Tuesday, January 30, 2007

A Wish For 2007: Less Truthiness

Another wish for 2007, this one from Dr. George Lord:

Getting to Truthiness Through Cherry Picking

Wikipedia, the source of modern neologisms, defines “cherry picking” as:
“In the literal case of harvesting cherries, or any other fruit, the picker would be expected to only select the ripest and healthiest fruits. An observer who only sees the selected fruit may thus wrongly conclude that most, or even all, of the fruit is in such good condition.

Thus, cherry picking is used metaphorically to indicate the act of pointing at individual cases which seem to confirm a particular position, while ignoring a significant portion of related cases that may contradict that position.”
It goes on to suggest that at times cherry picking is appropriate:
“When is cherry picking appropriate? When a person is assigned to advocate a particular position, then cherry picking is entirely appropriate. Lawyers in a criminal case are one such example, where it is assumed to be the responsibility of the opposing counsel to present any contrary data. However, when a person with a supposedly neutral position cherry picks, that is inappropriate. Examples would be journalists, scientists, and judges.”

Perplexed


Memphis and Memphians seem to be very adept at picking those cherries which present a negative view of the city. As one who regularly goes in and picks the stalls in which his horses reside, I think this might more accurately be called “horse apple picking” - choosing those statistics which serve to paint the smelliest picture of the community. In their heart, it appears that the attitude is “If anyone cannot do it, we can’t.” In other words, they seem to want to bash themselves and be bashed by others.

This is particularly perplexing to me as one who was in Memphis in the early eighties and has returned in the past year to a city which always held a soft spot in my heart. I like Memphis and I like most Memphians. While we have our share of problems, things look good compared to the 12 years I spent in Flint, Michigan.

I say all this to set up my complaint that there is a specific “cherry picked” statistic which has been driving me crazy since I returned: “Memphis has the highest infant mortality rate in the country.” This was repeated in the Commercial Appeal recently in an article on recently passed legislation to unearth the causes of infant mortality.

Please count me in as one who supports any research on infant mortality and ways to curb this tragedy. Over the years as I taught courses on Social Inequality at various universities around the country, I would frequently make the claim that if one wants a single indicator of the level of inequality in a country or community; one has only to look at the infant mortality rate. I believe that is still a fairly accurate statement.

Getting to the facts

The U.S is ranked 42nd on infant mortality among the 225 nations in the CIA database. If Shelby County were a nation, we would be ranked 82nd. While I won’t challenge the 1993 to 1998 data from which researchers claimed Memphis had the highest infant mortality rate (at 15.4 deaths per 1,000 live births), I will point out that much of what we know about infant mortality we know at the county level. It is at the county level that these statistics are most often reported. In 2004, the most recent year for which such statistics are available, Shelby County has the 18th highest infant mortality rate (12.8) of the 95 Tennessee counties. In fact nearby Tipton County is ranked 2nd, with 21.5 deaths per 1,000 live births.
None of these numbers are anything to be proud of.

However, there are many places with higher rates. Take Issaquena County in Mississippi, this delta county is just across the river from the northern most point of Louisiana. The region is well known to demographers and others who are trying to understand inequality; it is part of the “Black Belt,” that string of counties which was once the major home of the slave dominated plantation economy. The county had a 2004 Infant Mortality Rate of 71. This is of course a county where the per capita income for Black residents in 2000 was $6,813; just 54% of that for the same sub group in Memphis that year.

While I am pointing to the correlates of infant mortality, consider the following about Shelby County, in 2004: 52% of children are getting free and reduced lunches in the public schools (ranked 20th highest among TN counties), 19.6% of children are getting Families First Grants (ranked 1st among TN counties), 36% of children receive food stamps (11th among TN counties), and finally our per capita income was $35,237 (3rd highest in the state). While we have one of the highest per capita income levels in the state we also have very high rankings on indicators of poverty. In other words we have a very high level of inequality.

Finding the Causes

As CDC suggests in a 1999 report: “To develop effective strategies for the 21st century, studies of the underlying factors that contribute to morbidity and mortality should be conducted. These studies should include efforts to understand not only the biologic factors but also the social, economic, psychological, and environmental factors that contribute to maternal and infant deaths. A thorough review of the quality of health care and access to care for all women and infants is needed to avoid preventable mortality and morbidity and to develop public health programs that can eliminate racial/ethnic disparities in health. Preconception health services for all women of childbearing age, including healthy women who intend to become pregnant, and quality care during pregnancy, delivery, and the postpartum period are critical elements needed to improve maternal and infant outcomes.”

What is the Truth?

As Stephen Colbert suggests, we must be on the lookout for truthiness. “Truthiness is a satirical term popularized by Stephen Colbert in reference to the quality by which a person claims to know something intuitively, instinctively, or "from the gut" without regard to evidence, logic, intellectual examination, or actual facts (similar to the meaning of "bellyfeel", a Newspeak term from George Orwell's Nineteen Eighty-Four).” Saying we are the worst community in the country regarding infant mortality seems to be but one more way of bashing Memphis without getting at the truthiness of the statement.

I truly believe in data and good data is truly important if we want to move forward. The strongest, most effective priorities and strategies for Memphis are forged in open discussion and deliberations and rooted in the facts. This is not necessarily intended for us to reach consensus, but to arrive at sufficient agreement to enable action that will contribute to the common good. As we enter the twenty-first century, citizens have fewer and fewer public arenas for deliberation and for exercising our civic commitment.

If Smart Cities will allow it I hope to come back from time to time with an occasional “golden horse apple” award. This effort will be to point to the “horse apple” picking which occurs as part of the public discourse in the community. The attempted response will be to put the “horse apple” in the context of the whole basket of fruit we call home. If we are lucky, I will contribute to the elimination of truthiness in our discussions.

2 comments:

city watch said...

Dear George,
Could you explain the $35,237 per capita income for Shelby County. Is this for people or are businesses included? Is Fred Smith included? If there is an average of 2.3 people in each Shelby County household, then the "average household" would make $81,045. [Or is it "households on average"?]

Anonymous said...

Dear Chuck,

You ask a good question and embedded in the question is a very serious part of the answer. Yes Fred Smith is included in the numbers as are the other multi-millionaires who claim Shelby County as home.

The important thing in looking at income distributions is to understand that the correct measure of central tendency is the median not the mean, or average. The median is the number which half of the population is above and half is below. This helps mediate the fact that the distribution is highly skewed; that is very few at the top and many at the bottom.

I’m not sure where you got the $35,237, so I cannot respond to it. However, one of the most recent sources for such information on Shelby County, or any community in the US is the American Community Survey. The ACS is conducted at the mid point of each decade, between the decennial census. However, keep in mind that it is a sample and therefore subject to sampling error. You can go here http://factfinder.census.gov/home/saff/main.html?_lang=en and input any city, county, or state to get the estimates derived from this survey.

Estimates from the 2005 American Community Survey (number in parenthesis is the estimated margin of error):

Median household income for Memphis was $33,244 (+/- $1,411); for the US it was $46,242. Median family income Memphis $40,111 (+/- $2,265); for US it was $55,832. Per capita income in Memphis $20,279 (+/- $759); nationally it was $25,035. All amounts are in 2005 inflation adjusted dollars. The same numbers for Shelby County are Median household income = $40,839 (+/- $1,080), Median family income = $50,715 (+/- $1,422), and median per capita income = $24,352 (+/- $674).

Income is defined as:
"Total income" is the sum of the amounts reported separately for wages, salary, commissions, bonuses, or tips; self-employment income from own nonfarm or farm businesses, including proprietorships and partnerships; interest, dividends, net rental income, royalty income, or income from estates and trusts; Social Security or Railroad Retirement income; Supplemental Security Income (SSI); any public assistance or welfare payments from the state or local welfare office; retirement, survivor, or disability pensions; and any other sources of income received regularly such as Veterans' (VA) payments, unemployment compensation, child support, or alimony.

By the way, the average household size in Shelby County is 2.55 (+/- 0.03) and the average family size is 3.19 (+/- 0.07).


Hope this helps,

George