Science of Cities

How Can a City Measure Its Happiness?

Santa Monica will begin to survey residents about their well-being next month. (AP Photo/Richard Vogel)

At one time, questionnaires about well-being were the province of mental health professionals. But in recent years, a growing number of city governments have been getting into the game. Last year, Santa Monica, California won a Bloomberg Philanthropies Mayors’ Challenge grant to create a “Local Well-Being Index,” based in part on a survey it plans to administer next month. Other cities, including Seattle and Nevada City, Calif., are at various stages of implementing the idea as well (with different levels of direct municipal involvement). All share a goal that some see as unsuitable for government and others consider its fundamental task: to make citizens happier.

As more cities begin to undertake such surveys, it’s a good time to look to the municipal government that claims to be the pioneer: Somerville, Massachusetts. In the three years since Somerville launched its program, what has it learned? As a relative veteran, what lessons can it share with other cities that have taken up similar initiatives?

Somerville sent out its first questionnaire, which a city report calls “part psychology survey, and part customer satisfaction,” in 2011 as part of the city census. Inspired by the examples of Bhutan and the UK, both of which had begun to try to assess citizen well-being, Somerville’s staff plunged into the growing body of academic research on happiness, with additional input from Harvard psychologist Daniel Gilbert. They designed a survey that asks about general subjective well-being (both “How happy do you feel right now?” and “How satisfied are you with your life in general?”) as well as satisfaction specifically with Somerville as a place to live, including ratings of services such as schools and law enforcement.

The surveys yielded a variety of illuminating findings, including that city aesthetics seemed to play a surprisingly important role and residents were significantly happier with a single-stream recycling program.

But how exactly do you conduct such a survey to get the most accurate, representative answers?

The first time around, the city sent the survey to all of the city’s roughly 27,000 households. But the response rate was low — 22 percent. This is problematic because the people who have the time and inclination to fill out a survey might skew the results in a particular direction. “That’s when statisticians would definitely get nervous about is it representative, who is it leaving out?” says Daniel Hadley, director of SomerStat, the program that runs the initiative. They also conducted phone and web surveys to try to compensate for the low response rate.

This chart shows data from Somerville’s 2013 happiness survey.

For the second survey, in 2013, the city took a different approach. They randomly selected about 400 residents, and then doggedly followed up with them. First they mailed the survey. Then they mailed it again, this time with a two-dollar bill enclosed. A two-dollar bill? I asked Hadley. Does this make people feel obligated? Possibly, he said, or it might just catch their attention. “It was a small price to pay for a better response rate.” Then, finally, they sent out a reminder postcard. They got a response rate of 48 percent. This amounted to only 194 completed surveys, compared with more than 6,000 from the first year. But, though Hadley acknowledges it’s “totally non-intuitive,” statisticians will tell you that the smaller, random sample is likely to be more representative.

The results were still imperfect — African-American residents, for example, were under-represented. SomerStat tried to control for that in the statistical analysis, but they hope to make an effort to get more responses from black Somervillians next time. (As of yet, they have no definite plans for how to do so.)

Last fall, the city explored a partnership with the H(app)athon Project, which is building a smartphone app intended to facilitate happiness measurement. But Hadley told me the collaboration “kind of fizzled,” for the same reason that the first survey was flawed. At this stage, smartphone data would be unrepresentative, probably leaving out the elderly, for example.

In response to the survey results, Somerville has made some policy changes. Other than rolling out zero-sort recycling city-wide, they’ve also allocated more resources to the Traffic and Parking Department, and Hadley said they’ve begun to invest more in infrastructure in East Somerville, one of the less advantaged (and less happy) areas of town. But Hadley admits there are challenges with translating survey results into action. “One of the biggest lessons is just how nebulous a concept happiness is and how difficult it really is to influence through policy,” he says. “One of the things we’ve learned is that it’s easier to measure than manage happiness.”

Despite this caveat, Hadley stresses that the undertaking is eminently worthwhile, given the relative ease of conducting the surveys. “It’s not as hard as it seems to do a good, simple survey of your residents,” he says. “We did it all in-house and we did it all for under $4,000. It’s totally doable.” And the more cities that begin to do the surveys, the better, because they can compare results and learn from each other. For example, Somerville’s average rate of satisfaction was 7.5, but this number is hard to interpret without the context of responses from other cities.

Santa Monica, meanwhile, is planning to conduct a web-based survey in September. (They may supplement the web survey with other ways of reaching people.) Officials plan to combine the survey results with other existing data on public safety, enrollment in recreation classes and transit options in order to formulate the well-being index, which they will then share with citizens and use to shape policy.

Focusing on happiness may still sound like a strange exercise for government, at once too touchy-feely and too ambitious. But as more cities experiment and cross-pollinate ideas, they will presumably identify best practices and refine their methodology. And it may not sound so strange for long.

“Twenty years ago, sustainability was kind of a bit nebulous and a little crunchy,” says Julie Rusk, assistant director of Santa Monica’s Community and Cultural Services. Now, she said, it’s “become very mainstream. We kind of hope and expect that the science and research around well-being has a similar trajectory in the future to become mainstream.”

The Science of Cities column is made possible with the support of the John D. and Catherine T. MacArthur Foundation.

Tags: datahappinesspolls