Wednesday, September 17, 2008

The big fight: The Big Five or The Big Eight?

The big question of the day is whether to overthrow the last 40-yrs mature conventional wisdom that there are only five personalty traits or factors. The introduction of Big Five or FFM model of personality had spawned a big research paradigm and there are many independent confirmations; so before I try to throw the baby out with the bath-water, let me just say at the outset that just like the big five model is not incompatible with Eysneck's PEN model, so is my proposed eight factor model not inconsistent with the Big Five model- it just extends it and introduces a few new traits or dimensions. I have written in the past about personality, so it may help to read a few articles to know where I am coming from. I especially recommend this one related to perfectionism and personality.

First a quick review. the big Five personality dimensions are (in no particular order): Openness, Conscientiousness, Extraversion, Agreeableness and Neuroticism. It is to be recalled that these traits were discovered by lexical analysis of adjectives/ nouns and then doing factor analysis on the data to discover the minimum number of factors required to account for the variation in the data set. This way five factors have been discovered by exploratory Principal component analysis in many languages.

Recall that in PCA, factors are ordered. First factor is more important and can explain most of the variance. Second is less important/ responsible for variance and so on. It is my thesis that these personality traits would occur in a factor analysis in the order in which they evolve/develop , with the most evolved/ developed trait , which is most under hereditary control, reflected more in language and accounting for more variance in the data set.

Now, I do not have access to the original Goldberg or any McRae and Costa factor analysis results , so cannot say what the order of factors was. I propose, from my theoretical leanings and ordering of eight basic adaptive problems, the order would be (in order of less importance) : Neuroticism, conscientiousness, Extraversion, Agreeableness and Openness.

I have done some quick lookup on Google , but could not find much data related to how the five factors are ordered. One source I found, found support for NECAO ordering if only items from NEO-PI were analyzed; but NCEAO i.e. my order when additionally Zuckerman scales were also taken into account.

Let me delineate this further:

  1. Neuroticism (N): personality more focused towards solving the adaptive problem of avoiding predators. marked by negative emotionality, worry etc. Nettle calls these Worriers. The plot that works for them is 'overcoming the monster': everything apprised as a monster. primary mode of being: emotional.
  2. Conscientiousness (C): personality more focused on optimally finding and utilizing resources (or finding food) Nettle calls them controller. The plot that works for them is Rags to riches. How to become successful. Primary mode of being: motivational.
  3. Extraversion (E): personality more focused on forming alliances/friendships and thus issues of dominance- hierarchy. Nettle calls them wanderers. The plot for them a Quest, where they wander adn on the way make alliances/ friends to reach the illusive goal. The Journey , and the energy imbued with travel, becomes more important than the Goal. Primary mode of being: behavioral/social.
  4. Agreeableness (A): personality more focused on care of close ones; be it friends or children. Nettle calls them Empathizers. not sure if voyage and return is an apt plot for them. Primary mode of being: attachment/ care/ responsibility.
  5. Openness (O): Here I'll like to re-categorize this as rebelliousness vs social conformity: adaptive problem to be solved : who am I and who are like-minded people / roles that I should help. Nettle calls them Poets. The plot that works for them is comedy wherein one has to find true identities of disguised people or become reunited with twins/ kin etc. Primary mode of being: cognitive and self-appraisal
I'll like to add three more factors to the above based on CPS scales and any other theoretical as well as factor-analytic considerations:

6. Trust vs defensiveness (T)
: personality more focused on who can be trusted and who cannot. trying to see behind someones apparent persona.
7. Activity (Act) : personality more focused on being active, communicative, lively and humorous.
8. Masculinity- femininity (M-F): personality more focused on becoming desirable to the opposite sex.

The reasons I extend this are:

I) The evidence for Comrey Personality scales:
  1. Trust vs. Defensiveness ; A new factor (T above)
  2. Social Conformity vs. Rebelliousness: same as O
  3. Emotional Stability vs. Neuroticism: same as N
  4. Mental Toughness vs. Sensitivity: Same as Masculinity- femininity above
  5. Orderliness vs. Lack of Compulsion: Same as C
  6. Activity vs. Lack of Energy: A new factor (Act above)
  7. Extraversion vs. Introversion : same as E
  8. Empathy vs. Egocentrism : Same as A
II) when Nouns are factor analyzed we get eight factors (in the order of importance of factors)
  1. Social Unacceptability (Scum, tarsh, mororn): N (negative emotion)
  2. Intellect (philosopher, artist,nonconformist): C? (not conscientious person?)
  3. Egocentrism (snoop, busybody, know-it-all): E (not extarverted person)
  4. Ruggedness (tough, gentleman, fighter) : A (not agreeable person)
  5. Delinquency (law-breaker, googdy-ggody, innocent): O (more rebellious person)
  6. Attractiveness (babe , doll, hero) : T ??(a trustworthy person is attractive??)
  7. Liveliness (joker, chatterbox, loudmouth): Act ( a more active/ communicative person)
  8. Disorientation: (klutz, novice, daydreamer): M-F ? (are we being sexually selected for more 'orientation'?)
III) Goldberg, who had originally proposed the Big five has revised them to include two more ; he calls them Religiosity and WYSIWYG respectively.
Goldberg (Goldberg 1992b) has identified the “next two” factors that might be used to augment the big five. The first, tentatively called Religiosity, includes adjectives ranging from prayerful and reverent at the north pole to irreligious and unreligious at the south pole. The second, tentatively called what you see is what you get, includes adjectives ranging from undevious and unsly at the north pole to slick and aristocratic at the south pole. He goes on to point out that “there are no additional domains with anywhere near the breadth of the Big-Five factors”.


IV) Wikipedia entry says that others have also proposed more traits (and presumably also found in their factor analytic studies evidence for such traits)

Some psychologists have dissented from the model precisely because they feel it neglects other domains of personality, such as Religiosity, Manipulativeness/Machiavellianism, Honesty, Thriftiness, Conservativeness, Masculinity/Femininity, Snobbishness, Sense of humour, Identity, Self-concept, and Motivation.


See this also on personality research:

Saucier and Goldberg (1998) presented evidence that nearly all clusters of personality-relevant adjectives can be subsumed under the Big Five. Paunonen and Jackson (2000), however, argued that this study used too loose a criterion for inclusion in the Big Five--namely that the Big Five account for at least 9% of the variance in the adjective cluster. Reanalyzing the same data using a stricter criterion of 20% explained variance resulted in nine clusters of traits that fell outside of the Big Five: Religiosity, Honesty, Deceptiveness, Conservativeness, Conceit, Thirft, Humorousness, Sensuality, and Masculinity-Femininity. These analyses do not imply that the clusters are unrelated; for example, Honesty and Deceptiveness may be highly (negatively) related as opposite sides of the same dimension. Nevertheless, these results suggest that several important personality traits lie beyond the Big Five.


Considering all the above factors , especially keeping in mind the fact that Goldberg;'s new proposed religiosity may be more close to the now-traditional openness which I have re-characterized as rebellious- conformity ; and that Goldberg's unsly, slick, aristocratic and undevious may correspond to trust-defensiveness dimension; what we see is that the traditional intellect that Goldberg uses may be better thought of as Activity dimensions which relates to how lively, communicative and active a person is . Also he completely misses the last factor related to masculinity / femininity.

For the other proposed dimensions by Johnson et al , it is easy to see that religiosity can be subsumed under my definition of Opennnes; honesty/deceptiveness are opposite poles of the trust-defensiveness (T) trait); Conservativeness, Conceit, Thirft, Humorousness are better conceptualized as per me into Activity trait (Act) and that leaves us sensuality and Masculinity-Femininity as the last trait with these being two poles: at one end the role is more gender conformant; at the other it is more open and sensual in nature. If ever Humans speciate, it would because of this dimension!! It has been my thesis that we have been developing in diversity along these personality dimensions, but speciation would most likely happen only when assortattive mating and sexual selection acts at the eighths trait and the eighth trait is under more and more genetic control. Such a scenario may, thankfully, be far away!!

Sphere: Related Content

3 comments:

TheNerd said...

I'm so happy! You are the first person to finally say what I've been saying for years now: gender is a personality trait! I hope to see the male-female sliding scale as cannon in the future.

Sandy G said...

Hi Andrea,
Some other folks also have been saying the same thing: gender as a personality trait. I have just incorporated it in the big eight. Lets hope the model finds more takers in the future and is verifiable and spawns as big a research paradigm as the big five has had.

Lykaios said...

I think you are likely to be going beyond the data at your disposal. However, you are correct that the Big Five appear to be insufficient to explain existing personality variation. I would suggest reading up on the HEXACO model, which represents a rotation of the Five Factors that was unnoticed in most factor analytic studies for years. Specifically:

Agreeableness splits into

Emotionality (with Neuroticism), Honesty-Humility (Goldberg's WYSIWYG) and "new" Agreeableness (with Emotional Stability). Traits thought to be "beyond" the Big Five such as seductiveness, integrity, thrift, femininity, recklessnes, humorousness, and so forth are subsumed by the new Honesty-Humility and Emotionalty factors. Specifically, terms such as tough, manly, rugged, womanly, and feminine are found in Emotionality, so that your proposed masculinity/femininity factor is subsumed by emotionality. Integrity, sexual restraint, Machiavellianism, and Psychopathy are strong correlates of Honesty-Humility. And of course activity and energy level are (still) part of Extraversion and do not need to be accounted for by a separate activity factor as you propose.

For confirmation of these claims and a wealth of further information, try reading the articles that come up when performing a google search for the terms "hexaco" and "ashton." You may need access to scientific articles through a university or other organization to read the full articles, but the abstracts should at least give you a rough picture of the current state of research.