Data

load("merged_data.RData")

merged_data <- merged_data %>%
    mutate(new_class = case_when(is.na(featurew_class) ~ 1,
            featurew_class == "1" ~ 2,
            featurew_class == "2" ~ 0,
            featurew_class == "3" ~ 3,
            TRUE ~ 0),
        classtwo = ifelse(is.na(totalw_class), 0, totalw_class))

vac_comp_all

model <- glm(vac_comp_all ~factor(new_class)+sexident3+race+age, family = binomial(link = "probit"), data = merged_data)
summary(model)
## 
## Call:
## glm(formula = vac_comp_all ~ factor(new_class) + sexident3 + 
##     race + age, family = binomial(link = "probit"), data = merged_data)
## 
## Coefficients:
##                    Estimate Std. Error z value Pr(>|z|)   
## (Intercept)        -1.92036    0.69103  -2.779  0.00545 **
## factor(new_class)1 -0.30085    0.37233  -0.808  0.41909   
## factor(new_class)2  0.35168    0.37742   0.932  0.35144   
## factor(new_class)3  0.19203    0.35975   0.534  0.59349   
## sexident3          -0.11792    0.09268  -1.272  0.20324   
## race               -0.06161    0.03028  -2.034  0.04192 * 
## age                 0.03073    0.02527   1.216  0.22396   
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 647.77  on 1221  degrees of freedom
## Residual deviance: 621.46  on 1215  degrees of freedom
##   (5 observations deleted due to missingness)
## AIC: 635.46
## 
## Number of Fisher Scoring iterations: 6
odds_ratios <- exp(coef(model))
print(odds_ratios)
##        (Intercept) factor(new_class)1 factor(new_class)2 factor(new_class)3 
##          0.1465544          0.7401922          1.4214562          1.2117103 
##          sexident3               race                age 
##          0.8887632          0.9402535          1.0312042
exp(cbind(OR = coef(model), confint(model)))
## Waiting for profiling to be done...
##                           OR      2.5 %    97.5 %
## (Intercept)        0.1465544 0.03604343 0.5584559
## factor(new_class)1 0.7401922 0.37590092 1.6551975
## factor(new_class)2 1.4214562 0.71371803 3.2059712
## factor(new_class)3 1.2117103 0.63429105 2.6575621
## sexident3          0.8887632 0.73723757 1.0614893
## race               0.9402535 0.88409909 0.9961046
## age                1.0312042 0.98146154 1.0846031
model <- glm(vac_comp_all ~factor(classtwo)+sexident3+race+age,
    family = binomial(link = "probit"), data = merged_data)
summary(model)
## 
## Call:
## glm(formula = vac_comp_all ~ factor(classtwo) + sexident3 + race + 
##     age, family = binomial(link = "probit"), data = merged_data)
## 
## Coefficients:
##                   Estimate Std. Error z value Pr(>|z|)    
## (Intercept)       -2.17152    0.59429  -3.654 0.000258 ***
## factor(classtwo)1  0.50724    0.14030   3.615 0.000300 ***
## factor(classtwo)2  0.54515    0.16342   3.336 0.000850 ***
## sexident3         -0.11904    0.09270  -1.284 0.199072    
## race              -0.06288    0.03022  -2.080 0.037497 *  
## age                0.02869    0.02517   1.140 0.254337    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 647.77  on 1221  degrees of freedom
## Residual deviance: 622.88  on 1216  degrees of freedom
##   (5 observations deleted due to missingness)
## AIC: 634.88
## 
## Number of Fisher Scoring iterations: 6
odds_ratios <- exp(coef(model))
print(odds_ratios)
##       (Intercept) factor(classtwo)1 factor(classtwo)2         sexident3 
##         0.1140037         1.6606945         1.7248751         0.8877727 
##              race               age 
##         0.9390589         1.0291097
exp(cbind(OR = coef(model), confint(model)))
## Waiting for profiling to be done...
##                          OR      2.5 %    97.5 %
## (Intercept)       0.1140037 0.03463712 0.3633420
## factor(classtwo)1 1.6606945 1.26945383 2.2042810
## factor(classtwo)2 1.7248751 1.25491718 2.3859373
## sexident3         0.8877727 0.73637578 1.0603393
## race              0.9390589 0.88306612 0.9947287
## age               1.0291097 0.97966908 1.0821213

vac_init_all

model <- glm(vac_init_all ~factor(new_class)+sexident3+race+age, family = binomial(link = "probit"), data = merged_data)
summary(model)
## 
## Call:
## glm(formula = vac_init_all ~ factor(new_class) + sexident3 + 
##     race + age, family = binomial(link = "probit"), data = merged_data)
## 
## Coefficients:
##                     Estimate Std. Error z value Pr(>|z|)  
## (Intercept)        -0.874484   0.476255  -1.836   0.0663 .
## factor(new_class)1  0.264528   0.269531   0.981   0.3264  
## factor(new_class)2  0.580122   0.281583   2.060   0.0394 *
## factor(new_class)3  0.467867   0.266368   1.756   0.0790 .
## sexident3          -0.057956   0.059964  -0.967   0.3338  
## race               -0.027697   0.018840  -1.470   0.1415  
## age                 0.007293   0.017000   0.429   0.6679  
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 1542.4  on 1221  degrees of freedom
## Residual deviance: 1527.5  on 1215  degrees of freedom
##   (5 observations deleted due to missingness)
## AIC: 1541.5
## 
## Number of Fisher Scoring iterations: 4
odds_ratios <- exp(coef(model))
print(odds_ratios)
##        (Intercept) factor(new_class)1 factor(new_class)2 factor(new_class)3 
##          0.4170774          1.3028155          1.7862567          1.5965850 
##          sexident3               race                age 
##          0.9436913          0.9726831          1.0073195
exp(cbind(OR = coef(model), confint(model)))
## Waiting for profiling to be done...
##                           OR     2.5 %   97.5 %
## (Intercept)        0.4170774 0.1616093 1.058109
## factor(new_class)1 1.3028155 0.7798768 2.254622
## factor(new_class)2 1.7862567 1.0428225 3.160046
## factor(new_class)3 1.5965850 0.9621169 2.747146
## sexident3          0.9436913 0.8383069 1.061205
## race               0.9726831 0.9370776 1.009262
## age                1.0073195 0.9743462 1.041512
model <- glm(vac_init_all ~factor(classtwo)+sexident3+race+age, family = binomial(link = "probit"), data = merged_data)
summary(model)
## 
## Call:
## glm(formula = vac_init_all ~ factor(classtwo) + sexident3 + race + 
##     age, family = binomial(link = "probit"), data = merged_data)
## 
## Coefficients:
##                    Estimate Std. Error z value Pr(>|z|)  
## (Intercept)       -0.568985   0.395505  -1.439   0.1503  
## factor(classtwo)1  0.203705   0.085109   2.393   0.0167 *
## factor(classtwo)2  0.221161   0.106746   2.072   0.0383 *
## sexident3         -0.056179   0.059920  -0.938   0.3485  
## race              -0.028282   0.018788  -1.505   0.1322  
## age                0.005398   0.016969   0.318   0.7504  
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for binomial family taken to be 1)
## 
##     Null deviance: 1542.4  on 1221  degrees of freedom
## Residual deviance: 1532.0  on 1216  degrees of freedom
##   (5 observations deleted due to missingness)
## AIC: 1544
## 
## Number of Fisher Scoring iterations: 4
odds_ratios <- exp(coef(model))
print(odds_ratios)
##       (Intercept) factor(classtwo)1 factor(classtwo)2         sexident3 
##         0.5660996         1.2259363         1.2475238         0.9453696 
##              race               age 
##         0.9721141         1.0054130
exp(cbind(OR = coef(model), confint(model)))
## Waiting for profiling to be done...
##                          OR     2.5 %   97.5 %
## (Intercept)       0.5660996 0.2606018 1.226134
## factor(classtwo)1 1.2259363 1.0379249 1.449002
## factor(classtwo)2 1.2475238 1.0117580 1.537683
## sexident3         0.9453696 0.8399326 1.062936
## race              0.9721141 0.9366107 1.008586
## age               1.0054130 0.9725670 1.039462