News Spotlight Methodology More Spotlights

Most dangerous transport systems for women

Updated: Wed, 29 Oct 2014

DOWNLOADBack to top

For a PDF version of this methodology that includes the full poll questions, please click on the button:

SCOPEBack to top

The Thomson Reuters Foundation, the corporate charity of Thomson Reuters, the world’s largest news and information provider, has conducted a survey into women’s safety on transport in 15 of the world’s largest capitals, as defined by the United Nations, and New York, the most populous city in the United States.

The survey involved questioning women in each of the 16 cities, as well as experts focused on women’s rights, gender equality, urban planning and gender-friendly urban spaces. The survey was carried out in collaboration with a major UK polling company, YouGov.

POLLING METHODBack to top

The questionnaires for the two groups were drafted by the Thomson Reuters Foundation in collaboration with YouGov. YouGov conducted an online survey of between 380 and 513 women in each of the 16 cities. This sample size was deemed to be sufficient to provide statistically reliable data with a low margin of error.

YouGov is a pioneer of online market research and has a panel of more than 3.3 million people worldwide in 33 countries, representing all ages, socio-economic groups and other demographic types. Where YouGov did not have panels, partners were used and it ensured the quality of services was equal to those provided through its own panels.

The Thomson Reuters Foundation polled a minimum of 10 experts in each city except in Kuala Lumpur where nine experts were interviewed. These included women’s rights activists, architects and academics with an interest in gender equality and urban planning.

In total, 6,555 women and experts were surveyed.

RANKINGBack to top

The results of the polls were indexed and ranked by the Quantitative Research Team at StarMine, a division of Thomson Reuters whose researchers create analytics and quantitative models from large amounts of data.

SAMPLE SIZE AND RESPONDENTS’ POOLBack to top

Woman’s survey: The sample size was a minimum of 380 responses from each city. This was deemed to be sufficient to provide statistically reliable data with a five percent margin of error. All respondents were adult women (aged 18+) living in the cities on which they were polled. Fieldwork was undertaken from Aug. 27 – Sept. 8, 2014. The survey was carried out online.

YouGov attempted to interview a broadly representative sample based on age, whilst acknowledging that in many cities, only a small percentage of certain groups (especially the older age ranges) are online. Due to varied levels of internet penetration in each country applying weighting would have produced a high margin of error in certain markets, and therefore responses were allowed to fall out naturally. Due to these low levels of internet penetration in many countries, online panels generally have a higher average income and education level than the total population. For these reasons, no weighting has been applied to the figures.

Expert survey: The Thomson Reuters Foundation conducted this survey online from Aug. 27 – Oct. 13, 2014. Those interviewed included women’s rights activists, urban planners and architects working with a gender focus, social researchers with an interest in gender equality, academics focused on women’s rights, gender equality and violence against women and/or transport safety.

The questions addressed the same issues and were similar to the woman’s survey but experts were interviewed only on perception.

CRITERIA FOR CITY SELECTIONBack to top

The United Nations’ Population Division published population revisions figures in 2014.

The world’s 20 largest capitals were chosen, with Hong Kong excluded due to its status as a Special Administrative Region of China.

FURTHER EXCLUSIONSBack to top

The Thomson Reuters Foundation hired YouGov to conduct the popular poll on behalf of the Thomson Reuters Foundation.

Several cities were not feasible for online polling.

  • Baghdad. In recent months Iraq has suffered increasing instability due to war with the Islamic State militia. Due to this conflict Baghdad was excluded from the list.
  • Internet penetration in the Democratic Republic of Congo is 2.2 percent (according to the World Bank), therefore polling 380 women online in Kinshasa was unrealistic.
  • When polling globally, YouGov partners with three other polling companies. They use online panels where people can voluntarily join and receive invitations to participate. These polling companies could not find the sufficient number of women needed to poll Cairo, Dhaka, or Tehran.

A sample of a minimum of 380 respondents was needed to ensure a margin of error of ± 5  and guarantee a 95 percent confidence rate. A fewer number of respondents would have lowered the overall quality of the data.

The following is a list of capitals polled in order of population size.

  • Tokyo, Japan   
  • Delhi, India
  • Mexico City, Mexico
  • Beijing, China,
  • Buenos Aires, Argentina
  • Manila, Philippines
  • Moscow, Russia
  • Paris, France
  • London, United Kingdom
  • Jakarta, Indonesia
  • Seoul, South Korea
  • Lima, Peru
  • Bogota, Colombia
  • Bangkok, Thailand
  • Kuala Lumpur, Malaysia

EXCEPTIONSBack to top

New York

New York was included in the list for comparison reasons as the U.S. capital, Washington DC, was not in the list of largest capitals. New York is the most populous city in the United States with more than 8.4 million people and is home to the UN headquarters.

Cairo

The Thomson Reuters Foundation surveyed experts in Cairo, the world’s fifth largest capital, despite YouGov not being able to secure the necessary sample of women for a public poll. The results were not included in the overall ranking. Their rankings could have put the city in the top five most dangerous transport systems as deemed by experts.

DEFINITIONSBack to top

Transport

For the purpose of the poll, when referring to 'public transport', it means travelling on a bus, tram, metro or train or being at a station relating to one of these methods of transport. It doesn’t matter if these services are operated by state authorities or private companies.

Harassment

Apart from groping, there was no reference to any particular type of physical or verbal abuse. The survey was conducted in many different cultures and respondents were left to judge what would constitute harassment in their society.

ANALYSISBack to top

Six questions were taken into account for the overall ranking. These addressed:

  • travelling alone at night
  • the risk of being verbally harassed by men
  • the risk of being groped or subjected to other forms of physical harassment
  • trust that other passengers would assist a woman being physically or verbally abused
  • trust in authorities to investigate reports of sexual harassment or violence
  • availability of safe public transport.

The expert survey

All questions were ranked using a Likert scale. The left column shows the response and the right column the weight assigned to each answer where 5 stands for the worst outcome.

To what extent do you agree or disagree with the following statement? "It is safe for women to travel alone on public transport when it's dark."

Very safe: 1

Fairly safe: 2

Not very safe: 4

Not safe at all: 5

Undecided: 3

To what extent do you agree or disagree with the following statement? "Women are at risk of being verbally harassed by men when using public transport."

Strongly agree: 5

Tend to agree: 4

Neither agree nor disagree: 3

Tend to disagree: 2

Strongly disagree: 1

To what extent do you agree or disagree with the following statement? "Women are at risk of being groped or subjected to other forms of physical harassment on public transport."

Strongly agree: 5

Tend to agree: 4

Neither agree nor disagree: 3

Tend to disagree: 2

Strongly disagree: 1

How confident are you that other people would assist a woman who was being abused, either physically or verbally, on public transport?

Very confident: 1

Fairly confident: 2

Not very confident: 4

Not at all confident: 5

Undecided: 3

How confident are you that authorities would investigate if a woman reported being sexually harassed or attacked?

Very confident: 1

Fairly confident: 2

Not very confident: 4

Not at all confident: 5

Undecided: 3


To what extent do you agree or disagree with the following statement?  "Safe public transport is available in the city to which I am referring."

Strongly agree: 1

Tend to agree: 2

Neither agree nor disagree: 3

Tend to disagree: 4

Strongly disagree: 5

The woman’s survey

The Thomson Reuters Foundation used a mix of Likert scale and Boolean questions where 5 stands for the worst outcome and 1 for the best.

How safe, do you feel travelling alone at night in the city where you live?

Very safe: 1

Fairly safe: 2

Not very safe: 4

Not safe at all: 5

Undecided: 3

Have you been verbally harassed by men when using public transport?

Yes, I have: 1

No, I haven't: 0

Prefer not to say: excluded

Not applicable - I've never used public transport: excluded

Respondents choosing “Prefer not to say” and “Not applicable - I've never used public transport” were deducted from the original base number or respondents, leaving the question with a new base number. A new question score was calculated:

 

  • using this new smaller base number
  • multiplying the average score by 5, to convert it to an equivalent score on a Likert scale. Five has consistently been used as the maximum score for a negative meaning.

 

Have you been groped or experienced any other form of physical harassment when using public transport?

Yes, I have: 1

No, I haven't: 0

Prefer not to say: excluded

Not applicable - I've never used public transport: excluded

We applied the same scoring method as for the question above.

 

How confident are you that other people would come to your assistance if you were being abused, either physically or verbally, on public transport?

Very confident: 1

Fairly confident: 2

Not very confident: 4

Not confident at all: 5

Undecided: 3

How confident are you that authorities (e.g. police) would investigate it if you were to report that you had been sexually harassed or attacked whilst using public transport in the city where you live?

Very confident: 1

Fairly confident: 2

Not very confident: 4

Not confident at all: 5

Undecided: 3

To what extent do you agree or disagree with the following statement?

'Safe public transport is available in the city where I live.'

Strongly agree: 1

Tend to agree: 2

Neither agree nor disagree: 3

Tend to disagree: 4

Strongly disagree: 5

Don't know: excluded

 

For this last question, the number of respondents who answered “Don’t know” was deducted from the initial respondents’ base number. We then used the new smaller base number to calculate the mean.

We calculated an average score per question in the case of each city. We then calculated the average score per city using the six means.

INDEXING THE SCORESBack to top

StarMine, a Thomson Reuters company specialised in models and analytics, combined the results of the two surveys to create an index.

The first step was to create a scale from 0 - 100. All responses were re-scaled such that the safest raw score possible gets an Index score of 100, and the lowest raw score possible gets an Index score of 0.

As such, an index Score of 100 would translate to everyone saying the city is “very safe”, 75 = fairly safe, 50 = undecided/neutral, 25 = not very safe, 0 = not safe at all. Each city got an index score from the women’s  poll and the expert poll. The two scores were averaged into the Combined Index Score, which gave the final Safety Index ranking. The Index was constructed in this way to provide a metric that was clear, straightforward to interpret, and easy to translate to qualitative descriptions of safety.

TODAY'S TOP STORIES
Todays top stories

Child hunger and death rising in Zimbabwe due to drought

EDITORIAL PICKS
TOP SPOTLIGHT
NEWS IN PICTURES
Libyan forces fire weapons during a battle with IS fighters in Sirte
Members of the Italian Navy hold a child after disembarking in Palermo
People watch the flooded Jinsha River in Diqing, China
A young Palestinian crawls through a tunnel during a military-style exercise at a summer camp organised by the ...
TOP SLIDESHOW
LATEST SPOTLIGHTS