S. Studies that continue to seek a link between childhood SEP and adult BMI have tended to use "representative national samples," the authors explain. There is "regional variation in the SEP-adiposity relationship," they explain, that "may make a difference in a regionally constrained sample such as ours" (Trotter, p. 1092). "Regionally constrained" is a legitimate conclusion as to why there is such a variance between the Los Angeles sampling and the NHANES data collected between 1999 and 2004 and their own sampling. The NHANES reported a higher (31%) rate of obesity than the Los Angeles findings (20%). These authors admit the Los Angeles data is "regionally constrained."

Measurement: The reader loses faith in the veracity of this research when the authors explain (p. 1092) that there is "missing data" (up to 18% of data missing from "working class status"). Moreover, the "self-reported" weight of participants leaves a reader wondering just how far off those numbers were from what people actually weigh. There is a "downward bias," they explain, associated with "self-reported weight"; i.e., the heavier a person is, the more that person will tend to understate his or her weight. That's not scientific, but it's a fair assumption. To be empirically accurate, the authors should have put people on the scales, if that was practical to do.

Data Collection methods and analysis: Meantime, the authors mention that the "…growing prevalence of overweight and obesity may be eroding...

They say they believe their measures "are reliable" (p. 1093) but then there is no accounting for the change in parental status (divorce? Different parents?) since the child was 14.
Analysis: The authors report (p. 1093) that "future studies should aim for more precise measures of childhood SEP" and that is a justifiable conclusion. They state that their research is "subject to the problem of imprecision" which is honest. Also, they admit their research lacked the "mediational measures, such as childhood or family health behaviors or health status." That is like admitting that their entire approach was flawed.

Limited and justifiable conclusions: When authors say in one sentence they believe their data to be "reliable" and in the next sentence admit "imprecision," the contradiction raises serious questions of credibility. And when authors measure educational levels by jumping from a high school diploma to a college degree -- with no designation of "some college" or "associate degree" as a measure of some post high school education -- the reader wonders how well thought out this research must have been.

Works Cited

Trotter, LaTonya J., Bowen, Deborah J., and Beresford, Shirley a.A. (2010). Testing for Racial/Ethnic Differences in the Association Between Childhood Socioeconomic Position

and Adult…