Your choices of peer group, your outcomes

I know this is an extremely thin reed on which to rest a burden of speculation but this blog exists in order to exercise the discipline of thinking out loud and thinking outside the margins. So here we go.

There are three sociological findings in the past few decades for which the evidence seems reasonably strong but which strike me as somehow wrong.

1) What you read does not have a significantly measurable impact on you, particularly when controlling for pre-existing dispositions. In other words, the act of reading is useful to some degree, but it doesn't matter what you read. You can read gory adventure stories, soppy romantics, racy porn, manga, or comic books - none of it is predictive of how you behave. Reading a great deal is loosely associated with higher IQ but only loosely.

As a reading enthusiast, I instinctively want to believe that the more rigorous your reading, the more non-fiction, the better. I am also inclined to believe that reading is an improving activity and therefore it matters what you read. The evidence just doesn't seem to be there to support what I instinctively believe.

2) Once you control for genes, parents don't really matter in terms of predicting life outcomes for children. Children's peer groups are more predictive than children's home environment.

As an invested parent, I absolutely believe my efforts make a difference and yet much of the evidence over the past three decades seems to indicate that genes and peer groups are far more predictive than home environments.

3) Once you control for genes and behaviors, it doesn't matter which university you attend. Genes and behaviors are far more predictive of life outcomes than the prestige of the university. Students who have been admitted to Ivy Leagues but attend state universities instead have the same life outcomes (income, wealth accumulation, education attainment, health, social status, etc.) as those who actually attend the Ive League school to which they were admitted.

As a product of some reasonably prestigious schooling, I naturally want to believe that I got something special from my education there but the evidence suggests otherwise.

Taking a leaf from the White Queen's book, I have chosen to believe impossible things, in this case continuing with my instinctive beliefs while simultaneously acknowledging the data. It is not a natural position to take but is, in a sense, comforting.

Alice laughed. ‘There’s no use trying,’ she said: ‘one can’t believe impossible things.’

‘I daresay you haven’t had much practice,’ said the Queen. ‘When I was your age, I always did it for half-an-hour a day. Why, sometimes I’ve believed as many as six impossible things before breakfast.

Yesterday I came across Who Is the Fairest One of All? How Evolution Guides Peer and Media Influences on Female Body Dissatisfaction by Christopher J. Ferguson, Benjamin Winegard, and Bo M. Winegard. From the abstract:

Much attention has focused on the influence of media images of thin women on body dissatisfaction among female viewers. Disagreement exists regarding the nature of media influences, with meta-analytic results suggesting only small effect sizes. Fewer researchers have focused on the role of peer influences and peer competition on female body dissatisfaction. Furthermore, the relation between body dissatisfaction and eating disorders may be more complex than is often implied in the media effects literature. Links between body dissatisfaction and eating disorders may be overstated, and some eating disorders, primarily anorexia nervosa, may not always be motivated primarily by body dissatisfaction. The current paper discusses these issues from an evolutionary perspective, examining how sociocultural forces influence the intensity of female competition and how such competition effects body dissatisfaction.

People who want to control what is written and said want to justify that control by claiming harm from words and images. They have been at it for decades and it is a deep ideological conviction on their part that words can hurt and therefore must be controlled, First Amendment be damned.

Ferguson and the Winegards are looking at the scientific literature to assess the claim that media presentation is determinative of negative affects on women and their self-image. It is an interesting exercise but, as I said at the beginning, not especially rigorous. It is more suggestive than conclusive. A weak reed.

Their findings reject the premise of the speech regulators. There is little or no negative media impact on women's self-image.

1. Although research does suggest a small relationship between media exposure and body dissatisfaction, the effect size does not yet reach the level of practical significance (e.g., Ferguson, 2009). Furthermore, the weak effect sizes do not provide convincing evidence for causal effects. As such, we agree with Levine and Murnen (2009) that media effects are best considered as a variable risk factor rather than a causal risk factor (Kraemer et al., 1997).

2. Most women are relatively unaffected by thin models presented in the media. Those women most likely to be negatively influenced by such images appear to have preexisting body dissatisfaction concerns or high neuroticism. Other women, particularly those with positive body image, may experience positive effects following media consumption.

3. Given the available data, popular and even scientific analyses of body dissatisfaction (e.g., American Psychological Association, 2007; Wolf, 2002) often overemphasize the causal power of the media.

4. There is not a clear relationship between body dissatisfaction and eating disorders, especially AN [anorexia nervosa]. Many people who are unhappy about their bodies never develop eating disorders, and many of the forms of disordered eating that they do develop are much less severe than is clinical AN. As is noted above, historical analysis has suggested that AN can and has developed in individuals for whom body dissatisfaction was not a primary concern

But what really caught my attention was this more general summary of their research on the broader topic:

From the literature reviewed to this point we conclude several points:

1. Genetic effects on both eating disorders and body dissatisfaction are clearly the strongest effects, accounting for approximately 40% to 80% of the variance (Bulik, Sullivan, & Kendler, 1998; Keski-Rahkonen et al., 2005; Klump et al., 2001; Spanos, Burt, & Klump, 2010; Wade, Wilkinson, & Ben-Tovim, 2003) in such outcomes.

2. Among social factors, peer influences, both active and passive, exert the most powerful influence on body dissatisfaction.

3. Media effects on body dissatisfaction remain generally small and inconsistent, particularly when other factors, such as peer influence, are controlled.

All this is consistent with the three conundrums I mentioned at the beginning (genes matter more than peer groups, peer groups matter more than parents, reading content doesn't have much influence.) However, somehow the presentation of the findings presented a different way of looking at this which might bridge some of the apparent incompatibilities.

Perhaps parents do make a huge difference but in a fashion we haven't been investigating and one which is quite obvious when you consider it. Let's grant that peer groups have a much more powerful and direct influence on a child than their parents.

Who selects the peer groups? It is easy to say the child does and within limits that is true. But the parents determine the range of peer groups from whom the child can choose. The parents choose which neighborhoods to live in, which schools the child will attend, in which activities they will participate, whether and which church will be attended. These different choices determine what peer groups will be available from which to choose.

I would wager that we will find that parental influence is actually very strong but via direct and indirect mechanisms. Direct influence will be weak - no matter what you tell children, they will process the instruction and information in the way they wish. Past roughly five years of age, children spend the bulk of their waking hours with people other than their parents so it is natural that there will be substantial non-parental influence. But indirect parental influence will be strong - if you create the circumstances in which all the peer groups around a child have similar values and goals as your own, then the probability of a child growing up with your values increases dramatically. Not because they listened to you but because they listened to their peers whom you helped to select by the neighborhoods, schools, churches you chose for your children to attend.

It is the selection of peer groups which then becomes a strategic condition in the development of life outcomes. "A man is known by the company he keeps" is an age-old adage which probably has more implication than we acknowledge, especially when we consider the importance of the person being able to select the company he keeps. Your choices, your outcomes.