Logistic Regression (predicting voter choice 2024)

Abstract

Background: Zero-sum beliefs—the perception that one group’s gains necessarily result in another group’s losses—are important predictors of political attitudes. However, the referents for zero-sum beliefs as economic or social identity remain underexplored in relation to political ideology, party affiliation, and voting behavior in contemporary elections.

Method: We conducted a comprehensive analysis examining three dimensions of zero-sum beliefs (general, economic, and social identity). Using Kruskal-Wallis tests on eleven zero-sum beliefs, we investigated how political party affiliation and racial/ethnic identity influenced endorsement of zero-sum beliefs across multiple domains. Subsequently, we examined whether these zero-sum belief patterns predicted self-reported voting for Donald Trump versus Kamala Harris in the 2024 presidential election.

Results: Political party affiliation was a significant predictor for all eight zero-sum social identity beliefs, but none of the economic or general beliefs. Republican voters and certain racial/ethnic groups demonstrated higher endorsement of zero-sum social identity beliefs. A logistic regression shows that after controlling for political ideology, a composite of zero-sum social identity beliefs explains voting behavior in the 2024 presidential election, with stronger zero-sum social identity thinking associated with Trump support and lower zero-sum social identity beliefs predicting Harris support. Other sociodemographic factors and zero-sum economic thinking were not significant predictors.

Discussion: Zero-sum social identity beliefs may represent a competitive core belief underlying contemporary political party affiliation and candidate preference. These findings affirm prior work that zero-sum thinking about economics differ from social identities, with similar levels of agreement on zero-sum economic beliefs across political parties but significantly different levels of agreement on zero-sum social identity beliefs by party affiliation. To the best of our knowledge, this study is the first to show that zero-sum thinking about social identities predicts voter preference in the 2024 election. Ultimately, future work needs to examine how to reduce zero-sum social identity thinking.

Predicting Voting Behavior for 2024 Presidential Candidate

Factor Analysis of Zero-Sum Beliefs

In [1]:

Show the code


# Load necessary libraries
library(ggplot2)
library(dplyr)


Attaching package: 'dplyr'

The following objects are masked from 'package:stats':

    filter, lag

The following objects are masked from 'package:base':

    intersect, setdiff, setequal, union

Show the code


library(tidyr)
library(ggrain)

Registered S3 methods overwritten by 'ggpp':
  method                  from   
  heightDetails.titleGrob ggplot2
  widthDetails.titleGrob  ggplot2

Show the code


library(rmarkdown)
library(readr)
library(dplyr, warn.conflicts = FALSE)
library(haven)
library(rempsyc)

Suggested APA citation: Thériault, R. (2023). rempsyc: Convenience functions for psychology. 
Journal of Open Source Software, 8(87), 5466. https://doi.org/10.21105/joss.05466

Show the code


library(knitr)
library(broom)
library(ggdist)
library(devtools)

Loading required package: usethis

Show the code


library(apaTables)
library(ggpubr)
library(psych)


Attaching package: 'psych'

The following objects are masked from 'package:ggplot2':

    %+%, alpha

Show the code


library(forcats)
library(corrplot)

corrplot 0.95 loaded

In [2]:

Show the code

select_data <- read.csv("/cloud/project/data/select_data.csv")

In [3]:

Show the code


library(psych)
# create data frame of ZEROSUM variables for factor analysis
df.ZEROSUM <- select_data[, c("ZEROSUM_1", "ZEROSUM_2", "ZEROSUM_3", "ZEROSUM_4", 
                                  "ZEROSUM_5", "ZEROSUM_6", "ZEROSUM_7", "ZEROSUM_8", 
                                  "ZEROSUM_9", "ZEROSUM_10", "ZEROSUM_11")]

# Or using dplyr to select variables
# zerosum_vars <- select_data %>% select(ZEROSUM_1:ZEROSUM_11)

# Check the correlation matrix first
cor_matrix <- cor(df.ZEROSUM, use = "complete.obs")
print(cor_matrix)

            ZEROSUM_1    ZEROSUM_2   ZEROSUM_3   ZEROSUM_4    ZEROSUM_5
ZEROSUM_1  1.00000000  0.386159282  0.09223868  0.46244644  0.354780151
ZEROSUM_2  0.38615928  1.000000000  0.47893466  0.01604828 -0.007056022
ZEROSUM_3  0.09223868  0.478934659  1.00000000 -0.02418897 -0.109827301
ZEROSUM_4  0.46244644  0.016048284 -0.02418897  1.00000000  0.611174045
ZEROSUM_5  0.35478015 -0.007056022 -0.10982730  0.61117404  1.000000000
ZEROSUM_6  0.36699616  0.031303525 -0.17086105  0.60219942  0.607185097
ZEROSUM_7  0.12009911 -0.098678831 -0.14396820  0.34550888  0.536331658
ZEROSUM_8  0.25942078 -0.155377765 -0.12939334  0.49740256  0.586804352
ZEROSUM_9  0.41016910  0.078257987 -0.12138304  0.56204342  0.628375488
ZEROSUM_10 0.29170352 -0.114520512 -0.18994979  0.49480889  0.683232725
ZEROSUM_11 0.27789618 -0.019381988 -0.18865010  0.46048999  0.570020242
             ZEROSUM_6   ZEROSUM_7  ZEROSUM_8   ZEROSUM_9 ZEROSUM_10
ZEROSUM_1   0.36699616  0.12009911  0.2594208  0.41016910  0.2917035
ZEROSUM_2   0.03130352 -0.09867883 -0.1553778  0.07825799 -0.1145205
ZEROSUM_3  -0.17086105 -0.14396820 -0.1293933 -0.12138304 -0.1899498
ZEROSUM_4   0.60219942  0.34550888  0.4974026  0.56204342  0.4948089
ZEROSUM_5   0.60718510  0.53633166  0.5868044  0.62837549  0.6832327
ZEROSUM_6   1.00000000  0.55401598  0.5067006  0.58654354  0.6812058
ZEROSUM_7   0.55401598  1.00000000  0.4203826  0.38964490  0.6145775
ZEROSUM_8   0.50670056  0.42038261  1.0000000  0.52719441  0.5667796
ZEROSUM_9   0.58654354  0.38964490  0.5271944  1.00000000  0.6009058
ZEROSUM_10  0.68120581  0.61457749  0.5667796  0.60090582  1.0000000
ZEROSUM_11  0.67721379  0.51776718  0.4961602  0.52655236  0.6724939
            ZEROSUM_11
ZEROSUM_1   0.27789618
ZEROSUM_2  -0.01938199
ZEROSUM_3  -0.18865010
ZEROSUM_4   0.46048999
ZEROSUM_5   0.57002024
ZEROSUM_6   0.67721379
ZEROSUM_7   0.51776718
ZEROSUM_8   0.49616017
ZEROSUM_9   0.52655236
ZEROSUM_10  0.67249386
ZEROSUM_11  1.00000000

Show the code


# Determine number of factors using scree plot and parallel analysis
scree(df.ZEROSUM)

Show the code


fa.parallel(df.ZEROSUM, fa = "fa")

Parallel analysis suggests that the number of factors =  2  and the number of components =  NA

Show the code


# Run 2-factor factor analysis (adjust nfactors based on scree plot/parallel analysis)
fa_result <- fa(df.ZEROSUM, 
                nfactors = 2,  # adjust this number based on your analysis
                rotate = "promax",
                fm = "ml")  # maximum likelihood

Loading required namespace: GPArotation

Show the code


# View results
print(fa_result)

Factor Analysis using method =  ml
Call: fa(r = df.ZEROSUM, nfactors = 2, rotate = "promax", fm = "ml")
Standardized loadings (pattern matrix) based upon correlation matrix
             ML1   ML2   h2   u2 com
ZEROSUM_1   0.45  0.42 0.38 0.62 2.0
ZEROSUM_2   0.00  0.92 0.85 0.15 1.0
ZEROSUM_3  -0.17  0.51 0.29 0.71 1.2
ZEROSUM_4   0.69  0.06 0.48 0.52 1.0
ZEROSUM_5   0.81  0.00 0.66 0.34 1.0
ZEROSUM_6   0.82  0.03 0.67 0.33 1.0
ZEROSUM_7   0.64 -0.12 0.42 0.58 1.1
ZEROSUM_8   0.67 -0.15 0.48 0.52 1.1
ZEROSUM_9   0.74  0.09 0.56 0.44 1.0
ZEROSUM_10  0.84 -0.12 0.72 0.28 1.0
ZEROSUM_11  0.76 -0.03 0.57 0.43 1.0

                       ML1  ML2
SS loadings           4.71 1.36
Proportion Var        0.43 0.12
Cumulative Var        0.43 0.55
Proportion Explained  0.78 0.22
Cumulative Proportion 0.78 1.00

 With factor correlations of 
      ML1   ML2
ML1  1.00 -0.01
ML2 -0.01  1.00

Mean item complexity =  1.1
Test of the hypothesis that 2 factors are sufficient.

df null model =  55  with the objective function =  5.55 with Chi Square =  646.26
df of  the model are 34  and the objective function was  0.45 

The root mean square of the residuals (RMSR) is  0.04 
The df corrected root mean square of the residuals is  0.06 

The harmonic n.obs is  121 with the empirical chi square  26.71  with prob <  0.81 
The total n.obs was  122  with Likelihood Chi Square =  51.63  with prob <  0.027 

Tucker Lewis Index of factoring reliability =  0.951
RMSEA index =  0.065  and the 90 % confidence intervals are  0.023 0.1
BIC =  -111.7
Fit based upon off diagonal values = 0.99
Measures of factor score adequacy             
                                                   ML1  ML2
Correlation of (regression) scores with factors   0.96 0.93
Multiple R square of scores with factors          0.92 0.87
Minimum correlation of possible factor scores     0.84 0.73

Show the code


fa_result$loadings


Loadings:
           ML1    ML2   
ZEROSUM_1   0.448  0.424
ZEROSUM_2          0.922
ZEROSUM_3  -0.167  0.508
ZEROSUM_4   0.689       
ZEROSUM_5   0.810       
ZEROSUM_6   0.816       
ZEROSUM_7   0.639 -0.120
ZEROSUM_8   0.674 -0.149
ZEROSUM_9   0.742       
ZEROSUM_10  0.837 -0.124
ZEROSUM_11  0.757       

                 ML1   ML2
SS loadings    4.712 1.355
Proportion Var 0.428 0.123
Cumulative Var 0.428 0.551

Show the code


# Get factor scores
factor_scores <- fa_result$scores

The results of the factor analysis – an unsupervised machine learning technique – support a two factor model with a promax rotation. The first item loads equally on each factor and will not be included in the composite construction. Based on the items, we named the first factor as ZEROSUM_ECONOMIC and the second factor as ZEROSUM_IDENTITY to correspond with the two different referents of economic (e.g., wealth vs. poor) and social identity (e.g., racial minorities vs. white people), respectively.

In [4]:

Show the code


select_data <- select_data %>%
  mutate(
    ZEROSUM_ECONOMIC = (ZEROSUM_2 + ZEROSUM_3)/2,
    ZEROSUM_IDENTITY = (ZEROSUM_4 + ZEROSUM_5 + ZEROSUM_6 + ZEROSUM_7 + ZEROSUM_8 + ZEROSUM_9 + ZEROSUM_10 + ZEROSUM_11)/8
  )

In [5]:

Show the code

library(psych)
# Alpha for ZEROSUM_IDENTITY (8 items)
alpha_identity <- psych::alpha(select_data[, c("ZEROSUM_4", "ZEROSUM_5", "ZEROSUM_6", 
                                         "ZEROSUM_7", "ZEROSUM_8", "ZEROSUM_9", 
                                         "ZEROSUM_10", "ZEROSUM_11")])
print(alpha_identity)


Reliability analysis   
Call: psych::alpha(x = select_data[, c("ZEROSUM_4", "ZEROSUM_5", "ZEROSUM_6", 
    "ZEROSUM_7", "ZEROSUM_8", "ZEROSUM_9", "ZEROSUM_10", "ZEROSUM_11")])

  raw_alpha std.alpha G6(smc) average_r S/N   ase mean  sd median_r
      0.91      0.91    0.91      0.55  10 0.012  3.1 1.4     0.56

    95% confidence boundaries 
         lower alpha upper
Feldt     0.88  0.91  0.93
Duhachek  0.88  0.91  0.93

 Reliability if an item is dropped:
           raw_alpha std.alpha G6(smc) average_r S/N alpha se  var.r med.r
ZEROSUM_4       0.90      0.90    0.90      0.57 9.3    0.013 0.0064  0.57
ZEROSUM_5       0.89      0.89    0.89      0.54 8.2    0.015 0.0081  0.53
ZEROSUM_6       0.89      0.89    0.88      0.54 8.2    0.015 0.0077  0.54
ZEROSUM_7       0.91      0.91    0.90      0.58 9.6    0.013 0.0046  0.59
ZEROSUM_8       0.90      0.90    0.90      0.57 9.2    0.013 0.0083  0.59
ZEROSUM_9       0.90      0.90    0.89      0.56 8.8    0.014 0.0080  0.57
ZEROSUM_10      0.89      0.89    0.88      0.53 8.0    0.015 0.0066  0.54
ZEROSUM_11      0.90      0.90    0.89      0.55 8.6    0.014 0.0077  0.57

 Item statistics 
             n raw.r std.r r.cor r.drop mean  sd
ZEROSUM_4  121  0.73  0.73  0.68   0.64  3.1 1.7
ZEROSUM_5  122  0.83  0.84  0.82   0.77  2.8 1.8
ZEROSUM_6  121  0.84  0.83  0.82   0.78  3.4 2.1
ZEROSUM_7  120  0.70  0.70  0.64   0.61  3.5 1.8
ZEROSUM_8  121  0.73  0.74  0.68   0.65  2.9 1.6
ZEROSUM_9  121  0.77  0.77  0.73   0.69  2.6 1.7
ZEROSUM_10 122  0.85  0.85  0.84   0.80  3.2 1.7
ZEROSUM_11 121  0.79  0.79  0.75   0.72  3.2 1.7

Non missing response frequency for each item
              1    2    3    4    5    6    7 miss
ZEROSUM_4  0.28 0.11 0.19 0.20 0.14 0.07 0.02 0.01
ZEROSUM_5  0.36 0.09 0.20 0.16 0.07 0.09 0.02 0.00
ZEROSUM_6  0.31 0.08 0.12 0.14 0.13 0.14 0.07 0.01
ZEROSUM_7  0.21 0.12 0.11 0.22 0.22 0.06 0.06 0.02
ZEROSUM_8  0.31 0.08 0.21 0.21 0.13 0.03 0.02 0.01
ZEROSUM_9  0.43 0.12 0.14 0.13 0.11 0.04 0.02 0.01
ZEROSUM_10 0.27 0.10 0.19 0.17 0.16 0.11 0.01 0.00
ZEROSUM_11 0.25 0.16 0.15 0.21 0.11 0.11 0.02 0.01

In [6]:

Show the code


library(psych)

# Alpha for ZEROSUM_ECONOMIC (2 items)
alpha_economic <- psych::alpha(select_data[, c("ZEROSUM_2", "ZEROSUM_3")])
print(alpha_economic)


Reliability analysis   
Call: psych::alpha(x = select_data[, c("ZEROSUM_2", "ZEROSUM_3")])

  raw_alpha std.alpha G6(smc) average_r S/N   ase mean  sd median_r
      0.65      0.65    0.48      0.48 1.8 0.064  4.8 1.3     0.48

    95% confidence boundaries 
         lower alpha upper
Feldt     0.50  0.65  0.75
Duhachek  0.52  0.65  0.77

 Reliability if an item is dropped:
          raw_alpha std.alpha G6(smc) average_r  S/N alpha se var.r med.r
ZEROSUM_2      0.51      0.48    0.23      0.48 0.92       NA     0  0.48
ZEROSUM_3      0.45      0.48    0.23      0.48 0.92       NA     0  0.48

 Item statistics 
            n raw.r std.r r.cor r.drop mean  sd
ZEROSUM_2 121  0.87  0.86   0.6   0.48  4.8 1.6
ZEROSUM_3 121  0.85  0.86   0.6   0.48  4.9 1.5

Non missing response frequency for each item
             1    2    3    4    5    6    7 miss
ZEROSUM_2 0.04 0.06 0.09 0.21 0.29 0.15 0.17 0.01
ZEROSUM_3 0.02 0.06 0.06 0.21 0.29 0.19 0.17 0.01