# Read the data from indices_table.txt
df <- as.data.frame(readRDS("Israel Survey/data/il_pe.RDS"))

# Remove "Other" gender category (6 respondents) if it exists to prevent regression convergence issues
df <- df %>%
  filter(gender != "Other") %>%
  mutate(gender = droplevels(gender)) # Remove unused factor level 

wave_var <- "Wave"
wave_order <- c("First", "Second", "Third", "Fourth", "Fifth", "Sixth")

community_var <- "pe_religiosity"
community_order <- c("Secular", "Religious", "National Religious", "Ultra-Orthodox") # Secular is the reference level.

dimensions_order <- c("Overall", "Cognitive", "Behavioral", "Social")

# Calculate the Political Extremism Gauge Indices
indices_result <- af_gauge_indices(df, pop_var1 = wave_var, comm_var1 = community_var)
df <- indices_result$df

# Convert data to a more manageable format for analysis
df <- df %>%
  mutate(Wave = factor(Wave, levels = wave_order)) %>%
  mutate(!!sym(community_var) := factor(!!sym(community_var), levels = community_order)) %>%
  mutate(event_occurred = factor(1, levels = c(0, 1))) %>% # used for pairwise regression
  mutate( # used for cumulative event effect regression
    post_event1 = ifelse(Wave %in% c("Second", "Third", "Fourth", "Fifth", "Sixth"), 1, 0),
    post_event2 = ifelse(Wave %in% c("Third", "Fourth", "Fifth", "Sixth"), 1, 0),
    post_event3 = ifelse(Wave %in% c("Fourth", "Fifth", "Sixth"), 1, 0),
    post_event4 = ifelse(Wave %in% c("Fifth", "Sixth"), 1, 0),
    post_event5 = ifelse(Wave == "Sixth", 1, 0)
  ) %>%
  mutate(
    immediate_event1 = ifelse(Wave == "Second", 1, 0),    # Event 1 impact (First → Second)
    immediate_event2 = ifelse(Wave == "Third", 1, 0),     # Event 2 impact (Second → Third)
    immediate_event3 = ifelse(Wave == "Fourth", 1, 0),    # Event 3 impact (Third → Fourth)
    immediate_event4 = ifelse(Wave == "Fifth", 1, 0),     # Event 4 impact (Fourth → Fifth)
    immediate_event5 = ifelse(Wave == "Sixth", 1, 0)      # Event 5 impact (Fifth → Sixth)
  ) 


# Print Event Table 
event_table <- data.frame(
  event_name = c("Inland Terror", "Bennet Gov. Fall", "Judicial Reform", "Gallant Dismissal", "Oct. 7th War"),
  waves = c("1-2", "2-3", "3-4", "4-5", "5-6"),
  type = c("Security", "Political", "Political", "Political", "Security"),
  stringsAsFactors = FALSE
)

gt(event_table) %>%
  tab_header(
    title = md("**Event Table**"),
  )
Event Table
event_name waves type
Inland Terror 1-2 Security
Bennet Gov. Fall 2-3 Political
Judicial Reform 3-4 Political
Gallant Dismissal 4-5 Political
Oct. 7th War 5-6 Security
 

# add event_type and event_result info to df based on the event_table
df <- af_add_event_info(df, wave_var, wave_order, event_table, community_var)

# Create the wave list in the form
# list("Inland Terror"= c("First", "Second"), "Bennet Gov. Fall" = c("Second", "Third"), ...)
wave_list <- list()
for(i in 1:nrow(event_table)) {
  wave_range <- as.numeric(unlist(strsplit(event_table$waves[i], "-")))
  wave_list[[event_table$event_name[i]]] <- c(wave_order[wave_range[1]],              
                                                      wave_order[wave_range[2]])
}

# Create demographics regression formula part
demographics <- c("gender", "age_group") # , "education"
d_fmla <- paste(demographics, collapse = "+")

# Set display names for regression results 
display_names <- list(
  "Wave" = "Wave",
  "panel_wave" = "Panel Wave",
  "post_event1" = "Inland Terror", 
  "post_event2" = "Bennet Gov. Fall", 
  "post_event3" = "Judicial Reform", 
  "post_event4" = "Gallant Dismissal", 
  "post_event5" = "Oct. 7th War",
  "immediate_event1" = "Inland Terror", 
  "immediate_event2" = "Bennet Gov. Fall", 
  "immediate_event3" = "Judicial Reform", 
  "immediate_event4" = "Gallant Dismissal", 
  "immediate_event5" = "Oct. 7th War",
  "pe_religiosity" = "Religiosity",
  "event_occurred" = "Event Occured",
  "event_type" = "Event Type",
  "gender" = "Gender",
  "age_group" = "Age Group",
  "education" = "Education"
)

immediate_names = c(
  "immediate_event1" = "Inland Terror", 
  "immediate_event2" = "Bennet Gov. Fall", 
  "immediate_event3" = "Judicial Reform", 
  "immediate_event4" = "Gallant Dismissal", 
  "immediate_event5" = "Oct. 7th War"
)

coef_names <- c(
  "event_occurred1" = "Event Occurred",
  "pe_religiosityReligious" = "Religiosity[Religious]", 
  "pe_religiosityNational Religious" = "Religiosity[National Religious]",
  "pe_religiosityUltra-Orthodox" = "Religiosity[Ultra-Orthodox]",
  "event_occurred1:pe_religiosityReligious" = "Event Occurred : Religiosity[Religious]", 
  "event_occurred1:pe_religiosityNational Religious" = "Event Occurred : Religiosity[National Religious]",
  "event_occurred1:pe_religiosityUltra-Orthodox" = "Event Occurred : Religiosity[Ultra-Orthodox]"
)

1 Moderation by Religiosity

Create panel dataset

# Create Panel Dataset
df1 <- df$respondent_id[df$Wave == "Third"]
df2 <- df$respondent_id[df$Wave == "Fourth"]
panel_respondents <- intersect(df1, df2)

panel_df <- df %>%
  filter(Wave %in% c("Third", "Fourth")) %>%
  filter(respondent_id %in% panel_respondents) %>%
  mutate(event_occurred = factor(case_when(
    Wave == "Third" ~ 0,
    Wave == "Fourth" ~ 1
  ), levels = c(0, 1)))

p_data <- plm::pdata.frame(panel_df, index = c("respondent_id", "event_occurred"))

plm::pdim(p_data)

Balanced Panel: n = 671, T = 2, N = 1342

Create panel moderation models

# The within estimator will automatically drop the main effects of time-invariant variables 
# (like pe_religiosity, gender, age_group) but keeps the interactions panel_wave * pe_religiosity.

mp1 <- plm(pe_ideology ~ event_occurred * pe_religiosity + gender + age_group, data = p_data, model = "random")
mp2 <- plm(pe_violence ~ event_occurred * pe_religiosity + gender + age_group, data = p_data, model = "random")
mp3 <- plm(pe_intolerance ~ event_occurred * pe_religiosity + gender + age_group, data = p_data, model = "random")
mp4 <- plm(pe_overall ~ event_occurred * pe_religiosity + gender + age_group, data = p_data, model = "random")

Perform Pairwise Regression

formula_str <- paste("pe_ideology ~ event_occurred * pe_religiosity","+", d_fmla)
models <- af_wave_pair_regression(df, wave_var = wave_var, wave_list = wave_list, 
                                      formula_str = formula_str, regression_type = "OLS")
models[[3]] <- mp1 # Override with panel model

coef_table <- af_coef_and_ci_table(models, coef_names)
af_coef_and_ci_plot(coef_table, xpose = TRUE, title = "Moderation: Cognitive Dimension")

notes <- af_create_regression_notes(df, models = models, display_names = display_names,
                                   show_significance = TRUE, significance_levels = c(0.05, 0.01, 0.001))
cov_labels <- af_cov_names(df, models, display_names)
af_stargazer(models = models, cov_labels = cov_labels, notes = notes, 
             title = "Cognitive Political Extremism ~ Event Occurred x Religiosity + Demographics")
Cognitive Political Extremism ~ Event Occurred x Religiosity + Demographics
Dependent variable:
pe_ideology
OLS panel OLS
linear
Inland Terror Bennet Gov. Fall Judicial Reform Gallant Dismissal Oct. 7th War
(1) (2) (3) (4) (5)
Event Occured[1] -0.113 0.044 0.032 0.137 0.157
(0.100) (0.109) (0.085) (0.127) (0.114)
Religiosity[Religious] -0.303** -0.555*** -0.444** -0.472** -0.586***
(0.101) (0.092) (0.147) (0.147) (0.100)
Religiosity[National Religious] 0.811** 0.605* 0.002 -0.265 0.209
(0.257) (0.246) (0.299) (0.307) (0.234)
Religiosity[Ultra-Orthodox] 0.190 0.010 -0.346 -0.362 -0.043
(0.163) (0.151) (0.236) (0.240) (0.160)
Gender[Female] -0.522*** -0.528*** -0.659*** -0.566*** -0.457***
(0.067) (0.070) (0.125) (0.076) (0.071)
Age Group[31-45] -0.015 -0.123 -0.303 -0.076 0.002
(0.088) (0.092) (0.161) (0.099) (0.096)
Age Group[46-60] 0.046 0.025 0.074 0.113 0.249*
(0.093) (0.098) (0.172) (0.101) (0.098)
Age Group[60+] 0.395*** 0.458*** 0.400* 0.509*** 0.594***
(0.105) (0.106) (0.195) (0.130) (0.115)
Event Occured[1] × Religiosity[Religious] -0.260 0.054 0.021 -0.124 -0.191
(0.143) (0.156) (0.119) (0.178) (0.155)
Event Occured[1] × Religiosity[National Religious] -0.214 -0.819* -0.211 0.456 -0.529
(0.373) (0.366) (0.244) (0.384) (0.352)
Event Occured[1] × Religiosity[Ultra-Orthodox] -0.197 -0.334 0.054 0.294 -0.168
(0.231) (0.250) (0.192) (0.287) (0.246)
Observations 3,215 2,493 1,342 2,221 2,638
R2 0.051 0.067 0.045 0.060 0.065
Adjusted R2 0.048 0.063 0.038 0.055 0.061
Note: * p < 0.050; ** p < 0.010; *** p < 0.001. The reference category for Event Occured is ‘0’. The reference category for Religiosity is ‘Secular’. The reference category for Gender is ‘Male’. The reference category for Age Group is ‘18-30’. Standard errors in parentheses. 
formula_str <- paste("pe_violence ~ event_occurred * pe_religiosity","+", d_fmla)
models <- af_wave_pair_regression(df, wave_var = wave_var, wave_list = wave_list, 
                                      formula_str = formula_str, regression_type = "OLS") 
models[[3]] <- mp2 # Override with panel model

coef_table <- af_coef_and_ci_table(models, coef_names)
af_coef_and_ci_plot(coef_table, xpose = TRUE, title = "Moderation: Behavioral Dimension")

notes <- af_create_regression_notes(df, models = models, display_names = display_names,
                                   show_significance = TRUE, significance_levels = c(0.05, 0.01, 0.001))
cov_labels <- af_cov_names(df, models, display_names)
af_stargazer(models = models, cov_labels = cov_labels, notes = notes, 
             title = "Behavioral Political Extremism ~ Event Occurred x Religiosity + Demographics")
Behavioral Political Extremism ~ Event Occurred x Religiosity + Demographics
Dependent variable:
pe_violence
OLS panel OLS
linear
Inland Terror Bennet Gov. Fall Judicial Reform Gallant Dismissal Oct. 7th War
(1) (2) (3) (4) (5)
Event Occured[1] 0.111* -0.050 0.055 0.091 0.158**
(0.051) (0.063) (0.050) (0.067) (0.060)
Religiosity[Religious] 0.188*** 0.117* 0.251** 0.160* 0.035
(0.051) (0.053) (0.078) (0.078) (0.053)
Religiosity[National Religious] -0.144 -0.216 -0.258 -0.229 -0.287*
(0.131) (0.141) (0.159) (0.162) (0.124)
Religiosity[Ultra-Orthodox] -0.088 0.069 -0.159 -0.227 -0.171*
(0.083) (0.086) (0.125) (0.126) (0.085)
Gender[Female] -0.209*** -0.236*** -0.278*** -0.281*** -0.258***
(0.034) (0.040) (0.065) (0.040) (0.038)
Age Group[31-45] -0.158*** -0.188*** -0.079 -0.023 -0.070
(0.045) (0.053) (0.084) (0.052) (0.051)
Age Group[46-60] -0.213*** -0.284*** -0.414*** -0.270*** -0.211***
(0.047) (0.056) (0.089) (0.053) (0.052)
Age Group[60+] -0.334*** -0.354*** -0.235* -0.206** -0.256***
(0.053) (0.061) (0.101) (0.069) (0.061)
Event Occured[1] × Religiosity[Religious] -0.071 0.081 -0.095 -0.126 -0.171*
(0.073) (0.089) (0.070) (0.094) (0.082)
Event Occured[1] × Religiosity[National Religious] -0.068 -0.107 0.009 -0.057 -0.049
(0.189) (0.210) (0.143) (0.202) (0.187)
Event Occured[1] × Religiosity[Ultra-Orthodox] 0.164 -0.265 -0.077 0.055 -0.096
(0.117) (0.144) (0.112) (0.151) (0.131)
Observations 3,215 2,493 1,342 2,221 2,638
R2 0.032 0.038 0.048 0.049 0.036
Adjusted R2 0.029 0.034 0.040 0.044 0.032
Note: * p < 0.050; ** p < 0.010; *** p < 0.001. The reference category for Event Occured is ‘0’. The reference category for Religiosity is ‘Secular’. The reference category for Gender is ‘Male’. The reference category for Age Group is ‘18-30’. Standard errors in parentheses. 
formula_str <- paste("pe_intolerance ~ event_occurred * pe_religiosity","+", d_fmla)
models <- af_wave_pair_regression(df, wave_var = wave_var, wave_list = wave_list, 
                                      formula_str = formula_str, regression_type = "OLS")
models[[3]] <- mp3 # Override with panel model

coef_table <- af_coef_and_ci_table(models, coef_names)
af_coef_and_ci_plot(coef_table, xpose = TRUE, title = "Moderation: Social Dimension")

notes <- af_create_regression_notes(df, models = models, display_names = display_names,
                                   show_significance = TRUE, significance_levels = c(0.05, 0.01, 0.001))
cov_labels <- af_cov_names(df, models, display_names)
af_stargazer(models = models, cov_labels = cov_labels, notes = notes, 
             title = "Social Political Extremism ~ Event Occurred x Religiosity + Demographics")
Social Political Extremism ~ Event Occurred x Religiosity + Demographics
Dependent variable:
pe_intolerance
OLS panel OLS
linear
Inland Terror Bennet Gov. Fall Judicial Reform Gallant Dismissal Oct. 7th War
(1) (2) (3) (4) (5)
Event Occured[1] -0.049 0.243* -0.060 -0.213 0.061
(0.078) (0.097) (0.080) (0.111) (0.098)
Religiosity[Religious] 0.735*** 0.780*** 0.462*** 0.491*** 0.901***
(0.079) (0.081) (0.134) (0.129) (0.086)
Religiosity[National Religious] 0.838*** 1.325*** 0.217 0.293 1.276***
(0.201) (0.217) (0.273) (0.268) (0.201)
Religiosity[Ultra-Orthodox] 1.435*** 1.586*** 1.102*** 0.733*** 0.601***
(0.127) (0.133) (0.216) (0.210) (0.138)
Gender[Female] -0.131* -0.084 -0.222 -0.110 -0.052
(0.052) (0.061) (0.113) (0.067) (0.061)
Age Group[31-45] 0.180** 0.144 0.049 0.146 0.220**
(0.069) (0.081) (0.147) (0.087) (0.083)
Age Group[46-60] 0.257*** 0.236** 0.286 0.313*** 0.284***
(0.072) (0.086) (0.156) (0.089) (0.084)
Age Group[60+] 0.390*** 0.431*** 0.415* 0.419*** 0.366***
(0.082) (0.093) (0.177) (0.114) (0.099)
Event Occured[1] × Religiosity[Religious] 0.046 -0.351* 0.078 0.413** -0.284*
(0.112) (0.137) (0.113) (0.155) (0.133)
Event Occured[1] × Religiosity[National Religious] 0.482 -0.973** 0.073 0.983** -0.464
(0.291) (0.323) (0.232) (0.336) (0.304)
Event Occured[1] × Religiosity[Ultra-Orthodox] 0.144 -0.583** -0.358* -0.132 0.411
(0.180) (0.221) (0.182) (0.251) (0.212)
Observations 3,215 2,493 1,342 2,221 2,638
R2 0.108 0.089 0.033 0.066 0.067
Adjusted R2 0.105 0.085 0.025 0.062 0.063
Note: * p < 0.050; ** p < 0.010; *** p < 0.001. The reference category for Event Occured is ‘0’. The reference category for Religiosity is ‘Secular’. The reference category for Gender is ‘Male’. The reference category for Age Group is ‘18-30’. Standard errors in parentheses. 
formula_str <- paste("pe_overall ~ event_occurred * pe_religiosity","+", d_fmla)
models <- af_wave_pair_regression(df, wave_var = wave_var, wave_list = wave_list, 
                                      formula_str = formula_str, regression_type = "OLS") 
models[[3]] <- mp4 # Override with panel model

coef_table <- af_coef_and_ci_table(models, coef_names)
af_coef_and_ci_plot(coef_table, xpose = TRUE, title = "Moderation: Overall (Combined Dimensions)")

notes <- af_create_regression_notes(df, models = models, display_names = display_names,
                                   show_significance = TRUE, significance_levels = c(0.05, 0.01, 0.001))
cov_labels <- af_cov_names(df, models, display_names)
af_stargazer(models = models, cov_labels = cov_labels, notes = notes, 
             title = "Overall Political Extremism ~ Event Occurred * Religiosity + Demographics")
Overall Political Extremism ~ Event Occurred * Religiosity + Demographics
Dependent variable:
pe_overall
OLS panel OLS
linear
Inland Terror Bennet Gov. Fall Judicial Reform Gallant Dismissal Oct. 7th War
(1) (2) (3) (4) (5)
Event Occured[1] -0.052 0.130* 0.003 -0.018 0.144*
(0.052) (0.058) (0.046) (0.067) (0.059)
Religiosity[Religious] 0.189*** 0.098* 0.031 0.027 0.110*
(0.052) (0.049) (0.078) (0.077) (0.052)
Religiosity[National Religious] 0.571*** 0.621*** -0.046 -0.068 0.459***
(0.132) (0.131) (0.158) (0.161) (0.122)
Religiosity[Ultra-Orthodox] 0.616*** 0.624*** 0.242 0.018 0.183*
(0.084) (0.080) (0.125) (0.126) (0.083)
Gender[Female] -0.306*** -0.287*** -0.419*** -0.345*** -0.268***
(0.034) (0.037) (0.066) (0.040) (0.037)
Age Group[31-45] 0.049 -0.008 -0.121 0.021 0.074
(0.045) (0.049) (0.085) (0.052) (0.050)
Age Group[46-60] 0.116* 0.067 0.037 0.102 0.165**
(0.048) (0.052) (0.091) (0.053) (0.051)
Age Group[60+] 0.244*** 0.269*** 0.272** 0.308*** 0.316***
(0.054) (0.056) (0.103) (0.068) (0.060)
Event Occured[1] × Religiosity[Religious] -0.092 -0.130 0.026 0.079 -0.253**
(0.073) (0.083) (0.064) (0.093) (0.081)
Event Occured[1] × Religiosity[National Religious] 0.047 -0.721*** -0.009 0.519* -0.401*
(0.192) (0.195) (0.133) (0.202) (0.184)
Event Occured[1] × Religiosity[Ultra-Orthodox] 0.004 -0.425** -0.194 0.153 0.010
(0.119) (0.133) (0.104) (0.151) (0.128)
Observations 3,215 2,493 1,342 2,221 2,638
R2 0.069 0.063 0.045 0.049 0.043
Adjusted R2 0.066 0.059 0.037 0.045 0.039
Note: * p < 0.050; ** p < 0.010; *** p < 0.001. The reference category for Event Occured is ‘0’. The reference category for Religiosity is ‘Secular’. The reference category for Gender is ‘Male’. The reference category for Age Group is ‘18-30’. Standard errors in parentheses.

1.1 Multicollinearity (VIF) Test

formula_str <- paste("pe_overall ~ Wave * pe_religiosity","+", d_fmla)
model <- lm(formula_str, data=df, na.action = na.omit)

result <- af_vif_test(model, interactions = TRUE)

print(result$interpretation)

[1] “We use VIF to test for multicollinearity. The results indicate that all adjusted GVIF values are ≤ 2.is not a concern.adjusted GVIF value: 1.011.”


result$table
Multicollinearity Assessment (with Interactions)
Predictor Adjusted GVIF1
Wave 1.002
pe_religiosity 1.002
gender 1.011
age_group 1.011
1 Values > 2 indicate problematic multicollinearity

2 Robustness

2.1 Impact of Occurred Events

mp1 <- plm(pe_ideology ~ event_occurred, data = p_data, model = "within")
mp2 <- plm(pe_violence ~ event_occurred, data = p_data, model = "within")
mp3 <- plm(pe_intolerance ~ event_occurred, data = p_data, model = "within")
mp4 <- plm(pe_overall ~ event_occurred, data = p_data, model = "within")
formula_str <- "pe_ideology ~ event_occurred"
models <- af_wave_pair_regression(df, wave_var = wave_var, wave_list = wave_list, 
                                      formula_str = formula_str, regression_type = "OLS")
models[[3]] <- mp1 # Override with panel model
notes <- af_create_regression_notes(df, models = models, display_names = display_names,
                                   show_significance = TRUE, significance_levels = c(0.05, 0.01, 0.001))
cov_labels <- af_cov_names(df, models, display_names)
af_stargazer(models = models, cov_labels = cov_labels, notes = notes, 
             title = "Cognitive Political Extremism ~ Event Occurred")
Cognitive Political Extremism ~ Event Occurred
Dependent variable:
pe_ideology
OLS panel OLS
linear
Inland Terror Bennet Gov. Fall Judicial Reform Gallant Dismissal Oct. 7th War
(1) (2) (3) (4) (5)
Event Occured[1] -0.210** -0.047 0.035 0.106 0.121
(0.068) (0.074) (0.054) (0.084) (0.073)
Observations 3,215 2,493 1,342 2,221 2,638
R2 0.003 0.0002 0.001 0.001 0.001
Adjusted R2 0.003 -0.0002 -1.000 0.0003 0.001
Note: * p < 0.050; ** p < 0.010; *** p < 0.001. The reference category for Event Occured is ‘0’. Standard errors in parentheses. 
formula_str <- "pe_violence ~ event_occurred"
models <- af_wave_pair_regression(df, wave_var = wave_var, wave_list = wave_list, 
                                      formula_str = formula_str, regression_type = "OLS") 
models[[3]] <- mp2 # Override with panel model
notes <- af_create_regression_notes(df, models = models, display_names = display_names,
                                   show_significance = TRUE, significance_levels = c(0.05, 0.01, 0.001))
cov_labels <- af_cov_names(df, models, display_names)
af_stargazer(models = models, cov_labels = cov_labels, notes = notes, 
             title = "Behavioral Political Extremism ~ Event Occurred")
Behavioral Political Extremism ~ Event Occurred
Dependent variable:
pe_violence
OLS panel OLS
linear
Inland Terror Bennet Gov. Fall Judicial Reform Gallant Dismissal Oct. 7th War
(1) (2) (3) (4) (5)
Event Occured[1] 0.075* -0.038 0.005 0.054 0.034
(0.034) (0.042) (0.032) (0.044) (0.038)
Observations 3,215 2,493 1,342 2,221 2,638
R2 0.001 0.0003 0.00004 0.001 0.0003
Adjusted R2 0.001 -0.0001 -1.001 0.0002 -0.0001
Note: * p < 0.050; ** p < 0.010; *** p < 0.001. The reference category for Event Occured is ‘0’. Standard errors in parentheses. 
formula_str <- "pe_intolerance ~ event_occurred"
models <- af_wave_pair_regression(df, wave_var = wave_var, wave_list = wave_list, 
                                      formula_str = formula_str, regression_type = "OLS")
models[[3]] <- mp3 # Override with panel model
notes <- af_create_regression_notes(df, models = models, display_names = display_names,
                                   show_significance = TRUE, significance_levels = c(0.05, 0.01, 0.001))
cov_labels <- af_cov_names(df, models, display_names)
af_stargazer(models = models, cov_labels = cov_labels, notes = notes, 
             title = "Social Political Extremism ~ Event Occurred")
Social Political Extremism ~ Event Occurred
Dependent variable:
pe_intolerance
OLS panel OLS
linear
Inland Terror Bennet Gov. Fall Judicial Reform Gallant Dismissal Oct. 7th War
(1) (2) (3) (4) (5)
Event Occured[1] 0.0004 -0.001 -0.057 0.000 0.000
(0.055) (0.066) (0.052) (0.073) (0.063)
Observations 3,215 2,493 1,342 2,221 2,638
R2 0.000 0.00000 0.002 0.000 0.000
Adjusted R2 -0.0003 -0.0004 -0.998 -0.0005 -0.0004
Note: * p < 0.050; ** p < 0.010; *** p < 0.001. The reference category for Event Occured is ‘0’. Standard errors in parentheses. 
formula_str <- "pe_overall ~ event_occurred"
models <- af_wave_pair_regression(df, wave_var = wave_var, wave_list = wave_list, 
                                      formula_str = formula_str, regression_type = "OLS") 
models[[3]] <- mp4 # Override with panel model
notes <- af_create_regression_notes(df, models = models, display_names = display_names,
                                   show_significance = TRUE, significance_levels = c(0.05, 0.01, 0.001))
cov_labels <- af_cov_names(df, models, display_names)
af_stargazer(models = models, cov_labels = cov_labels, notes = notes, 
             title = "Overall Political Extremism ~ Event Occurred")
Overall Political Extremism ~ Event Occurred
Dependent variable:
pe_overall
OLS panel OLS
linear
Inland Terror Bennet Gov. Fall Judicial Reform Gallant Dismissal Oct. 7th War
(1) (2) (3) (4) (5)
Event Occured[1] -0.078* -0.013 -0.006 0.050 0.057
(0.035) (0.039) (0.030) (0.044) (0.038)
Observations 3,215 2,493 1,342 2,221 2,638
R2 0.002 0.00004 0.0001 0.001 0.001
Adjusted R2 0.001 -0.0004 -1.001 0.0001 0.001
Note: * p < 0.050; ** p < 0.010; *** p < 0.001. The reference category for Event Occured is ‘0’. Standard errors in parentheses.

2.2 Six-Waves Analysis

2.2.1 Immediate Events

Immediate effects Shows the direct impact of each event on the next wave. They are Non-cumulative i.e. effects don’t build upon each other.

The coefficient of immediate_event_i should be interpreted as follows:

  • immediate_event1: How much response changed from First to Second wave (due to Event 1)
  • immediate_event2: How much response changed from Second to Third wave (due to Event 2)
# Create immediate event regression formula part
immediate_fmla <- "immediate_event1 + immediate_event2 + immediate_event3 + immediate_event4 + immediate_event5"

formula_str <- paste("pe_overall ~",immediate_fmla)
m1 <- lm(formula_str, data=df, na.action = na.omit)

formula_str <- paste("pe_ideology ~",immediate_fmla)
m2 <- lm(formula_str, data=df, na.action = na.omit)

formula_str <- paste("pe_violence ~",immediate_fmla)
m3 <- lm(formula_str, data=df, na.action = na.omit)

formula_str <- paste("pe_intolerance ~",immediate_fmla)
m4 <- lm(formula_str, data=df, na.action = na.omit)

models <- list("Overall" = m1, "Cognitive" = m2, "Behavioral" = m3, "Social" = m4)

coef_table <- af_coef_and_ci_table(models, immediate_names)
af_coef_and_ci_plot(coef_table, xpose = TRUE, title = "Immediate Effects")

notes <- af_create_regression_notes(df, models = models, display_names = display_names,
                                   show_significance = TRUE, significance_levels = c(0.05, 0.01, 0.001))
cov_labels <- af_cov_names(df, models, display_names)
af_stargazer(models = models, cov_labels = cov_labels, notes = notes, 
             title = "Political Extremism ~ Immediate Event")
Political Extremism ~ Immediate Event
Dependent variable:
pe_overall pe_ideology pe_violence pe_intolerance
Overall Cognitive Behavioral Social
(1) (2) (3) (4)
Inland Terror -0.078* -0.210** 0.075* 0.0004
(0.034) (0.066) (0.034) (0.056)
Bennet Gov. Fall -0.091* -0.257** 0.036 -0.001
(0.041) (0.078) (0.040) (0.066)
Judicial Reform -0.089* -0.227** -0.002 -0.001
(0.044) (0.085) (0.044) (0.072)
Gallant Dismissal -0.039 -0.121 0.052 -0.001
(0.035) (0.067) (0.035) (0.057)
Oct. 7th War 0.018 0.0001 0.086* -0.001
(0.038) (0.073) (0.038) (0.062)
Observations 7,436 7,436 7,436 7,436
R2 0.002 0.003 0.001 0.00000
Adjusted R2 0.001 0.002 0.001 -0.001
Note: * p < 0.050; ** p < 0.010; *** p < 0.001. Standard errors in parentheses.

2.2.2 Religiosity Moderation of Immediate Events

# Create immediate event interaction with Religiosity regression formula part
int_fmla <- paste0("(", immediate_fmla,") * pe_religiosity +", d_fmla)

formula_str <- paste("pe_overall ~",int_fmla)
m1 <- lm(formula_str, data=df, na.action = na.omit)

formula_str <- paste("pe_ideology ~",int_fmla)
m2 <- lm(formula_str, data=df, na.action = na.omit)

formula_str <- paste("pe_violence ~",int_fmla)
m3 <- lm(formula_str, data=df, na.action = na.omit)

formula_str <- paste("pe_intolerance ~",int_fmla)
m4 <- lm(formula_str, data=df, na.action = na.omit)

models <- list(m1, m2, m3, m4)

notes <- af_create_regression_notes(df, models = models, display_names = display_names,
                                   show_significance = TRUE, significance_levels = c(0.05, 0.01, 0.001))
cov_labels <- af_cov_names(df, models, display_names)
af_stargazer(models = models, cov_labels = cov_labels, notes = notes, 
             title = "Political Extremism ~ Immediate Event * Religiosity + Demographics")
Political Extremism ~ Immediate Event * Religiosity + Demographics
Dependent variable:
pe_overall pe_ideology pe_violence pe_intolerance
(1) (2) (3) (4)
Inland Terror -0.057 -0.126 0.111* -0.049
(0.050) (0.097) (0.051) (0.081)
Bennet Gov. Fall 0.072 -0.081 0.062 0.187
(0.061) (0.117) (0.061) (0.098)
Judicial Reform 0.069 -0.057 0.056 0.187
(0.066) (0.127) (0.066) (0.106)
Gallant Dismissal 0.046 0.073 0.138** -0.026
(0.053) (0.101) (0.053) (0.085)
Oct. 7th War 0.187** 0.235* 0.302*** 0.022
(0.058) (0.111) (0.058) (0.093)
Religiosity[Religious] 0.190*** -0.304** 0.191*** 0.734***
(0.051) (0.098) (0.051) (0.082)
Religiosity[National Religious] 0.572*** 0.812** -0.137 0.838***
(0.129) (0.249) (0.130) (0.208)
Religiosity[Ultra-Orthodox] 0.619*** 0.198 -0.084 1.436***
(0.082) (0.157) (0.082) (0.131)
Gender[Female] -0.310*** -0.516*** -0.238*** -0.122***
(0.022) (0.043) (0.022) (0.036)
Age Group[31-45] 0.034 -0.050 -0.117*** 0.184***
(0.029) (0.057) (0.030) (0.047)
Age Group[46-60] 0.122*** 0.133* -0.257*** 0.278***
(0.031) (0.059) (0.031) (0.049)
Age Group[60+] 0.283*** 0.494*** -0.302*** 0.389***
(0.035) (0.068) (0.035) (0.057)
Inland Terror × Religiosity[Religious] -0.090 -0.251 -0.075 0.046
(0.072) (0.138) (0.072) (0.115)
Inland Terror × Religiosity[National Religious] 0.052 -0.201 -0.071 0.482
(0.187) (0.361) (0.188) (0.301)
Inland Terror × Religiosity[Ultra-Orthodox] 0.010 -0.177 0.159 0.145
(0.116) (0.223) (0.117) (0.186)
Bennet Gov. Fall × Religiosity[Religious] -0.218* -0.201 0.004 -0.300*
(0.086) (0.165) (0.086) (0.138)
Bennet Gov. Fall × Religiosity[National Religious] -0.668*** -1.023** -0.184 -0.478
(0.197) (0.380) (0.199) (0.317)
Bennet Gov. Fall × Religiosity[Ultra-Orthodox] -0.416** -0.518* -0.105 -0.435*
(0.137) (0.264) (0.138) (0.221)
Judicial Reform × Religiosity[Religious] -0.171 -0.179 -0.035 -0.247
(0.093) (0.179) (0.093) (0.149)
Judicial Reform × Religiosity[National Religious] -0.641** -1.080** -0.090 -0.546
(0.208) (0.400) (0.209) (0.334)
Judicial Reform × Religiosity[Ultra-Orthodox] -0.608*** -0.568 -0.156 -0.706**
(0.151) (0.291) (0.152) (0.243)
Gallant Dismissal × Religiosity[Religious] -0.083 -0.291* -0.159* 0.166
(0.073) (0.141) (0.074) (0.118)
Gallant Dismissal × Religiosity[National Religious] -0.120 -0.620 -0.158 0.436
(0.178) (0.343) (0.179) (0.286)
Gallant Dismissal × Religiosity[Ultra-Orthodox] -0.446*** -0.262 -0.092 -0.840***
(0.117) (0.225) (0.117) (0.188)
Oct. 7th War × Religiosity[Religious] -0.326*** -0.466** -0.330*** -0.104
(0.080) (0.154) (0.080) (0.128)
Oct. 7th War × Religiosity[National Religious] -0.508** -1.135** -0.204 -0.005
(0.189) (0.365) (0.191) (0.305)
Oct. 7th War × Religiosity[Ultra-Orthodox] -0.429*** -0.430 -0.204 -0.405*
(0.127) (0.245) (0.128) (0.205)
Observations 7,436 7,436 7,436 7,436
R2 0.059 0.060 0.038 0.078
Adjusted R2 0.055 0.057 0.035 0.075
Note: * p < 0.050; ** p < 0.010; *** p < 0.001. The reference category for Religiosity is ‘Secular’. The reference category for Gender is ‘Male’. The reference category for Age Group is ‘18-30’. Standard errors in parentheses.

2.2.3 Religiosity and Event Type

We now test whether the interaction of event type and Religiosity impacts political extremism. Note that event type has multicolinearity with Wave (or immediate event) since there is only one event type per each wave.

# Create immediate event interaction with Religiosity regression formula part
int_fmla <- paste0("event_type * pe_religiosity +", d_fmla)

formula_str <- paste("pe_overall ~",int_fmla)
m1 <- lm(formula_str, data=df, na.action = na.omit)

formula_str <- paste("pe_ideology ~",int_fmla)
m2 <- lm(formula_str, data=df, na.action = na.omit)

formula_str <- paste("pe_violence ~",int_fmla)
m3 <- lm(formula_str, data=df, na.action = na.omit)

formula_str <- paste("pe_intolerance ~",int_fmla)
m4 <- lm(formula_str, data=df, na.action = na.omit)

models <- list(m1, m2, m3, m4)

notes <- af_create_regression_notes(df, models = models, display_names = display_names,
                                   show_significance = TRUE, significance_levels = c(0.05, 0.01, 0.001))
cov_labels <- af_cov_names(df, models, display_names)
af_stargazer(models = models, cov_labels = cov_labels, notes = notes, 
             title = "Political Extremism ~ Event Type * Religiosity + Demographics")
Political Extremism ~ Event Type * Religiosity + Demographics
Dependent variable:
pe_overall pe_ideology pe_violence pe_intolerance
(1) (2) (3) (4)
Event Type[Security] 0.036 0.010 0.183*** -0.021
(0.046) (0.088) (0.046) (0.073)
Event Type[Political] 0.059 -0.002 0.097* 0.086
(0.045) (0.086) (0.045) (0.072)
Religiosity[Religious] 0.190*** -0.304** 0.191*** 0.735***
(0.051) (0.098) (0.051) (0.082)
Religiosity[National Religious] 0.572*** 0.812** -0.137 0.837***
(0.129) (0.249) (0.130) (0.208)
Religiosity[Ultra-Orthodox] 0.619*** 0.198 -0.084 1.434***
(0.082) (0.157) (0.082) (0.132)
Gender[Female] -0.316*** -0.518*** -0.238*** -0.133***
(0.022) (0.043) (0.022) (0.036)
Age Group[31-45] 0.033 -0.051 -0.113*** 0.178***
(0.029) (0.057) (0.030) (0.047)
Age Group[46-60] 0.123*** 0.139* -0.255*** 0.274***
(0.031) (0.059) (0.031) (0.049)
Age Group[60+] 0.283*** 0.498*** -0.297*** 0.379***
(0.035) (0.068) (0.035) (0.056)
Event Type[Security] × Religiosity[Religious] -0.179** -0.326** -0.175** -0.014
(0.064) (0.124) (0.065) (0.103)
Event Type[Political] × Religiosity[Religious] -0.140* -0.237* -0.084 -0.059
(0.063) (0.121) (0.063) (0.101)
Event Type[Security] × Religiosity[National Religious] -0.199 -0.624* -0.115 0.247
(0.161) (0.311) (0.162) (0.260)
Event Type[Political] × Religiosity[National Religious] -0.418** -0.863** -0.151 -0.083
(0.153) (0.295) (0.154) (0.247)
Event Type[Security] × Religiosity[Ultra-Orthodox] -0.165 -0.267 0.013 -0.085
(0.103) (0.199) (0.104) (0.166)
Event Type[Political] × Religiosity[Ultra-Orthodox] -0.472*** -0.395* -0.107 -0.702***
(0.101) (0.194) (0.101) (0.162)
Observations 7,436 7,436 7,436 7,436
R2 0.054 0.057 0.036 0.073
Adjusted R2 0.052 0.055 0.034 0.071
Note: * p < 0.050; ** p < 0.010; *** p < 0.001. The reference category for Event Type is ‘No-Event’. The reference category for Religiosity is ‘Secular’. The reference category for Gender is ‘Male’. The reference category for Age Group is ‘18-30’. Standard errors in parentheses.

2.2.4 Cummulative Events

We further test whether there is a cumulative impact for the events on political extremism. Cumulative effects captures how events build upon each other over time. Each coefficient isolates the unique contribution of that specific event.

If post_event1 = 0.5 and post_event2 = 0.3, then:

  • After Event 1: response increases by 0.5
  • After Event 2: response increases by additional 0.3 (total = 0.8)
  • Wave “Third” respondents have 0.8 higher response than Wave “First”
# Create post event regression formula part
post_fmla <- "post_event1 + post_event2 + post_event3 + post_event4 + post_event5"

formula_str <- paste("pe_overall ~",post_fmla)
m1 <- lm(formula_str, data=df, na.action = na.omit)

formula_str <- paste("pe_ideology ~",post_fmla)
m2 <- lm(formula_str, data=df, na.action = na.omit)

formula_str <- paste("pe_violence ~",post_fmla)
m3 <- lm(formula_str, data=df, na.action = na.omit)

formula_str <- paste("pe_intolerance ~",post_fmla)
m4 <- lm(formula_str, data=df, na.action = na.omit)

models <- list(m1, m2, m3, m4)

notes <- af_create_regression_notes(df, models = models, display_names = display_names,
                                   show_significance = TRUE, significance_levels = c(0.05, 0.01, 0.001))
cov_labels <- af_cov_names(df, models, display_names)
af_stargazer(models = models, cov_labels = cov_labels, notes = notes, 
             title = "Cumulative Post-Event Impact: Political Extremism ~ Event")
Cumulative Post-Event Impact: Political Extremism ~ Event
Dependent variable:
pe_overall pe_ideology pe_violence pe_intolerance
(1) (2) (3) (4)
Inland Terror -0.078* -0.210** 0.075* 0.0004
(0.034) (0.066) (0.034) (0.056)
Bennet Gov. Fall -0.013 -0.047 -0.038 -0.001
(0.041) (0.078) (0.040) (0.066)
Judicial Reform 0.001 0.030 -0.039 -0.000
(0.049) (0.095) (0.049) (0.080)
Gallant Dismissal 0.050 0.106 0.054 0.000
(0.044) (0.086) (0.044) (0.072)
Oct. 7th War 0.057 0.121 0.034 -0.000
(0.038) (0.074) (0.038) (0.062)
Observations 7,436 7,436 7,436 7,436
R2 0.002 0.003 0.001 0.00000
Adjusted R2 0.001 0.002 0.001 -0.001
Note: * p < 0.050; ** p < 0.010; *** p < 0.001. Standard errors in parentheses.