Zero-sum mindset predicts general beliefs that the world is scarce, competitive, unimprovable, interconnected, hierarchical, dangerous, and unfair.
The zero-sum mindset is not reducible to any single primal world belief or personality trait.
# Load required libraries
library(glmnet)
library(here)
library(readr)
library(tidyverse)
library(apaTables)s6 <- read_csv(here("data", "s6.csv"))s6_subset <- select(s6, zsm_mean, p_changing, p_interconnected, p_abundant, p_hierarchical, p_just, p_cooperative, p_improvable , p_harmless, O, C, E, A, N)apaTables::apa.cor.table(s6_subset)##
##
## Means, standard deviations, and correlations with confidence intervals
##
##
## Variable M SD 1 2 3
## 1. zsm_mean 2.67 1.02
##
## 2. p_changing 3.49 0.81 .01
## [-.08, .10]
##
## 3. p_interconnected 3.47 0.97 -.11* .25**
## [-.20, -.02] [.17, .33]
##
## 4. p_abundant 3.20 0.89 -.35** -.12** .15**
## [-.42, -.27] [-.21, -.04] [.06, .23]
##
## 5. p_hierarchical 2.50 0.96 .26** .06 -.07
## [.18, .34] [-.02, .15] [-.15, .02]
##
## 6. p_just 2.44 0.95 -.15** -.11* .04
## [-.24, -.06] [-.20, -.03] [-.05, .13]
##
## 7. p_cooperative 2.51 1.05 -.37** -.25** .05
## [-.44, -.29] [-.33, -.17] [-.04, .14]
##
## 8. p_improvable 3.32 0.80 -.30** .03 .31**
## [-.38, -.22] [-.06, .12] [.23, .38]
##
## 9. p_harmless 2.13 1.00 -.23** -.32** .04
## [-.31, -.14] [-.40, -.24] [-.04, .13]
##
## 10. O 3.96 0.80 -.09* .12** .27**
## [-.18, -.01] [.03, .21] [.18, .34]
##
## 11. C 3.33 0.98 -.05 -.08 .05
## [-.14, .04] [-.17, .01] [-.03, .14]
##
## 12. E 2.71 0.95 -.03 -.10* .08
## [-.12, .05] [-.19, -.02] [-.01, .17]
##
## 13. A 3.76 0.63 -.32** .05 .19**
## [-.39, -.23] [-.04, .13] [.10, .27]
##
## 14. N 3.06 0.97 .09 .15** -.02
## [-.00, .17] [.07, .24] [-.11, .07]
##
## 4 5 6 7 8 9
##
##
##
##
##
##
##
##
##
##
##
## -.06
## [-.15, .03]
##
## .53** .15**
## [.46, .59] [.06, .24]
##
## .49** -.33** .41**
## [.42, .55] [-.41, -.25] [.33, .48]
##
## .57** -.03 .47** .39**
## [.50, .62] [-.11, .06] [.40, .54] [.32, .47]
##
## .38** -.18** .34** .58** .32**
## [.30, .45] [-.27, -.10] [.26, .41] [.52, .63] [.24, .40]
##
## .11* -.18** -.06 .06 .18** -.05
## [.02, .20] [-.26, -.09] [-.15, .03] [-.03, .14] [.10, .27] [-.13, .04]
##
## .14** .12** .19** .04 .17** .02
## [.06, .23] [.04, .21] [.11, .28] [-.05, .12] [.09, .26] [-.07, .11]
##
## .28** .04 .25** .13** .28** .14**
## [.20, .36] [-.05, .13] [.17, .33] [.05, .22] [.20, .36] [.06, .23]
##
## .23** -.16** .19** .29** .28** .28**
## [.15, .31] [-.25, -.08] [.10, .27] [.21, .37] [.20, .36] [.19, .36]
##
## -.38** -.03 -.35** -.26** -.33** -.31**
## [-.45, -.30] [-.11, .06] [-.42, -.27] [-.34, -.17] [-.41, -.25] [-.39, -.23]
##
## 10 11 12 13
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
## -.01
## [-.10, .08]
##
## .20** .33**
## [.11, .28] [.25, .41]
##
## .16** .16** .15**
## [.07, .24] [.07, .24] [.07, .24]
##
## -.09 -.47** -.50** -.25**
## [-.17, .00] [-.53, -.40] [-.57, -.44] [-.33, -.16]
##
##
## Note. M and SD are used to represent mean and standard deviation, respectively.
## Values in square brackets indicate the 95% confidence interval.
## The confidence interval is a plausible range of population correlations
## that could have caused the sample correlation (Cumming, 2014).
## * indicates p < .05. ** indicates p < .01.
## test <- lm(scale(s6$zsm_mean) ~ scale(s6$p_abundant) + scale(s6$p_cooperative) + scale(s6$p_hierarchical) + scale(s6$p_improvable) + scale(s6$p_just) + scale(s6$p_changing) + scale(s6$p_harmless) + scale(s6$p_interconnected) + scale(s6$E) + scale(s6$N) + scale(s6$A) + scale(s6$C) + scale(s6$O))
summary(test)##
## Call:
## lm(formula = scale(s6$zsm_mean) ~ scale(s6$p_abundant) + scale(s6$p_cooperative) +
## scale(s6$p_hierarchical) + scale(s6$p_improvable) + scale(s6$p_just) +
## scale(s6$p_changing) + scale(s6$p_harmless) + scale(s6$p_interconnected) +
## scale(s6$E) + scale(s6$N) + scale(s6$A) + scale(s6$C) + scale(s6$O))
##
## Residuals:
## Min 1Q Median 3Q Max
## -2.2682 -0.5857 -0.0849 0.5091 4.1309
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) -2.051e-15 3.899e-02 0.000 1.000000
## scale(s6$p_abundant) -2.248e-01 5.400e-02 -4.163 3.71e-05 ***
## scale(s6$p_cooperative) -1.782e-01 5.622e-02 -3.170 0.001619 **
## scale(s6$p_hierarchical) 1.564e-01 4.481e-02 3.489 0.000528 ***
## scale(s6$p_improvable) -1.249e-01 5.299e-02 -2.358 0.018788 *
## scale(s6$p_just) 6.807e-02 5.173e-02 1.316 0.188847
## scale(s6$p_changing) -3.928e-02 4.427e-02 -0.887 0.375435
## scale(s6$p_harmless) 9.798e-03 5.209e-02 0.188 0.850881
## scale(s6$p_interconnected) 1.788e-02 4.389e-02 0.407 0.683926
## scale(s6$E) 6.048e-02 4.706e-02 1.285 0.199308
## scale(s6$N) -9.199e-02 5.285e-02 -1.740 0.082432 .
## scale(s6$A) -1.932e-01 4.355e-02 -4.436 1.13e-05 ***
## scale(s6$C) -5.803e-02 4.592e-02 -1.264 0.206943
## scale(s6$O) 7.467e-03 4.315e-02 0.173 0.862697
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.8709 on 485 degrees of freedom
## Multiple R-squared: 0.2613, Adjusted R-squared: 0.2415
## F-statistic: 13.19 on 13 and 485 DF, p-value: < 2.2e-16apa.reg.table(test)##
##
## Regression results using scale(s6$zsm_mean) as the criterion
##
##
## Predictor b b_95%_CI beta beta_95%_CI sr2
## (Intercept) -0.00 [-0.08, 0.08]
## scale(s6$p_abundant) -0.22** [-0.33, -0.12] -0.22 [-0.33, -0.12] .03
## scale(s6$p_cooperative) -0.18** [-0.29, -0.07] -0.18 [-0.29, -0.07] .02
## scale(s6$p_hierarchical) 0.16** [0.07, 0.24] 0.16 [0.07, 0.24] .02
## scale(s6$p_improvable) -0.12* [-0.23, -0.02] -0.12 [-0.23, -0.02] .01
## scale(s6$p_just) 0.07 [-0.03, 0.17] 0.07 [-0.03, 0.17] .00
## scale(s6$p_changing) -0.04 [-0.13, 0.05] -0.04 [-0.13, 0.05] .00
## scale(s6$p_harmless) 0.01 [-0.09, 0.11] 0.01 [-0.09, 0.11] .00
## scale(s6$p_interconnected) 0.02 [-0.07, 0.10] 0.02 [-0.07, 0.10] .00
## scale(s6$E) 0.06 [-0.03, 0.15] 0.06 [-0.03, 0.15] .00
## scale(s6$N) -0.09 [-0.20, 0.01] -0.09 [-0.20, 0.01] .00
## scale(s6$A) -0.19** [-0.28, -0.11] -0.19 [-0.28, -0.11] .03
## scale(s6$C) -0.06 [-0.15, 0.03] -0.06 [-0.15, 0.03] .00
## scale(s6$O) 0.01 [-0.08, 0.09] 0.01 [-0.08, 0.09] .00
##
##
##
## sr2_95%_CI r Fit
##
## [.00, .05] -.35**
## [-.00, .03] -.37**
## [-.00, .04] .26**
## [-.01, .02] -.30**
## [-.01, .01] -.15**
## [-.00, .01] .01
## [-.00, .00] -.23**
## [-.00, .00] -.11*
## [-.01, .01] -.03
## [-.01, .01] .09
## [.00, .06] -.32**
## [-.01, .01] -.05
## [-.00, .00] -.09*
## R2 = .261**
## 95% CI[.18,.30]
##
##
## Note. A significant b-weight indicates the beta-weight and semi-partial correlation are also significant.
## b represents unstandardized regression weights. beta indicates the standardized regression weights.
## sr2 represents the semi-partial correlation squared. r represents the zero-order correlation.
## Square brackets are used to enclose the lower and upper limits of a confidence interval.
## * indicates p < .05. ** indicates p < .01.
## #adds demographics
test <- lm(scale(s6$zsm_mean) ~ scale(s6$p_abundant) + scale(s6$p_cooperative) + scale(s6$p_hierarchical) + scale(s6$p_improvable) + scale(s6$p_just) + scale(s6$p_changing) + scale(s6$p_harmless) + scale(s6$p_interconnected) + scale(s6$E) + scale(s6$N) + scale(s6$A) + scale(s6$C) + scale(s6$O) + scale(s6$age) + scale(s6$income) + scale(s6$education) + s6$sex + s6$race)
summary(test)##
## Call:
## lm(formula = scale(s6$zsm_mean) ~ scale(s6$p_abundant) + scale(s6$p_cooperative) +
## scale(s6$p_hierarchical) + scale(s6$p_improvable) + scale(s6$p_just) +
## scale(s6$p_changing) + scale(s6$p_harmless) + scale(s6$p_interconnected) +
## scale(s6$E) + scale(s6$N) + scale(s6$A) + scale(s6$C) + scale(s6$O) +
## scale(s6$age) + scale(s6$income) + scale(s6$education) +
## s6$sex + s6$race)
##
## Residuals:
## Min 1Q Median 3Q Max
## -2.3097 -0.5205 -0.0687 0.4702 4.1150
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.16073 0.14946 1.075 0.282752
## scale(s6$p_abundant) -0.21259 0.05625 -3.779 0.000179 ***
## scale(s6$p_cooperative) -0.18589 0.05887 -3.158 0.001697 **
## scale(s6$p_hierarchical) 0.14183 0.04627 3.065 0.002306 **
## scale(s6$p_improvable) -0.11024 0.05557 -1.984 0.047887 *
## scale(s6$p_just) 0.05111 0.05417 0.944 0.345918
## scale(s6$p_changing) -0.07132 0.04627 -1.542 0.123895
## scale(s6$p_harmless) 0.02243 0.05446 0.412 0.680583
## scale(s6$p_interconnected) 0.02259 0.04573 0.494 0.621659
## scale(s6$E) 0.08260 0.04926 1.677 0.094250 .
## scale(s6$N) -0.07208 0.05769 -1.249 0.212136
## scale(s6$A) -0.18903 0.04586 -4.122 4.48e-05 ***
## scale(s6$C) -0.02784 0.04851 -0.574 0.566265
## scale(s6$O) -0.00646 0.04541 -0.142 0.886927
## scale(s6$age) -0.07631 0.04574 -1.668 0.095944 .
## scale(s6$income) -0.06987 0.04444 -1.572 0.116594
## scale(s6$education) -0.01249 0.04292 -0.291 0.771130
## s6$sexMale 0.05663 0.08980 0.631 0.528589
## s6$raceBlack/African American 0.09222 0.20932 0.441 0.659741
## s6$raceLatinx -0.05839 0.20398 -0.286 0.774806
## s6$raceMixed -0.47407 0.22590 -2.099 0.036412 *
## s6$raceOther 0.17377 0.32559 0.534 0.593801
## s6$raceWhite -0.24689 0.15221 -1.622 0.105490
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.8706 on 448 degrees of freedom
## (28 observations deleted due to missingness)
## Multiple R-squared: 0.2908, Adjusted R-squared: 0.256
## F-statistic: 8.349 on 22 and 448 DF, p-value: < 2.2e-16s6_subset <- na.omit(s6_subset)set.seed(1)
train_index <- sample(1:nrow(s6_subset), 0.7 * nrow(s6_subset))
train_data <- s6_subset[train_index, ]
test_data <- s6_subset[-train_index, ]class(train_data$zsm_mean)## [1] "numeric"#x <- as.matrix(train_data[, -1])
x <- select(train_data, -zsm_mean)
y <- train_data$zsm_mean # Assuming "zsm_mean" is the first column in the subset# Impute missing values in x
x <- makeX(x)
lasso_model <- cv.glmnet(x, y, alpha = 1)
# Find the optimal lambda value
opt_lambda <- lasso_model$lambda.min
# Fit the final Lasso model with optimal lambda
lasso_final <- glmnet(x, y, alpha = 1, lambda = opt_lambda)
summary(lasso_final)## Length Class Mode
## a0 1 -none- numeric
## beta 13 dgCMatrix S4
## df 1 -none- numeric
## dim 2 -none- numeric
## lambda 1 -none- numeric
## dev.ratio 1 -none- numeric
## nulldev 1 -none- numeric
## npasses 1 -none- numeric
## jerr 1 -none- numeric
## offset 1 -none- logical
## call 5 -none- call
## nobs 1 -none- numericlasso_final$beta## 13 x 1 sparse Matrix of class "dgCMatrix"
## s0
## p_changing -0.03824982
## p_interconnected .
## p_abundant -0.26236402
## p_hierarchical 0.22241089
## p_just 0.03399125
## p_cooperative -0.12775129
## p_improvable -0.11830800
## p_harmless .
## O .
## C .
## E 0.09907955
## A -0.32067195
## N -0.05097192# Extract selected features
selected_features <- coef(lasso_final)[-1, ]
selected_features <- selected_features[selected_features != 0]Abundant, Cooperative, Changing, Hierarchical, Improvable, Just, Extraversion, Neuroticism, Agreeableness,
test <- lm(s6$zsm_mean ~ s6$A + s6$p_abundant + s6$p_cooperative + s6$p_hierarchical + s6$p_improvable + s6$p_just +s6$p_changing + s6$E + s6$N)
summary(test)##
## Call:
## lm(formula = s6$zsm_mean ~ s6$A + s6$p_abundant + s6$p_cooperative +
## s6$p_hierarchical + s6$p_improvable + s6$p_just + s6$p_changing +
## s6$E + s6$N)
##
## Residuals:
## Min 1Q Median 3Q Max
## -2.3595 -0.5924 -0.0917 0.5197 4.2032
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 5.22284 0.45583 11.458 < 2e-16 ***
## s6$A -0.31482 0.06903 -4.561 6.45e-06 ***
## s6$p_abundant -0.25204 0.06162 -4.090 5.04e-05 ***
## s6$p_cooperative -0.16435 0.05050 -3.255 0.001214 **
## s6$p_hierarchical 0.15829 0.04680 3.382 0.000777 ***
## s6$p_improvable -0.15149 0.06425 -2.358 0.018778 *
## s6$p_just 0.06769 0.05459 1.240 0.215590
## s6$p_changing -0.04264 0.05191 -0.822 0.411726
## s6$E 0.05960 0.04923 1.211 0.226627
## s6$N -0.07213 0.05096 -1.415 0.157622
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.884 on 489 degrees of freedom
## Multiple R-squared: 0.2583, Adjusted R-squared: 0.2446
## F-statistic: 18.92 on 9 and 489 DF, p-value: < 2.2e-16apaTables::apa.reg.table(test, filename = "Primals, Personality, Regression.doc")##
##
## Regression results using s6$zsm_mean as the criterion
##
##
## Predictor b b_95%_CI beta beta_95%_CI sr2 sr2_95%_CI
## (Intercept) 5.22** [4.33, 6.12]
## s6$A -0.31** [-0.45, -0.18] -0.19 [-0.28, -0.11] .03 [.01, .06]
## s6$p_abundant -0.25** [-0.37, -0.13] -0.22 [-0.33, -0.11] .03 [.00, .05]
## s6$p_cooperative -0.16** [-0.26, -0.07] -0.17 [-0.27, -0.07] .02 [-.00, .04]
## s6$p_hierarchical 0.16** [0.07, 0.25] 0.15 [0.06, 0.24] .02 [-.00, .04]
## s6$p_improvable -0.15* [-0.28, -0.03] -0.12 [-0.22, -0.02] .01 [-.01, .02]
## s6$p_just 0.07 [-0.04, 0.17] 0.06 [-0.04, 0.16] .00 [-.00, .01]
## s6$p_changing -0.04 [-0.14, 0.06] -0.03 [-0.12, 0.05] .00 [-.00, .01]
## s6$E 0.06 [-0.04, 0.16] 0.06 [-0.03, 0.15] .00 [-.00, .01]
## s6$N -0.07 [-0.17, 0.03] -0.07 [-0.16, 0.03] .00 [-.01, .01]
##
##
##
## r Fit
##
## -.32**
## -.35**
## -.37**
## .26**
## -.30**
## -.15**
## .01
## -.03
## .09
## R2 = .258**
## 95% CI[.18,.31]
##
##
## Note. A significant b-weight indicates the beta-weight and semi-partial correlation are also significant.
## b represents unstandardized regression weights. beta indicates the standardized regression weights.
## sr2 represents the semi-partial correlation squared. r represents the zero-order correlation.
## Square brackets are used to enclose the lower and upper limits of a confidence interval.
## * indicates p < .05. ** indicates p < .01.
##