1 Load Libraries

library(psych)
library(dplyr)
## 
## Attaching package: 'dplyr'
## The following objects are masked from 'package:stats':
## 
##     filter, lag
## The following objects are masked from 'package:base':
## 
##     intersect, setdiff, setequal, union
library(expss)
## Loading required package: maditr
## 
## To get total summary skip 'by' argument: take_all(mtcars, mean)
## 
## Attaching package: 'maditr'
## The following objects are masked from 'package:dplyr':
## 
##     between, coalesce, first, last
## 
## Use 'expss_output_rnotebook()' to display tables inside R Notebooks.
##  To return to the console output, use 'expss_output_default()'.
## 
## Attaching package: 'expss'
## The following objects are masked from 'package:dplyr':
## 
##     compute, contains, na_if, recode, vars, where
library(sjPlot)

2 Load Data

df <- read.csv(file="data_study2_clean.csv", header=T)

Variables Analyzed are: Gender, Third Person Effect, Source Credibility Gender = Q4, Q4_7_TEXT Third Person Effect: Q28_1, Q28_2, Q28_3, Q28_5 Source Credibility Competence: Q27_1, Q27_2, Q27_3, Q27_4, Q27_5, Q27_6 Source Credibility Trustworthiness: Q94_1, Q94_2, Q94_3

3 Descriptives Check

3.1 Gender

# 1 == Woman
# 2 == Man
# 3 == Agender
# 4 == Non-binary
# 5 == Cisgender
# 6 == Transgender
# 7 == Another gender not listed

table(df$Q4, useNA = "always")
## 
##     1   1,5   1,6     2 2,4,6   2,5     6  <NA> 
##    94     2     1    83     1     1     2     0
df$gen[is.na(df$Q4)] <- NA
df$gen[df$Q4 == 1] <- "W"
df$gen[df$Q4 == 2] <- "M"
df$gen[df$Q4 == 3] <- "N"
df$gen[df$Q4 == 4] <- "N"
df$gen[df$Q4 == 5] <- "A"
df$gen[df$Q4 == 6] <- "T"
df$gen[df$Q4 == 7] <- "A"
df$gen[df$Q4 == "1,5"] <- "W"
df$gen[df$Q4 == "1,6"] <- "TW"
df$gen[df$Q4 == "2,4,6"] <- NA
df$gen[df$Q4 == "2,5"] <- "M"
table(df$gen, useNA = "always")
## 
##    M    T   TW    W <NA> 
##   84    2    1   96    1
df$gen2 <- df$gen
df$gen2[df$gen2 == "TW"] <- "W"
df$gen2[df$gen2 == "T"] <- NA
table(df$gen2, useNA = "always")
## 
##    M    W <NA> 
##   84   97    3
table(df$Q4, df$gen, useNA = "always")
##        
##          M  T TW  W <NA>
##   1      0  0  0 94    0
##   1,5    0  0  0  2    0
##   1,6    0  0  1  0    0
##   2     83  0  0  0    0
##   2,4,6  0  0  0  0    1
##   2,5    1  0  0  0    0
##   6      0  2  0  0    0
##   <NA>   0  0  0  0    0

3.2 Third Person Effect

hist(df$Q28_1)

describe(df$Q28_1)
##    vars   n mean   sd median trimmed  mad min max range  skew kurtosis  se
## X1    1 184 6.08 2.75      6    6.09 2.97   0  11    11 -0.09     -0.8 0.2
hist(df$Q28_2)

describe(df$Q28_2)
##    vars   n mean   sd median trimmed  mad min max range skew kurtosis  se
## X1    1 184 5.27 2.73    5.5    5.26 3.71   0  11    11 0.03    -0.91 0.2
hist(df$Q28_3)

describe(df$Q28_3)
##    vars   n mean  sd median trimmed  mad min max range  skew kurtosis  se
## X1    1 184  6.5 2.7      7    6.64 2.97   0  11    11 -0.44     -0.5 0.2
hist(df$Q28_5)

describe(df$Q28_5)
##    vars   n mean   sd median trimmed  mad min max range  skew kurtosis   se
## X1    1 184 6.99 2.11      7    7.11 1.48   1  11    10 -0.56     0.18 0.16

3.3 Source Credibility Competence

hist(df$Q27_1)

describe(df$Q27_1)
##    vars   n mean   sd median trimmed  mad min max range  skew kurtosis   se
## X1    1 184 4.92 1.18      5    4.95 1.48   2   7     5 -0.23    -0.32 0.09
hist(df$Q27_2)

describe(df$Q27_2)
##    vars   n mean   sd median trimmed  mad min max range  skew kurtosis   se
## X1    1 184 3.94 1.44      4    3.95 1.48   1   7     6 -0.05    -0.46 0.11
hist(df$Q27_3)

describe(df$Q27_3)
##    vars   n mean   sd median trimmed  mad min max range  skew kurtosis  se
## X1    1 184 3.87 1.36      4    3.86 1.48   1   7     6 -0.02    -0.28 0.1
hist(df$Q27_4)

describe(df$Q27_4)
##    vars   n mean   sd median trimmed  mad min max range  skew kurtosis   se
## X1    1 184 4.64 1.43      5    4.66 1.48   1   7     6 -0.25    -0.59 0.11
hist(df$Q27_5)

describe(df$Q27_5)
##    vars   n mean   sd median trimmed  mad min max range  skew kurtosis  se
## X1    1 184 5.01 1.36      5    5.07 1.48   1   7     6 -0.36    -0.32 0.1
hist(df$Q27_6)

describe(df$Q27_6)
##    vars   n mean   sd median trimmed  mad min max range  skew kurtosis   se
## X1    1 184 4.98 1.21      5    4.99 1.48   2   7     5 -0.12     -0.3 0.09

3.4 Source Credibility Trustworthiness

hist(df$Q94_1)

describe(df$Q94_1)
##    vars   n mean   sd median trimmed  mad min max range  skew kurtosis  se
## X1    1 184  4.7 1.37      5    4.71 1.48   1   7     6 -0.19    -0.61 0.1
hist(df$Q94_2)

describe(df$Q94_2)
##    vars   n mean  sd median trimmed  mad min max range skew kurtosis   se
## X1    1 184 4.41 1.5      4    4.43 1.48   1   7     6 -0.1    -0.61 0.11
hist(df$Q94_3)

describe(df$Q94_3)
##    vars   n mean   sd median trimmed  mad min max range  skew kurtosis  se
## X1    1 184 4.59 1.32      5    4.63 1.48   1   7     6 -0.27    -0.22 0.1

4 Composites

4.1 Third Person Effect

Overall variable reflecting how an individual is influenced by what they read and how they perceive others to be influenced

df$tpe_self <- (df$Q28_1 + df$Q28_2)/2
df$tpe_othr <- (df$Q28_3 + df$Q28_5)/2
hist(df$tpe_self)

describe(df$tpe_self)
##    vars   n mean   sd median trimmed  mad min max range  skew kurtosis   se
## X1    1 184 5.68 2.51      6     5.7 2.97   0  11    11 -0.05    -0.71 0.19
hist(df$tpe_othr)

describe(df$tpe_othr)
##    vars   n mean   sd median trimmed  mad min max range  skew kurtosis   se
## X1    1 184 6.74 2.11      7    6.84 2.22 0.5  11  10.5 -0.46    -0.15 0.16

4.2 Source Credibility Competence

Overall variable regarding the perceived competence of the authors of the presented social media post

df$source_comp <- (df$Q27_1 + df$Q27_2 + df$Q27_3 + df$Q27_4 + df$Q27_5 + df$Q27_6)/6
hist(df$source_comp)

describe(df$source_comp)
##    vars   n mean   sd median trimmed  mad  min max range  skew kurtosis   se
## X1    1 184 4.56 1.17   4.58    4.57 1.36 1.67   7  5.33 -0.12    -0.54 0.09

4.3 Source Credibility Trustworthiness

Overall variable regarding the perceived trustworthiness of the authors of the presented social media post

df$source_trust <- (df$Q94_1 + df$Q94_2 + df$Q94_3)/3
hist(df$source_trust)

describe(df$source_trust)
##    vars   n mean   sd median trimmed  mad  min max range skew kurtosis  se
## X1    1 184 4.57 1.33   4.67    4.58 1.48 1.33   7  5.67 -0.1    -0.58 0.1

5 t-Test

5.1 TPE Self

Gender differences as to how individuals were influenced by social media post; no gender diff found

t.test(df$tpe_self ~ df$gen2, data = df)
## 
##  Welch Two Sample t-test
## 
## data:  df$tpe_self by df$gen2
## t = 0.21139, df = 163.98, p-value = 0.8328
## alternative hypothesis: true difference in means between group M and group W is not equal to 0
## 95 percent confidence interval:
##  -0.6628197  0.8217544
## sample estimates:
## mean in group M mean in group W 
##        5.708333        5.628866

5.2 TPE Others

Gender differences as to how individuals perceived others were influenced by social media post; no gender diff found

t.test(df$tpe_othr ~ df$gen2, data = df)
## 
##  Welch Two Sample t-test
## 
## data:  df$tpe_othr by df$gen2
## t = 0.86164, df = 171.85, p-value = 0.3901
## alternative hypothesis: true difference in means between group M and group W is not equal to 0
## 95 percent confidence interval:
##  -0.3530402  0.9000456
## sample estimates:
## mean in group M mean in group W 
##        6.886905        6.613402

5.3 Source Credibility Competence

Gender differences and Competence; no gender diff found

t.test(df$source_comp ~ df$gen2, data = df)
## 
##  Welch Two Sample t-test
## 
## data:  df$source_comp by df$gen2
## t = 0.069677, df = 174.14, p-value = 0.9445
## alternative hypothesis: true difference in means between group M and group W is not equal to 0
## 95 percent confidence interval:
##  -0.3303427  0.3545205
## sample estimates:
## mean in group M mean in group W 
##        4.577381        4.565292

5.4 Source Credibility Trustworthiness

Gender differences and Trustworthiness; no gender diff found (men slightly higher than women)

t.test(df$source_trust ~ df$gen2, data = df) 
## 
##  Welch Two Sample t-test
## 
## data:  df$source_trust by df$gen2
## t = 1.0392, df = 175.03, p-value = 0.3001
## alternative hypothesis: true difference in means between group M and group W is not equal to 0
## 95 percent confidence interval:
##  -0.1851660  0.5970462
## sample estimates:
## mean in group M mean in group W 
##        4.690476        4.484536

5.5 TPE Paired

Differences in participants’ self and other scores; third person effect present (significant difference between groups, confirming that tpe is occurring/present without regard to gender)

t.test(df$tpe_self, df$tpe_othr, paired = TRUE)
## 
##  Paired t-test
## 
## data:  df$tpe_self and df$tpe_othr
## t = -9.0498, df = 183, p-value < 2.2e-16
## alternative hypothesis: true mean difference is not equal to 0
## 95 percent confidence interval:
##  -1.3007631 -0.8351065
## sample estimates:
## mean difference 
##       -1.067935
describe(subset(df, select = c(tpe_self, tpe_othr)))
##          vars   n mean   sd median trimmed  mad min max range  skew kurtosis
## tpe_self    1 184 5.68 2.51      6    5.70 2.97 0.0  11  11.0 -0.05    -0.71
## tpe_othr    2 184 6.74 2.11      7    6.84 2.22 0.5  11  10.5 -0.46    -0.15
##            se
## tpe_self 0.19
## tpe_othr 0.16

5.6 Third Person Effect

Calculated difference score (higher score indicates more tpe for the individual; above 0 = tpe present and below 0 = no tpe present)

df <- df %>%
  mutate(Q28_6 = case_when(
    Q28_1 == Q28_3 ~ 0,
    Q28_1 <  Q28_3 ~ 1,
    Q28_1 >  Q28_3 ~ -1
  )) %>%
  mutate(Q28_7 = case_when(
    Q28_2 == Q28_5 ~ 0,
    Q28_2 <  Q28_5 ~ 1,
    Q28_2 >  Q28_5 ~ -1
  ))
table(df$Q28_6)
## 
## -1  0  1 
## 45 65 74
table(df$Q28_7)
## 
##  -1   0   1 
##  26  32 126
df$tpe_diff <- (df$Q28_6 + df$Q28_7)

hist(df$tpe_self)

describe(df$tpe_self)
##    vars   n mean   sd median trimmed  mad min max range  skew kurtosis   se
## X1    1 184 5.68 2.51      6     5.7 2.97   0  11    11 -0.05    -0.71 0.19
hist(df$tpe_othr)

describe(df$tpe_othr)
##    vars   n mean   sd median trimmed  mad min max range  skew kurtosis   se
## X1    1 184 6.74 2.11      7    6.84 2.22 0.5  11  10.5 -0.46    -0.15 0.16
hist(df$tpe_diff)

describe(df$tpe_diff)
##    vars   n mean   sd median trimmed  mad min max range  skew kurtosis   se
## X1    1 184  0.7 1.16      1    0.81 1.48  -2   2     4 -0.51    -0.59 0.09
cross_cases(df, Q28_1, Q28_3)
 Q28_3 
 0   1   2   3   4   5   6   7   8   9   10   11 
 Q28_1 
   0  2
   1  3 2 1 1
   2  2 2 1 3 3 1
   3  2 3 3 1 3 1 1 1
   4  1 1 4 5 1 5 3 3
   5  1 2 4 4 3 1 1
   6  3 2 7 8 2 1
   7  1 2 2 6 6 4 1
   8  1 3 3 13 8
   9  2 4 8 2 2
   10  1 4
   11  2 2 1 8
   #Total cases  3 7 8 11 14 16 23 26 31 25 9 11 
cross_cases(df, Q28_2, Q28_5)
 Q28_5 
 1   2   3   4   5   6   7   8   9   10   11 
 Q28_2 
   0  1 1 1
   1  1 1 2 3 3 2 3 1
   2  2 2 2 2 2 3 2
   3  1 1 2 5 5 3 4 2
   4  1 2 5 4 3 1
   5  1 3 2 5 6 2
   6  1 9 11 4 2 1 1
   7  1 1 2 6 3 6 3
   8  1 3 5 3 4
   9  1 3 2 3 4
   10  3 3 1 2
   11  1 1 1
   #Total cases  2 6 6 7 16 29 39 36 24 14 5

6 Moderated Regression

6.1 Gender Differences in Third Person Effect

Regression (Reg1) between perceptions of TPE and trustworthiness: relationship found (higher trustworthiness indicates less TPE); Reg2 indicates no gender difference present Regression (Reg3) between TPE and competence: relationship found (higher competence indicates less TPE); Reg4 indicates gender difference present Plots comparing men and women: men perceive increased competence as indicating less TPE, a pattern not present in women

t.test(tpe_diff ~ gen2, data = df)
## 
##  Welch Two Sample t-test
## 
## data:  tpe_diff by gen2
## t = -0.56176, df = 169.49, p-value = 0.575
## alternative hypothesis: true difference in means between group M and group W is not equal to 0
## 95 percent confidence interval:
##  -0.4415442  0.2459134
## sample estimates:
## mean in group M mean in group W 
##       0.6547619       0.7525773
df2 <- subset(df, select = c(tpe_self, tpe_othr, tpe_diff, source_trust, source_comp)) 
corr.test(df2)
## Call:corr.test(x = df2)
## Correlation matrix 
##              tpe_self tpe_othr tpe_diff source_trust source_comp
## tpe_self         1.00     0.77    -0.38         0.60        0.69
## tpe_othr         0.77     1.00     0.12         0.49        0.52
## tpe_diff        -0.38     0.12     1.00        -0.24       -0.32
## source_trust     0.60     0.49    -0.24         1.00        0.83
## source_comp      0.69     0.52    -0.32         0.83        1.00
## Sample Size 
## [1] 184
## Probability values (Entries above the diagonal are adjusted for multiple tests.) 
##              tpe_self tpe_othr tpe_diff source_trust source_comp
## tpe_self            0     0.00     0.00            0           0
## tpe_othr            0     0.00     0.11            0           0
## tpe_diff            0     0.11     0.00            0           0
## source_trust        0     0.00     0.00            0           0
## source_comp         0     0.00     0.00            0           0
## 
##  To see confidence intervals of the correlations, print with the short=FALSE option
reg1 <- lm(tpe_diff ~ source_trust, data = df)
summary(reg1)
## 
## Call:
## lm(formula = tpe_diff ~ source_trust, data = df)
## 
## Residuals:
##    Min     1Q Median     3Q    Max 
## -3.321 -0.822  0.071  1.035  1.820 
## 
## Coefficients:
##              Estimate Std. Error t value Pr(>|t|)    
## (Intercept)   1.67779    0.29889   5.613  7.3e-08 ***
## source_trust -0.21395    0.06288  -3.402 0.000822 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 1.129 on 182 degrees of freedom
## Multiple R-squared:  0.0598, Adjusted R-squared:  0.05463 
## F-statistic: 11.58 on 1 and 182 DF,  p-value: 0.0008218
plot(df$tpe_diff, df$source_trust)

reg2 <- lm(tpe_diff ~ source_trust*gen2, data = df)
summary(reg2)
## 
## Call:
## lm(formula = tpe_diff ~ source_trust * gen2, data = df)
## 
## Residuals:
##     Min      1Q  Median      3Q     Max 
## -3.1040 -0.8130  0.1038  1.1038  1.8536 
## 
## Coefficients:
##                    Estimate Std. Error t value Pr(>|t|)    
## (Intercept)         2.10612    0.45177   4.662 6.15e-06 ***
## source_trust       -0.30943    0.09269  -3.338  0.00103 ** 
## gen2W              -0.79420    0.60684  -1.309  0.19232    
## source_trust:gen2W  0.18470    0.12690   1.455  0.14732    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 1.126 on 177 degrees of freedom
##   (3 observations deleted due to missingness)
## Multiple R-squared:  0.07114,    Adjusted R-squared:  0.05539 
## F-statistic: 4.518 on 3 and 177 DF,  p-value: 0.00443
reg3 <- lm(tpe_diff ~ source_comp, data = df)
summary(reg3)
## 
## Call:
## lm(formula = tpe_diff ~ source_comp, data = df)
## 
## Residuals:
##     Min      1Q  Median      3Q     Max 
## -3.2948 -0.7329  0.0415  0.9666  1.9074 
## 
## Coefficients:
##             Estimate Std. Error t value Pr(>|t|)    
## (Intercept)  2.13105    0.32914   6.475 8.58e-10 ***
## source_comp -0.31360    0.06994  -4.484 1.29e-05 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 1.105 on 182 degrees of freedom
## Multiple R-squared:  0.09948,    Adjusted R-squared:  0.09453 
## F-statistic: 20.11 on 1 and 182 DF,  p-value: 1.295e-05
reg4 <- lm(tpe_diff ~ source_comp*gen2, data = df)
summary(reg4)
## 
## Call:
## lm(formula = tpe_diff ~ source_comp * gen2, data = df)
## 
## Residuals:
##     Min      1Q  Median      3Q     Max 
## -3.0930 -0.7045  0.0863  1.0163  1.9005 
## 
## Coefficients:
##                   Estimate Std. Error t value Pr(>|t|)    
## (Intercept)         2.6785     0.4854   5.518 1.20e-07 ***
## source_comp        -0.4421     0.1027  -4.303 2.78e-05 ***
## gen2W              -1.1074     0.6686  -1.656   0.0995 .  
## source_comp:gen2W   0.2628     0.1418   1.854   0.0655 .  
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 1.101 on 177 degrees of freedom
##   (3 observations deleted due to missingness)
## Multiple R-squared:  0.1116, Adjusted R-squared:  0.09656 
## F-statistic: 7.413 on 3 and 177 DF,  p-value: 0.000105
plot_model(reg4, type = "int")
## Some of the focal terms are of type `character`. This may lead to
##   unexpected results. It is recommended to convert these variables to
##   factors before fitting the model.
##   The following variables are of type character: `gen2`

plot_model(reg2, type = "int")
## Some of the focal terms are of type `character`. This may lead to
##   unexpected results. It is recommended to convert these variables to
##   factors before fitting the model.
##   The following variables are of type character: `gen2`