Country Voter ID Laws

Dataset notes:

  1. Data for compulsory national ID requirements comes from the Polity data source (Marshall & Gurr, 2020) and World Bank data.
  2. Voter registration laws were mined from a report by James and Bernal (2014).
  3. The data for registered voter turnout and compulsory voting was gathered from the International IDEA online database (International IDEA, 2021a) and merged with the CVIL.
ItemDesc DV More info Relationship with MD Direction Holds with Tightness? Interacts with Tightness?”
id_type ID type Voter ID law type; “1: Voter must give basic personal details. 2: Voter is requested/required to present proof of identification or voter registration card, not photo ID. 3: Voter to present photo identification and/or other biometric data. Marginal Negative No No
reg_law Regulation “1: Lassie-faire 2: Assisted 3: Automatic” Not Significant NA NA No
min_id Min. # of Ids What is the minimum number of IDs required by a jurisdictions electoral law Significant Negative No No
reg_law Number of IDs How many different IDs are allowed for voters to prove their identity? Not Significant NA NA No
cmp_id Compulsory voting Does the country have compulsory voting? From: Source: https://www.idea.int/datatools/question-view/576/ Marginal Positive No No
num_id Compulsory ID Does the country have compulsory national ID cards? (The World Bank 2021) Significant Negative No No

Analyses

Main Effect

ID type (Higher numbers = stricter)

## formula: id_type ~ PC1_SDS
## data:    cntryvoter_data
## 
##  link  threshold nobs logLik AIC   niter max.grad cond.H 
##  logit flexible  45   -23.86 53.71 5(0)  2.55e-09 7.2e+01
## 
## Coefficients:
##         Estimate Std. Error z value Pr(>|z|)  
## PC1_SDS  -0.1679     0.0978   -1.72    0.086 .
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Threshold coefficients:
##     Estimate Std. Error z value
## 1|2   -3.194      0.758   -4.21
## 2|3   -1.553      0.420   -3.70
## (2 observations deleted due to missingness)

Registration Law Type (Higher numbers = stricter)

## formula: reg_law ~ PC1_SDS
## data:    cntryvoter_data
## 
##  link  threshold nobs logLik AIC   niter max.grad cond.H 
##  logit flexible  45   -41.47 88.93 4(1)  4.62e-08 5.3e+01
## 
## Coefficients:
##         Estimate Std. Error z value Pr(>|z|)
## PC1_SDS  -0.0726     0.0624   -1.16     0.24
## 
## Threshold coefficients:
##     Estimate Std. Error z value
## 1|2    0.302      0.314    0.96
## 2|3    0.786      0.333    2.36
## (2 observations deleted due to missingness)

Number of ID cards

## 
## Call:
## lm(formula = num_id ~ PC1_SDS, data = cntryvoter_data)
## 
## Residuals:
##    Min     1Q Median     3Q    Max 
##  -5.51  -2.56  -1.34   0.45  49.07 
## 
## Coefficients:
##             Estimate Std. Error t value Pr(>|t|)  
## (Intercept)    3.301      1.234    2.67    0.011 *
## PC1_SDS        0.305      0.260    1.17    0.248  
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 8 on 41 degrees of freedom
##   (4 observations deleted due to missingness)
## Multiple R-squared:  0.0324, Adjusted R-squared:  0.00875 
## F-statistic: 1.37 on 1 and 41 DF,  p-value: 0.248

Minimum number of IDs

## 
## Call:
## lm(formula = min_id ~ PC1_SDS, data = cntryvoter_data)
## 
## Residuals:
##     Min      1Q  Median      3Q     Max 
## -0.8394 -0.1239  0.0135  0.0685  1.0608 
## 
## Coefficients:
##             Estimate Std. Error t value            Pr(>|t|)    
## (Intercept)  0.98682    0.04481   22.02 <0.0000000000000002 ***
## PC1_SDS     -0.02314    0.00944   -2.45               0.019 *  
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 0.29 on 41 degrees of freedom
##   (4 observations deleted due to missingness)
## Multiple R-squared:  0.128,  Adjusted R-squared:  0.106 
## F-statistic:    6 on 1 and 41 DF,  p-value: 0.0186

Compulsory voting (binary)

## 
## Call:
## glm(formula = cmp_vt ~ PC1_SDS, family = "binomial", data = cntryvoter_data)
## 
## Coefficients:
##             Estimate Std. Error z value Pr(>|z|)  
## (Intercept)  -0.8769     0.3414   -2.57    0.010 *
## PC1_SDS       0.1277     0.0771    1.66    0.098 .
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 54.104  on 44  degrees of freedom
## Residual deviance: 51.090  on 43  degrees of freedom
##   (2 observations deleted due to missingness)
## AIC: 55.09
## 
## Number of Fisher Scoring iterations: 4
## (Intercept)     PC1_SDS 
##        0.42        1.14
##             2.5 % 97.5 %
## (Intercept)  0.20   0.79
## PC1_SDS      0.98   1.34

Compulsory ID (binary)

## 
## Call:
## glm(formula = cmp_id ~ PC1_SDS, family = "binomial", data = cntryvoter_data)
## 
## Coefficients:
##             Estimate Std. Error z value Pr(>|z|)  
## (Intercept)    6.869      3.227    2.13    0.033 *
## PC1_SDS       -1.114      0.525   -2.12    0.034 *
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 26.9964  on 44  degrees of freedom
## Residual deviance:  9.8857  on 43  degrees of freedom
##   (2 observations deleted due to missingness)
## AIC: 13.89
## 
## Number of Fisher Scoring iterations: 9
## (Intercept)     PC1_SDS 
##      961.80        0.33
##              2.5 %      97.5 %
## (Intercept) 18.603 53012241.84
## PC1_SDS      0.062        0.67

Tightness (Control)

ID type (Higher numbers = stricter)

## formula: id_type ~ PC1_SDS + Tightness_adjusted_scale
## data:    cntryvoter_data
## 
##  link  threshold nobs logLik AIC   niter max.grad cond.H 
##  logit flexible  22   -4.38  16.76 8(0)  4.17e-08 7.2e+03
## 
## Coefficients:
##                          Estimate Std. Error z value Pr(>|z|)  
## PC1_SDS                     -1.96       1.18   -1.66    0.096 .
## Tightness_adjusted_scale    -1.65       4.17   -0.40    0.693  
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Threshold coefficients:
##     Estimate Std. Error z value
## 1|2   -16.55      10.39   -1.59
## 2|3   -11.21       8.65   -1.30
## (25 observations deleted due to missingness)

Registration Law Type (Higher numbers = stricter)

## formula: reg_law ~ PC1_SDS + Tightness_adjusted_scale
## data:    cntryvoter_data
## 
##  link  threshold nobs logLik AIC   niter max.grad cond.H 
##  logit flexible  22   -17.38 42.77 5(1)  4.67e-10 2.0e+03
## 
## Coefficients:
##                          Estimate Std. Error z value Pr(>|z|)
## PC1_SDS                    -0.212      0.163   -1.30     0.19
## Tightness_adjusted_scale    0.509      2.012    0.25     0.80
## 
## Threshold coefficients:
##     Estimate Std. Error z value
## 1|2     1.33       4.01    0.33
## 2|3     1.82       4.03    0.45
## (25 observations deleted due to missingness)

Number of ID cards

## 
## Call:
## lm(formula = num_id ~ PC1_SDS + Tightness_adjusted_scale, data = cntryvoter_data)
## 
## Residuals:
##    Min     1Q Median     3Q    Max 
##  -9.53  -3.96  -2.15   0.29  45.33 
## 
## Coefficients:
##                          Estimate Std. Error t value Pr(>|t|)
## (Intercept)                 6.719     21.384    0.31     0.76
## PC1_SDS                     0.978      0.866    1.13     0.27
## Tightness_adjusted_scale   -1.793     10.936   -0.16     0.87
## 
## Residual standard error: 12 on 18 degrees of freedom
##   (26 observations deleted due to missingness)
## Multiple R-squared:  0.0807, Adjusted R-squared:  -0.0214 
## F-statistic: 0.79 on 2 and 18 DF,  p-value: 0.469

Minimum number of IDs

## 
## Call:
## lm(formula = min_id ~ PC1_SDS + Tightness_adjusted_scale, data = cntryvoter_data)
## 
## Residuals:
##     Min      1Q  Median      3Q     Max 
## -0.8570 -0.0784 -0.0315  0.0504  0.9533 
## 
## Coefficients:
##                          Estimate Std. Error t value Pr(>|t|)  
## (Intercept)                1.6008     0.5875    2.72    0.014 *
## PC1_SDS                   -0.0292     0.0238   -1.23    0.235  
## Tightness_adjusted_scale  -0.2930     0.3004   -0.98    0.342  
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 0.32 on 18 degrees of freedom
##   (26 observations deleted due to missingness)
## Multiple R-squared:  0.0949, Adjusted R-squared:  -0.00571 
## F-statistic: 0.943 on 2 and 18 DF,  p-value: 0.408

Compulsory voting (binary)

## 
## Call:
## glm(formula = cmp_vt ~ PC1_SDS + Tightness_adjusted_scale, family = "binomial", 
##     data = cntryvoter_data)
## 
## Coefficients:
##                          Estimate Std. Error z value Pr(>|z|)
## (Intercept)                3.9532     3.8934    1.02     0.31
## PC1_SDS                    0.0761     0.1621    0.47     0.64
## Tightness_adjusted_scale  -2.3322     2.0114   -1.16     0.25
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 29.767  on 21  degrees of freedom
## Residual deviance: 27.427  on 19  degrees of freedom
##   (25 observations deleted due to missingness)
## AIC: 33.43
## 
## Number of Fisher Scoring iterations: 4

Compulsory ID (binary)

## 
## Call:
## glm(formula = cmp_id ~ PC1_SDS + Tightness_adjusted_scale, family = "binomial", 
##     data = cntryvoter_data)
## 
## Coefficients:
##                          Estimate Std. Error z value Pr(>|z|)
## (Intercept)                -20.64      36.72   -0.56     0.57
## PC1_SDS                     -2.97       3.85   -0.77     0.44
## Tightness_adjusted_scale    20.32      31.30    0.65     0.52
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 13.4040  on 21  degrees of freedom
## Residual deviance:  5.4516  on 19  degrees of freedom
##   (25 observations deleted due to missingness)
## AIC: 11.45
## 
## Number of Fisher Scoring iterations: 11

Tightness (Moderator)

ID type (Higher numbers = stricter)

## formula: id_type ~ PC1_SDS * Tightness_adjusted_scale
## data:    cntryvoter_data
## 
##  link  threshold nobs logLik AIC   niter max.grad cond.H 
##  logit flexible  22   -0.00  10.00 26(0) 3.12e-09 4.9e+07
## 
## Coefficients:
##                                  Estimate Std. Error z value Pr(>|z|)
## PC1_SDS                             -2841         NA      NA       NA
## Tightness_adjusted_scale            -4160         NA      NA       NA
## PC1_SDS:Tightness_adjusted_scale     1206         NA      NA       NA
## 
## Threshold coefficients:
##     Estimate Std. Error z value
## 1|2   -11240         NA      NA
## 2|3    -9844         NA      NA
## (25 observations deleted due to missingness)

Registration Law Type (Higher numbers = stricter)

## formula: reg_law ~ PC1_SDS * Tightness_adjusted_scale
## data:    cntryvoter_data
## 
##  link  threshold nobs logLik AIC   niter max.grad cond.H 
##  logit flexible  22   -16.39 42.78 5(1)  1.34e-08 1.2e+04
## 
## Coefficients:
##                                  Estimate Std. Error z value Pr(>|z|)
## PC1_SDS                             2.181      1.753    1.24     0.21
## Tightness_adjusted_scale            1.113      2.216    0.50     0.62
## PC1_SDS:Tightness_adjusted_scale   -1.154      0.848   -1.36     0.17
## 
## Threshold coefficients:
##     Estimate Std. Error z value
## 1|2     3.14       4.51    0.70
## 2|3     3.67       4.54    0.81
## (25 observations deleted due to missingness)

Number of ID cards

## 
## Call:
## lm(formula = num_id ~ PC1_SDS * Tightness_adjusted_scale, data = cntryvoter_data)
## 
## Residuals:
##    Min     1Q Median     3Q    Max 
## -12.78  -3.19  -1.49   0.92  41.24 
## 
## Coefficients:
##                                  Estimate Std. Error t value Pr(>|t|)
## (Intercept)                         -6.86      23.36   -0.29     0.77
## PC1_SDS                             13.10       9.25    1.42     0.17
## Tightness_adjusted_scale             3.84      11.54    0.33     0.74
## PC1_SDS:Tightness_adjusted_scale    -5.87       4.46   -1.32     0.21
## 
## Residual standard error: 11 on 17 degrees of freedom
##   (26 observations deleted due to missingness)
## Multiple R-squared:  0.166,  Adjusted R-squared:  0.0185 
## F-statistic: 1.13 on 3 and 17 DF,  p-value: 0.366

Minimum number of IDs

## 
## Call:
## lm(formula = min_id ~ PC1_SDS * Tightness_adjusted_scale, data = cntryvoter_data)
## 
## Residuals:
##     Min      1Q  Median      3Q     Max 
## -0.8171 -0.0839 -0.0199  0.0685  0.9596 
## 
## Coefficients:
##                                  Estimate Std. Error t value Pr(>|t|)  
## (Intercept)                        1.7679     0.6675    2.65    0.017 *
## PC1_SDS                           -0.1784     0.2643   -0.68    0.509  
## Tightness_adjusted_scale          -0.3623     0.3298   -1.10    0.287  
## PC1_SDS:Tightness_adjusted_scale   0.0722     0.1274    0.57    0.578  
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 0.32 on 17 degrees of freedom
##   (26 observations deleted due to missingness)
## Multiple R-squared:  0.112,  Adjusted R-squared:  -0.0451 
## F-statistic: 0.712 on 3 and 17 DF,  p-value: 0.558

Compulsory voting (binary)

## 
## Call:
## glm(formula = cmp_vt ~ PC1_SDS * Tightness_adjusted_scale, family = "binomial", 
##     data = cntryvoter_data)
## 
## Coefficients:
##                                  Estimate Std. Error z value Pr(>|z|)
## (Intercept)                          7.62       5.49    1.39     0.16
## PC1_SDS                             -2.59       2.16   -1.20     0.23
## Tightness_adjusted_scale            -4.02       2.81   -1.43     0.15
## PC1_SDS:Tightness_adjusted_scale     1.32       1.08    1.22     0.22
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 29.767  on 21  degrees of freedom
## Residual deviance: 25.551  on 18  degrees of freedom
##   (25 observations deleted due to missingness)
## AIC: 33.55
## 
## Number of Fisher Scoring iterations: 5

Compulsory ID (binary)

## 
## Call:
## glm(formula = cmp_id ~ PC1_SDS * Tightness_adjusted_scale, family = "binomial", 
##     data = cntryvoter_data)
## 
## Coefficients:
##                                  Estimate Std. Error z value Pr(>|z|)
## (Intercept)                           748     693062       0        1
## PC1_SDS                              -282     255626       0        1
## Tightness_adjusted_scale             -284     274318       0        1
## PC1_SDS:Tightness_adjusted_scale      128     117247       0        1
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 13.4039882833665  on 21  degrees of freedom
## Residual deviance:  0.0000000020133  on 18  degrees of freedom
##   (25 observations deleted due to missingness)
## AIC: 8
## 
## Number of Fisher Scoring iterations: 25