by eila in Civil and Human Rights, Economy & Poverty, Media, Somerville By The Numbers
Posted on September 7, 2011 at 7:33 pm
Last Modified on September 10, 2011 at 11:57 am
loading…
September 6, 2011: The City of Somerville has generously published the raw data spreadsheets from the much-touted “Happiness Survey” online.
The Community Access Project heard from low-income, disabled and senior residents, who felt excluded. Instead of receiving a Happiness Survey included with our Street Census, we received a sheet of information heralding the new bike sharrows.
How were the families and individuals who received this written survey through the mail chosen? [update: see comment #4]
When those respondents are sorted by annual incomes reported, we discover that nearly 60% reported annual incomes of $50,000- plus.
More startling: over one-quarter of the Somerville Happiness Survey responses come from households with incomes of $100,000 and up.
Here’s the breakdown:
Total # of responses: 6,168
25.7% : annual income of $100,000 and up.
8.6% : annual income of$50,000 – $59,999.
8.4% chose not to respond to this question.
8.3% : annual income of $40,000 – $49,999.
7.5% : annual income of $60,000 – $69,999.
7.4% : annual income of $30,000 – $39,000.
6.8% : annual income of $20,000 – $29,000.
6.4% : annual income of $70,000 = $79,999.
5.7% : annual income of $80,000 – $89,999.
5.7% : annual income of $10,000 – $19,999.
5.3% : annual income of $90,000 – $99,999.
4.1% : annual income of Less than $10,000.
A pie chart with a spreadsheet is online at http://tinyurl.com/3gcen7w
loading…
Did the city comment on how it selected which censuses (censii?) included the happiness survey?
loading...
I wrote a long comment, but it didn’t get published. This comment is to test whether or not the blog works.
loading...
OK, so it seems to be working now. First, thanks for your interest in the wellbeing data. As far as I know, a survey went out with every single census form. We knew there would be an inherent bias, though (because certain people are more likely to send back the survey than others), so we also did a phone and email survey of around 300 randomly-selected residents. That group is closer to Somerville overall, which has a median salary of approx. $70k: http://bit.ly/pV9npw.
In addition, we used survey weights to help correct for sampling bias. This allowed us to arrive at results that more closely resembled the diverse population of Somerville. You can read about our methodology in the report.
In sum, the data are not perfect, but they give a pretty good snapshot of the city. In the future, I encourage you to call me if you have questions about the data we publish. Thanks!
Daniel (617)625-6600
loading...
Dear Daniel, Thank you for your response, your ongoing great work as ResiStat Coordinator (and, the logo you designed for the Young Somerville Advisory Group is very fresh).
It seems nearly impossible for this “happiness survey” to resemble the diverse population of the Ville, considering our reality of hidden, unvoiced and invisibilized populations, such as many public housing residents, group quarters residents, residents who have been waiting unbearably long times for USCIS to act on their applications… etc..
In this post, I’m pointing out that residents with low-incomes have expressed feeling excluded – and we are wondering if, indeed, some or many of us may have been selected out of receiving the “Somerville Happiness survey” in our mailboxes.
SO, first, your office may not have all the information regarding who received the surveys? I know of residents who first received a mistakenly addressed census form without the survey; and then the corrected form arrived several weeks later- still, without a survey included.
This happened to residents living in rental buildings with less than 8 units, so it doesn’t seem to have been because of M.G.L. Ch. 51 § 10A. (http://www.malegislature.gov/Laws/GeneralLaws/PartI/TitleVIII/Chapter51/Section10a)
That statute, relevant to the annual street census, defines“multi-dwelling unit” as: “shall mean condominiums of any size and any residential apartment complex consisting of more than eight rental dwelling units” and further states: “In the case of the multi-dwelling unit , the owner or his designated representative at such place… shall provide the required information.”
So here’s the first question: In Somerville, do residents living in public housing units each receive an annual street census form?, or does the manager of each public housing building fill it out for the residents, instead?
loading...
Eila,
Thanks again for your interest in this topic. I’m glad you are engaged and keeping tabs on the data.
I spoke to our Elections department and you are correct: census forms were not mailed to individual public housing units (nor did they go to units in nursing homes or the YMCA). On the other hand, a census form went to each unit in 8+ unit buildings.
The other thing Elections told me is that a survey went out with each census form in the first mailing. Subsequent mailings did not contain the survey, which may be the reason certain people felt they had been excluded.
I’m glad you brought the public housing issue to my attention. Elections staff told me about that at one point, but it slipped my mind.
With all that said, I still think the 6,000+ responses we received from the census survey are reasonably representative of the city as a whole. The reason I think this is because after we applied survey weights to remedy the demographic imbalance, the average responses to our key questions (how happy are you, etc.) were strikingly similar to the phone and email responses, which were randomly pulled from the population and likely included people from public housing.
Still, I think your point is a fair one. One cannot say that our data is perfect, nor that it reflects the opinions of all city residents. It is good to view the data with a healthy dose of skepticism.
Reaching out to underrepresented populations is definitely one of SomerStat/ResiStat’s goals. We will be working harder next time (hopefully with more resources) to collect a better sample. Thanks again for your feedback!
Daniel
loading...
Running the numbers on your spreadsheet, Daniel, it appears 28% of people in the follow-up survey are >100K. But the numbers are still very consistent with a median income of 70K in the city, in both surveys. About half of respondents were above that number, half below. I don’t see the problem here. Weighting of course improves things, but this set of numbers does not necessarily exhibit evidence of extreme bias.
loading...
Actually, running the numbers from that follow up phone/email/Facebook survey- which only included 9 people who said they were living in public housing, btw- we get the following:
Out of 560 responses:
133 people would not answer the income question.
123 reported income btn. $20,000 and $60,000.
126 reported an income between $60,000 and $100,000 and 126 reported an income between $100,000 and $200,000.
The outliers are those reporting an income above $200,000 (N =34), and those reporting incomes below $20,000 (N =44).
So, it appears that the most frequently occurring value in this set of data was: those that did not report their incomes!
loading...
The importance of the happiness question is that it was asked, and will be asked again, and again. The value of the initial data is that people answered, regardless of income. The point of the question is to build a common community, and to identify problems worth that community’s effort to resolve. Far too often we get distracted with the nuance and forget these points.
That issue of community might be even more pointed if we survey at the ballot box next year. A few elections go I remember encountering some neighbors after voting and asking if they’d voted. “Oh, didn’t you notice, we never vote!” was their response. “OK, more for me,” was mine.
Happiness takes many forms – for some it’s voting, and for others is not-voting. The point of a government – any government – asking about happiness is to engage the larger, uninvolved population in its community role, to build that community and care for whether neighbors care about each other. For some that will mean voting, and for others it means they trust my values enough to let me vote on their behalf. Pretty dangerous, given the divisions in the nation; pretty safe, given that we live in the most densely populated city in the Northeast.
loading...
Thanks again, Eila. The problem of underrepresentation of certain groups in our survey data is real, and I am glad others are aware of it. Weighting the data can help, but it is not a perfect solution. Part of the reason we released the data is to invite this type of scrutiny to our process and results. Between now and our next survey, I will be researching how to better reach underrepresented groups. If you have any suggestions, I welcome them.
Joe, I agree that this is about building community as much as it is about collecting data. We really hope that happiness and wellbeing become communal priorities.
loading...
the point of the post was not about income- it was about who was offered the opportunity to have a voice, during this much-touted “community-wide” process.
The fact of the matter is that, of the approximately 32,500 households in Somerville, over 10% did NOT receive this “Happiness Survey” in their mailboxes.
That’s people, not numbers. It’s a vibrant tribe of neighbors, mainly elderly, disabled, “Special Needs,” and lowest-income families living in public housing units. It’s about 3,226 households, plus folks living at the various assisted living facilities scattered throughout the city, plus folks housed at the Y- and who else? that continues to fall through the cracks?
This was absolutely NOT the fault of the excellent staffers at Somer- and Resi-Stat. They do the hard work; and, in this case, made an understandable mistake.
“We need to change our mind-set in how we serve people,” said Mayor Joseph A. Curtatone. (May 1, 2011, page A1, New York Times: How Happy Are You? In a Boston Suburb, It’s a Census Question.)
It’s hard for me to believe that the Mayor was unaware that residents with the lowest income, most substandard housing conditions, and least opportunities- were not even slightly represented in this much-touted “Happiness Survey.” Sure, the ordinary public doesn’t know how the annual street census forms are distributed and completed- but the Mayor, the Elections Commissioner, and anyone who decides what’s sent out with those forms- knows it well. Every year, any policy that results from the annual street surveys utterly leaves out the needs of our lowest income residents.
So, that’s a pretty pony: “Happiness comes in many forms-for some it’s voting, and for others is not-voting.” But In Somerville, as I’ve shown with photos that are undeniable- even many voting precincts are still locked to people who cannot use stairs. Will the Teele Square Fire Station be available for ALL voters to readily use, this Monday? Or is it OK with the rest of the community to look on, while disabled members are expected to jump through hoops just to get inside?
All it takes for thousands of community members to be isolated, marginalized, discriminated against, and kept outside the fence, looking in- is for everyone else to pretend it’s about numbers, not people… and to remain mum.
loading...
In my above comment I mentioned the Ward 7 primaries coming up this Tuesday- but I said “this Monday,” instead. My mistake, the Ward 7 primaries are Tuesday September 13. As Joe says, voting is a choice to be involved…
In Ward 7, residents living at at least 3 housing developments were excluded from the Happiness Survey:
- Clarendon Hill, aka “North Street”
-Visiting Nurses Assisted Living Facility at 405 Alewife Brook Parkway
-Weston Manor- 15 Weston St.
Thank you, Daniel, for forthrightly acknowledging the very real problem of underrepresentation of certain groups. Your sincere commitment to the community is obvious.
loading...
this so called survey was not anonymous as it claims to be. the survey was mailed with a census form that required you to give your name and address. the city only provided one envelope. so, if you filled out both forms and mailed them back in the one envelope, they knew who filled out the”anonymous” happiness survey.
therefor, many citizens did not send back the form because they understood there was no anonymity. i believe that the survey was developed in part for political purposes. the chief executive of this city is known to have a ‘big book’ which contains among other things. names of supporters, patronage employees and people helped by the city during the last 8 years. i believe the people who responded positively to the survey are now listed in the “big book”as friends of the chief executive who would most likely vote for him.
also, this list could be given to friendly incumbents to use in their campaigns.
are you happy now that you know that you were duped by your government? kevin crowley
loading...
Hi Kevin,
I appreciate your concern, but in fact the survey was anonymous. Staff from the Elections Department scrupulously separated the surveys from the census envelopes before giving us the forms. Furthermore, as project manager, I am the only person who has seen the complete dataset. We are only interested in larger trends – not individual responses.
Thanks
loading...
I’m bound to be happier than you are, since you’ve been so eager to identify yourself as enemy to those same people. Wow, such insights!
loading...
News flash: Rich people live in Somerville, too!
56% of the mailed surveys who responded to the income question had income below 70k (the median income). This is an outrage! It should have been only 50%!
“the point of the post was not about income-” ummm…., if it wasn’t the point of the post, it WAS the point of the boldfacing in the post, right?
loading...
That’s right Bill- plenty of wealthy people have addresses in Somerville, and there’s certainly nothing wrong with that. And yes, the blog discusses income disparities as a signifier of who was excluded from participating in the survey.
Did you notice how many news stories generated since April about the Happiness Survey gave the impression that everyone in the community had a chance to represent? (For example, see “One, big happy family of 80,000″ – Boston Globe, August 28, 2011 )
Does it concern you that the public officials, who were aware of the unintentional media misrepresentations while they were being published, chose to not correct the record?
The stories and Final report give similar impressions of which gaps were analyzed, “After correcting for underrepresented groups, such as young people and minorities, a plurality of respondents scored themselves an 8 or higher.”
Fact is, over 10% of us- mainly lowest income elderly, disabled and families living in public housing- did not get to score at all.
That’s not merely “some” of us.
Even when SomerStat did the follow-up phone and email survey:
out of those 560 residents surveyed,
only 9 reported living in public housing.
The statistical weighting was done PRIOR to a fuller awareness of who was not represented. I believe Daniel H. of SomerStat, who has promised to research how to better reach underrepresented groups, and am very grateful for his forthright acknowledgement of these issues!
But, I’m guessing, most residents who were left out may not yet be feeling the love.
Sept. 22: Boston Globe correspondent Danielle Dreilinger has published “Activist raises doubts about Somerville happiness survey” at http://www.boston.com/yourtown/news/somerville/2011/09/activist_raises_doubts_about_s.html?camp=localsearch:on:twit:HLsomerville&dlvrit=155349
loading...