The Love Value of Food - page 2

Currently 0/5 Stars.
1
2
3
4
5

You may use the stars on the left to rate and leave feedback for the current article. No registration is required. Waiting for 5 votes

Part 2: Why Use South African Data?

One of the first questions that comes to mind when reading the research is why South African data was used to test the "robustness" of results drawn from US data. A friend suggests:

South African data was used because there's lots of it (we scientists tend to take our data from wherever we can get it), and because, unlike the US data, the South African household expenditure data is broken down by category of food. The researchers wanted to see if the extra money biological mothers spend on food is for nutritional or junk foods.

On the other hand, the South African data has less detailed information on parental relationships, including information only on whether the biological mother is present in the household or not.

and goes on to note that from the online research there appear to be instances where the researchers ignore the results of their own data analysis when they do not support the study's premise.

Observations

More of this Feature
• Part 1: Press, Study Slam Adoptive Parents
• Part 3: Our Response

Elsewhere on the Web • Mothers & Others: Health
• The Study - Full Text

[I want to thank a friend, who prefers to remain unnamed, for professional opinion and insight into research techniques and data interpretation. I am not a research scientist, and it has been extremely valuable to have input from a data analyst - a PhD scientist at a national laboratory with 14 years professional experience, more than seven of which are in heavy-duty data analysis with large datasets.]

The following are just a few observations, and they are based on the online version of the study, without access to raw data used by the researchers.

The US data were for 1968 through 1985. That means the data are 15-32 years old. Attitudes about adoption, step, and foster parenting have evolved significantly so even if the finding is valid for the period 1968-1985, that doesn't mean it's valid for 2000. One way to explore this would be to look at the data for separate years to see if there's a time trend in the finding.
South African data was for 1995.
The study did not find that adoptive, step, and foster parents feed their children less. It only found that there is a small, but statistically significant, difference in what non-biological households spend on food. It could very well be that they feed their children exactly the same but that they're better bargain hunters, or less inclined to purchase expensive junk foods, etc. which, to this writer, seems a reasonable possibility.
The study assumes that - whatever the reason for less expenditure by adoptive, step, and foster parents - the reason is the same for all three groups. I see no reason to assume that this is the case.
The researchers claim to have somehow checked for and eliminated the possibility that stepmothers spend less on food because the child's food is sometimes being paid for by the non-custodial parent. I didn't quite understand their check, but it didn't sound convincing that they've eliminated this possibility.
The authors admit the possibility that spending less on food might actually be good for kids if the parents are spending less on unhealthy foods. They then claim to have eliminated this possibility by filling in their US data with South African data. Even if we accept their conclusions for South Africa, they do not in any way eliminate the possibility that US non-biological parents spend less on food because they buy less junk food and more healthy, less expensive fruits and vegetables.

Next page > A Community Response > Page 1, 2, 3

Add Your Comments!

We want to know what you think. Your comments are important to us and the other readers. You are what makes this site special.