travel route, she used data from two mobile applications – SafetiPin and Safecity.

SafetiPin furnishes data gathered through safety audits of Delhi’s localities. The audits involve scoring an area out of three on each of nine parameters – availability of light, openness (whether you can see ahead and the area around), visibility (whether there are windows and street vendors), presence of others, of security guards or the police, availability of a path to walk or run on, availability of public transport, whether all genders can be seen in the area and general feeling. Borker used data from 26,500 audits conducted between November 2013 to January 2016.

SafeCity includes testimonies from those who have been harassed in public spaces, and the mode of transport is mentioned. Borker used 5,500 such crowd-sourced reports.

The application data was used to score each route for safety and then those scores were used to compute a “unit of safety” for every “choice set”. To put that unit into perspective, Borker uses district-level data on rape from the National Crime Records Bureau. The paper says each unit of travel safety while walking “is equivalent to a 3.1 percent decrease in the rapes reported annually”.

Buses unsafe

The Delhi Metro, predictably, is considered the safest mode of travel by women students although about 16% of the harassment incidents were on the metro. 86% of the women who use it travel in the women’s compartment.

Unsurprisingly, buses are considered the least secure and according to the study, 40% of the harassment incidents “mention…a bus or the people in it”. Still, 38% students – 33% of them women – cover some portion of the distance to college by public or private buses but men are more likely to use them than women.

A large number, 68%, walk for some distance. But here too, more men (71%) walk than women (66%).