look at demographics as controls, have seen trust in sci and analytical media consumption seen, but not sure where. What is the confounding factor (may be education?) check this!

DESCRIPTIVE STATS

1. demographics

a. sample size

table(!is.na(d$participant))

## 
## TRUE 
## 3860

table(!is.na(d$participant) & d$wave_1 == 1)

## 
## TRUE 
## 3626

table(!is.na(d$participant) & d$wave_2 == 2)

## 
## TRUE 
## 2749

table(!is.na(d$participant) & d$wave_3 == 3)

## 
## TRUE 
## 1655

b. age

describe(d$age)

describe(d$age[(!is.na(d$participant) & d$wave_1 == 1)])

describe(d$age[(!is.na(d$participant) & d$wave_2 == 2)])

describe(d$age[(!is.na(d$participant) & d$wave_3 == 3)])

c. gender

t <- table(d$gender_factor[(!is.na(d$participant))])
round(prop.table(t), 3)

## 
## custom female   male 
##  0.002  0.503  0.495

t <- table(d$gender_factor[(!is.na(d$participant) & d$wave_1 == 1)])
round(prop.table(t), 3)

## 
## custom female   male 
##  0.002  0.503  0.495

t <- table(d$gender_factor[(!is.na(d$participant) & d$wave_2 == 2)])
round(prop.table(t), 3)

## 
## custom female   male 
##  0.001  0.477  0.522

t <- table(d$gender_factor[(!is.na(d$participant) & d$wave_3 == 3)])
round(prop.table(t), 3)

## 
## custom female   male 
##  0.001  0.466  0.533

d. ethnicity

(t <- table(d$race_factor))

## 
##   asian   black   latin natAmer   other  pacIsl   white 
##     293     446     288      27      47      16    2224

round(prop.table(t), 3)

## 
##   asian   black   latin natAmer   other  pacIsl   white 
##   0.088   0.133   0.086   0.008   0.014   0.005   0.666

(t <- table(d$race_factor[d$wave_1 == 1]))

## 
##   asian   black   latin natAmer   other  pacIsl   white 
##     293     446     288      27      47      16    2224

round(prop.table(t), 3)

## 
##   asian   black   latin natAmer   other  pacIsl   white 
##   0.088   0.133   0.086   0.008   0.014   0.005   0.666

(t <- table(d$race_factor[d$wave_2 == 2]))

## 
##   asian   black   latin natAmer   other  pacIsl   white 
##     236     308     197      21      32      10    1725

round(prop.table(t), 3)

## 
##   asian   black   latin natAmer   other  pacIsl   white 
##   0.093   0.122   0.078   0.008   0.013   0.004   0.682

(t <- table(d$race_factor[d$wave_3 == 3]))

## 
##   asian   black   latin natAmer   other  pacIsl   white 
##     128     177     114       7      14       4    1059

round(prop.table(t), 3)

## 
##   asian   black   latin natAmer   other  pacIsl   white 
##   0.085   0.118   0.076   0.005   0.009   0.003   0.705

e. party identity

(t <- table(d$party_factor))

## 
##    Democrat Independent  Republican 
##        1533         674        1238

round(prop.table(t), 3)

## 
##    Democrat Independent  Republican 
##       0.445       0.196       0.359

(t <- table(d$party_factor[d$wave_1 == 1]))

## 
##    Democrat Independent  Republican 
##        1533         674        1238

round(prop.table(t), 3)

## 
##    Democrat Independent  Republican 
##       0.445       0.196       0.359

(t <- table(d$party_factor[d$wave_2 == 2]))

## 
##    Democrat Independent  Republican 
##        1127         452         968

round(prop.table(t), 3)

## 
##    Democrat Independent  Republican 
##       0.442       0.177       0.380

(t <- table(d$party_factor[d$wave_3 == 3]))

## 
##    Democrat Independent  Republican 
##         680         260         581

round(prop.table(t), 3)

## 
##    Democrat Independent  Republican 
##       0.447       0.171       0.382

f. education

describe(d$education)

describe(d$education[d$wave_1 == 1])

describe(d$education[d$wave_2 == 2])

describe(d$education[d$wave_3 == 3])

2. individual difference measures

a. media exposure

Not included here, but asked in all 3 waves, media exposure asks participants to “Consider each of the media sources below. In general, how much do you get news about Covid-19 from each source?” on a scale from 1 (Not at all) to 3 (Somewhat) to 5 (A great deal).

b. analytical media index

i. wave 1

Collected in wave 1 (July - August 2020), the first step to creating the analytical media index is multiplying individual analytic thinking scores for each media outlet by participant rated exposure to that outlet, Then taking the proportion of each of these products (i.e., dividing by the 12 possible US outlets to be exposed to)

foxAnalyticalIndex = (foxNewsExposure x foxAnalyticalScore) (foxAnalyticalIndex + cnnAnalyticalIndex + msnbcAnalyticalIndex + …) / 12 total outlets

describe(d$index_ANexp_w1)

ii. wave 2

Collected in wave 2 (November 2020), the creation of this index follows the same path as above with the exception of using wave 2 updated media analytical thinking scores and participant media exposure ratings.

describe(d$index_ANexp_w2)

describe(w2$analytic)

iii. wave 3

Collected in wave 3 (March 2022), the creation of this index follows the same path as above using updated participant media exposure ratings, but with the exception of using the average of wave 1 and wave 2 media analytical thinking scores.

describe(d$index_ANexp_w3)

c. symbolic ideology

Asked during our wave 1 survey (July-August 2020), symbolic ideology is an average of three items. They ask “How liberal/conservative…” (1) in general, (2) on social issues, and (3) on economic issues. Participants answered on a scale from -3 (Very liberal) to 0 (Moderate) to +3 (Very conservative).

## 
## Reliability analysis   
## Call: psych::alpha(x = data.frame(d$symbolic_beliefs_1, d$symbolic_beliefs_2, 
##     d$symbolic_beliefs_3), cumulative = F, na.rm = T, delete = T)
## 
##   raw_alpha std.alpha G6(smc) average_r S/N    ase mean  sd median_r
##       0.95      0.95    0.93      0.85  17 0.0016 0.11 1.6     0.86
## 
##  lower alpha upper     95% confidence boundaries
## 0.94 0.95 0.95 
## 
##  Reliability if an item is dropped:
##                      raw_alpha std.alpha G6(smc) average_r  S/N alpha se var.r
## d.symbolic_beliefs_1      0.89      0.89    0.80      0.80  8.2   0.0035    NA
## d.symbolic_beliefs_2      0.92      0.92    0.86      0.86 12.1   0.0025    NA
## d.symbolic_beliefs_3      0.94      0.94    0.89      0.89 17.0   0.0018    NA
##                      med.r
## d.symbolic_beliefs_1  0.80
## d.symbolic_beliefs_2  0.86
## d.symbolic_beliefs_3  0.89
## 
##  Item statistics 
##                         n raw.r std.r r.cor r.drop   mean  sd
## d.symbolic_beliefs_1 3462  0.97  0.97  0.95   0.92  0.081 1.7
## d.symbolic_beliefs_2 3463  0.95  0.95  0.91   0.88 -0.041 1.7
## d.symbolic_beliefs_3 3463  0.93  0.93  0.88   0.85  0.292 1.7
## 
## Non missing response frequency for each item
##                        -3   -2   -1    0    1    2    3 miss
## d.symbolic_beliefs_1 0.07 0.13 0.08 0.38 0.10 0.13 0.10  0.1
## d.symbolic_beliefs_2 0.09 0.14 0.12 0.33 0.10 0.13 0.09  0.1
## d.symbolic_beliefs_3 0.06 0.11 0.09 0.35 0.13 0.16 0.11  0.1

d. trust in science

Asked during our wave 3 survey (March 2022), trust in science is an average of 15-items. Examples range from trust in the scientific method (e.g., “We cannot trust science because it moves too slowly”) to trust in scientists themselves (e.g., “We should trust that scientists are being ethical in their work”). Participants answered on a scale from 1 (strongly disagree) to 3 (neutral) to 5 (strongly agree).

## 
## Reliability analysis   
## Call: psych::alpha(x = data.frame(d$sciTrust_1, d$sciTrust_2, d$sciTrust_3, 
##     d$sciTrust_4, d$sciTrust_5, d$sciTrust_6, d$sciTrust_7, d$sciTrust_8, 
##     d$sciTrust_9, d$sciTrust_10, d$sciTrust_11, d$sciTrust_12, 
##     d$sciTrust_13, d$sciTrust_14, d$sciTrust_15), cumulative = F, 
##     na.rm = T, delete = T)
## 
##   raw_alpha std.alpha G6(smc) average_r S/N    ase mean   sd median_r
##       0.94      0.94    0.95       0.5  15 0.0015  3.5 0.82      0.5
## 
##  lower alpha upper     95% confidence boundaries
## 0.94 0.94 0.94 
## 
##  Reliability if an item is dropped:
##               raw_alpha std.alpha G6(smc) average_r S/N alpha se var.r med.r
## d.sciTrust_1       0.94      0.94    0.95      0.51  15   0.0015 0.021  0.51
## d.sciTrust_2       0.93      0.93    0.95      0.49  14   0.0016 0.022  0.49
## d.sciTrust_3       0.93      0.93    0.95      0.50  14   0.0016 0.022  0.49
## d.sciTrust_4       0.93      0.93    0.95      0.50  14   0.0016 0.022  0.49
## d.sciTrust_5       0.93      0.93    0.95      0.49  14   0.0016 0.022  0.49
## d.sciTrust_6       0.94      0.93    0.95      0.51  14   0.0015 0.021  0.53
## d.sciTrust_7       0.93      0.93    0.95      0.50  14   0.0015 0.021  0.51
## d.sciTrust_8       0.94      0.94    0.95      0.51  15   0.0015 0.020  0.53
## d.sciTrust_9       0.94      0.94    0.95      0.51  15   0.0015 0.019  0.53
## d.sciTrust_10      0.94      0.93    0.95      0.51  14   0.0015 0.021  0.53
## d.sciTrust_11      0.93      0.93    0.95      0.49  13   0.0016 0.023  0.48
## d.sciTrust_12      0.93      0.93    0.95      0.50  14   0.0016 0.022  0.49
## d.sciTrust_13      0.93      0.93    0.95      0.49  13   0.0016 0.021  0.49
## d.sciTrust_14      0.93      0.93    0.95      0.50  14   0.0016 0.022  0.49
## d.sciTrust_15      0.93      0.93    0.95      0.51  14   0.0016 0.022  0.49
## 
##  Item statistics 
##                  n raw.r std.r r.cor r.drop mean   sd
## d.sciTrust_1  1077  0.65  0.64  0.60   0.59  3.3 1.16
## d.sciTrust_2  1077  0.81  0.80  0.79   0.78  3.2 1.23
## d.sciTrust_3  1075  0.78  0.77  0.75   0.74  3.6 1.11
## d.sciTrust_4  1075  0.72  0.70  0.68   0.66  3.2 1.19
## d.sciTrust_5  1075  0.81  0.80  0.79   0.77  3.5 1.19
## d.sciTrust_6  1077  0.68  0.70  0.68   0.63  3.7 1.00
## d.sciTrust_7  1071  0.72  0.74  0.73   0.68  3.6 1.01
## d.sciTrust_8  1073  0.64  0.66  0.64   0.58  3.6 1.06
## d.sciTrust_9  1076  0.62  0.64  0.62   0.57  3.5 0.96
## d.sciTrust_10 1077  0.68  0.69  0.68   0.63  3.5 1.06
## d.sciTrust_11 1075  0.83  0.83  0.81   0.80  3.5 1.19
## d.sciTrust_12 1075  0.76  0.75  0.73   0.71  3.2 1.19
## d.sciTrust_13 1075  0.84  0.83  0.83   0.81  3.3 1.17
## d.sciTrust_14 1072  0.73  0.72  0.69   0.68  3.2 1.19
## d.sciTrust_15 1073  0.70  0.70  0.68   0.66  3.8 1.04
## 
## Non missing response frequency for each item
##                  1    2    3    4    5 miss
## d.sciTrust_1  0.08 0.16 0.30 0.30 0.16 0.72
## d.sciTrust_2  0.11 0.17 0.27 0.28 0.16 0.72
## d.sciTrust_3  0.05 0.10 0.28 0.32 0.24 0.72
## d.sciTrust_4  0.09 0.18 0.32 0.24 0.18 0.72
## d.sciTrust_5  0.07 0.12 0.27 0.29 0.25 0.72
## d.sciTrust_6  0.04 0.07 0.26 0.41 0.22 0.72
## d.sciTrust_7  0.04 0.07 0.29 0.40 0.19 0.72
## d.sciTrust_8  0.06 0.08 0.27 0.41 0.18 0.72
## d.sciTrust_9  0.04 0.06 0.39 0.36 0.14 0.72
## d.sciTrust_10 0.05 0.09 0.30 0.38 0.18 0.72
## d.sciTrust_11 0.07 0.13 0.25 0.31 0.23 0.72
## d.sciTrust_12 0.09 0.17 0.31 0.26 0.16 0.72
## d.sciTrust_13 0.08 0.16 0.31 0.27 0.18 0.72
## d.sciTrust_14 0.09 0.18 0.31 0.25 0.17 0.72
## d.sciTrust_15 0.04 0.05 0.26 0.33 0.32 0.72

e. science literacy (might cut)

Asked during out wave 3 survey (March 2022), science literacy is a proportion of correct/incorrect answers pertaining to 6-items identifying basic scientific understanding (e.g., “The center of the earth is very hot”, “Electrons are smaller than atoms”).

f. cognitive reflection test

Asked during our wave 3 survey (March 2022), the CRT is a proportion of correct/incorrect answers that leans into system 1 intuitive answer (wrong) over deliberate system 2 answer (correct answers; e.g., “A farmer had 15 sheep and all but 8 died. How many are left?”).

3. LIWC media measures

a. Analytical thinking wave 1

Collected during wave 1 (July - August 2020), media outlet LIWC analytic thinking scores captures the degree to which people use words that suggest formal, logical, and hierarchical thinking patterns. Column two is raw scores, and column three is standardized scores.

ggplot(data = w1, aes(x = mediaOutlet, y = analytic.s, fill = mediaOutlet)) + 
  ggtitle("Wave 1 Analytical Thinking Standardized") +
  geom_bar(stat = "identity", position = position_dodge()) + 
  geom_text(aes(label = mediaOutlet), vjust = 1.6, color = "black", position = position_dodge(0.9), size = 3) + 
  theme_minimal() + 
  coord_cartesian(ylim = c(-3, 3))

ggplot(data = w1, aes(x = mediaOutlet, y = analytic, fill = mediaOutlet)) + 
  ggtitle("Wave 1 Analytical Thinking Raw") +
  geom_bar(stat = "identity", position = position_dodge()) + 
  geom_text(aes(label = mediaOutlet), vjust = 1.6, color = "black", position = position_dodge(0.9), size = 3) + 
  theme_minimal() + 
  coord_cartesian(ylim = c(1, 100))

w1.x <- w1[,2:4]
w1.x[, 2:3] <- round(w1.x[, 2:3], 2)
w1.x

b. Analytical thinking wave 2

Collected during wave 2 (November 2020), media outlet LIWC analytic thinking scores captures the degree to which people use words that suggest formal, logical, and hierarchical thinking patterns. Column two is raw scores, and column three is standardized scores.

ggplot(data = w2, aes(x = mediaOutlet, y = analytic.s, fill = mediaOutlet)) + 
  ggtitle("Wave 2 + Analytical Thinking Standardized") +
  geom_bar(stat = "identity", position = position_dodge()) + 
  geom_text(aes(label = mediaOutlet), vjust = 1.6, color = "black", position = position_dodge(0.9), size = 3) + 
  theme_minimal() + 
  coord_cartesian(ylim = c(-3, 3))

ggplot(data = w2, aes(x = mediaOutlet, y = analytic, fill = mediaOutlet)) + 
  ggtitle("Wave 2 + Analytical Thinking Raw") +
  geom_bar(stat = "identity", position = position_dodge()) + 
  geom_text(aes(label = mediaOutlet), vjust = 1.6, color = "black", position = position_dodge(0.9), size = 3) + 
  theme_minimal() + 
  coord_cartesian(ylim = c(1, 100))

w2.x <- w2[,2:4]
w2.x[, 2:3] <- round(w2.x[, 2:3], 2)
w2.x

4. outcome measures

a. wave 1 vaccine attitudes

Collected during wave 1 (July - August 2020), participants were asked “would you get a Covid-19 vaccine?” and answered on a scale from -3 (Definitely would not get it) to 0 (Undecided) to +3 (Definitely would get it).

describe(d$vaxxIntentions_w1)

b. wave 2 vaccine attitudes

Collected during wave 2 (November 2020), participants were asked “If a Covid-19 vaccine were available today, would you get it?” and answered on a scale from -3 (Definitely would not get it) to 0 (Completely undecided) to +3 (Definitely would get it).

describe(d$vaxxIntentions_w2)

c. average of wave 1 + wave 2 vaccine attitudes

colnames(d)[which(names(d) == "avgvaxxIntentions")] <- "avgVaxxIntentions"
describe(d$avgVaxxIntentions)

d. correlation between wave 1 and wave 2 vaccine attitudes

cor.test(d$vaxxIntentions_w1, d$vaxxIntentions_w2)

## 
##  Pearson's product-moment correlation
## 
## data:  d$vaxxIntentions_w1 and d$vaxxIntentions_w2
## t = 53.555, df = 2493, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
##  0.7126403 0.7491665
## sample estimates:
##       cor 
## 0.7314276

e. vaccine behavior

Collected during wave 3 (March 2022), vaccine behavior is a collective measure of branched questions given to participants.

1 = not vaccinated

2 = partially vaccinated (e.g., 1 Moderna shot, but not the 2nd)

3 = fully vaccinated (e.g., 1 Johnson & Johnson shot, or 2 Modern/Pfizer shots)

4 = fully vaccinated and boosted

describe(d$vaxxBehavior)

5. pairwise correlation table for IVs, DVs, and covariates

corr <- data.frame(d$age)
colnames(corr)[colnames(corr)=="d.age"] <- "age"
corr$edu <- d$education
corr$female_1 <- d$female_1 #leave in here even if not covariate used
corr$white_1 <- d$white_1
corr$ideology <- d$ideology
corr$index_w1 <- d$index_ANexp_w1
corr$index_w2 <- d$index_ANexp_w2
corr$index_w3<- d$index_ANexp_w3
corr$vaxxIntentions <- d$avgVaxxIntentions
corr$vaxxBehavior<- d$vaxxBehavior

corr2 <- cor(corr, use = "pairwise")

ggcorrplot(corr2, type = "lower",
   lab = TRUE, title = "pairwise correlations")

6. ANALYSES

Study 2: US only (November 2020)

Key aims: 1) Replication, proximity vaccine, 2) stronger causal evidence w/ longitudinal data

a. vaxxIntentions ~ index.w2 * ideology (replication)

summary(m1 <- lm(vaxxIntentions_w2 ~ index_ANexp_w2.z * ideology.z + age.z + education.z + white_.5, data = d))

## 
## Call:
## lm(formula = vaxxIntentions_w2 ~ index_ANexp_w2.z * ideology.z + 
##     age.z + education.z + white_.5, data = d)
## 
## Residuals:
##     Min      1Q  Median      3Q     Max 
## -4.8466 -1.4224  0.0855  1.6579  4.3979 
## 
## Coefficients:
##                             Estimate Std. Error t value Pr(>|t|)    
## (Intercept)                  0.27442    0.04504   6.093 1.28e-09 ***
## index_ANexp_w2.z             0.42426    0.04302   9.862  < 2e-16 ***
## ideology.z                  -0.15717    0.04176  -3.764 0.000171 ***
## age.z                        0.37731    0.04275   8.825  < 2e-16 ***
## education.z                  0.14276    0.04531   3.150 0.001650 ** 
## white_.5                     0.38117    0.09147   4.167 3.19e-05 ***
## index_ANexp_w2.z:ideology.z  0.04115    0.03771   1.091 0.275241    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 2.01 on 2441 degrees of freedom
##   (1412 observations deleted due to missingness)
## Multiple R-squared:  0.08306,    Adjusted R-squared:  0.08081 
## F-statistic: 36.85 on 6 and 2441 DF,  p-value: < 2.2e-16

tab_model(m1,
          show.df = T, 
          show.ci = .95,
          show.se = T,
          show.stat = T,
          string.stat = "t",
          string.se="SE",
          string.est = "Est",
          digits = 3)

	vaxxIntentions_w2
Predictors	Est	SE	CI	t	p	df
(Intercept)	0.274	0.045	0.186 – 0.363	6.093	<0.001	2441.000
index_ANexp_w2.z	0.424	0.043	0.340 – 0.509	9.862	<0.001	2441.000
ideology.z	-0.157	0.042	-0.239 – -0.075	-3.764	<0.001	2441.000
age.z	0.377	0.043	0.293 – 0.461	8.825	<0.001	2441.000
education.z	0.143	0.045	0.054 – 0.232	3.150	0.002	2441.000
white_.5	0.381	0.091	0.202 – 0.561	4.167	<0.001	2441.000
index_ANexp_w2.z * ideology.z	0.041	0.038	-0.033 – 0.115	1.091	0.275	2441.000
Observations	2448
R² / R² adjusted	0.083 / 0.081

b. vaxxIntentions.w2 ~ index.w2 * all demographics (replication)

summary(m1 <- lm(vaxxIntentions_w2 ~ index_ANexp_w2.z * (ideology.z + age.z + education.z + white_.5), data = d))

## 
## Call:
## lm(formula = vaxxIntentions_w2 ~ index_ANexp_w2.z * (ideology.z + 
##     age.z + education.z + white_.5), data = d)
## 
## Residuals:
##     Min      1Q  Median      3Q     Max 
## -4.9372 -1.3793  0.0754  1.6636  4.8060 
## 
## Coefficients:
##                              Estimate Std. Error t value Pr(>|t|)    
## (Intercept)                   0.26650    0.04632   5.754 9.81e-09 ***
## index_ANexp_w2.z              0.40611    0.04387   9.258  < 2e-16 ***
## ideology.z                   -0.17549    0.04235  -4.143 3.54e-05 ***
## age.z                         0.37513    0.04272   8.781  < 2e-16 ***
## education.z                   0.15714    0.04649   3.380 0.000735 ***
## white_.5                      0.39014    0.09239   4.223 2.50e-05 ***
## index_ANexp_w2.z:ideology.z   0.04899    0.03809   1.286 0.198507    
## index_ANexp_w2.z:age.z       -0.13619    0.04446  -3.063 0.002216 ** 
## index_ANexp_w2.z:education.z -0.04115    0.03856  -1.067 0.286010    
## index_ANexp_w2.z:white_.5     0.01528    0.08687   0.176 0.860437    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 2.006 on 2438 degrees of freedom
##   (1412 observations deleted due to missingness)
## Multiple R-squared:  0.08729,    Adjusted R-squared:  0.08393 
## F-statistic: 25.91 on 9 and 2438 DF,  p-value: < 2.2e-16

tab_model(m1,
          show.df = T, 
          show.ci = .95,
          show.se = T,
          show.stat = T,
          string.stat = "t",
          string.se="SE",
          string.est = "Est",
          digits = 3)

	vaxxIntentions_w2
Predictors	Est	SE	CI	t	p	df
(Intercept)	0.266	0.046	0.176 – 0.357	5.754	<0.001	2438.000
index_ANexp_w2.z	0.406	0.044	0.320 – 0.492	9.258	<0.001	2438.000
ideology.z	-0.175	0.042	-0.259 – -0.092	-4.143	<0.001	2438.000
age.z	0.375	0.043	0.291 – 0.459	8.781	<0.001	2438.000
education.z	0.157	0.046	0.066 – 0.248	3.380	0.001	2438.000
white_.5	0.390	0.092	0.209 – 0.571	4.223	<0.001	2438.000
index_ANexp_w2.z * ideology.z	0.049	0.038	-0.026 – 0.124	1.286	0.199	2438.000
index_ANexp_w2.z * age.z	-0.136	0.044	-0.223 – -0.049	-3.063	0.002	2438.000
index_ANexp_w2.z * education.z	-0.041	0.039	-0.117 – 0.034	-1.067	0.286	2438.000
index_ANexp_w2.z * white_.5	0.015	0.087	-0.155 – 0.186	0.176	0.860	2438.000
Observations	2448
R² / R² adjusted	0.087 / 0.084

c. mediation path: index.w2 –> expertTrust –> vaxxIntentions.w2

i. controlling for ideology

vaxxIntentions.w2 ~ inde.w2.z + ideology.z + age.z + white_.5 + education.z

trustExpert.w2 ~ index.w2.z + ideology.z + age.z + white_.5 + education.z

vaxxIntentions.w2 ~ trustExpert.w2.z + index.w2.z + ideology.z + age.z + white_.5 + education.z

indirect effect: https://quantpsy.org/sobel/sobel.htm

#model 1
m2.xy <- lm(vaxxIntentions_w2 ~ index_ANexp_w2.z + ideology.z + age.z + white_.5 + education.z, data = d)

# model 2
m2.xm <- lm(trustExpert_w2 ~ index_ANexp_w2.z + ideology.z + age.z + white_.5 + education.z, data = d)

# model 3
m2.xmy <- lm(vaxxIntentions_w2 ~ trustExpert_w2.z + index_ANexp_w2.z + ideology.z + age.z + white_.5 + education.z, data = d)

tab_model(m2.xy, m2.xm, m2.xmy, 
          show.ci = .95,
          show.stat = T,
          string.stat = "t",
          string.se="SE",
          string.est = "Est",
          digits = 3)

	vaxxIntentions_w2				trustExpert_w2				vaxxIntentions_w2
Predictors	Est	CI	t	p	Est	CI	t	p	Est	CI	t	p
(Intercept)	0.265	0.178 – 0.352	5.995	<0.001	1.491	1.433 – 1.550	50.135	<0.001	0.278	0.193 – 0.362	6.460	<0.001
index_ANexp_w2.z	0.422	0.337 – 0.506	9.817	<0.001	0.304	0.247 – 0.360	10.497	<0.001	0.316	0.232 – 0.400	7.397	<0.001
ideology.z	-0.156	-0.238 – -0.074	-3.739	<0.001	-0.402	-0.457 – -0.347	-14.312	<0.001	-0.016	-0.099 – 0.067	-0.382	0.702
age.z	0.374	0.290 – 0.458	8.771	<0.001	0.210	0.153 – 0.266	7.304	<0.001	0.300	0.218 – 0.383	7.167	<0.001
white_.5	0.376	0.197 – 0.555	4.115	<0.001	0.250	0.130 – 0.371	4.069	<0.001	0.288	0.114 – 0.463	3.235	0.001
education.z	0.142	0.053 – 0.231	3.140	0.002	0.109	0.050 – 0.169	3.585	<0.001	0.105	0.018 – 0.191	2.367	0.018
trustExpert_w2.z									0.516	0.432 – 0.601	11.948	<0.001
Observations	2448				2449				2448
R² / R² adjusted	0.083 / 0.081				0.158 / 0.156				0.133 / 0.131

ii. interaction with ideology

vaxxIntentions.w2 ~ inde.w2.z * ideology.z + age.z + white_.5 + education.z

trustExpert.w2 ~ index.w2.z * ideology.z + age.z + white_.5 + education.z

vaxxIntentions.w2 ~ (trustExpert.w2.z + index.w2.z) * ideology.z + age.z + white_.5 + education.z

#model 1
m2.xy <- lm(vaxxIntentions_w2 ~ index_ANexp_w2.z * ideology.z + age.z + white_.5 + education.z, data = d)

# model 2
m2.xm <- lm(trustExpert_w2 ~ index_ANexp_w2.z * ideology.z + age.z + white_.5 + education.z, data = d)

# model 3
m2.xmy <- lm(vaxxIntentions_w2 ~ (trustExpert_w2.z + index_ANexp_w2.z) * ideology.z + age.z + white_.5 + education.z, data = d)

tab_model(m2.xy, m2.xm, m2.xmy, 
          show.ci = .95,
          show.stat = T,
          string.stat = "t",
          string.se="SE",
          string.est = "Est",
          digits = 3)

	vaxxIntentions_w2				trustExpert_w2				vaxxIntentions_w2
Predictors	Est	CI	t	p	Est	CI	t	p	Est	CI	t	p
(Intercept)	0.274	0.186 – 0.363	6.093	<0.001	1.518	1.459 – 1.577	50.259	<0.001	0.268	0.181 – 0.356	6.001	<0.001
index_ANexp_w2.z	0.424	0.340 – 0.509	9.862	<0.001	0.311	0.254 – 0.367	10.769	<0.001	0.317	0.233 – 0.401	7.400	<0.001
ideology.z	-0.157	-0.239 – -0.075	-3.764	<0.001	-0.405	-0.460 – -0.350	-14.470	<0.001	-0.011	-0.094 – 0.072	-0.259	0.796
age.z	0.377	0.293 – 0.461	8.825	<0.001	0.219	0.162 – 0.275	7.626	<0.001	0.297	0.215 – 0.380	7.047	<0.001
white_.5	0.381	0.202 – 0.561	4.167	<0.001	0.265	0.145 – 0.385	4.321	<0.001	0.283	0.108 – 0.458	3.167	0.002
education.z	0.143	0.054 – 0.232	3.150	0.002	0.111	0.051 – 0.170	3.639	<0.001	0.103	0.016 – 0.189	2.319	0.020
index_ANexp_w2.z * ideology.z	0.041	-0.033 – 0.115	1.091	0.275	0.114	0.065 – 0.164	4.527	<0.001	0.014	-0.061 – 0.090	0.374	0.709
trustExpert_w2.z									0.527	0.440 – 0.614	11.855	<0.001
trustExpert_w2.z * ideology.z									-0.047	-0.128 – 0.035	-1.120	0.263
Observations	2448				2449				2448
R² / R² adjusted	0.083 / 0.081				0.165 / 0.163				0.134 / 0.131

iii. interaction with all demographics

vaxxIntentions.w2 ~ inde.w2.z * (ideology.z + age.z + white_.5 + education.z)

trustExpert.w2 ~ index.w2.z * (ideology.z + age.z + white_.5 + education.z)

vaxxIntentions.w2 ~ (trustExpert.w2.z + index.w2.z) * (ideology.z + age.z + white_.5 + education.z)

#model 1
m2.xy <- lm(vaxxIntentions_w2 ~ index_ANexp_w2.z * (ideology.z + age.z + white_.5 + education.z), data = d)

# model 2
m2.xm <- lm(trustExpert_w2 ~ index_ANexp_w2.z * (ideology.z + age.z + white_.5 + education.z), data = d)

# model 3
m2.xmy <- lm(vaxxIntentions_w2 ~ (trustExpert_w2.z + index_ANexp_w2.z) * (ideology.z + age.z + white_.5 + education.z), data = d)

tab_model(m2.xy, m2.xm, m2.xmy, 
          show.ci = .95,
          show.stat = T,
          string.stat = "t",
          string.se="SE",
          string.est = "Est",
          digits = 3)

	vaxxIntentions_w2				trustExpert_w2				vaxxIntentions_w2
Predictors	Est	CI	t	p	Est	CI	t	p	Est	CI	t	p
(Intercept)	0.266	0.176 – 0.357	5.754	<0.001	1.535	1.474 – 1.596	49.378	<0.001	0.246	0.155 – 0.337	5.286	<0.001
index_ANexp_w2.z	0.406	0.320 – 0.492	9.258	<0.001	0.308	0.250 – 0.365	10.448	<0.001	0.300	0.215 – 0.386	6.881	<0.001
ideology.z	-0.175	-0.259 – -0.092	-4.143	<0.001	-0.392	-0.448 – -0.336	-13.802	<0.001	-0.027	-0.112 – 0.058	-0.625	0.532
age.z	0.375	0.291 – 0.459	8.781	<0.001	0.217	0.161 – 0.273	7.576	<0.001	0.297	0.214 – 0.381	7.007	<0.001
white_.5	0.390	0.209 – 0.571	4.223	<0.001	0.243	0.122 – 0.365	3.925	<0.001	0.300	0.123 – 0.477	3.323	0.001
education.z	0.157	0.066 – 0.248	3.380	0.001	0.099	0.038 – 0.160	3.167	0.002	0.126	0.037 – 0.215	2.765	0.006
index_ANexp_w2.z * ideology.z	0.049	-0.026 – 0.124	1.286	0.199	0.106	0.055 – 0.156	4.129	<0.001	0.023	-0.053 – 0.099	0.596	0.551
index_ANexp_w2.z * age.z	-0.136	-0.223 – -0.049	-3.063	0.002	0.039	-0.019 – 0.098	1.323	0.186	-0.129	-0.218 – -0.040	-2.836	0.005
index_ANexp_w2.z * white_.5	0.015	-0.155 – 0.186	0.176	0.860	0.108	-0.007 – 0.222	1.848	0.065	-0.071	-0.240 – 0.099	-0.815	0.415
index_ANexp_w2.z * education.z	-0.041	-0.117 – 0.034	-1.067	0.286	0.029	-0.022 – 0.079	1.103	0.270	-0.058	-0.132 – 0.016	-1.534	0.125
trustExpert_w2.z									0.502	0.410 – 0.594	10.724	<0.001
trustExpert_w2.z * ideology.z									-0.060	-0.144 – 0.024	-1.408	0.159
trustExpert_w2.z * age.z									-0.056	-0.147 – 0.034	-1.217	0.224
trustExpert_w2.z * white_.5									0.209	0.029 – 0.389	2.281	0.023
trustExpert_w2.z * education.z									0.055	-0.025 – 0.136	1.343	0.179
Observations	2448				2449				2448
R² / R² adjusted	0.087 / 0.084				0.167 / 0.164				0.142 / 0.137

d. Longitudinal: vaxxIntentions.w2 ~ index.w2.z * ideology.z + age.z + edu.z + white_.5 + index.w1.z + vaxxIntentions.w1.z

summary(mb <- lm(vaxxIntentions_w2 ~ index_ANexp_w2.z * ideology.z + age.z + education.z + white_.5 + index_ANexp_w1.z + vaxxIntentions_w1.z, data = d))

## 
## Call:
## lm(formula = vaxxIntentions_w2 ~ index_ANexp_w2.z * ideology.z + 
##     age.z + education.z + white_.5 + index_ANexp_w1.z + vaxxIntentions_w1.z, 
##     data = d)
## 
## Residuals:
##     Min      1Q  Median      3Q     Max 
## -5.1857 -0.8324  0.0467  0.9249  5.4213 
## 
## Coefficients:
##                             Estimate Std. Error t value Pr(>|t|)    
## (Intercept)                  0.27268    0.03187   8.556  < 2e-16 ***
## index_ANexp_w2.z             0.15245    0.04124   3.697 0.000223 ***
## ideology.z                   0.11536    0.03012   3.831 0.000131 ***
## age.z                        0.08994    0.03085   2.915 0.003586 ** 
## education.z                 -0.01642    0.03223  -0.510 0.610442    
## white_.5                     0.13952    0.06496   2.148 0.031825 *  
## index_ANexp_w1.z            -0.05699    0.04088  -1.394 0.163390    
## vaxxIntentions_w1.z          1.52534    0.03100  49.211  < 2e-16 ***
## index_ANexp_w2.z:ideology.z -0.01819    0.02679  -0.679 0.497250    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 1.422 on 2439 degrees of freedom
##   (1412 observations deleted due to missingness)
## Multiple R-squared:  0.5412, Adjusted R-squared:  0.5397 
## F-statistic: 359.7 on 8 and 2439 DF,  p-value: < 2.2e-16

tab_model(mb,
          show.ci = .95,
          show.stat = T,
          string.stat = "t",
          string.se="SE",
          string.est = "Est",
          digits = 3)

	vaxxIntentions_w2
Predictors	Est	CI	t	p
(Intercept)	0.273	0.210 – 0.335	8.556	<0.001
index_ANexp_w2.z	0.152	0.072 – 0.233	3.697	<0.001
ideology.z	0.115	0.056 – 0.174	3.831	<0.001
age.z	0.090	0.029 – 0.150	2.915	0.004
education.z	-0.016	-0.080 – 0.047	-0.510	0.610
white_.5	0.140	0.012 – 0.267	2.148	0.032
index_ANexp_w1.z	-0.057	-0.137 – 0.023	-1.394	0.163
vaxxIntentions_w1.z	1.525	1.465 – 1.586	49.211	<0.001
index_ANexp_w2.z * ideology.z	-0.018	-0.071 – 0.034	-0.679	0.497
Observations	2448
R² / R² adjusted	0.541 / 0.540

e. Longitudinal: trustExpert.w2 ~ index.w2.z * ideology.z + age.z + white_.5 + edu.z + index.w1.z + trustExpert.w1.z (causal)

summary(m.s2.b <- lm(trustExpert_w2 ~ index_ANexp_w2.z * ideology.z + age.z + education.z + white_.5 + index_ANexp_w1.z + trustExpert_w1.z, data = d))

## 
## Call:
## lm(formula = trustExpert_w2 ~ index_ANexp_w2.z * ideology.z + 
##     age.z + education.z + white_.5 + index_ANexp_w1.z + trustExpert_w1.z, 
##     data = d)
## 
## Residuals:
##     Min      1Q  Median      3Q     Max 
## -5.9300 -0.4943  0.1684  0.6092  3.9935 
## 
## Coefficients:
##                             Estimate Std. Error t value Pr(>|t|)    
## (Intercept)                  1.48200    0.02430  60.984  < 2e-16 ***
## index_ANexp_w2.z             0.11050    0.03139   3.520 0.000439 ***
## ideology.z                  -0.19312    0.02325  -8.305  < 2e-16 ***
## age.z                        0.08264    0.02339   3.533 0.000419 ***
## education.z                  0.01120    0.02459   0.455 0.648901    
## white_.5                     0.10984    0.04955   2.217 0.026733 *  
## index_ANexp_w1.z             0.05347    0.03117   1.715 0.086389 .  
## trustExpert_w1.z             0.89656    0.02483  36.111  < 2e-16 ***
## index_ANexp_w2.z:ideology.z  0.05412    0.02044   2.648 0.008157 ** 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 1.084 on 2440 degrees of freedom
##   (1411 observations deleted due to missingness)
## Multiple R-squared:  0.4605, Adjusted R-squared:  0.4587 
## F-statistic: 260.3 on 8 and 2440 DF,  p-value: < 2.2e-16

tab_model(m.s2.b,
          show.df = T,
          show.ci = .95,
          show.se = T,
          show.stat = T,
          string.stat = "t",
          string.se="SE",
          string.est = "Est",
          digits = 3)

	trustExpert_w2
Predictors	Est	SE	CI	t	p	df
(Intercept)	1.482	0.024	1.434 – 1.530	60.984	<0.001	2440.000
index_ANexp_w2.z	0.111	0.031	0.049 – 0.172	3.520	<0.001	2440.000
ideology.z	-0.193	0.023	-0.239 – -0.148	-8.305	<0.001	2440.000
age.z	0.083	0.023	0.037 – 0.129	3.533	<0.001	2440.000
education.z	0.011	0.025	-0.037 – 0.059	0.455	0.649	2440.000
white_.5	0.110	0.050	0.013 – 0.207	2.217	0.027	2440.000
index_ANexp_w1.z	0.053	0.031	-0.008 – 0.115	1.715	0.086	2440.000
trustExpert_w1.z	0.897	0.025	0.848 – 0.945	36.111	<0.001	2440.000
index_ANexp_w2.z * ideology.z	0.054	0.020	0.014 – 0.094	2.648	0.008	2440.000
Observations	2449
R² / R² adjusted	0.460 / 0.459

Study 3: US vaxx Behavior (March 2022)

a. vaxxBehavior ~ index.w3.z * ideology.z + edu.z + age.z + white_.5 (replication)

summary(m.s3.a <- lm(vaxxBehavior ~ index_ANexp_w3.z * ideology.z + age.z + education.z + white_.5, data = d))

## 
## Call:
## lm(formula = vaxxBehavior ~ index_ANexp_w3.z * ideology.z + age.z + 
##     education.z + white_.5, data = d)
## 
## Residuals:
##     Min      1Q  Median      3Q     Max 
## -3.1044 -0.3664  0.3046  0.7190  2.4337 
## 
## Coefficients:
##                             Estimate Std. Error t value Pr(>|t|)    
## (Intercept)                  3.20751    0.03947  81.268  < 2e-16 ***
## index_ANexp_w3.z             0.28388    0.03601   7.884 8.33e-15 ***
## ideology.z                  -0.23941    0.03328  -7.194 1.24e-12 ***
## age.z                        0.25291    0.03667   6.897 9.44e-12 ***
## education.z                  0.14832    0.03723   3.983 7.29e-05 ***
## white_.5                    -0.04062    0.07848  -0.518  0.60485    
## index_ANexp_w3.z:ideology.z  0.09972    0.03139   3.176  0.00154 ** 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 1.065 on 990 degrees of freedom
##   (2863 observations deleted due to missingness)
## Multiple R-squared:  0.1755, Adjusted R-squared:  0.1705 
## F-statistic: 35.13 on 6 and 990 DF,  p-value: < 2.2e-16

tab_model(m.s3.a, 
          show.df = T, 
          show.ci = .95,
          show.se = T,
          show.stat = T,
          string.stat = "t",
          string.se="SE",
          string.est = "Est",
          digits = 3)

	vaxxBehavior
Predictors	Est	SE	CI	t	p	df
(Intercept)	3.208	0.039	3.130 – 3.285	81.268	<0.001	990.000
index_ANexp_w3.z	0.284	0.036	0.213 – 0.355	7.884	<0.001	990.000
ideology.z	-0.239	0.033	-0.305 – -0.174	-7.194	<0.001	990.000
age.z	0.253	0.037	0.181 – 0.325	6.897	<0.001	990.000
education.z	0.148	0.037	0.075 – 0.221	3.983	<0.001	990.000
white_.5	-0.041	0.078	-0.195 – 0.113	-0.518	0.605	990.000
index_ANexp_w3.z * ideology.z	0.100	0.031	0.038 – 0.161	3.176	0.002	990.000
Observations	997
R² / R² adjusted	0.176 / 0.171

b. Longitudinal: vaxxBehavior ~ index.w3.z * ideology.z + edu.z + age.z + white_.5 + index.avg.w1w2.z + vaxxIntentions.avg.w1w2.z (causal)

summary(m.s3.b <- lm(vaxxBehavior ~ index_ANexp_w3.z * ideology.z + age.z + education.z + white_.5 +  avgANexp_w1w2.z + avgVaxxIntentions.z, data = d))

## 
## Call:
## lm(formula = vaxxBehavior ~ index_ANexp_w3.z * ideology.z + age.z + 
##     education.z + white_.5 + avgANexp_w1w2.z + avgVaxxIntentions.z, 
##     data = d)
## 
## Residuals:
##     Min      1Q  Median      3Q     Max 
## -3.5013 -0.4539  0.1884  0.6307  2.1048 
## 
## Coefficients:
##                             Estimate Std. Error t value Pr(>|t|)    
## (Intercept)                  3.08361    0.03806  81.012  < 2e-16 ***
## index_ANexp_w3.z             0.19972    0.04692   4.256 2.29e-05 ***
## ideology.z                  -0.17276    0.03143  -5.497 5.02e-08 ***
## age.z                        0.13209    0.03508   3.765 0.000177 ***
## education.z                  0.11695    0.03536   3.307 0.000979 ***
## white_.5                    -0.09751    0.07397  -1.318 0.187752    
## avgANexp_w1w2.z             -0.01496    0.04737  -0.316 0.752268    
## avgVaxxIntentions.z          0.25332    0.01695  14.950  < 2e-16 ***
## index_ANexp_w3.z:ideology.z  0.05306    0.02952   1.797 0.072638 .  
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 0.9492 on 907 degrees of freedom
##   (2944 observations deleted due to missingness)
## Multiple R-squared:  0.337,  Adjusted R-squared:  0.3312 
## F-statistic: 57.64 on 8 and 907 DF,  p-value: < 2.2e-16

tab_model(m.s3.b,
          show.df = T, 
          show.ci = .95,
          show.se = T,
          show.stat = T,
          string.stat = "t",
          string.se="SE",
          string.est = "Est",
          digits = 3)

	vaxxBehavior
Predictors	Est	SE	CI	t	p	df
(Intercept)	3.084	0.038	3.009 – 3.158	81.012	<0.001	907.000
index_ANexp_w3.z	0.200	0.047	0.108 – 0.292	4.256	<0.001	907.000
ideology.z	-0.173	0.031	-0.234 – -0.111	-5.497	<0.001	907.000
age.z	0.132	0.035	0.063 – 0.201	3.765	<0.001	907.000
education.z	0.117	0.035	0.048 – 0.186	3.307	0.001	907.000
white_.5	-0.098	0.074	-0.243 – 0.048	-1.318	0.188	907.000
avgANexp_w1w2.z	-0.015	0.047	-0.108 – 0.078	-0.316	0.752	907.000
avgVaxxIntentions.z	0.253	0.017	0.220 – 0.287	14.950	<0.001	907.000
index_ANexp_w3.z * ideology.z	0.053	0.030	-0.005 – 0.111	1.797	0.073	907.000
Observations	916
R² / R² adjusted	0.337 / 0.331

c. mediation path: index.w3 –> trust in science –> vaxx Behavior

sobel test: https://quantpsy.org/sobel/sobel.htm

i. controlling for ideology

vaxxBehavior ~ index.w3.z + ideology.z + age.z + white_.5 + education.z

trustSci ~ index.w3.z + ideology.z + age.z + white_.5 + education.z

vaxxBehavior ~ trustSci.z + index.w3.z + ideology.z + age.z + white_.5 + education.z

summary(m4.xy <- lm(vaxxBehavior ~ index_ANexp_w3.z + ideology.z + age.z + white_.5 + education.z, data = d))

## 
## Call:
## lm(formula = vaxxBehavior ~ index_ANexp_w3.z + ideology.z + age.z + 
##     white_.5 + education.z, data = d)
## 
## Residuals:
##     Min      1Q  Median      3Q     Max 
## -3.3665 -0.3807  0.3173  0.7568  2.3831 
## 
## Coefficients:
##                  Estimate Std. Error t value Pr(>|t|)    
## (Intercept)       3.18244    0.03885  81.920  < 2e-16 ***
## index_ANexp_w3.z  0.27464    0.03605   7.618 6.02e-14 ***
## ideology.z       -0.24213    0.03342  -7.246 8.65e-13 ***
## age.z             0.24371    0.03672   6.637 5.26e-11 ***
## white_.5         -0.04550    0.07883  -0.577    0.564    
## education.z       0.15482    0.03735   4.145 3.68e-05 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 1.07 on 991 degrees of freedom
##   (2863 observations deleted due to missingness)
## Multiple R-squared:  0.1671, Adjusted R-squared:  0.1629 
## F-statistic: 39.77 on 5 and 991 DF,  p-value: < 2.2e-16

summary(m4.xm <- lm(trustSci ~ index_ANexp_w3.z + ideology.z + age.z + white_.5 + education.z, data = d))

## 
## Call:
## lm(formula = trustSci ~ index_ANexp_w3.z + ideology.z + age.z + 
##     white_.5 + education.z, data = d)
## 
## Residuals:
##      Min       1Q   Median       3Q      Max 
## -2.37123 -0.45216  0.00732  0.51357  1.78295 
## 
## Coefficients:
##                  Estimate Std. Error t value Pr(>|t|)    
## (Intercept)       3.40205    0.02616 130.047  < 2e-16 ***
## index_ANexp_w3.z  0.05183    0.02427   2.135 0.032971 *  
## ideology.z       -0.36310    0.02244 -16.181  < 2e-16 ***
## age.z             0.10092    0.02477   4.075 4.98e-05 ***
## white_.5          0.23917    0.05308   4.506 7.41e-06 ***
## education.z       0.08401    0.02505   3.353 0.000828 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 0.7177 on 987 degrees of freedom
##   (2867 observations deleted due to missingness)
## Multiple R-squared:  0.2461, Adjusted R-squared:  0.2423 
## F-statistic: 64.43 on 5 and 987 DF,  p-value: < 2.2e-16

summary(m4.xmy <- lm(vaxxBehavior ~ trustSci.z + index_ANexp_w3.z + ideology.z +  age.z + white_.5 + education.z, data = d))

## 
## Call:
## lm(formula = vaxxBehavior ~ trustSci.z + index_ANexp_w3.z + ideology.z + 
##     age.z + white_.5 + education.z, data = d)
## 
## Residuals:
##     Min      1Q  Median      3Q     Max 
## -3.6413 -0.4671  0.2718  0.6970  2.1198 
## 
## Coefficients:
##                  Estimate Std. Error t value Pr(>|t|)    
## (Intercept)       3.20689    0.03736  85.836  < 2e-16 ***
## trustSci.z        0.36369    0.03725   9.764  < 2e-16 ***
## index_ANexp_w3.z  0.24583    0.03482   7.060 3.13e-12 ***
## ideology.z       -0.08459    0.03603  -2.348 0.019092 *  
## age.z             0.20110    0.03558   5.651 2.08e-08 ***
## white_.5         -0.15989    0.07651  -2.090 0.036901 *  
## education.z       0.11879    0.03590   3.309 0.000969 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 1.022 on 984 degrees of freedom
##   (2869 observations deleted due to missingness)
## Multiple R-squared:  0.2414, Adjusted R-squared:  0.2368 
## F-statistic: 52.19 on 6 and 984 DF,  p-value: < 2.2e-16

tab_model(m4.xy, m4.xm, m4.xmy,
          show.df = T, 
          show.ci = .95,
          show.se = T,
          show.stat = T,
          string.stat = "t",
          string.se="SE",
          string.est = "Est",
          digits = 3)

	vaxxBehavior						trustSci						vaxxBehavior
Predictors	Est	SE	CI	t	p	df	Est	SE	CI	t	p	df	Est	SE	CI	t	p	df
(Intercept)	3.182	0.039	3.106 – 3.259	81.920	<0.001	991.000	3.402	0.026	3.351 – 3.453	130.047	<0.001	987.000	3.207	0.037	3.134 – 3.280	85.836	<0.001	984.000
index_ANexp_w3.z	0.275	0.036	0.204 – 0.345	7.618	<0.001	991.000	0.052	0.024	0.004 – 0.099	2.135	0.033	987.000	0.246	0.035	0.178 – 0.314	7.060	<0.001	984.000
ideology.z	-0.242	0.033	-0.308 – -0.177	-7.246	<0.001	991.000	-0.363	0.022	-0.407 – -0.319	-16.181	<0.001	987.000	-0.085	0.036	-0.155 – -0.014	-2.348	0.019	984.000
age.z	0.244	0.037	0.172 – 0.316	6.637	<0.001	991.000	0.101	0.025	0.052 – 0.150	4.075	<0.001	987.000	0.201	0.036	0.131 – 0.271	5.651	<0.001	984.000
white_.5	-0.045	0.079	-0.200 – 0.109	-0.577	0.564	991.000	0.239	0.053	0.135 – 0.343	4.506	<0.001	987.000	-0.160	0.077	-0.310 – -0.010	-2.090	0.037	984.000
education.z	0.155	0.037	0.082 – 0.228	4.145	<0.001	991.000	0.084	0.025	0.035 – 0.133	3.353	0.001	987.000	0.119	0.036	0.048 – 0.189	3.309	0.001	984.000
trustSci.z													0.364	0.037	0.291 – 0.437	9.764	<0.001	984.000
Observations	997						993						991
R² / R² adjusted	0.167 / 0.163						0.246 / 0.242						0.241 / 0.237

ii. interaction with ideology

vaxxBehavior ~ index.w3.z * ideology.z + age.z + white_.5 + education.z

trustSci ~ index.w3.z * ideology.z + age.z + white_.5 + education.z

vaxxBehavior ~ (trustSci.z + index.w3.z) * ideology.z + age.z + white_.5 + education.z

summary(m4.xy <- lm(vaxxBehavior ~ index_ANexp_w3.z * ideology.z + age.z + white_.5 + education.z, data = d))

## 
## Call:
## lm(formula = vaxxBehavior ~ index_ANexp_w3.z * ideology.z + age.z + 
##     white_.5 + education.z, data = d)
## 
## Residuals:
##     Min      1Q  Median      3Q     Max 
## -3.1044 -0.3664  0.3046  0.7190  2.4337 
## 
## Coefficients:
##                             Estimate Std. Error t value Pr(>|t|)    
## (Intercept)                  3.20751    0.03947  81.268  < 2e-16 ***
## index_ANexp_w3.z             0.28388    0.03601   7.884 8.33e-15 ***
## ideology.z                  -0.23941    0.03328  -7.194 1.24e-12 ***
## age.z                        0.25291    0.03667   6.897 9.44e-12 ***
## white_.5                    -0.04062    0.07848  -0.518  0.60485    
## education.z                  0.14832    0.03723   3.983 7.29e-05 ***
## index_ANexp_w3.z:ideology.z  0.09972    0.03139   3.176  0.00154 ** 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 1.065 on 990 degrees of freedom
##   (2863 observations deleted due to missingness)
## Multiple R-squared:  0.1755, Adjusted R-squared:  0.1705 
## F-statistic: 35.13 on 6 and 990 DF,  p-value: < 2.2e-16

summary(m4.xm <- lm(trustSci ~ index_ANexp_w3.z * ideology.z + age.z + white_.5 + education.z, data = d))

## 
## Call:
## lm(formula = trustSci ~ index_ANexp_w3.z * ideology.z + age.z + 
##     white_.5 + education.z, data = d)
## 
## Residuals:
##      Min       1Q   Median       3Q      Max 
## -2.37912 -0.47334  0.00095  0.51059  1.79660 
## 
## Coefficients:
##                             Estimate Std. Error t value Pr(>|t|)    
## (Intercept)                  3.41722    0.02659 128.500  < 2e-16 ***
## index_ANexp_w3.z             0.05824    0.02429   2.398  0.01666 *  
## ideology.z                  -0.36155    0.02236 -16.166  < 2e-16 ***
## age.z                        0.10713    0.02477   4.324 1.68e-05 ***
## white_.5                     0.24315    0.05290   4.596 4.87e-06 ***
## education.z                  0.07943    0.02501   3.176  0.00154 ** 
## index_ANexp_w3.z:ideology.z  0.06015    0.02092   2.875  0.00413 ** 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 0.7151 on 986 degrees of freedom
##   (2867 observations deleted due to missingness)
## Multiple R-squared:  0.2523, Adjusted R-squared:  0.2478 
## F-statistic: 55.47 on 6 and 986 DF,  p-value: < 2.2e-16

summary(m4.xmy <- lm(vaxxBehavior ~ (trustSci.z + index_ANexp_w3.z) * ideology.z +  age.z + white_.5 + education.z, data = d))

## 
## Call:
## lm(formula = vaxxBehavior ~ (trustSci.z + index_ANexp_w3.z) * 
##     ideology.z + age.z + white_.5 + education.z, data = d)
## 
## Residuals:
##     Min      1Q  Median      3Q     Max 
## -3.3584 -0.4085  0.2579  0.6639  2.2270 
## 
## Coefficients:
##                             Estimate Std. Error t value Pr(>|t|)    
## (Intercept)                  3.25066    0.03990  81.472  < 2e-16 ***
## trustSci.z                   0.35106    0.03729   9.413  < 2e-16 ***
## index_ANexp_w3.z             0.24866    0.03487   7.132 1.92e-12 ***
## ideology.z                  -0.09022    0.03596  -2.509  0.01226 *  
## age.z                        0.21282    0.03563   5.974 3.24e-09 ***
## white_.5                    -0.14752    0.07629  -1.934  0.05343 .  
## education.z                  0.11733    0.03582   3.276  0.00109 ** 
## trustSci.z:ideology.z        0.06048    0.02973   2.034  0.04219 *  
## index_ANexp_w3.z:ideology.z  0.06935    0.03044   2.278  0.02294 *  
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 1.018 on 982 degrees of freedom
##   (2869 observations deleted due to missingness)
## Multiple R-squared:  0.2494, Adjusted R-squared:  0.2433 
## F-statistic: 40.79 on 8 and 982 DF,  p-value: < 2.2e-16

tab_model(m4.xy, m4.xm, m4.xmy,
          show.df = T, 
          show.ci = .95,
          show.se = T,
          show.stat = T,
          string.stat = "t",
          string.se="SE",
          string.est = "Est",
          digits = 3)

	vaxxBehavior						trustSci						vaxxBehavior
Predictors	Est	SE	CI	t	p	df	Est	SE	CI	t	p	df	Est	SE	CI	t	p	df
(Intercept)	3.208	0.039	3.130 – 3.285	81.268	<0.001	990.000	3.417	0.027	3.365 – 3.469	128.500	<0.001	986.000	3.251	0.040	3.172 – 3.329	81.472	<0.001	982.000
index_ANexp_w3.z	0.284	0.036	0.213 – 0.355	7.884	<0.001	990.000	0.058	0.024	0.011 – 0.106	2.398	0.017	986.000	0.249	0.035	0.180 – 0.317	7.132	<0.001	982.000
ideology.z	-0.239	0.033	-0.305 – -0.174	-7.194	<0.001	990.000	-0.362	0.022	-0.405 – -0.318	-16.166	<0.001	986.000	-0.090	0.036	-0.161 – -0.020	-2.509	0.012	982.000
age.z	0.253	0.037	0.181 – 0.325	6.897	<0.001	990.000	0.107	0.025	0.059 – 0.156	4.324	<0.001	986.000	0.213	0.036	0.143 – 0.283	5.974	<0.001	982.000
white_.5	-0.041	0.078	-0.195 – 0.113	-0.518	0.605	990.000	0.243	0.053	0.139 – 0.347	4.596	<0.001	986.000	-0.148	0.076	-0.297 – 0.002	-1.934	0.053	982.000
education.z	0.148	0.037	0.075 – 0.221	3.983	<0.001	990.000	0.079	0.025	0.030 – 0.129	3.176	0.002	986.000	0.117	0.036	0.047 – 0.188	3.276	0.001	982.000
index_ANexp_w3.z * ideology.z	0.100	0.031	0.038 – 0.161	3.176	0.002	990.000	0.060	0.021	0.019 – 0.101	2.875	0.004	986.000	0.069	0.030	0.010 – 0.129	2.278	0.023	982.000
trustSci.z													0.351	0.037	0.278 – 0.424	9.413	<0.001	982.000
trustSci.z * ideology.z													0.060	0.030	0.002 – 0.119	2.034	0.042	982.000
Observations	997						993						991
R² / R² adjusted	0.176 / 0.171						0.252 / 0.248						0.249 / 0.243

iii. interaction with all demographics

vaxxBehavior ~ index.w3.z * (ideology.z + age.z + white_.5 + education.z)

trustSci ~ index.w3.z * (ideology.z + age.z + white_.5 + education.z)

vaxxBehavior ~ (trustSci.z + index.w3.z) * (ideology.z + age.z + white_.5 + education.z)

summary(m4.xy <- lm(vaxxBehavior ~ index_ANexp_w3.z * (ideology.z + age.z + white_.5 + education.z), data = d))

## 
## Call:
## lm(formula = vaxxBehavior ~ index_ANexp_w3.z * (ideology.z + 
##     age.z + white_.5 + education.z), data = d)
## 
## Residuals:
##     Min      1Q  Median      3Q     Max 
## -3.0256 -0.3790  0.3015  0.7011  2.6801 
## 
## Coefficients:
##                              Estimate Std. Error t value Pr(>|t|)    
## (Intercept)                   3.19525    0.04100  77.942  < 2e-16 ***
## index_ANexp_w3.z              0.29030    0.03674   7.901 7.35e-15 ***
## ideology.z                   -0.24195    0.03368  -7.184 1.33e-12 ***
## age.z                         0.25623    0.03674   6.975 5.59e-12 ***
## white_.5                     -0.02206    0.07976  -0.277 0.782216    
## education.z                   0.17429    0.03905   4.464 8.99e-06 ***
## index_ANexp_w3.z:ideology.z   0.10814    0.03190   3.390 0.000727 ***
## index_ANexp_w3.z:age.z       -0.03086    0.03697  -0.835 0.404019    
## index_ANexp_w3.z:white_.5    -0.02191    0.07405  -0.296 0.767372    
## index_ANexp_w3.z:education.z -0.06362    0.03046  -2.088 0.037017 *  
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 1.064 on 987 degrees of freedom
##   (2863 observations deleted due to missingness)
## Multiple R-squared:   0.18,  Adjusted R-squared:  0.1725 
## F-statistic: 24.07 on 9 and 987 DF,  p-value: < 2.2e-16

summary(m4.xm <- lm(trustSci ~ index_ANexp_w3.z * (ideology.z + age.z + white_.5 + education.z), data = d))

## 
## Call:
## lm(formula = trustSci ~ index_ANexp_w3.z * (ideology.z + age.z + 
##     white_.5 + education.z), data = d)
## 
## Residuals:
##      Min       1Q   Median       3Q      Max 
## -2.27291 -0.45392  0.01196  0.50327  1.75854 
## 
## Coefficients:
##                              Estimate Std. Error t value Pr(>|t|)    
## (Intercept)                   3.43187    0.02760 124.350  < 2e-16 ***
## index_ANexp_w3.z              0.05198    0.02468   2.106  0.03546 *  
## ideology.z                   -0.34743    0.02258 -15.385  < 2e-16 ***
## age.z                         0.10575    0.02475   4.274 2.11e-05 ***
## white_.5                      0.22241    0.05360   4.149 3.62e-05 ***
## education.z                   0.08016    0.02611   3.070  0.00220 ** 
## index_ANexp_w3.z:ideology.z   0.05047    0.02112   2.390  0.01705 *  
## index_ANexp_w3.z:age.z        0.07852    0.02501   3.139  0.00174 ** 
## index_ANexp_w3.z:white_.5     0.05573    0.04980   1.119  0.26334    
## index_ANexp_w3.z:education.z -0.01544    0.02035  -0.759  0.44816    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 0.7113 on 983 degrees of freedom
##   (2867 observations deleted due to missingness)
## Multiple R-squared:  0.2625, Adjusted R-squared:  0.2557 
## F-statistic: 38.87 on 9 and 983 DF,  p-value: < 2.2e-16

summary(m4.xmy <- lm(vaxxBehavior ~ (trustSci.z + index_ANexp_w3.z) * (ideology.z +  age.z + white_.5 + education.z), data = d))

## 
## Call:
## lm(formula = vaxxBehavior ~ (trustSci.z + index_ANexp_w3.z) * 
##     (ideology.z + age.z + white_.5 + education.z), data = d)
## 
## Residuals:
##     Min      1Q  Median      3Q     Max 
## -3.1670 -0.3951  0.2540  0.6355  2.4172 
## 
## Coefficients:
##                              Estimate Std. Error t value Pr(>|t|)    
## (Intercept)                   3.21304    0.04205  76.417  < 2e-16 ***
## trustSci.z                    0.31276    0.04254   7.352 4.14e-13 ***
## index_ANexp_w3.z              0.25661    0.03570   7.187 1.31e-12 ***
## ideology.z                   -0.09222    0.03599  -2.562  0.01055 *  
## age.z                         0.22432    0.03578   6.269 5.43e-10 ***
## white_.5                     -0.12841    0.07738  -1.659  0.09734 .  
## education.z                   0.14379    0.03795   3.789  0.00016 ***
## trustSci.z:ideology.z         0.05128    0.03090   1.659  0.09738 .  
## trustSci.z:age.z             -0.02619    0.03589  -0.730  0.46571    
## trustSci.z:white_.5           0.19147    0.07865   2.435  0.01509 *  
## trustSci.z:education.z        0.05201    0.03688   1.410  0.15883    
## index_ANexp_w3.z:ideology.z   0.07267    0.03102   2.343  0.01935 *  
## index_ANexp_w3.z:age.z       -0.04254    0.03691  -1.153  0.24933    
## index_ANexp_w3.z:white_.5    -0.09030    0.07239  -1.247  0.21255    
## index_ANexp_w3.z:education.z -0.05653    0.02907  -1.945  0.05212 .  
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 1.013 on 976 degrees of freedom
##   (2869 observations deleted due to missingness)
## Multiple R-squared:  0.2612, Adjusted R-squared:  0.2506 
## F-statistic: 24.65 on 14 and 976 DF,  p-value: < 2.2e-16

tab_model(m4.xy, m4.xm, m4.xmy,
          show.df = T, 
          show.ci = .95,
          show.se = T,
          show.stat = T,
          string.stat = "t",
          string.se="SE",
          string.est = "Est",
          digits = 3)

	vaxxBehavior						trustSci						vaxxBehavior
Predictors	Est	SE	CI	t	p	df	Est	SE	CI	t	p	df	Est	SE	CI	t	p	df
(Intercept)	3.195	0.041	3.115 – 3.276	77.942	<0.001	987.000	3.432	0.028	3.378 – 3.486	124.350	<0.001	983.000	3.213	0.042	3.131 – 3.296	76.417	<0.001	976.000
index_ANexp_w3.z	0.290	0.037	0.218 – 0.362	7.901	<0.001	987.000	0.052	0.025	0.004 – 0.100	2.106	0.035	983.000	0.257	0.036	0.187 – 0.327	7.187	<0.001	976.000
ideology.z	-0.242	0.034	-0.308 – -0.176	-7.184	<0.001	987.000	-0.347	0.023	-0.392 – -0.303	-15.385	<0.001	983.000	-0.092	0.036	-0.163 – -0.022	-2.562	0.011	976.000
age.z	0.256	0.037	0.184 – 0.328	6.975	<0.001	987.000	0.106	0.025	0.057 – 0.154	4.274	<0.001	983.000	0.224	0.036	0.154 – 0.295	6.269	<0.001	976.000
white_.5	-0.022	0.080	-0.179 – 0.134	-0.277	0.782	987.000	0.222	0.054	0.117 – 0.328	4.149	<0.001	983.000	-0.128	0.077	-0.280 – 0.023	-1.659	0.097	976.000
education.z	0.174	0.039	0.098 – 0.251	4.464	<0.001	987.000	0.080	0.026	0.029 – 0.131	3.070	0.002	983.000	0.144	0.038	0.069 – 0.218	3.789	<0.001	976.000
index_ANexp_w3.z * ideology.z	0.108	0.032	0.046 – 0.171	3.390	0.001	987.000	0.050	0.021	0.009 – 0.092	2.390	0.017	983.000	0.073	0.031	0.012 – 0.134	2.343	0.019	976.000
index_ANexp_w3.z * age.z	-0.031	0.037	-0.103 – 0.042	-0.835	0.404	987.000	0.079	0.025	0.029 – 0.128	3.139	0.002	983.000	-0.043	0.037	-0.115 – 0.030	-1.153	0.249	976.000
index_ANexp_w3.z * white_.5	-0.022	0.074	-0.167 – 0.123	-0.296	0.767	987.000	0.056	0.050	-0.042 – 0.153	1.119	0.263	983.000	-0.090	0.072	-0.232 – 0.052	-1.247	0.213	976.000
index_ANexp_w3.z * education.z	-0.064	0.030	-0.123 – -0.004	-2.088	0.037	987.000	-0.015	0.020	-0.055 – 0.024	-0.759	0.448	983.000	-0.057	0.029	-0.114 – 0.001	-1.945	0.052	976.000
trustSci.z													0.313	0.043	0.229 – 0.396	7.352	<0.001	976.000
trustSci.z * ideology.z													0.051	0.031	-0.009 – 0.112	1.659	0.097	976.000
trustSci.z * age.z													-0.026	0.036	-0.097 – 0.044	-0.730	0.466	976.000
trustSci.z * white_.5													0.191	0.079	0.037 – 0.346	2.435	0.015	976.000
trustSci.z * education.z													0.052	0.037	-0.020 – 0.124	1.410	0.159	976.000
Observations	997						993						991
R² / R² adjusted	0.180 / 0.172						0.262 / 0.256						0.261 / 0.251

d. robustness checks

i. does analytical media w3 predict trust in science wave 3 controlling for consumption wave 1 + 2?

summary(m3.xm <- lm(trustSci ~ index_ANexp_w3.z * ideology.z + age.z + white_.5 + education.z + avgANexp_w1w2.z, data = d))

## 
## Call:
## lm(formula = trustSci ~ index_ANexp_w3.z * ideology.z + age.z + 
##     white_.5 + education.z + avgANexp_w1w2.z, data = d)
## 
## Residuals:
##     Min      1Q  Median      3Q     Max 
## -2.3380 -0.4842  0.0160  0.5048  1.8045 
## 
## Coefficients:
##                             Estimate Std. Error t value Pr(>|t|)    
## (Intercept)                  3.41108    0.02809 121.414  < 2e-16 ***
## index_ANexp_w3.z             0.01538    0.03548   0.434 0.664702    
## ideology.z                  -0.35266    0.02363 -14.923  < 2e-16 ***
## age.z                        0.10459    0.02613   4.002 6.79e-05 ***
## white_.5                     0.25488    0.05594   4.556 5.93e-06 ***
## education.z                  0.08956    0.02665   3.360 0.000812 ***
## avgANexp_w1w2.z              0.07031    0.03562   1.974 0.048678 *  
## index_ANexp_w3.z:ideology.z  0.04431    0.02236   1.982 0.047818 *  
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 0.7166 on 904 degrees of freedom
##   (2948 observations deleted due to missingness)
## Multiple R-squared:  0.2538, Adjusted R-squared:  0.248 
## F-statistic: 43.92 on 7 and 904 DF,  p-value: < 2.2e-16

ii. science literacy

model 1: vaxxBehavior ~ index_ANexp_w3.z + sciLit.z + ideology.z + age.z + white_.5 + education.z

model 2: vaxxBehavior ~ index_ANexp_w3.z * sciLit.z + ideology.z + age.z + white_.5 + education.z

model 3: vaxxBehavior ~ index_ANexp_w3.z * (sciLit.z + ideology.z) + age.z + white_.5 + education.z

summary(m1 <- lm(vaxxBehavior ~ index_ANexp_w3.z + sciLit.z + ideology.z + age.z + white_.5 + education.z, data = d))

## 
## Call:
## lm(formula = vaxxBehavior ~ index_ANexp_w3.z + sciLit.z + ideology.z + 
##     age.z + white_.5 + education.z, data = d)
## 
## Residuals:
##     Min      1Q  Median      3Q     Max 
## -3.3311 -0.3940  0.3155  0.7533  2.3642 
## 
## Coefficients:
##                  Estimate Std. Error t value Pr(>|t|)    
## (Intercept)       3.18017    0.03907  81.407  < 2e-16 ***
## index_ANexp_w3.z  0.27587    0.03620   7.620 5.95e-14 ***
## sciLit.z          0.02980    0.03513   0.848    0.396    
## ideology.z       -0.24696    0.03350  -7.372 3.55e-13 ***
## age.z             0.24849    0.03680   6.752 2.49e-11 ***
## white_.5         -0.05272    0.07993  -0.660    0.510    
## education.z       0.15042    0.03780   3.979 7.42e-05 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 1.07 on 987 degrees of freedom
##   (2866 observations deleted due to missingness)
## Multiple R-squared:  0.1699, Adjusted R-squared:  0.1648 
## F-statistic: 33.67 on 6 and 987 DF,  p-value: < 2.2e-16

summary(m2 <- lm(vaxxBehavior ~ index_ANexp_w3.z * sciLit.z + ideology.z + age.z + white_.5 + education.z, data = d))

## 
## Call:
## lm(formula = vaxxBehavior ~ index_ANexp_w3.z * sciLit.z + ideology.z + 
##     age.z + white_.5 + education.z, data = d)
## 
## Residuals:
##     Min      1Q  Median      3Q     Max 
## -3.4071 -0.3853  0.3237  0.7532  2.3603 
## 
## Coefficients:
##                           Estimate Std. Error t value Pr(>|t|)    
## (Intercept)                3.17848    0.03913  81.229  < 2e-16 ***
## index_ANexp_w3.z           0.27253    0.03645   7.476 1.68e-13 ***
## sciLit.z                   0.03497    0.03572   0.979    0.328    
## ideology.z                -0.24791    0.03353  -7.395 3.03e-13 ***
## age.z                      0.24914    0.03682   6.767 2.26e-11 ***
## white_.5                  -0.05670    0.08010  -0.708    0.479    
## education.z                0.15206    0.03787   4.016 6.37e-05 ***
## index_ANexp_w3.z:sciLit.z -0.02378    0.02971  -0.800    0.424    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 1.07 on 986 degrees of freedom
##   (2866 observations deleted due to missingness)
## Multiple R-squared:  0.1704, Adjusted R-squared:  0.1645 
## F-statistic: 28.94 on 7 and 986 DF,  p-value: < 2.2e-16

summary(m3 <- lm(vaxxBehavior ~ index_ANexp_w3.z * (sciLit.z + ideology.z) + age.z + white_.5 + education.z, data = d))

## 
## Call:
## lm(formula = vaxxBehavior ~ index_ANexp_w3.z * (sciLit.z + ideology.z) + 
##     age.z + white_.5 + education.z, data = d)
## 
## Residuals:
##     Min      1Q  Median      3Q     Max 
## -3.1559 -0.3888  0.3080  0.7145  2.4073 
## 
## Coefficients:
##                             Estimate Std. Error t value Pr(>|t|)    
## (Intercept)                  3.20367    0.03973  80.638  < 2e-16 ***
## index_ANexp_w3.z             0.28150    0.03639   7.736 2.53e-14 ***
## sciLit.z                     0.03768    0.03557   1.060 0.289631    
## ideology.z                  -0.24522    0.03338  -7.346 4.27e-13 ***
## age.z                        0.25847    0.03676   7.031 3.83e-12 ***
## white_.5                    -0.05265    0.07973  -0.660 0.509242    
## education.z                  0.14541    0.03774   3.853 0.000124 ***
## index_ANexp_w3.z:sciLit.z   -0.02886    0.02962  -0.974 0.330067    
## index_ANexp_w3.z:ideology.z  0.10100    0.03146   3.211 0.001368 ** 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 1.065 on 985 degrees of freedom
##   (2866 observations deleted due to missingness)
## Multiple R-squared:  0.179,  Adjusted R-squared:  0.1724 
## F-statistic: 26.85 on 8 and 985 DF,  p-value: < 2.2e-16

tab_model(m1, m2, m3,
          show.df = T, 
          show.ci = .95,
          show.se = T,
          show.stat = T,
          string.stat = "t",
          string.se="SE",
          string.est = "Est",
          digits = 3)

	vaxxBehavior						vaxxBehavior						vaxxBehavior
Predictors	Est	SE	CI	t	p	df	Est	SE	CI	t	p	df	Est	SE	CI	t	p	df
(Intercept)	3.180	0.039	3.104 – 3.257	81.407	<0.001	987.000	3.178	0.039	3.102 – 3.255	81.229	<0.001	986.000	3.204	0.040	3.126 – 3.282	80.638	<0.001	985.000
index_ANexp_w3.z	0.276	0.036	0.205 – 0.347	7.620	<0.001	987.000	0.273	0.036	0.201 – 0.344	7.476	<0.001	986.000	0.282	0.036	0.210 – 0.353	7.736	<0.001	985.000
sciLit.z	0.030	0.035	-0.039 – 0.099	0.848	0.396	987.000	0.035	0.036	-0.035 – 0.105	0.979	0.328	986.000	0.038	0.036	-0.032 – 0.107	1.060	0.290	985.000
ideology.z	-0.247	0.033	-0.313 – -0.181	-7.372	<0.001	987.000	-0.248	0.034	-0.314 – -0.182	-7.395	<0.001	986.000	-0.245	0.033	-0.311 – -0.180	-7.346	<0.001	985.000
age.z	0.248	0.037	0.176 – 0.321	6.752	<0.001	987.000	0.249	0.037	0.177 – 0.321	6.767	<0.001	986.000	0.258	0.037	0.186 – 0.331	7.031	<0.001	985.000
white_.5	-0.053	0.080	-0.210 – 0.104	-0.660	0.510	987.000	-0.057	0.080	-0.214 – 0.100	-0.708	0.479	986.000	-0.053	0.080	-0.209 – 0.104	-0.660	0.509	985.000
education.z	0.150	0.038	0.076 – 0.225	3.979	<0.001	987.000	0.152	0.038	0.078 – 0.226	4.016	<0.001	986.000	0.145	0.038	0.071 – 0.219	3.853	<0.001	985.000
index_ANexp_w3.z * sciLit.z							-0.024	0.030	-0.082 – 0.035	-0.800	0.424	986.000	-0.029	0.030	-0.087 – 0.029	-0.974	0.330	985.000
index_ANexp_w3.z * ideology.z													0.101	0.031	0.039 – 0.163	3.211	0.001	985.000
Observations	994						994						994
R² / R² adjusted	0.170 / 0.165						0.170 / 0.165						0.179 / 0.172

iii. CRT

model 1: vaxxBehavior ~ index.w3.z + CRT.z + ideology.z + age.z + white_.5 + education.z

model 2: vaxxBehavior ~ index.w3.z * CRT.z + ideology.z + age.z + white_.5 + education.z

model 3: vaxxBehavior ~ index.w3.z * (CRT.z + ideology.z) + age.z + white_.5 + education.z

summary(m1 <- lm(vaxxBehavior ~ index_ANexp_w3.z + CRT.z + ideology.z + age.z + white_.5 + education.z, data = d))

## 
## Call:
## lm(formula = vaxxBehavior ~ index_ANexp_w3.z + CRT.z + ideology.z + 
##     age.z + white_.5 + education.z, data = d)
## 
## Residuals:
##     Min      1Q  Median      3Q     Max 
## -3.2731 -0.3902  0.3159  0.7503  2.3534 
## 
## Coefficients:
##                  Estimate Std. Error t value Pr(>|t|)    
## (Intercept)       3.18589    0.03945  80.749  < 2e-16 ***
## index_ANexp_w3.z  0.28719    0.03728   7.704 3.24e-14 ***
## CRT.z             0.10622    0.03635   2.922 0.003555 ** 
## ideology.z       -0.24332    0.03362  -7.237 9.29e-13 ***
## age.z             0.24690    0.03696   6.681 4.00e-11 ***
## white_.5         -0.07271    0.08060  -0.902 0.367224    
## education.z       0.13278    0.03830   3.467 0.000549 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 1.068 on 973 degrees of freedom
##   (2880 observations deleted due to missingness)
## Multiple R-squared:  0.1748, Adjusted R-squared:  0.1697 
## F-statistic: 34.36 on 6 and 973 DF,  p-value: < 2.2e-16

summary(m2 <- lm(vaxxBehavior ~ index_ANexp_w3.z * CRT.z + ideology.z + age.z + white_.5 + education.z, data = d))

## 
## Call:
## lm(formula = vaxxBehavior ~ index_ANexp_w3.z * CRT.z + ideology.z + 
##     age.z + white_.5 + education.z, data = d)
## 
## Residuals:
##     Min      1Q  Median      3Q     Max 
## -3.2504 -0.3908  0.3134  0.7530  2.3476 
## 
## Coefficients:
##                        Estimate Std. Error t value Pr(>|t|)    
## (Intercept)             3.18867    0.04076  78.232  < 2e-16 ***
## index_ANexp_w3.z        0.29110    0.03992   7.293 6.29e-13 ***
## CRT.z                   0.10755    0.03669   2.932 0.003451 ** 
## ideology.z             -0.24223    0.03387  -7.152 1.68e-12 ***
## age.z                   0.24681    0.03698   6.675 4.15e-11 ***
## white_.5               -0.07429    0.08085  -0.919 0.358379    
## education.z             0.13204    0.03841   3.438 0.000611 ***
## index_ANexp_w3.z:CRT.z  0.01051    0.03831   0.274 0.783830    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 1.069 on 972 degrees of freedom
##   (2880 observations deleted due to missingness)
## Multiple R-squared:  0.1749, Adjusted R-squared:  0.169 
## F-statistic: 29.43 on 7 and 972 DF,  p-value: < 2.2e-16

summary(m3 <- lm(vaxxBehavior ~ index_ANexp_w3.z * (CRT.z + ideology.z) + age.z + white_.5 + education.z, data = d))

## 
## Call:
## lm(formula = vaxxBehavior ~ index_ANexp_w3.z * (CRT.z + ideology.z) + 
##     age.z + white_.5 + education.z, data = d)
## 
## Residuals:
##     Min      1Q  Median      3Q     Max 
## -2.9773 -0.4140  0.3023  0.7127  2.3953 
## 
## Coefficients:
##                             Estimate Std. Error t value Pr(>|t|)    
## (Intercept)                  3.21593    0.04142  77.648  < 2e-16 ***
## index_ANexp_w3.z             0.30245    0.03987   7.585 7.76e-14 ***
## CRT.z                        0.11256    0.03654   3.080  0.00212 ** 
## ideology.z                  -0.23796    0.03373  -7.055 3.27e-12 ***
## age.z                        0.25677    0.03692   6.954 6.51e-12 ***
## white_.5                    -0.07240    0.08045  -0.900  0.36838    
## education.z                  0.12397    0.03830   3.237  0.00125 ** 
## index_ANexp_w3.z:CRT.z       0.01096    0.03812   0.288  0.77369    
## index_ANexp_w3.z:ideology.z  0.10293    0.03165   3.252  0.00119 ** 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 1.063 on 971 degrees of freedom
##   (2880 observations deleted due to missingness)
## Multiple R-squared:  0.1838, Adjusted R-squared:  0.1771 
## F-statistic: 27.33 on 8 and 971 DF,  p-value: < 2.2e-16

tab_model(m1, m2, m3,
          show.df = T, 
          show.ci = .95,
          show.se = T,
          show.stat = T,
          string.stat = "t",
          string.se="SE",
          string.est = "Est",
          digits = 3)

	vaxxBehavior						vaxxBehavior						vaxxBehavior
Predictors	Est	SE	CI	t	p	df	Est	SE	CI	t	p	df	Est	SE	CI	t	p	df
(Intercept)	3.186	0.039	3.108 – 3.263	80.749	<0.001	973.000	3.189	0.041	3.109 – 3.269	78.232	<0.001	972.000	3.216	0.041	3.135 – 3.297	77.648	<0.001	971.000
index_ANexp_w3.z	0.287	0.037	0.214 – 0.360	7.704	<0.001	973.000	0.291	0.040	0.213 – 0.369	7.293	<0.001	972.000	0.302	0.040	0.224 – 0.381	7.585	<0.001	971.000
CRT.z	0.106	0.036	0.035 – 0.178	2.922	0.004	973.000	0.108	0.037	0.036 – 0.180	2.932	0.003	972.000	0.113	0.037	0.041 – 0.184	3.080	0.002	971.000
ideology.z	-0.243	0.034	-0.309 – -0.177	-7.237	<0.001	973.000	-0.242	0.034	-0.309 – -0.176	-7.152	<0.001	972.000	-0.238	0.034	-0.304 – -0.172	-7.055	<0.001	971.000
age.z	0.247	0.037	0.174 – 0.319	6.681	<0.001	973.000	0.247	0.037	0.174 – 0.319	6.675	<0.001	972.000	0.257	0.037	0.184 – 0.329	6.954	<0.001	971.000
white_.5	-0.073	0.081	-0.231 – 0.085	-0.902	0.367	973.000	-0.074	0.081	-0.233 – 0.084	-0.919	0.358	972.000	-0.072	0.080	-0.230 – 0.085	-0.900	0.368	971.000
education.z	0.133	0.038	0.058 – 0.208	3.467	0.001	973.000	0.132	0.038	0.057 – 0.207	3.438	0.001	972.000	0.124	0.038	0.049 – 0.199	3.237	0.001	971.000
index_ANexp_w3.z * CRT.z							0.011	0.038	-0.065 – 0.086	0.274	0.784	972.000	0.011	0.038	-0.064 – 0.086	0.288	0.774	971.000
index_ANexp_w3.z * ideology.z													0.103	0.032	0.041 – 0.165	3.252	0.001	971.000
Observations	980						980						980
R² / R² adjusted	0.175 / 0.170						0.175 / 0.169						0.184 / 0.177

7. Figures

wave 1

wave 2

#run model
m.w2 <- lm(vaxxIntentions_w2 ~ index_ANexp_w2.z * ideology.z + age.z + white_.5 + education.z, data = d)

#create plot
p <- plot_model(m.w2, type = "pred", 
                terms = c("index_ANexp_w2.z", "ideology.z [-1, 1]")) +
  ggtitle("") + 
  ylab("") +
  xlab("") +
  theme_minimal() +
  theme(panel.grid.major = element_blank(), 
        panel.grid.minor = element_blank(),
        panel.background = element_blank(), 
        axis.line = element_line(colour = "black"),
        legend.position = "none", 
        legend.background = element_rect(fill = "white", color = "white"),
        legend.title = element_blank())

p.w2 <-  p + scale_color_manual(labels = NULL, 
                       values = c("blue", "red")) +
  scale_fill_manual(values = c("blue", "red")) +
  scale_x_continuous(breaks = c(-1, -.5, 0, .5, 1, 1.5, 2, 2.5, 3), 
                     limits = c(-1, 3)) +
  scale_y_continuous(breaks = c(-3, -2, -1, 0, 1, 2, 3), 
                     limits = c(-3, 3),
                     labels = c("Definitely
                                would not", 
                                "Probably
                                would not",
                                "Leaning toward
                                would not",
                                "Undecided",
                                "Leaning toward
                                would",
                                "Probably
                                would",
                                "Definitely
                                would"))

## Scale for 'colour' is already present. Adding another scale for 'colour',
## which will replace the existing scale.

## Scale for 'y' is already present. Adding another scale for 'y', which will
## replace the existing scale.

p.w2

## Warning: Removed 4 row(s) containing missing values (geom_path).

wave 3

#run model
m.w3 <- lm(vaxxBehavior ~ index_ANexp_w3.z * ideology.z +
             age.z + white_.5 + education.z, data = d)

#create plot
p <- plot_model(m.w3, type = "pred", 
                terms = c("index_ANexp_w3.z", "ideology.z [-1, 1]")) +
  ggtitle("") + 
  ylab("") +
  xlab("Exposure to Media Outlets x Media Analytical Thinking") +
  theme_minimal() +
  theme(panel.grid.major = element_blank(), 
        panel.grid.minor = element_blank(),
        panel.background = element_blank(), 
        axis.line = element_line(colour = "black"),
        legend.position = "none", 
        legend.background = element_rect(fill = "white", color = "white"),
        legend.title = element_blank())

p.w3 <- p +
  scale_color_manual(labels = NULL, 
                       values = c("blue", "red")) +
  scale_fill_manual(values = c("blue", "red")) +
  scale_x_continuous(breaks = c(-1, -1, -.5, 0, .5, 1, 1.5, 2, 2.5, 3),
                     limits = c(-1, 3)) +
  scale_y_continuous(breaks = c(1, 2, 3, 4), 
                     limits = c(1, 5),
                     labels = c("Not
                                vaccinated
                                ", 
                                "Partially
                                vaccinated
                                ", 
                                "Fully
                                vaccinated
                                ",
                                "Vaccinated
                                and boosted
                                "))

## Scale for 'colour' is already present. Adding another scale for 'colour',
## which will replace the existing scale.

## Scale for 'y' is already present. Adding another scale for 'y', which will
## replace the existing scale.

p.w3

## Warning: Removed 4 row(s) containing missing values (geom_path).

wave 2 long

#run model
m.w2.long <- lm(vaxxIntentions_w2 ~ index_ANexp_w2.z * ideology.z + 
                  index_ANexp_w1.z + vaxxIntentions_w1.z + 
                  education.z + age.z + white_.5, data = d)

#create plot
p <- plot_model(m.w2.long, type = "pred", 
                terms = c("index_ANexp_w2.z", "ideology.z [-1, 1]")) +
  ggtitle("") + 
  ylab("") +
  xlab("Exposure to Media Outlets x Media Analytical Thinking") +
  theme_minimal() +
  theme(panel.grid.major = element_blank(), 
        panel.grid.minor = element_blank(),
        panel.background = element_blank(), 
        axis.line = element_line(colour = "black"),
        legend.position = "none", 
        legend.background = element_rect(fill = "white", color = "white"),
        legend.title = element_blank(),
        axis.text.x = element_text(angle = 0, vjust = .5, hjust = .5))

p.w2.long <- p + scale_color_manual(labels = NULL, 
                       values = c("blue", "red")) +
  scale_fill_manual(values = c("blue", "red")) +
  scale_x_continuous(breaks = c(-1, -1, -.5, 0, .5, 1, 1.5, 2, 2.5, 3, 3.5), #-1.13,    3.42    
                     limits = c(-1, 3.5)) +
  scale_y_continuous(breaks = c(-3, -2, -1, 0, 1, 2, 3), 
                     limits = c(-3, 3),
                     labels = c("Definitely
                                would not", 
                                "Probably
                                would not",
                                "Leaning toward
                                would not",
                                "Undecided",
                                "Leaning toward
                                would",
                                "Probably
                                would",
                                "Definitely
                                would"))

## Scale for 'colour' is already present. Adding another scale for 'colour',
## which will replace the existing scale.

## Scale for 'y' is already present. Adding another scale for 'y', which will
## replace the existing scale.

p.w2.long

## Warning: Removed 4 row(s) containing missing values (geom_path).

ggarrange(p.w1, p.w2, p.w3, p.w2.long, 
          labels = c("       Wave 1: Vaccine Intentions      ", 
                     "       Wave 2: Vaccine Intentions      ", 
                     "       Wave 3: Vaccination Behaviors  ", 
                     "       Wave 2 Long: Vaccine Intentions "),
          ncol = 2, nrow = 2, align = "hv",
          common.legend = T)

## Warning: Removed 4 row(s) containing missing values (geom_path).

## Warning: Removed 4 row(s) containing missing values (geom_path).

## Warning: Removed 4 row(s) containing missing values (geom_path).

## Warning: Removed 4 row(s) containing missing values (geom_path).

## Warning: Removed 4 row(s) containing missing values (geom_path).

Covid Media Manuscript Analyses

Dani Grant

DESCRIPTIVE STATS

1. demographics

a. sample size

b. age

c. gender

d. ethnicity

e. party identity

f. education

2. individual difference measures

a. media exposure

b. analytical media index

i. wave 1

ii. wave 2

iii. wave 3

c. symbolic ideology

d. trust in science

e. science literacy (might cut)

f. cognitive reflection test

3. LIWC media measures

a. Analytical thinking wave 1

b. Analytical thinking wave 2

4. outcome measures

a. wave 1 vaccine attitudes

b. wave 2 vaccine attitudes

c. average of wave 1 + wave 2 vaccine attitudes

d. correlation between wave 1 and wave 2 vaccine attitudes

e. vaccine behavior

5. pairwise correlation table for IVs, DVs, and covariates

6. ANALYSES

Study 2: US only (November 2020)

a. vaxxIntentions ~ index.w2 * ideology (replication)

b. vaxxIntentions.w2 ~ index.w2 * all demographics (replication)

c. mediation path: index.w2 –> expertTrust –> vaxxIntentions.w2

i. controlling for ideology

ii. interaction with ideology

iii. interaction with all demographics

d. Longitudinal: vaxxIntentions.w2 ~ index.w2.z * ideology.z + age.z + edu.z + white_.5 + index.w1.z + vaxxIntentions.w1.z

e. Longitudinal: trustExpert.w2 ~ index.w2.z * ideology.z + age.z + white_.5 + edu.z + index.w1.z + trustExpert.w1.z (causal)

Study 3: US vaxx Behavior (March 2022)

a. vaxxBehavior ~ index.w3.z * ideology.z + edu.z + age.z + white_.5 (replication)

b. Longitudinal: vaxxBehavior ~ index.w3.z * ideology.z + edu.z + age.z + white_.5 + index.avg.w1w2.z + vaxxIntentions.avg.w1w2.z (causal)

c. mediation path: index.w3 –> trust in science –> vaxx Behavior

i. controlling for ideology

ii. interaction with ideology

iii. interaction with all demographics

d. robustness checks

i. does analytical media w3 predict trust in science wave 3 controlling for consumption wave 1 + 2?

ii. science literacy

iii. CRT

7. Figures

wave 1

wave 2

wave 3

wave 2 long