- Image via Wikipedia
In my last post I had mentioned how Seligman and Peterson have tried to correlate their structure of human virtues/character strengths with work of other researchers like the universal dimensions of human mate preferences discovered by Buss et al. Today I wish to discuss in detail the universal dimensions of human mate preferences discovered by Buss et al.
Buss et al looked at data , using an 18 item preference ratings archival database, of about ten thousand people, from various cultures across the globe, and used the analysis strategy outlined by Bond to take care of different sample size from different cultures. they then applied the Principal component analysis to the refined data so obtained to determine the underlying structure of the mate preferences.
Their PCA analysis led to discovery of four dimensions all of which could be quantified as bipolar dimensions with one pole representing a different construct and another representing a sort-of-but-not-really opposed construct. For eg.,the first factor included loadings from ‘‘good financial prospects’’ (-0.65), ‘‘favorable social status or ratings’’ (-0.62), and ‘‘ambition and industriousness’’ (-0.41), each of which loaded negatively. The component also included ‘‘mutual attraction—love’’ (0.49), which loaded positively. They thus labeled this component ‘‘Love vs. Status/Resources.’’
Similarly the 3 other components were labeled “Dependable/Stable vs. Good Looks/Health”; “Education/Intelligence vs. Desire for Home/Children”; and “Sociability vs. Similar Religion”.
To my naive mind all of these bipolar dimensions seem to be separate constructs in themselves and I cannot fathom why dependable/stable should be taken as opposed to good looks/health. to me they seem sort of independent constructs. I would rather view the findings as eight separate poles than 4 bipolar dimensions with each dimension conflating two constructs/factors.
The paper immediately drew to my mind this paper, by Haslam et al, that while finding the underlying structure of positive characteristics, found three consistent bipolar dimensions using multi dimensional scaling. However, when the same data was subjected to cluster analysis, 6 factors were or clusters were apparent , each cluster being the pole of a single bipolar dimension. These 6 factors were “self-control,” “love,” “wisdom,” “drive,” “vivacity,” and “collaboration” which to my mind seems to map onto the virtues of self-restraint/temperance strengths, interpersonal or humanity strengths, intellectual or wisdom strengths, courage or emotional strengths , activity or vitality strengths and justice or civic strengths. Of course I think their MDS missed a fourth dimension which would have led to 8 clusters , the 2 remaining being religious and transcendence strengths.
Returning back to our current paper on universal mate preferences, I would like to break up the 4 dimensions into 8 factors and present them in a staged developmental order. It would be worthwhile to note that the two opposed dimensions are usually two adjacent stages following each other and may indeed reflect some conflict in mind of people as to which stage of mate preference to prefer based on their evolved natures . Here goes:
- first stage: Physical/biological : good looks/health
- second stage: will, restraint and control: dependable/stable.
- third stage: dominance/hierarchy, friends and foes: status/ resources.
- fourth stage: interpersonal: Love
- fifth stage: cognitive: education/ intelligence
- sixth stage: intimacy: desire for home/children.
- seventh stage: communicative/ generative: sociability
- eights stage: integrity, ingroup/outgroup: similar vs dissimilar religion.
Of course this is not the first time I have tried to put Buss’s findings in a 8 stage model; earlier I had tried to put his views on personality in a eight fold structure- whereby the last three stages of reproduction/evolution may be now characterized as biological, linguistic and cultural evolution. Anyway getting back to universal mate preferences, I can see that eight fold structure is found in the mate preferences too depending on which stage of preferences you have evolved/developed.
SHACKELFORD, T., SCHMITT, D., & BUSS, D. (2005). Universal dimensions of human mate preferences Personality and Individual Differences, 39 (2), 447-458 DOI: 10.1016/j.paid.2005.01.023
Haslam, N., Bain, P., & Neal, D. (2004). The Implicit Structure of Positive Characteristics Personality and Social Psychology Bulletin, 30 (4), 529-541 DOI: 10.1177/0146167203261893
Pingback: Tweets that mention universal human mate preference: four dimensions or eight factors? | The Mouse Trap -- Topsy.com
Pingback: uberVU - social comments
Its difficult to interpret Buss et al’s results b/c they used principle components analysis which has been demonstrated over and over again as a poor technique for determining the underlying structure of data.
Hi Mark,
again I agree, but still believe that a confirmatory factor analysis grounded in theory can help confirm underlying structure.
BTW, you may like this article on factor analysis’ limitations and need for theory: http://repository.upenn.edu/cgi/viewcontent.cgi?article=1015&context=marketing_papers
Makes sense. It not only does it come across properly in theory but also plays out well in reality. Interesting choice of picture for the blog post – seems to avoid the first stage, “good looks/health”!