MPQ, simulated

Quarto

Quarto enables you to weave together content and executable code into a finished document. To learn more about Quarto see https://quarto.org.

Running Code

When you click the Render button a document will be generated that includes both content and the output of embedded code. You can embed code like this:

#doesn’t hurt to check, still new at this.

edited DF
vars n mean sd median trimmed mad min max range skew kurtosis se
MPQ_POSITIVE_EMOTIONALITY 1 166 56.61 8.14 58.51 57.31 8.11 29.79 70.82 41.02 -0.74 0.22 0.63
MPQ_NEGATIVE_EMOTIONALITY 2 166 46.45 9.26 44.58 45.31 9.05 34.41 83.90 49.49 1.15 1.40 0.72
MPQ_CONSTRAINT 3 166 49.39 7.93 50.48 49.77 8.23 24.12 65.04 40.92 -0.44 -0.09 0.62
Original DF
vars n mean sd median trimmed mad min max range skew kurtosis se
MPQ_POSITIVE_EMOTIONALITY 1 501 56.67 8.13 58.51 57.39 8.11 29.79 70.82 41.02 -0.75 0.25 0.36
MPQ_NEGATIVE_EMOTIONALITY 2 501 46.47 9.21 44.58 45.32 9.05 34.41 83.90 49.49 1.15 1.44 0.41
MPQ_CONSTRAINT 3 501 49.45 7.92 50.48 49.85 8.23 24.12 65.04 40.92 -0.45 -0.08 0.35
Mean Personality By Total Weapon Use
total_weapon_use n Mean Positive Emotionality Mean Negative Emotionality Mean Contstraint
0 30 57.21 46.84 50.11
1 60 56.88 45.78 50.24
2 44 56.16 46.62 48.76
3 24 57.48 47.29 47.53
4 7 50.69 47.10 48.50
6 1 63.29 42.55 56.03

there is a slight difference in means. this is caused by one of the participants being recorded as both an officer and a civilian. I am going to remove this participant

weapon use has uneven groups.

Recoded Mean Personality By Total Weapon Use
total_weapon_use n Mean Positive Emotionality Mean Negative Emotionality Mean Contstraint
0 30 57.21 46.84 50.11
1 60 56.88 45.78 50.24
2 44 56.16 46.62 48.76
3 32 56.18 47.10 48.01

binary outcome variable creation

Replication Results
Estimate Std. Error z value Pr(>|z|)
(Intercept) 3.958 1.444 2.740 0.006
MPQ_POSITIVE_EMOTIONALITY -0.017 0.016 -1.053 0.292
MPQ_NEGATIVE_EMOTIONALITY -0.012 0.013 -0.854 0.393
MPQ_CONSTRAINT -0.028 0.015 -1.827 0.068
SAMPLEOfficer 1.727 0.331 5.217 0.000

here are the authors’ results… verify them in attached pdf.

Original Findings
Personality_Factors b SE
Positive Emotionality -0.024 0.014
Negative Emotionality 0.003 0.012
Constraint -0.022 0.015

Let’s explore the MPQ more.

this person (cite) found that MPQ can be used to measure antisocial traits. The sub-factors associated with these findings are intuitive. I’m. going to use are Aggression, Social Potency, Alienation, Low social closeness and low harm avoidance.

But first, lets verify that creating a new super factor is possible with the current data. The MPQ sub factors are standardized, so I should be able to find the mean score across sub factors constituting super factors, and check these against the original to ensure this is a viable strategy

Negative Emotionality: stress reaction + alienation + aggression

[1] 0.9948554

Positive Emotionality: well being + social potency + achievement + social closeness

[1] 0.9948554

Constraint : MPQ_CONTROL+MPQ_HARM_AVOIDANCE+ MPQ_TRADITIONALISM

[1] 0.9948554

Correlations between the super factors that I calculated and the original super factors. Now, I’ll build a new superfactor based on the literature of correlated traits.

Low social closeness and low harm avoidance are associated with antisocial traits. The current scale measures high harm avoidance and high social closeness. Because these are standardized scores, I can subtract the scores from 100 to approximate being low in the trait in question. I do this with harm avoidance, social closeness, and stress reaction.

Now, I calculate each participant’s average score across these traits.

There is no indication of severe multi-colinearity. The highest correlation is constraint at -0.55

Now, I’m going to re run the first analysis including this new super factor.

Replication Results
Estimate Std. Error z value Pr(>|z|)
(Intercept) -2.593 2.663 -0.974 0.330
MPQ_ANTISOCIAL 0.108 0.037 2.893 0.004
MPQ_POSITIVE_EMOTIONALITY -0.020 0.016 -1.236 0.217
MPQ_NEGATIVE_EMOTIONALITY -0.014 0.014 -1.007 0.314
MPQ_CONSTRAINT -0.003 0.018 -0.159 0.873
SAMPLEOfficer 1.768 0.338 5.236 0.000 
plot(anti_MODEL)

total_model <-lm(total_weapon_use~MPQ_ANTISOCIAL+ MPQ_POSITIVE_EMOTIONALITY+ 
                         MPQ_NEGATIVE_EMOTIONALITY+SAMPLE,
                       data=OnlyOne)



summary(total_model)

Call:
lm(formula = total_weapon_use ~ MPQ_ANTISOCIAL + MPQ_POSITIVE_EMOTIONALITY + 
    MPQ_NEGATIVE_EMOTIONALITY + SAMPLE, data = OnlyOne)

Residuals:
    Min      1Q  Median      3Q     Max 
-1.7440 -0.7503 -0.1652  0.7881  1.8971 

Coefficients:
                           Estimate Std. Error t value Pr(>|t|)    
(Intercept)               -0.161471   0.683630  -0.236  0.81338    
MPQ_ANTISOCIAL             0.032846   0.010849   3.028  0.00259 ** 
MPQ_POSITIVE_EMOTIONALITY -0.005285   0.005374  -0.983  0.32587    
MPQ_NEGATIVE_EMOTIONALITY  0.000644   0.004760   0.135  0.89244    
SAMPLEOfficer              0.532255   0.088861   5.990 4.04e-09 ***
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

Residual standard error: 0.9647 on 496 degrees of freedom
Multiple R-squared:  0.08245,   Adjusted R-squared:  0.07505 
F-statistic: 11.14 on 4 and 496 DF,  p-value: 1.157e-08
weapon_anti_MODEL<-glm(weapon~ MPQ_ANTISOCIAL+SAMPLE, family = binomial,
                       data=OnlyOne)

summary(weapon_anti_MODEL)

Call:
glm(formula = weapon ~ MPQ_ANTISOCIAL + SAMPLE, family = binomial, 
    data = OnlyOne)

Coefficients:
               Estimate Std. Error z value Pr(>|z|)    
(Intercept)    -4.40662    1.66306  -2.650 0.008056 ** 
MPQ_ANTISOCIAL  0.10618    0.03226   3.291 0.000998 ***
SAMPLEOfficer   1.71306    0.33070   5.180 2.22e-07 ***
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

(Dispersion parameter for binomial family taken to be 1)

    Null deviance: 471.79  on 500  degrees of freedom
Residual deviance: 427.68  on 498  degrees of freedom
AIC: 433.68

Number of Fisher Scoring iterations: 5
anti_total_model <-lm(total_weapon_use~MPQ_ANTISOCIAL+SAMPLE,
                       data=OnlyOne)

summary(anti_total_model)

Call:
lm(formula = total_weapon_use ~ MPQ_ANTISOCIAL + SAMPLE, data = OnlyOne)

Residuals:
    Min      1Q  Median      3Q     Max 
-1.7712 -0.7675 -0.1794  0.7673  1.9569 

Coefficients:
               Estimate Std. Error t value Pr(>|t|)    
(Intercept)    -0.46845    0.56305  -0.832  0.40581    
MPQ_ANTISOCIAL  0.03360    0.01076   3.123  0.00189 ** 
SAMPLEOfficer   0.52801    0.08865   5.956 4.89e-09 ***
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

Residual standard error: 0.9637 on 498 degrees of freedom
Multiple R-squared:  0.08052,   Adjusted R-squared:  0.07683 
F-statistic: 21.81 on 2 and 498 DF,  p-value: 8.353e-10