Saturday, July 08, 2006

Happiness and assortative mating for personality   posted by agnostic @ 7/08/2006 04:51:00 PM

I'm sure I'm not the only introverted neurotic who's ever thought that these personality traits would forever doom me to date my fellow melancholics. So here's some good news: at the present time in the US and Britain, there is no assortative mating for any of the Big Five personality traits. Eaves et al (1998) (pdf) investigated the non-additive effects of genes on Extraversion and Neuroticism, making the assumption of random mating. They had access to a large data set (N = 20,554) on these traits in American twins and their spouses, so they were in good position to check for assortative mating -- zippo. They also cite an English study of 889 spouse pairs done by two of the co-authors and Eysenck for a 1989 book, which also found no assortative mating for E and N.

A later study by Eaves et al (1999) (pdf) looked at biological and non-biological modes of transmission for personality traits and social / political attitudes, examining 29,691 American twins, 4391 of whose spouses also provided data. They were scored for the previously mentioned traits E and N, as well as Psychoticism, which due to its pronounced skew in the population and far greater ratio of non-additive to additive genetic causes compared to E and N (see Table 12 of the pdf), is most likely the multiplicative result of two independent traits: Agreeableness and Conscientiousness, in Big Five terms. Assortative mating for these three (or four) personality traits in this large sample? Nil. However, there was substantial assortative mating for attitudes (e.g., toward taxes, the military, and so on). The upshot is that if you thought those hotheaded, exhibitionist, sensation-seekers would never give you the time of day on account of your quietness, anxiety, and moodiness -- think again, you damned pessimist! On the other hand, they won't necessarily be magnetically drawn to you either...

All right, but would you be happier if you married someone similar to you, whether based on attitudes or personality? A recent study by Luo and Klohnen (2005) (pdf) looked at 291 American newlyweds to see whether there was assortative mating in the sample as a whole, and whether spousal similarity for attitudes and/or personality correlated with marital satisfaction. On measures of attitudes, there was substantially greater similarity than expected, replicating the Eaves et al (1999) finding. There was very weak positive assortment for the overarching personality measures of the Positive and Negative Affect Schedule (PANAS) and Emotional expression. On the component traits like Neuroticism or Openness, there was no positive assortment -- indeed, there was even extremely weak negative assortment for Extraversion, though this was only significant at the level of p less than 0.1. So, overall, these data replicate the findings for lack of assortative mating for personality reported in both of the Eaves et al studies mentioned above.

As for happiness, Tables 3 through 8 in the pdf show that, in general, similarity in attitudes had little to no effect (highest r = 0.16) and, to the limited extent that it matters at all, appears to matter somewhat more for the husband's satisfaction than the wive's. In contrast, the personality traits relating to attachment (e.g., Agreeableness) were often at or just above the weak-moderate border of r = 0.3 and appear to matter somewhat more for the wive's satisfaction than the husband's. For the Big Five factors, similarity in Agreeableness matters most, followed by Openness and Neuroticism, while similarity in Extraversion and Conscientiousness were not significant in association with marital satisfaction.

Makes sense, sort of: if you've committed to marriage, you're not exactly going to be carrying on with all sorts of strange men or women -- you'll be with each other most of the time -- so Extraversion differences shouldn't matter. Agreeableness measures how nurturing & empathetic vs antagonistic & suspicious one is, so again, no surprise that that's the most important Big Five factor for marital satisfaction. Also, if you're emotionally unstable, you want someone who will understand your frustration over what a more stable person would write off as a trifle, which would only frustrate you even more! And vice versa, if you're emotionally stable, you don't want to hear your partner's incessant complaints and worries which you'll discover you can do nothing to diminish. I'm not sure what to make of Openness playing a role -- for Closed people, more things are "out of bounds" for discussion, so relations could become plagued by what the Closed partner sees as the Open partner's flagrant disregard of deserved taboo barriers. The one finding that surprised me was that similarity in Conscientiousness had no effect on marital happiness for either partner -- I would've surely thought that if the wife had a strong work ethic while the husband was lazy, that would create instant friction. But itt may be that only one facet of Conscientiousness (say, work ethic) has an effect on happiness, while the other facets (like punctuality, preferring order & structure) don't have an effect -- the slob views the neat freak's quirks as tolerable, perhaps endearing, but not a deal-breaker.

These are the kinds of things they should teach in health class in high school -- no one's engaging in sexual acts other than the tiny clique of cool kids, and only a negligible minority are doing hard drugs (well, at a good school anyway). Knowing whether or not So-and-So would date you due to personality differences, or whether you'd be happier with someone similar to you on some social attitude or personality trait -- that's actually worth the time and effort of class attendance. It would also clear up a lot of speculation on who dates who and what makes for a satisfying relationship, and when you're an adolescent, you don't have time to do a mini research project to find out: you want someone to teach you rather than figure it out for yourself, perhaps the hard way.