Stockton: Fat City? Probably not

2010 was not a good year for the City of Stockton in terms of titles as we were named most miserable, most illiterate* and most obese. If you look at these dubious distinctions, you probably come to the conclusion that Stockton is not a nice place to live. How could three reputable studies all separately come to the conclusions that Stockton is the worst in the country for something? To the untrained eye, being at the top (or bottom) of these lists is hard proof that reinforces negative stereotypes about Stockton. But when you take closer looks at these studies, you begin to realize that these lists are not always what they are cracked up to be. I would like to set the record straight: Stockton is not the worst city in the country for ANY of the aforementioned titles.

Today, I will tackle Gallup’s obesity rankings from 2009 which had Stockton ranked as the most obese metro area in the country. First and foremost, it should be noted that since our title of most obese city was bestowed upon us in 2010, we have fallen dramatically in more recent rankings. From a high of 34.6% obese in 2010, we now have an obesity rate of 25%. This puts Stockton right out of the top 100 entirely. This is great news, but it calls in to question the methodology used by Gallup to come up with these ratings because, to me at least, it seems like a stretch to think that a city with the highest obesity rate in the nation could reduce its obesity rate by a whopping 10% in just two years.

Another red flag is that the rankings themselves vary widely from year to year. For example, in three years, 23 different cities have been featured in the top ten for obesity rates, with only two areas making the list all three years. Take a look:

Metropolitan Statistical Area (Obesity rate)

2009

2010

2011

Stockton (34.6)

Evansville, IN-KY (37.8)

McAllen-Edinburg-Mission, TX (38.8)

Montgomery, AL (34.6)

Hagerstown-Martinsburg, MD-WV (36.5)

Binghamton, NY (37.6)

Visalia-Porterville, CA (34.1)

Huntington-Ashland, WV-KY-OH (34.6)

Huntington-Ashland, WV-KY-OH (36)

York-Hanover, PA (34)

McAllen-Edinburg-Mission, TX (34.5)

Rockford, IL (35.5)

Flint, MI (33.9)

Reading, PA- (33.8)

Beaumont-Port Arthur, TX (33.8)

McAllen-Edinburg-Mission, TX (33.7)

Montgomery, AL (33.4)

Charleston, WV (33.8)

Bakersfield, CA (33.6)

Harrisburg-Carlisle, PA (33)

Lakeland-Winter Haven, FL (33.5)

Lynchburg, VA (33)

Cedar Rapids, IA (32.9)

Topeka, KS (33.3)

Huntington-Ashland, WV-KY-OH (33)

Shreveport-Bossier City, LA (32.5)

Kennewick-Pasco-Richland, WA (33.2)

Kingsport-Bristol, TN-VA (32.9)

Beaumont-Port Arthur, TX (32.4)

Reading, PA (32.7)

With a condition like obesity, I suspect that the percentage of people who are obese in a city would not fluctuate greatly year to year. However, the way the cities are ranked here seems to indicate that cities seem to be prone to annual swings in obesity. This really makes no sense: How could the most obese city in the nation just drop off from the top ten list in the following year, as Stockton in 2009 and Evansville in 2010 did?

To get to the bottom of this, I contacted the good people at Gallup to see if they could help me understand their methodology for their Well-Being Index (which you can play around with here). Here is what they told me:

“Gallup requires at least n=300 cases for MSA-level reporting.  Each MSA has a unique sample size, so the expected error range for any given proportion will vary according to the response itself as well as the sample size for the population.  (In other words, the expected error is not the same for each reported city.)  In Stockton’s case, we get between n=550-600 in one year of measurement, so our obesity results for it are pretty stable: about +/-4.4% with 95% confidence, design effect included.”

What all this wonky nonsense means is that Gallup makes sure that they are collecting enough samples from Stockton to make sure that their analysis is sound. With the number of Stocktonians polled, they can predict with 95% confidence that their data is within a 4.4% margin of error. In this business, this result is known as “highly significant,” or in layman’s terms, very accurate.

So was Stockton really the most obese city in the country in 2009? I still say no. Despite the high confidence levels that Gallup uses, there is still a margin for error that is large enough to explain, at least in part, some of the variation from year to year. With a 4.4% margin of error, it is plausible that Stockton was not quite as obese as the data pointed out in 2009. Just a 2 point change, within the margin of error, would have knocked Stockton out of the top ten easily that year. I believe that this high rating in 2009 is an aberration that can be contributed to this small margin of error. To be sure, Stockton’s obesity rate was high 2009, but I don’t  think it was the worst in the country, especially when you compare Stockton’s 2009 rate to more recent years, it appears that 2009 is an outlier. The cities at the top are so tightly bunched, separated by just fractions of a percent, that these small margins of error can explain the seemingly strange variations in the top ten every year.

In conclusion, Gallup does have strong data, but to use their well-being index as a year-to-year ranking system of obese cities is a bit misleading. Were we obese in 2009? Yes. Were we the most obese? Maybe not. Even if Stockton were the most obese in 2009, we should really be looking at the trends. As of 2011, Stockton is now around the national average for obesity rates, which still is not desirable, but at least we are trending downward. A 10% drop in obesity rates in just three years is a positive step forward, no matter how you look at it.

*the most illiterate title was actually bestowed in 2007. My mistake.

Advertisements

Tags: ,

Categories: Community Commentary

Author:David A. Garcia

David A. Garcia founded SCL in March of 2012. He holds degrees from UCLA as well as Johns Hopkins University and currently works as the Chief Operating Officer at Ten Space in Downtown Stockton, and previously worked as a researcher/analyst for a congressional agency in Washington DC. The views expressed here are solely of the author.

3 Comments on “Stockton: Fat City? Probably not”

  1. Sean
    April 24, 2012 at 8:13 am #

    Sounds suspicious to Sean…if we take what Gallup is saying at face value every public health researcher in the country should be beating a path to Stockon’s front door to try and understand such a dramatic drop in obesity rates. Also, with a 4.4% confidence interval Stockon could have been at 30.2% for 2009 and 29.4% in 2011 which sounds much more reasonable

Trackbacks/Pingbacks

  1. Forbes’ folly: why the Most Miserable City rankings are useless « Stockton City Limits - May 22, 2012

    […] recent weeks I have addressed how Stockton is depicted in national “rankings.” I looked at Gallup’s obesity index as well as a study ranking Stockton at the bottom of the “Most Literate Cities” list. In both […]

  2. Fighting obesity with smart growth: how Stockton’s sprawl contributes to the city’s high obesity rate « Stockton City Limits - June 21, 2012

    […] obesity is a major problem. While we were “named” most obese in 2010 (a title I debunked here) our current obesity rate is about 25%, right around the national average, which is not by any […]

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: