Thursday, September 17, 2009

Adoption and parental investment

It has long been known that children are likelier to be abused, neglected, or murdered by stepparents than by birth parents. This kind of genetic discrimination seems consistent with kin selection theory: parents are expected to care more for children who share kinship with them, as opposed to a purely legal and social relationship (Daly & Wilson, 1980).

If stepchildren are mistreated because they are not kin, we should see the same mistreatment of adopted children. To test this hypothesis, Gibson (2009) surveyed parents with at least one genetic and one adopted child over the age of 22, the idea being to compare the two groups of children for total parental investment. Contrary to expectation, the parents invested more in their adopted children than in their own:


This study categorically fails to support the hypothesis that parents bias investment toward genetically related children. Every case of significant differential investment was biased toward adoptees. Parents were more likely to provide preschool, private tutoring, summer school, cars, rent, personal loans and time with sports to adopted children (Gibson, 2009).

Why? One can imagine the parents making no distinction, but why would they discriminate against their own children? The answer seems to be that the adopted siblings made greater demands.

Adoptees were more likely than genetic offspring to have ever received public assistance, been divorced or been arrested. They also completed fewer years of schooling and were more likely to have ever required professional treatment for mental health, alcohol and drug issues.

… The current study may demonstrate cases where “the squeaky wheel gets the grease.” Summer school and private tutors are often remedial, and the fact that adopted children were more likely to receive them suggests they required them more often than genetic ones. The same can be said for rent, treatment and public assistance. Adoptees may have more difficulty establishing themselves relative to genetic children, and the fact that they divorce more often suggests they also have more difficulty staying established. Addiction and divorce may put adoptees in situations that require more parental investment. Parents provide more for adoptees not because they favor them, but because they need the help more often.
(Gibson, 2009)

For many behavioral traits, adoptees seem to differ genetically not only from their adoptive parents but also from the general population:

This supports other research showing that, compared to genetic children, American adoptees have a higher overall risk of contact with mental health professionals, specifically for eating disorders, learning disabilities, personality disorders and attention deficit hyperactivity disorder … They also have lower achievement and more problems in school, abuse drugs and alcohol more, and fight with or lie to parents more than genetic children …

… Adoptees may be genetically predisposed to negative outcomes at higher rates than the general population. Genetic factors clearly contribute to alcohol and drug addiction, as well as to some mental disorders like attention deficit hyperactivity disorder and schizophrenia …. An association between nonviolent criminality has been found between European adoptees and their genetic parents … Furthermore, research with Swedish adoptees suggests 55-60% of their educational performance is explained by genetic factors, and that the number of years of school adoptees complete is significantly related to how many years their genetic mothers completed ...
(Gibson, 2009).

All of this may explain why parents invest more in adopted children than in their own. But why do any parents adopt? Doesn’t such a decision, in itself, contradict kin selection theory?

The contradiction may be more apparent than real. Most adoptive parents have fertility problems and cannot have children on their own. Their only other option is to remain childless.

It may be that adopting fulfills a common instinct to reproduce and parents do it because it produces positive emotions. When people cannot have children biologically, adoption gives them a way to fulfill the “drive” to parent, maladaptive or not. (Gibson, 2009)

References

Daly, M., & M. Wilson. (1980). Discriminative parental solicitude: A biological perspective. Journal of Marriage and the Family, 2, 277-288.

Gibson, K. (2009). Differential parental investment in families with both adopted and genetic children, Evolution and Human Behavior, 30, 184-189.

16 comments:

John Smyth said...

I'm guessing stepparents end up stepparents less often out of a totally free choice compared to adopting parents. Adopting parents deliberately want a child to take care of, and the same psychological bonds can occur with an adopted child as with a genetic child. A stepparent may be more likely to end up taking care of a child by circumstance, and not had the ability or desire to bond as well (on average). The main difference I see between genetic and adopted is the adopted might have spent less time together, thus less time to bond, especially missing out on pregnancy and the first few years of attachment.

Tod said...

My impression is that the average SES of those who adopt despite already having a genetic child of their own is especially high. The adopted child is being held to a high standard in comparisons with the genetic child of such adoptive parents I think.

Women who gave their baby up for adoption would be likely to have been in an unusual emotional state during their pregnancy. Would fetal stress be a non genetic explaination for difficulties as adults ?


Not realy relevant to this post but I found this surprising

"The proportion of people who never married and the age at first marriage increased in rural Ireland [...] In 1851, 11% of the population were never married at 45–54 years and this percentage increased steadily over time to 34% for men and 25% for women in 1936.
In rural Ireland wealthy heads of households were more likely to be celibate than occupiers of small holdings".
Kent 2002. In the moral climate of Ireland at that time choosing to remain unmarried meant most of these men were chooosing to be truly celibate.

Peter Frost said...

John,

I agree. Another factor is that stepparents tend to be a select group. They are probably more altruistic than parents in general. Stepparents probably represent a more random sample of the general population.

Tod,

Fetal stress might be an explanation. There's also alcohol and substance abuse -- both of which are common among mothers who give up their children for adoption. But all of these environmental factors are confounded with genetic factors. People who end up in dysfunctional relationships tend to be less discerning than other people. The same goes for people who abuse alcohol. Genetically, they're not a random sample of the general population.

Peter Frost said...

John,

I should have written 'adoptive parents tend to be a select group'.

Anonymous said...

A bit on another topic, but regarding the overrepresentation of adoptees in the juvenile justice system, psychiatric treatment centres, do you feel that this is due to the supposed "genealogical bewilderment" theory (that adoptees are psychologically traumatized by not being raised by their birth parents) or to genetic and prenatal factors? I strongly suspect it's the second (genetic factors), but I like to know others' opinions and explanations.

Anonymous said...

Jokes of the right kind, nicely told, can do more to make aware questions of civics, opinion, and creative writings than any number of bovine arguments.

Anonymous said...

But right away I arrange rush at to put faith that the whole domain is an puzzle, a benign problem that is made regretful not later than our own mad strive to explicate it as in spite of it had an underlying truth.

Anonymous said...

I don't like the earshot of all those lists he's making - it's like intriguing too many notes at school; you sense you've achieved something when you haven't.

Anonymous said...

In everyone's existence, at some pass‚, our inner foment goes out. It is then blow up into zeal beside an contend with with another hominoid being. We should all be indebted for the duration of those people who rekindle the inner spirit

Rudy Owens said...

The study you mention fails to mention the sample size of the survey in the abstract that is available to the public. I won't be buying the study, as I can't access research in the for-profit academic publishing world. I appreciate you quoting this study. But from your quotes, a reader can't tell if this study has any reliability because we do not know about the validity of the data collection methods. Was it random? Self-selective? Phone survey? Internet survey? Office survey? Without that information, all of the findings should be treated extremely cautiously. I see no evidence here yet that disproves the population-based, epidemiological evidence Daly and Wilson gathered from a records database of nearly 90,000 cases of abuse. In that study, we have a true population sample. Methods matter, and so do the sample sizes.

Peter Frost said...

Rudy,

Please show a bit of initiative. If you do a search on Google Scholar, you'll find a PDF of the entire article. That took me less than thirty seconds.

Gibson's study doesn't really contradict findings by Daly and Wilson. Parents invest more in adopted children because such children make more demands on parental resources.

Rudy Owens said...

Dear Peter, I did search for it, and it's truly behind a fee-based firewall (http://www.ehbonline.org/article/S1090-5138%2809%2900003-8/pdf). However, I searched again and found a free version. https://www.researchgate.net/publication/228470797_Differential_parent_investment_in_families_with_both_adopted_and_genetic_children.

A best practice when discussing research is to discuss that research's statistical validity, methods, etc. Even basic blogs do that. The comment I made is completely on point, and since you didn't include a link to the article, you didn't contribute to the education of the reader or help them make an informed decision.

Actually, the author claims his research disproves what Daly and Wilson found: "This study categorically fails to support the hypothesis that parents bias investment toward genetically related children." You simply parroted what the author claims he found in his study, which has a lot of problems and should not be accepted because of a weak study. Your sweeping statement hasn't been proven at all.

For the record, this is very small sample size (a total of 126 (42%) of 300 of surveys were returned). That state of the survey isn't listed. We have no details on the type of adoption agency, or the parents, or ethnicities of the families, etc. So I'd say is doesn't matter what Gibson found from his same of less than 126 surveys, because it's not a very strong study to make broad generalizations about large populations of people like adoptive parents. And he tried to do that.

So I'm not going to place much weight in what is said here, relative to what is published by Daly and Wilson. http://www.cep.ucsb.edu/buller/cinderella%20effect%20facts.pdf

Sorry you have problems with asking for basic methodologies. You seem to have a very strong reaction to someone asking for what should be the first thing a person shares with research. Very odd that? Curious why? Are you by chance a supporter of international adoptions and the adoption industry? I would almost say given your tone and how you responded, you not be revealing that bias. For the record, I am adopted, so yes, I have a very informed perspective.

Cheers mate.

Peter Frost said...

I'm not a supporter of international adoptions or the adoption industry. You're making an awful lot of wrong assumptions about me, and those assumptions are fueling a lot of animus on your part. Please be civil. Words like "parrot" are intended to get people riled.

This is a general rule of life. Whenever you have a difference of opinion with someone, please be civil and remember there are two sides to every question. In this case, the differences between my position and yours are trivial.

A few points:

- In recent years, I've begun linking to non-firewall articles, largely because some people can't be bothered using Google Scholar to find them.

- Comparing Gibson with Daly and Wilson is like comparing apples and oranges. Daly and Wilson did their work on non-biological parents in general. In most cases, the non-biological father is the mother's second husband or boyfriend, i.e., a stepfather. This is a very different situation from that of parents who consciously choose to adopt children. I agree that such parents often become ambivalent towards their adopted children later on, but at least initially they make an effort to do their best, and that means assigning more resources to adopted children because such children tend to make more demands than natural children do.

- Gibson explains at length that "Parents provide more for adoptees not because they favor them, but because they need the help more often." So she's not challenging Daly and Wilson's general finding. Let me quote Gibson's conclusion in full:

"The current study may demonstrate cases where “the squeaky wheel gets the grease.” Summer school and private tutors are often remedial, and the fact that adopted children were more likely to receive them suggests they required them more often than genetic ones. The same can be said for rent, treatment and public assistance. Adoptees may have more difficulty establishing themselves relative to genetic children, and the fact that they divorce more often suggests they also have more difficulty staying established. Addiction and divorce may put adoptees in situations that require more parental investment. Parents provide more for adoptees not because they favor them, but because they need the help more often."

Rudy Owens said...

Thanks for your reply to my discussion, Peter.

The main point, which I will repeat one more time, is that a study with an N=126 has low statistical power (for the benefit of readers, please go here for why that matters: http://www.nature.com/nrn/journal/v14/n5/full/nrn3475.html).

Thus a study with this sample size cannot be used to make any sweeping conclusions about wider populations in the millions. Sure, Gibson's study certainly is quotable for you and others who publish. But it isn't good science and a poorly designed study. Thus, it does not contribute to the work Daly and Wilson have done on parental investment theory using a sample size of N=87,789--very strong power, AND population based.

This is important because bad and poorly designed studies can be picked up and cited in research, pop-lit publications, and the media when they appear to be valid, but lead to harmful outcomes. I'm sure you are familiar with the utter junk-science research that Dr. Wakefield did that was proven fraudulent on MMR vaccines/autism (http://www.cnn.com/2011/HEALTH/01/05/autism.vaccines/). It was exactly the kind of study (N=12) that should never have been published by the Lancet or quoted, but was--repeatedly, spurring on vaccination deniers who don't understand science and who encouraged policy actions that have caused real harm and death to kids and communities, but also to our public health system.

I say, first things first--if the study is not valid and has very low power, make it clear and talk about the methodologies, not the quotable and fun conclusions that aren't backed up. That should be the job of the researcher first and foremost, and the peer review panel next. But bloggers have roles too. I don't assume the media will be responsible. They will always grab the sexy headline, like, eating dark chocolate makes you skinny, or dark beer makes you smarter.

Thanks. Rudy Owens

Peter Frost said...

Rudy,

I addressed your main point in my previous comment. Do I need to repeat "one last time"? You're comparing apples and oranges. Gibson studied families that had both adopted children and natural children. Daly and Wilson studied the broader question of stepparents and stepchildren. Most stepfathers have ambivalent feelings towards their stepchildren from the very beginning. In contrast, all adoptive parents make a conscious and sustained choice to adopt children. So your comparison between N=126 and N=87,789 is misleading.

There is no universal rule as to what constitutes a correct sample size. A sample size of 20,000 may be useless if sources of sampling bias haven't been controlled. Many significant findings have been made with sample sizes of only a hundred or so. I'm sorry, but you need to educate yourself on this point.

Rudy Owens said...

Peter, I actually hope you read the article I quoted on statistical power. I also hope the readers of this blog do as well. I'm checking out of this conversation, but I appreciate having the chance to provide any future reader with links to relevant information, including the original study so they can review this material themselves and see the issues that our detailed above. I trust the intelligence of smart readers who understand basic statistics and population studies. Best.