Covid-19_NSF_RAPID_US_Wave2_BriefReport
- 1. Models: Does media perception predict vote confidence?
- 2. Does partisanship predict vote confidence?
- 3. Models: Do media perception and partisanship predict vote confidence?
- 4. Models: Do media perception and partisanship interact to predict vote confidence?
- 5. Model Testing: Does the model predicting vote confidence by partisanship and media consumption explain more variance than the model without media consumption in it?
- 6. Figure 1: Vote Confidence by partisanship
- 7. Stats for Figure 1
- 8. Figure 2: Media Perception box plot
- 9. Stats for Figure 2
# continuous party
d$partyCont <- NA
d$partyCont[d$demStrength == 1] <- -3
d$partyCont[d$demStrength == 2] <- -2
d$partyCont[d$partyClose == 1] <- -1
d$partyCont[d$partyClose == 3] <- 0
d$partyCont[d$repStrength == 1] <- 3
d$partyCont[d$repStrength == 2] <- 2
d$partyCont[d$partyClose == 2] <- 1
# policy group proposer
d$Policy_Group <- NA
d$Policy_Group[d$FL_71_DO == 'proportional_dem' | d$FL_71_DO == 'US1st_dem'] <- "Democratic"
d$Policy_Group[d$FL_71_DO == 'proportional_rep' | d$FL_71_DO == 'US1st_rep'] <- "Republican"
d$Policy_Group[d$FL_71_DO == 'proportional_bipart' | d$FL_71_DO == 'US1st_bipart'] <- "Bipartisan"
d$Policy_Group[d$FL_71_DO == 'proportional_expert' | d$FL_71_DO == 'US1st_expert'] <- "Expert"
d$Policy_Group <- factor(d$Policy_Group, levels = c('Democratic', 'Republican', 'Bipartisan','Expert'))
# policy frame
d$Policy_Frame <- NA
d$Policy_Frame[d$FL_71_DO == 'proportional_dem' | d$FL_71_DO == 'proportional_rep' | d$FL_71_DO == 'proportional_bipart' | d$FL_71_DO == 'proportional_expert'] <- "Proportional"
d$Policy_Frame[d$FL_71_DO == 'US1st_dem' | d$FL_71_DO == 'US1st_rep' | d$FL_71_DO == 'US1st_bipart' | d$FL_71_DO == 'US1st_expert'] <- "US1st"
# party factor
d$party_factor <- NA
d$party_factor[d$partyCont < 0] <- 'Democrat'
d$party_factor[d$partyCont == 0] <- 'Independent'
d$party_factor[d$partyCont > 0] <- 'Republican'
## Order of timing var
d$election_timing <- factor(d$election_timing, levels = c('Pre-election', 'During-election','Post-election'))
d$party_factor <- factor(d$party_factor, levels = c('Democrat', 'Republican','Independent'))### Partisan framing codes
# Code 1: Left vs. Right
d$fDem_Rep <- NA
d$fDem_Rep[d$Policy_Group == 'Democratic'] <- -.5
d$fDem_Rep[d$Policy_Group == 'Republican'] <- .5
d$fDem_Rep[d$Policy_Group == 'Bipartisan' | d$Policy_Group == 'Expert'] <- 0
# Code 2: Bi vs. Exp
d$fBi_Exp <- NA
d$fBi_Exp[d$Policy_Group == 'Democratic' | d$Policy_Group == 'Republican'] <- 0
d$fBi_Exp[d$Policy_Group == 'Bipartisan'] <- -.5
d$fBi_Exp[d$Policy_Group == 'Expert'] <- .5
# Code 3 = Dem and Rep vs. Exp and Bi
d$fParties_BiExp <- NA
d$fParties_BiExp[d$Policy_Group == 'Democratic' | d$Policy_Group == 'Republican'] <- -.5
d$fParties_BiExp[d$Policy_Group == 'Bipartisan' | d$Policy_Group == 'Expert'] <- .5
### Policy framing codes
# contrasts
d$fProp_US <- NA
d$fProp_US[d$Policy_Frame == 'Proportional'] <- -.5
d$fProp_US[d$Policy_Frame == 'US1st'] <- .5
#dummies
d$fProportional <- NA
d$fProportional[d$Policy_Frame == 'Proportional'] <- 0
d$fProportional[d$Policy_Frame == 'US1st'] <- 1
d$fUS <- NA
d$fUS[d$Policy_Frame == 'Proportional'] <- 1
d$fUS[d$Policy_Frame == 'US1st'] <- 0
### Timing codes
## Contrast
d$tPre_Post <- NA
d$tPre_Post[d$election_timing == 'Pre-election'] <- -.5
d$tPre_Post[d$election_timing == 'During-election'] <- 0
d$tPre_Post[d$election_timing == 'Post-election'] <- .5
d$tDuring_Not <- NA
d$tDuring_Not[d$election_timing == 'Pre-election'] <- .33
d$tDuring_Not[d$election_timing == 'During-election'] <- -.67
d$tDuring_Not[d$election_timing == 'Post-election'] <- .33
# Dummy
# Post!
d$tPostP <- NA
d$tPostP[d$election_timing == 'Pre-election'] <- 1
d$tPostP[d$election_timing == 'During-election'] <- 0
d$tPostP[d$election_timing == 'Post-election'] <- 0
d$tPostD <- NA
d$tPostD[d$election_timing == 'Pre-election'] <- 0
d$tPostD[d$election_timing == 'During-election'] <- 1
d$tPostD[d$election_timing == 'Post-election'] <- 0
# During!
d$tDurPre <- NA
d$tDurPre[d$election_timing == 'Pre-election'] <- 1
d$tDurPre[d$election_timing == 'During-election'] <- 0
d$tDurPre[d$election_timing == 'Post-election'] <- 0
d$tDurPost <- NA
d$tDurPost[d$election_timing == 'Pre-election'] <- 0
d$tDurPost[d$election_timing == 'During-election'] <- 0
d$tDurPost[d$election_timing == 'Post-election'] <- 1
### Party Factor
d$pDem_Rep <- NA
d$pDem_Rep[d$party_factor == 'Democrat'] <- -.5
d$pDem_Rep[d$party_factor == 'Independent'] <- 0
d$pDem_Rep[d$party_factor == 'Republican'] <- .5
d$pInd_Not <- NA
d$pInd_Not[d$party_factor == 'Democrat'] <- .33
d$pInd_Not[d$party_factor == 'Independent'] <- -.67
d$pInd_Not[d$party_factor == 'Republican'] <- .33
### centering
d$vaxxAtt.c <- d$vaxxAttitudes - mean(d$vaxxAttitudes, na.rm = T)
d$ownvote.c <- d$ownvote_conf - mean(d$ownvote_conf, na.rm = T)
d$overallvote.c <- d$overallvote_conf - mean(d$overallvote_conf, na.rm = T)
d$govtPolEff.c <- d$polEfficacy_1 - mean(d$polEfficacy_1, na.rm = T)
d$electPolEff.c <- d$polEfficacy_2 - mean(d$polEfficacy_2, na.rm = T)
# Media Measures - All, including Fox
d$mediaExposure_all <- rowMeans(d[,c("mediaExposure_1","mediaExposure_2","mediaExposure_3","mediaExposure_4","mediaExposure_5","mediaExposure_6","mediaExposure_7","mediaExposure_8","mediaExposure_9","mediaExposure_10","mediaExposure_11","mediaExposure_12","mediaExposure_13","mediaExposure_14","mediaExposure_15")], na.rm = T)
d$mediaTrust_all <- rowMeans(d[,c("mediaTrust_1","mediaTrust_2","mediaTrust_3","mediaTrust_4","mediaTrust_5","mediaTrust_6","mediaTrust_7","mediaTrust_8","mediaTrust_9","mediaTrust_10","mediaTrust_11","mediaTrust_12","mediaTrust_13","mediaTrust_14","mediaTrust_15")], na.rm = T)
# Media Measures - Excluding Fox
d$mediaExposure <- rowMeans(d[,c("mediaExposure_1","mediaExposure_2","mediaExposure_3","mediaExposure_4","mediaExposure_6","mediaExposure_7","mediaExposure_8","mediaExposure_9","mediaExposure_10","mediaExposure_11","mediaExposure_12","mediaExposure_13","mediaExposure_14","mediaExposure_15")], na.rm = T)
d$mediaTrust <- rowMeans(d[,c("mediaTrust_1","mediaTrust_2","mediaTrust_3","mediaTrust_4","mediaTrust_6","mediaTrust_7","mediaTrust_8","mediaTrust_9","mediaTrust_10","mediaTrust_11","mediaTrust_12","mediaTrust_13","mediaTrust_14","mediaTrust_15")], na.rm = T)
### Composite Media variables
#average own and overall vote
d$voteLegit <- (d$ownvote_conf + d$overallvote_conf)/2
d$voteLegit.c <- d$voteLegit - mean(d$voteLegit, na.rm = T)
#recenter media/fox trust to match media/fox exposure scale
d$mediaTrust5 <- d$mediaTrust_5 + 3
d$mediaTrust <- d$mediaTrust + 3
d$foxPerception <- (d$mediaExposure_5 + d$mediaTrust5)/2
d$mediaPerception <- (d$mediaExposure + d$mediaTrust)/2
d$foxPerception.c <- d$foxPerception - mean(d$foxPerception, na.rm = T)
d$mediaPerception.c <- d$mediaPerception - mean(d$mediaPerception, na.rm = T)
#### Indices
Proportional <- d$Policy_Frame == 'Proportional'
US1st <- d$Policy_Frame == 'US1st'
Pre <- d$election_timing == 'Pre-election'
During <- d$election_timing == 'During-election'
Post <- d$election_timing == 'Post-election'
#dummy codes for party ID
d$pDemR[d$party_factor == 'Democrat'] <- 0
d$pDemR[d$party_factor == 'Republican'] <- 1
d$pDemR[d$party_factor == 'Independent'] <- 0
d$pDemI[d$party_factor == 'Democrat'] <- 0
d$pDemI[d$party_factor == 'Republican'] <- 0
d$pDemI[d$party_factor == 'Independent'] <- 1
d$pRepD[d$party_factor == 'Democrat'] <- 1
d$pRepD[d$party_factor == 'Republican'] <- 0
d$pRepD[d$party_factor == 'Independent'] <- 0
d$pRepI[d$party_factor == 'Democrat'] <- 0
d$pRepI[d$party_factor == 'Republican'] <- 0
d$pRepI[d$party_factor == 'Independent'] <- 1
# mean center fox exposure
d$foxExposure.c <- d$mediaExposure_5 - mean(d$mediaExposure_5, na.rm = T)
# mean center other media exposure
d$otherMediaExposure.c <- d$mediaExposure - mean(d$mediaExposure, na.rm = T)
#create difference score
d$diffFoxOther <- d$mediaExposure_5 - d$mediaExposure
#reverse code biden to trump rating win --> trump definitely win = 1; biden definitely win = 9
d$electPredict_T_B2 <- 10 - d$electPredict_B_T
d$electPredict_T_B3 <- 10 - d$electPredict_B_T.1
#combine to make 1 column
d$electPredictTB <- ifelse(!is.na(d$electPredict_T_B), d$electPredict_T_B,
ifelse(!is.na(d$electPredict_T_B2), d$electPredict_T_B2,
ifelse(!is.na(d$electPredict_T_B3), d$electPredict_T_B3, NA)))
d$electPredictTB.plot <- d$electPredictTB - 5 1. Models: Does media perception predict vote confidence?
a. Vote confidence by Fox News perception
vote.fox <- lm(voteLegit ~ foxPerception, data = d)
tab_model(vote.fox)| Â | voteLegit | ||
|---|---|---|---|
| Predictors | Estimates | CI | p |
| (Intercept) | 3.98 | 3.82 – 4.13 | <0.001 |
| foxPerception | -0.20 | -0.26 – -0.15 | <0.001 |
| Observations | 1209 | ||
| R2 / R2 adjusted | 0.038 / 0.037 | ||
b. Vote confidence by Other Media perception
vote.othMedia <- lm(voteLegit ~ mediaPerception, data = d)
tab_model(vote.othMedia)| Â | voteLegit | ||
|---|---|---|---|
| Predictors | Estimates | CI | p |
| (Intercept) | 1.76 | 1.57 – 1.96 | <0.001 |
| mediaPerception | 0.70 | 0.63 – 0.78 | <0.001 |
| Observations | 1209 | ||
| R2 / R2 adjusted | 0.216 / 0.216 | ||
c. Vote confidence by Fox & Other Media perception
vote.media <- lm(voteLegit ~ foxPerception + mediaPerception, data = d)
tab_model(vote.media)| Â | voteLegit | ||
|---|---|---|---|
| Predictors | Estimates | CI | p |
| (Intercept) | 2.25 | 2.02 – 2.48 | <0.001 |
| foxPerception | -0.21 | -0.26 – -0.15 | <0.001 |
| mediaPerception | 0.70 | 0.63 – 0.78 | <0.001 |
| Observations | 1209 | ||
| R2 / R2 adjusted | 0.255 / 0.254 | ||
2. Does partisanship predict vote confidence?
a. Vote confidence by partisanship
vote.party <- lm(voteLegit ~ (pDem_Rep + pInd_Not), data = d)
tab_model(vote.party)| Â | voteLegit | ||
|---|---|---|---|
| Predictors | Estimates | CI | p |
| (Intercept) | 3.34 | 3.27 – 3.42 | <0.001 |
| pDem_Rep | -1.27 | -1.41 – -1.13 | <0.001 |
| pInd_Not | 0.56 | 0.38 – 0.74 | <0.001 |
| Observations | 1208 | ||
| R2 / R2 adjusted | 0.232 / 0.231 | ||
summary(vote.party)##
## Call:
## lm(formula = voteLegit ~ (pDem_Rep + pInd_Not), data = d)
##
## Residuals:
## Min 1Q Median 3Q Max
## -3.1667 -0.8920 0.1080 0.8333 2.1080
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 3.34407 0.03644 91.78 < 2e-16 ***
## pDem_Rep -1.27466 0.07151 -17.82 < 2e-16 ***
## pInd_Not 0.56142 0.09055 6.20 7.73e-10 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.138 on 1205 degrees of freedom
## (1268 observations deleted due to missingness)
## Multiple R-squared: 0.2324, Adjusted R-squared: 0.2311
## F-statistic: 182.4 on 2 and 1205 DF, p-value: < 2.2e-163. Models: Do media perception and partisanship predict vote confidence?
a. Vote confidence by Fox News perception & partisan ID
vote.fox.party <- lm(voteLegit ~ foxPerception + (pDem_Rep + pInd_Not), data = d)
tab_model(vote.fox.party)| Â | voteLegit | ||
|---|---|---|---|
| Predictors | Estimates | CI | p |
| (Intercept) | 3.47 | 3.32 – 3.62 | <0.001 |
| foxPerception | -0.05 | -0.11 – 0.00 | 0.057 |
| pDem_Rep | -1.22 | -1.37 – -1.07 | <0.001 |
| pInd_Not | 0.58 | 0.40 – 0.76 | <0.001 |
| Observations | 1208 | ||
| R2 / R2 adjusted | 0.235 / 0.233 | ||
b. Vote confidence by Other Media perception
vote.othMedia.party <- lm(voteLegit ~ mediaPerception + (pDem_Rep + pInd_Not), data = d)
tab_model(vote.othMedia.party)| Â | voteLegit | ||
|---|---|---|---|
| Predictors | Estimates | CI | p |
| (Intercept) | 2.20 | 1.99 – 2.41 | <0.001 |
| mediaPerception | 0.47 | 0.39 – 0.56 | <0.001 |
| pDem_Rep | -0.85 | -1.00 – -0.70 | <0.001 |
| pInd_Not | 0.53 | 0.36 – 0.70 | <0.001 |
| Observations | 1208 | ||
| R2 / R2 adjusted | 0.308 / 0.306 | ||
c. Vote confidence by Fox & Other Media perception
vote.media.party <- lm(voteLegit ~ foxPerception + mediaPerception + (pDem_Rep + pInd_Not), data = d)
tab_model(vote.media.party)| Â | voteLegit | ||
|---|---|---|---|
| Predictors | Estimates | CI | p |
| (Intercept) | 2.40 | 2.17 – 2.62 | <0.001 |
| foxPerception | -0.13 | -0.18 – -0.07 | <0.001 |
| mediaPerception | 0.51 | 0.43 – 0.60 | <0.001 |
| pDem_Rep | -0.69 | -0.85 – -0.52 | <0.001 |
| pInd_Not | 0.56 | 0.39 – 0.73 | <0.001 |
| Observations | 1208 | ||
| R2 / R2 adjusted | 0.320 / 0.318 | ||
summary(vote.media.party)##
## Call:
## lm(formula = voteLegit ~ foxPerception + mediaPerception + (pDem_Rep +
## pInd_Not), data = d)
##
## Residuals:
## Min 1Q Median 3Q Max
## -3.3634 -0.7997 0.1389 0.7738 2.8733
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.39699 0.11331 21.155 < 2e-16 ***
## foxPerception -0.12519 0.02737 -4.575 5.27e-06 ***
## mediaPerception 0.51459 0.04188 12.286 < 2e-16 ***
## pDem_Rep -0.68835 0.08434 -8.161 8.29e-16 ***
## pInd_Not 0.56140 0.08566 6.554 8.28e-11 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.072 on 1203 degrees of freedom
## (1268 observations deleted due to missingness)
## Multiple R-squared: 0.32, Adjusted R-squared: 0.3177
## F-statistic: 141.5 on 4 and 1203 DF, p-value: < 2.2e-164. Models: Do media perception and partisanship interact to predict vote confidence?
a. Vote confidence by Fox News perception & partisan ID
vote.fox.party.int <- lm(voteLegit ~ foxPerception * (pDem_Rep + pInd_Not), data = d)
tab_model(vote.fox.party.int)| Â | voteLegit | ||
|---|---|---|---|
| Predictors | Estimates | CI | p |
| (Intercept) | 3.50 | 3.33 – 3.67 | <0.001 |
| foxPerception | -0.08 | -0.15 – -0.01 | 0.025 |
| pDem_Rep | -1.57 | -1.89 – -1.25 | <0.001 |
| pInd_Not | 0.34 | -0.09 – 0.78 | 0.119 |
| foxPerception * pDem_Rep | 0.14 | 0.02 – 0.26 | 0.021 |
| foxPerception * pInd_Not | 0.09 | -0.09 – 0.27 | 0.318 |
| Observations | 1208 | ||
| R2 / R2 adjusted | 0.239 / 0.236 | ||
b. Vote confidence by Other Media perception
vote.othMedia.party.int <- lm(voteLegit ~ mediaPerception * (pDem_Rep + pInd_Not), data = d)
tab_model(vote.othMedia.party.int)| Â | voteLegit | ||
|---|---|---|---|
| Predictors | Estimates | CI | p |
| (Intercept) | 2.17 | 1.94 – 2.39 | <0.001 |
| mediaPerception | 0.49 | 0.40 – 0.58 | <0.001 |
| pDem_Rep | -1.00 | -1.48 – -0.53 | <0.001 |
| pInd_Not | 0.84 | 0.29 – 1.39 | 0.003 |
|
mediaPerception * pDem_Rep |
0.05 | -0.12 – 0.23 | 0.549 |
|
mediaPerception * pInd_Not |
-0.13 | -0.34 – 0.09 | 0.253 |
| Observations | 1208 | ||
| R2 / R2 adjusted | 0.309 / 0.306 | ||
c. Vote confidence by Fox & Other Media perception
vote.media.party.int <- lm(voteLegit ~ foxPerception * mediaPerception * (pDem_Rep + pInd_Not), data = d)
tab_model(vote.media.party.int)| Â | voteLegit | ||
|---|---|---|---|
| Predictors | Estimates | CI | p |
| (Intercept) | 2.16 | 1.68 – 2.64 | <0.001 |
| foxPerception | -0.15 | -0.34 – 0.04 | 0.128 |
| mediaPerception | 0.72 | 0.52 – 0.92 | <0.001 |
| pDem_Rep | -1.69 | -2.70 – -0.68 | 0.001 |
| pInd_Not | 0.19 | -0.95 – 1.34 | 0.740 |
|
foxPerception * mediaPerception |
-0.04 | -0.10 – 0.03 | 0.276 |
| foxPerception * pDem_Rep | 0.27 | -0.11 – 0.65 | 0.166 |
| foxPerception * pInd_Not | 0.50 | 0.03 – 0.97 | 0.035 |
|
mediaPerception * pDem_Rep |
0.38 | -0.04 – 0.79 | 0.078 |
|
mediaPerception * pInd_Not |
-0.14 | -0.61 – 0.33 | 0.551 |
|
(foxPerception mediaPerception) pDem_Rep |
-0.09 | -0.23 – 0.04 | 0.178 |
|
(foxPerception mediaPerception) pInd_Not |
-0.07 | -0.23 – 0.09 | 0.396 |
| Observations | 1208 | ||
| R2 / R2 adjusted | 0.335 / 0.329 | ||
5. Model Testing: Does the model predicting vote confidence by partisanship and media consumption explain more variance than the model without media consumption in it?
a. ANOVA testing model 2a against model 3c
anova(vote.party, vote.media.party)6. Figure 1: Vote Confidence by partisanship
a. Combining own and nationwide vote conf. into long df
(voteconf <- as.data.frame(cbind(d[d$election_timing != "Pre-election",]$party_factor, d[d$election_timing != "Pre-election",]$ownvote_conf, d[d$election_timing != "Pre-election",]$overallvote_conf)))voteconf <- as.data.frame(cbind(voteconf, d[d$election_timing != "Pre-election",]$election_timing))
names(voteconf) <- c("party_factor", "Own","Nationwide", "election_timing")
voteconf$party_factor <- recode_factor(voteconf$party_factor, `1` = "Democrat", `2` = "Republican", `3` = "Independent")
voteconffull.df <- tidyr::gather(voteconf, Vote_Type, Confidence, Own:Nationwide, factor_key=TRUE)
voteconf.df <- voteconffull.df[voteconffull.df$election_timing != "Pre-election",]b. Violin & box plot of the two vote confidence measures
voteConf.party_plot <- ggplot(voteconf.df[!is.na(voteconf.df$party_factor),],
aes(x = election_timing, y = Confidence)) +
geom_violin(alpha = .6, aes(fill = party_factor)) +
# geom_boxplot(position = position_dodge(.9), width = .1, outlier.colour = NA) +
geom_point(stat = 'summary', fun = 'mean', size = 1) +
geom_path(stat = 'summary', fun = 'mean', aes(group = 1, col = party_factor)) +
stat_summary(fun.data = mean_cl_normal, geom = "errorbar", position = position_dodge(.9), width=.1, fun.args = list(mult = 1)) +
facet_grid(Vote_Type ~ party_factor)
voteConf.party_plot +
scale_fill_manual(values = c('dodgerblue','red3','orchid4')) +
scale_color_manual(values = c('dodgerblue','red3','orchid4')) +
xlab("Participant Partisan ID") +
ylab("Vote Legitimacy") +
theme_classic()## Warning: Removed 39 rows containing non-finite values (stat_ydensity).## Warning: Removed 39 rows containing non-finite values (stat_summary).
## Warning: Removed 39 rows containing non-finite values (stat_summary).
## Warning: Removed 39 rows containing non-finite values (stat_summary).
7. Stats for Figure 1
a. Correlation between two vote conf. measures
cor.test(d$ownvote_conf, d$overallvote_conf)##
## Pearson's product-moment correlation
##
## data: d$ownvote_conf and d$overallvote_conf
## t = 39.423, df = 1207, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.7245086 0.7738872
## sample estimates:
## cor
## 0.7502422b. Descriptives for Vote Confidence by party ID
aggregate(d$voteLegit, list(d$party_factor, d$election_timing), FUN = mean, na.rm = T)c. Interactive model of partisanship and election timing
# Timing contrast JUST for vote confidence as DV (only during & post available)
d$tDuring_Post <- NA
d$tDuring_Post[d$election_timing == 'During-election'] <- -.5
d$tDuring_Post[d$election_timing == 'Post-election'] <- .5
# Overall Model
vote.party.time <- lm(voteLegit ~ (pDem_Rep + pInd_Not) * tDuring_Post, data = d)
summary(vote.party.time)##
## Call:
## lm(formula = voteLegit ~ (pDem_Rep + pInd_Not) * tDuring_Post,
## data = d)
##
## Residuals:
## Min 1Q Median 3Q Max
## -3.3636 -0.8636 0.0760 0.6364 2.2146
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 3.33537 0.03622 92.097 < 2e-16 ***
## pDem_Rep -1.25428 0.07106 -17.652 < 2e-16 ***
## pInd_Not 0.54940 0.09002 6.103 1.40e-09 ***
## tDuring_Post 0.08277 0.07243 1.143 0.253
## pDem_Rep:tDuring_Post -0.64791 0.14212 -4.559 5.66e-06 ***
## pInd_Not:tDuring_Post 0.09974 0.18003 0.554 0.580
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.127 on 1202 degrees of freedom
## (1268 observations deleted due to missingness)
## Multiple R-squared: 0.248, Adjusted R-squared: 0.2448
## F-statistic: 79.27 on 5 and 1202 DF, p-value: < 2.2e-16# At level of Dems
vote.party.time.Dems <- lm(voteLegit ~ (pDemR + pDemI) * tDuring_Post, data = d)
summary(vote.party.time.Dems)##
## Call:
## lm(formula = voteLegit ~ (pDemR + pDemI) * tDuring_Post, data = d)
##
## Residuals:
## Min 1Q Median 3Q Max
## -3.3636 -0.8636 0.0760 0.6364 2.2146
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 4.14382 0.04799 86.356 < 2e-16 ***
## pDemR -1.25428 0.07106 -17.652 < 2e-16 ***
## pDemI -1.17654 0.09562 -12.304 < 2e-16 ***
## tDuring_Post 0.43964 0.09597 4.581 5.11e-06 ***
## pDemR:tDuring_Post -0.64791 0.14212 -4.559 5.66e-06 ***
## pDemI:tDuring_Post -0.42369 0.19124 -2.215 0.0269 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.127 on 1202 degrees of freedom
## (1268 observations deleted due to missingness)
## Multiple R-squared: 0.248, Adjusted R-squared: 0.2448
## F-statistic: 79.27 on 5 and 1202 DF, p-value: < 2.2e-16# At level of Reps
vote.party.time.Reps <- lm(voteLegit ~ (pRepD + pRepI) * tDuring_Post, data = d)
summary(vote.party.time.Reps)##
## Call:
## lm(formula = voteLegit ~ (pRepD + pRepI) * tDuring_Post, data = d)
##
## Residuals:
## Min 1Q Median 3Q Max
## -3.3636 -0.8636 0.0760 0.6364 2.2146
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.88953 0.05241 55.135 < 2e-16 ***
## pRepD 1.25428 0.07106 17.652 < 2e-16 ***
## pRepI 0.07774 0.09791 0.794 0.4274
## tDuring_Post -0.20827 0.10482 -1.987 0.0471 *
## pRepD:tDuring_Post 0.64791 0.14212 4.559 5.66e-06 ***
## pRepI:tDuring_Post 0.22422 0.19583 1.145 0.2525
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.127 on 1202 degrees of freedom
## (1268 observations deleted due to missingness)
## Multiple R-squared: 0.248, Adjusted R-squared: 0.2448
## F-statistic: 79.27 on 5 and 1202 DF, p-value: < 2.2e-168. Figure 2: Media Perception box plot
a. creating df. with only media measures, party ID, and election timing
(media <- as.data.frame(cbind(d[,c(146,141,84:98,100:114)])))media[,18:32] <- media[,18:32] + 3
media$media_1 <- rowMeans(media[,c(3,18)], na.rm = T)
media$media_2 <- rowMeans(media[,c(4,19)], na.rm = T)
media$media_3 <- rowMeans(media[,c(5,20)], na.rm = T)
media$media_4 <- rowMeans(media[,c(6,21)], na.rm = T)
media$media_5 <- rowMeans(media[,c(7,22)], na.rm = T)
media$media_6 <- rowMeans(media[,c(8,23)], na.rm = T)
media$media_7 <- rowMeans(media[,c(9,24)], na.rm = T)
media$media_8 <- rowMeans(media[,c(10,25)], na.rm = T)
media$media_9 <- rowMeans(media[,c(11,26)], na.rm = T)
media$media_10 <- rowMeans(media[,c(12,27)], na.rm = T)
media$media_11 <- rowMeans(media[,c(13,28)], na.rm = T)
media$media_12 <- rowMeans(media[,c(14,29)], na.rm = T)
media$media_13 <- rowMeans(media[,c(15,30)], na.rm = T)
media$media_14 <- rowMeans(media[,c(16,31)], na.rm = T)
media$media_15 <- rowMeans(media[,c(17,32)], na.rm = T)
media.df <- tidyr::gather(media, Source, Perception, media_1:media_15, factor_key=TRUE)
media.df$Source <- recode_factor(media.df$Source, "media_1" = "NYTimes", "media_2" = "WSJ", "media_3" = "WashPost", "media_4" = "USAToday", "media_5" = "Fox News", "media_6" = "CNN", "media_7" = "MSNBC", "media_8" = "Yahoo", "media_9" = "HuffPost", "media_10" = "AOL", "media_11" = "NPR", "media_12" = "ABC", "media_13" = "NBC", "media_14" = "CBS", "media_15" = "PBS")
media.df$Source <- factor(media.df$Source, levels = c("Fox News", "ABC", "AOL", "CBS", "CNN", "HuffPost", "MSNBC", "NBC", "NPR", "NYTimes", "PBS", "USAToday", "WashPost", "WSJ", "Yahoo"))b. Boxplot of media measures
media_plot <- ggplot(media.df[!is.na(media.df$party_factor),],
aes(x = Source, y = Perception, fill = party_factor)) +
geom_bar(stat = 'summary', fun = 'mean', position = position_dodge(.9), alpha = .9) +
stat_summary(fun.data = mean_cl_normal, geom = "errorbar", position = position_dodge(.9), width=.1, fun.args = list(mult = 1))
media_plot +
scale_fill_manual("Participant Partisan ID", values = c("dodgerblue", "red3", "orchid4")) +
xlab("Media Source") +
ylab("Perception of Media Source") +
coord_cartesian(ylim = c(1,4)) +
theme_minimal() +
theme(axis.text.x = element_text(angle = 45, hjust = 1)) ## Warning: Removed 330 rows containing non-finite values (stat_summary).
## Warning: Removed 330 rows containing non-finite values (stat_summary).
9. Stats for Figure 2
a. Correlation between media measures
cor.test(d$mediaExposure, d$mediaTrust)##
## Pearson's product-moment correlation
##
## data: d$mediaExposure and d$mediaTrust
## t = 35.364, df = 2413, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.5573583 0.6099292
## sample estimates:
## cor
## 0.5842563