31 ms·

The Central Limit Theorem is one big reason to expect that traits have a normal distribution. Wikipedia has a good explanation. [1] and [2] [1] https://en.wiki

by todd8 6y ago

The Central Limit Theorem is one big reason to expect that traits have a normal distribution. Wikipedia has a good explanation. [1] and [2] [1] https://en.wikipedia.org/wiki/Central_limit_theorem https://en.wikipedia.org/wiki/Central_limit_theorem [2] https://en.wikipedia.org/wiki/De_Moivre–Laplace_theorem https://en.wikipedia.org/wiki/De_Moivre–Laplace_theorem

criddell 6y ago
How would the central limit theorem apply to seemingly bimodal distributions like left vs right handedness?
- todd8 6y ago
  With respect to personality, I believe that there are likely a number of independent factors, perhaps hundreds, that affect in some way where one’s position lies on each of the axes of personality. These factors could be particular childhood traumas, birth order, parent’s behavior/addiction, childhood friends, and genetic factors. Even if each of these numerous random variable is binomial the central limit theorem applies.
  - criddell 6y ago
    This article [1] about what determines a persons handedness starts out: > Like most aspects of human behavior, handedness is a complex trait that appears to be influenced by multiple factors, including genetics, environment, and chance. It goes on to say that there are up to 40 genes that contribute. It seems like it's complex enough that the central limit theorem should apply yet handedness is definitely a bimodal distribution. [1]: https://medlineplus.gov/genetics/understanding/traits/handedness/ https://medlineplus.gov/genetics/understanding/traits/handed...
    - heliodor 6y ago
      Handedness is not what people care about in discussing personality assessment. Check out the big five model: https://en.m.wikipedia.org/wiki/Big_Five_personality_traits https://en.m.wikipedia.org/wiki/Big_Five_personality_traits Neuroticism, openness, agreeableness, conscientiousness, extraversion. On each dimension, many people are average and a few are on each extreme. At least that's the world I've observed. It's not a world full of extreme introverts and extreme extroverts. Most people are in the middle of that scale.
      - criddell 6y ago
        When I did a Myers Briggs test almost 40 years ago, I think I came up INTP (or something like that). I never interpreted the I to be extreme introvert, although I'm definitely on that side of the scale. I wonder how many people score exactly on the mid-point and when that happens, what do they do?
runarberg 6y ago
Central limit theorem does not apply unless each experiment is independent. This is not true of the concepts which psychometrician operationally define including personality factors and intelligence. As an example: Human height is not normally distributed. It is multi-modal based on geographical location, socioeconomic status and sex. There are genetic and environmental factors that makes it so that each childbirth is not an independent event marking a new height. The same holds for personality traits (if they exists) and intelligence (if it exists as described by IQ tests). The field of psychometrics has a problem with assigning normal distribution to everything even though there is reason to belief things are not normally distributed.

by todd8 6y ago

criddell 6y ago

How would the central limit theorem apply to seemingly bimodal distributions like left vs right handedness?

todd8 6y ago
With respect to personality, I believe that there are likely a number of independent factors, perhaps hundreds, that affect in some way where one’s position lies on each of the axes of personality. These factors could be particular childhood traumas, birth order, parent’s behavior/addiction, childhood friends, and genetic factors. Even if each of these numerous random variable is binomial the central limit theorem applies.
- criddell 6y ago
  This article [1] about what determines a persons handedness starts out: > Like most aspects of human behavior, handedness is a complex trait that appears to be influenced by multiple factors, including genetics, environment, and chance. It goes on to say that there are up to 40 genes that contribute. It seems like it's complex enough that the central limit theorem should apply yet handedness is definitely a bimodal distribution. [1]: https://medlineplus.gov/genetics/understanding/traits/handedness/ https://medlineplus.gov/genetics/understanding/traits/handed...
  - heliodor 6y ago
    Handedness is not what people care about in discussing personality assessment. Check out the big five model: https://en.m.wikipedia.org/wiki/Big_Five_personality_traits https://en.m.wikipedia.org/wiki/Big_Five_personality_traits Neuroticism, openness, agreeableness, conscientiousness, extraversion. On each dimension, many people are average and a few are on each extreme. At least that's the world I've observed. It's not a world full of extreme introverts and extreme extroverts. Most people are in the middle of that scale.
    - criddell 6y ago
      When I did a Myers Briggs test almost 40 years ago, I think I came up INTP (or something like that). I never interpreted the I to be extreme introvert, although I'm definitely on that side of the scale. I wonder how many people score exactly on the mid-point and when that happens, what do they do?

runarberg 6y ago

Central limit theorem does not apply unless each experiment is independent. This is not true of the concepts which psychometrician operationally define including personality factors and intelligence. As an example: Human height is not normally distributed. It is multi-modal based on geographical location, socioeconomic status and sex. There are genetic and environmental factors that makes it so that each childbirth is not an independent event marking a new height. The same holds for personality traits (if they exists) and intelligence (if it exists as described by IQ tests). The field of psychometrics has a problem with assigning normal distribution to everything even though there is reason to belief things are not normally distributed.