R Notebook
library(tidyverse)## ── Attaching packages ──────────────────────────────────────────── tidyverse 1.3.0 ──## ✓ ggplot2 3.3.2 ✓ purrr 0.3.4
## ✓ tibble 3.0.3 ✓ dplyr 1.0.0
## ✓ tidyr 1.1.0 ✓ stringr 1.4.0
## ✓ readr 1.3.1 ✓ forcats 0.5.0## ── Conflicts ─────────────────────────────────────────────── tidyverse_conflicts() ──
## x dplyr::filter() masks stats::filter()
## x dplyr::lag() masks stats::lag()library(corrr)
library(ggplot2)
library(lme4)## Loading required package: Matrix##
## Attaching package: 'Matrix'## The following objects are masked from 'package:tidyr':
##
## expand, pack, unpacklibrary(lmerTest)##
## Attaching package: 'lmerTest'## The following object is masked from 'package:lme4':
##
## lmer## The following object is masked from 'package:stats':
##
## steplibrary(ggeffects)
library(sjPlot)setwd("/Volumes/Research Project/Preference Falsification/Analysis/")
indDiffDf <- read.csv("./output/PFindDiffDf.csv")
longDf <- read.csv("./output/PFfullDf.csv")
longDf$partyN <- as.factor(longDf$partyN)
longDf$Rep <- as.factor(longDf$Rep)
contrasts(longDf$Rep) <- contr.sum(3)
longDf$RepN <- as.factor(longDf$RepN)
longDf$RepN <- relevel(longDf$RepN,"In")
longDf$Info <- as.factor(longDf$Info)
contrasts(longDf$Info) <- contr.sum(2)
longDf$partyN <- as.factor(longDf$partyN)
contrasts(longDf$partyN) <- contr.sum(2)
demDf <- subset(longDf, partyN == "Dem")
repDf <- subset(longDf, partyN == "Rep")Correlation Matrix of Individual Differences with Subject-Level Falsification Variable
Falsification is correlation between the extent to which an individual moves away from ingroup and towards outgroup (i.e. negative correlation; r-to-z transformed).
Negative correlations below are more falsification (moving away from your own party and towards opposite party) is associated with the individual difference. Positive correlations below mean more consistently evaluating with ingroup and unlike outgroup is associated with individual differences.
Individual Differences: affPol - Affective Polarization SGO.wt - We They overlap SGO.st - Self They overlap SGO.sw - Self We overlap WSC - Willingness to Self-Censor Extra - Extraversion Neur - Neuroticism MLAM - Approval Motivation SCS.sa - Self-Consciousness social anxiety subscale SCS.pub - Public self-consciousness SCS.priv - Private slef-consciousness NFC - Need for Cognition CSEid - Identity importance CSEpub - Public collective self-esteem CSEpriv - Private collective self-esteem
For entire sample
corDf <- indDiffDf[7:22]
x <- corDf %>%
correlate() %>%
focus(falsify)##
## Correlation method: 'pearson'
## Missing treated using: 'pairwise.complete.obs'corDf <- x[order(-x$falsify),]
corDfParticipants whose party aligned with the researcher’s political affiliation
corDfCong <- subset(indDiffDf, as.character(Rep) == as.character(partyN) )
corDfCong <- corDfCong[7:22]
x <- corDfCong %>%
correlate() %>%
focus(falsify)##
## Correlation method: 'pearson'
## Missing treated using: 'pairwise.complete.obs'corDfCong <- x[order(-x$falsify),]
corDfCongParticipants whose party did not align with researcher
corDfIncong <- subset(indDiffDf, Rep == "Dem" & partyN == "Rep" | Rep == "Rep" & partyN == "Dem")
corDfIncong <- corDfIncong[7:22]
x <- corDfIncong %>%
correlate() %>%
focus(falsify)##
## Correlation method: 'pearson'
## Missing treated using: 'pairwise.complete.obs'corDfIncong <- x[order(-x$falsify),]
corDfIncongParticipants who believed they were interacting with non-partisan
corDfNonPart <- subset(indDiffDf, Rep == "Non")
corDfNonPart <- corDfNonPart[7:22]
x <- corDfNonPart %>%
correlate() %>%
focus(falsify)##
## Correlation method: 'pearson'
## Missing treated using: 'pairwise.complete.obs'corDfNonPart <- x[order(-x$falsify),]
corDfNonPartParticipants who believed they were interacting non-partisan or opposite party
corDfNonPartOppPart <- subset(indDiffDf, Rep == "Non" | Rep == "Dem" & partyN == "Rep" | Rep == "Rep" & partyN == "Dem")
corDfNonPartOppPart <- corDfNonPartOppPart[7:22]
x <- corDfNonPartOppPart %>%
correlate() %>%
focus(falsify)##
## Correlation method: 'pearson'
## Missing treated using: 'pairwise.complete.obs'corDfNonPartOppPart <- x[order(-x$falsify),]
corDfNonPartOppPartSeems like strongest correlations by far is people moving more towards ingroup and less away from outgroup (i.e. less falsifying) who are higher in affective polarization.
The best correlations we have for falsification are willingness to self-censor, approval motivation, and self-they overlap, but they’re not incredibly strong correlations.
Testing issue-specific, level 1 effects
Effect of issue-specific honesty
m <- lmer( scale(toInG) ~ scale(honest) + ( scale(honest)| subID) + (1 | issues), data = longDf, control=lmerControl(optimizer="bobyqa",
optCtrl=list(maxfun=2e5)))
summary(m)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: scale(toInG) ~ scale(honest) + (scale(honest) | subID) + (1 |
## issues)
## Data: longDf
## Control: lmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 2e+05))
##
## REML criterion at convergence: 99791.5
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -3.4892 -0.7307 0.1818 0.7484 2.7253
##
## Random effects:
## Groups Name Variance Std.Dev. Corr
## subID (Intercept) 0.13444 0.3667
## scale(honest) 0.04283 0.2070 0.67
## issues (Intercept) 0.04944 0.2223
## Residual 0.77228 0.8788
## Number of obs: 37917, groups: subID, 384; issues, 100
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) -7.044e-04 2.941e-02 2.443e+02 -0.024 0.9809
## scale(honest) -4.516e-02 2.503e-02 1.427e+02 -1.805 0.0732 .
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr)
## scale(hnst) 0.181plot_model(m, type = "pred", terms = c("honest"))
m <- lmer( scale(eval) ~ scale(honest) + ( scale(honest)| subID) + (1 | issues), data = longDf, control=lmerControl(optimizer="bobyqa",
optCtrl=list(maxfun=2e5)))
summary(m)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: scale(eval) ~ scale(honest) + (scale(honest) | subID) + (1 |
## issues)
## Data: longDf
## Control: lmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 2e+05))
##
## REML criterion at convergence: 96267.9
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -3.2263 -0.7008 0.1549 0.7186 3.1313
##
## Random effects:
## Groups Name Variance Std.Dev. Corr
## subID (Intercept) 0.038737 0.19682
## scale(honest) 0.007631 0.08735 -0.40
## issues (Intercept) 0.147447 0.38399
## Residual 0.714690 0.84539
## Number of obs: 37917, groups: subID, 384; issues, 100
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) -0.001242 0.039928 111.510467 -0.031 0.975
## scale(honest) 0.306828 0.038900 100.609705 7.888 3.84e-12 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr)
## scale(hnst) -0.012Participants falsified more for issues that people expect others would be less honest about. *marginal
Effect of issue-specific reputational cost
m <- lmer( scale(toInG) ~ scale(reput) + ( scale(reput)| subID) + ( 1 | issues), data = longDf, control=lmerControl(optimizer="bobyqa",
optCtrl=list(maxfun=2e5)))
summary(m)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: scale(toInG) ~ scale(reput) + (scale(reput) | subID) + (1 | issues)
## Data: longDf
## Control: lmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 2e+05))
##
## REML criterion at convergence: 101125.4
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -3.0278 -0.7614 0.1748 0.7826 2.5890
##
## Random effects:
## Groups Name Variance Std.Dev. Corr
## subID (Intercept) 0.134153 0.36627
## scale(reput) 0.007382 0.08592 0.63
## issues (Intercept) 0.051248 0.22638
## Residual 0.808221 0.89901
## Number of obs: 37917, groups: subID, 384; issues, 100
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) -6.554e-04 2.972e-02 2.388e+02 -0.022 0.982
## scale(reput) 1.501e-02 2.352e-02 1.051e+02 0.638 0.525
##
## Correlation of Fixed Effects:
## (Intr)
## scale(rept) 0.074m <- lmer( scale(eval) ~ scale(reput) + ( scale(reput)| subID) + ( 1 | issues), data = longDf, control=lmerControl(optimizer="bobyqa",
optCtrl=list(maxfun=2e5)))
summary(m)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: scale(eval) ~ scale(reput) + (scale(reput) | subID) + (1 | issues)
## Data: longDf
## Control: lmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 2e+05))
##
## REML criterion at convergence: 94065.8
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -3.3515 -0.6814 0.1463 0.7029 3.8152
##
## Random effects:
## Groups Name Variance Std.Dev. Corr
## subID (Intercept) 0.03940 0.1985
## scale(reput) 0.05866 0.2422 0.51
## issues (Intercept) 0.24204 0.4920
## Residual 0.66344 0.8145
## Number of obs: 37917, groups: subID, 384; issues, 100
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) -0.001212 0.050404 106.361953 -0.024 0.981
## scale(reput) -0.038318 0.050901 110.540682 -0.753 0.453
##
## Correlation of Fixed Effects:
## (Intr)
## scale(rept) 0.025No effect of reputation level 1 by itself.
Effect of issue-specifc political orientation
m <- lmer( scale(toInG) ~ scale(politic) + ( scale(politic) | subID) + ( 1 | issues), data = longDf, control=lmerControl(optimizer="bobyqa",
optCtrl=list(maxfun=2e5)))
summary(m)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: scale(toInG) ~ scale(politic) + (scale(politic) | subID) + (1 |
## issues)
## Data: longDf
## Control: lmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 2e+05))
##
## REML criterion at convergence: 98140
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -3.6756 -0.7196 0.1402 0.7432 2.9108
##
## Random effects:
## Groups Name Variance Std.Dev. Corr
## subID (Intercept) 0.13493 0.3673
## scale(politic) 0.08080 0.2842 0.63
## issues (Intercept) 0.05168 0.2273
## Residual 0.73404 0.8568
## Number of obs: 37917, groups: subID, 384; issues, 100
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) -7.188e-04 2.979e-02 2.393e+02 -0.024 0.981
## scale(politic) -2.217e-03 2.732e-02 1.814e+02 -0.081 0.935
##
## Correlation of Fixed Effects:
## (Intr)
## scale(pltc) 0.210m <- lmer( scale(eval) ~ scale(politic) + ( scale(politic) | subID) + ( 1 | issues), data = longDf, control=lmerControl(optimizer="bobyqa",
optCtrl=list(maxfun=2e5)))
summary(m)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: scale(eval) ~ scale(politic) + (scale(politic) | subID) + (1 |
## issues)
## Data: longDf
## Control: lmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 2e+05))
##
## REML criterion at convergence: 83593.7
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -4.1916 -0.6122 0.0859 0.6549 4.2358
##
## Random effects:
## Groups Name Variance Std.Dev. Corr
## subID (Intercept) 0.04105 0.2026
## scale(politic) 0.22493 0.4743 0.53
## issues (Intercept) 0.22164 0.4708
## Residual 0.49398 0.7028
## Number of obs: 37917, groups: subID, 384; issues, 100
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) -0.00138 0.04834 107.54658 -0.029 0.97727
## scale(politic) 0.14732 0.05306 153.42824 2.777 0.00618 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr)
## scale(pltc) 0.052No effect of politi by itself
Effect of issue-specific breadth
m <- lmer( scale(toInG) ~ scale(breadth) + ( scale(breadth) | subID) + ( 1| issues), data = longDf, control=lmerControl(optimizer="bobyqa",
optCtrl=list(maxfun=2e5)))## boundary (singular) fit: see ?isSingularsummary(m)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: scale(toInG) ~ scale(breadth) + (scale(breadth) | subID) + (1 |
## issues)
## Data: longDf
## Control: lmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 2e+05))
##
## REML criterion at convergence: 101224
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.9573 -0.7608 0.1770 0.7832 2.6666
##
## Random effects:
## Groups Name Variance Std.Dev. Corr
## subID (Intercept) 0.134001 0.36606
## scale(breadth) 0.001646 0.04057 -1.00
## issues (Intercept) 0.050940 0.22570
## Residual 0.813962 0.90220
## Number of obs: 37917, groups: subID, 384; issues, 100
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) -7.235e-04 2.966e-02 2.394e+02 -0.024 0.981
## scale(breadth) 2.349e-02 2.313e-02 9.957e+01 1.016 0.312
##
## Correlation of Fixed Effects:
## (Intr)
## scal(brdth) -0.056
## convergence code: 0
## boundary (singular) fit: see ?isSingularm <- lmer( scale(eval) ~ scale(breadth) + ( scale(breadth) | subID) + ( 1| issues), data = longDf, control=lmerControl(optimizer="bobyqa",
optCtrl=list(maxfun=2e5)))## boundary (singular) fit: see ?isSingularsummary(m)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: scale(eval) ~ scale(breadth) + (scale(breadth) | subID) + (1 |
## issues)
## Data: longDf
## Control: lmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 2e+05))
##
## REML criterion at convergence: 96465.8
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -3.2597 -0.7106 0.1603 0.7226 3.1374
##
## Random effects:
## Groups Name Variance Std.Dev. Corr
## subID (Intercept) 3.861e-02 0.196502
## scale(breadth) 5.679e-05 0.007536 -1.00
## issues (Intercept) 2.397e-01 0.489547
## Residual 7.223e-01 0.849910
## Number of obs: 37917, groups: subID, 384; issues, 100
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) -0.001278 0.050162 106.266009 -0.025 0.980
## scale(breadth) -0.061615 0.049151 98.016328 -1.254 0.213
##
## Correlation of Fixed Effects:
## (Intr)
## scal(brdth) -0.002
## convergence code: 0
## boundary (singular) fit: see ?isSingularNo effect of breadth of issue by itself.
Effect of issue-specific changeability
m <- lmer(scale(toInG)~ scale(change) + ( scale(change) | subID) + ( 1| issues), data = longDf, control=lmerControl(optimizer="bobyqa",
optCtrl=list(maxfun=2e5)))
summary(m)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: scale(toInG) ~ scale(change) + (scale(change) | subID) + (1 |
## issues)
## Data: longDf
## Control: lmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 2e+05))
##
## REML criterion at convergence: 99837
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -3.1493 -0.7273 0.1548 0.7630 3.0647
##
## Random effects:
## Groups Name Variance Std.Dev. Corr
## subID (Intercept) 0.13376 0.3657
## scale(change) 0.03994 0.1998 0.78
## issues (Intercept) 0.05105 0.2259
## Residual 0.77532 0.8805
## Number of obs: 37917, groups: subID, 384; issues, 100
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) -8.234e-04 2.965e-02 2.392e+02 -0.028 0.978
## scale(change) 2.466e-02 2.520e-02 1.382e+02 0.978 0.330
##
## Correlation of Fixed Effects:
## (Intr)
## scale(chng) 0.199m <- lmer( scale(eval) ~ scale(change) + ( scale(change) | subID) + ( 1| issues), data = longDf, control=lmerControl(optimizer="bobyqa",
optCtrl=list(maxfun=2e5)))
summary(m)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: scale(eval) ~ scale(change) + (scale(change) | subID) + (1 |
## issues)
## Data: longDf
## Control: lmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 2e+05))
##
## REML criterion at convergence: 96204.7
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -3.2695 -0.7068 0.1479 0.7086 3.4968
##
## Random effects:
## Groups Name Variance Std.Dev. Corr
## subID (Intercept) 0.03861 0.1965
## scale(change) 0.01052 0.1026 0.42
## issues (Intercept) 0.20137 0.4487
## Residual 0.71176 0.8437
## Number of obs: 37917, groups: subID, 384; issues, 100
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) -0.001414 0.046185 107.846126 -0.031 0.976
## scale(change) 0.203434 0.045386 100.651330 4.482 1.96e-05 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr)
## scale(chng) 0.010No effect of changeability by itself.
Hypothesis 1: Between-subjects reputation main effect
No issue-specific features
For everyone
m <- lmer(scale(toInG) ~ RepN + ( 1 | subID) + (1 | issues), data = longDf, control=lmerControl(optimizer="bobyqa",
optCtrl=list(maxfun=2e5)))
summary(m)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: scale(toInG) ~ RepN + (1 | subID) + (1 | issues)
## Data: longDf
## Control: lmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 2e+05))
##
## REML criterion at convergence: 101293.7
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.9808 -0.7633 0.1782 0.7850 2.6156
##
## Random effects:
## Groups Name Variance Std.Dev.
## subID (Intercept) 0.13214 0.3635
## issues (Intercept) 0.05094 0.2257
## Residual 0.81563 0.9031
## Number of obs: 37917, groups: subID, 384; issues, 100
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 0.04576 0.03997 428.70987 1.145 0.2529
## RepNNon -0.02517 0.04639 381.11367 -0.543 0.5878
## RepNOut -0.11807 0.04722 381.04611 -2.500 0.0128 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) RepNNn
## RepNNon -0.587
## RepNOut -0.577 0.497plot_model(m, type = "pred", terms = c("RepN"))
Participants falsify more when they believe they are interacting with outgroup. Also when they think they are interacting with non-partisan
Further analyses of hypothesis 1; incorporating issue-specific features
Political
# Approval evaluations
m <- lmer( scale(toInG) ~ RepN * scale(politic) + ( scale(politic) | subID) + (1|issues), data = longDf, control=lmerControl(optimizer="bobyqa",
optCtrl=list(maxfun=2e5)))
summary(m)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: scale(toInG) ~ RepN * scale(politic) + (scale(politic) | subID) +
## (1 | issues)
## Data: longDf
## Control: lmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 2e+05))
##
## REML criterion at convergence: 98152.4
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -3.6779 -0.7201 0.1393 0.7431 2.9101
##
## Random effects:
## Groups Name Variance Std.Dev. Corr
## subID (Intercept) 0.13313 0.3649
## scale(politic) 0.08053 0.2838 0.62
## issues (Intercept) 0.05168 0.2273
## Residual 0.73404 0.8568
## Number of obs: 37917, groups: subID, 384; issues, 100
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 0.045528 0.040082 426.678605 1.136 0.257
## RepNNon -0.024609 0.046420 381.099334 -0.530 0.596
## RepNOut -0.117964 0.047250 381.040374 -2.497 0.013 *
## scale(politic) 0.019995 0.034619 345.663406 0.578 0.564
## RepNNon:scale(politic) -0.007659 0.036715 380.486068 -0.209 0.835
## RepNOut:scale(politic) -0.061120 0.037371 380.396318 -1.636 0.103
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) RepNNn RepNOt scl(p) RNN:()
## RepNNon -0.586
## RepNOut -0.575 0.497
## scale(pltc) 0.361 -0.312 -0.306
## RpNNn:scl() -0.341 0.582 0.289 -0.536
## RpNOt:scl() -0.335 0.289 0.582 -0.527 0.497# Difference from ingroup
m <- lmer( scale(eval) ~ RepN * scale(politic) + ( scale(politic) | subID) + (1|issues), data = longDf, control=lmerControl(optimizer="bobyqa",
optCtrl=list(maxfun=2e5)))
summary(m)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: scale(eval) ~ RepN * scale(politic) + (scale(politic) | subID) +
## (1 | issues)
## Data: longDf
## Control: lmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 2e+05))
##
## REML criterion at convergence: 83607.5
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -4.1910 -0.6116 0.0853 0.6551 4.2369
##
## Random effects:
## Groups Name Variance Std.Dev. Corr
## subID (Intercept) 0.04106 0.2026
## scale(politic) 0.22291 0.4721 0.53
## issues (Intercept) 0.22164 0.4708
## Residual 0.49398 0.7028
## Number of obs: 37917, groups: subID, 384; issues, 100
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 0.01746 0.05073 129.72988 0.344 0.73133
## RepNNon -0.01970 0.02657 380.40487 -0.741 0.45889
## RepNOut -0.03766 0.02705 380.29542 -1.392 0.16459
## scale(politic) 0.20180 0.06311 269.30915 3.197 0.00155 **
## RepNNon:scale(politic) -0.03320 0.05910 381.02567 -0.562 0.57468
## RepNOut:scale(politic) -0.13443 0.06016 380.99957 -2.234 0.02603 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) RepNNn RepNOt scl(p) RNN:()
## RepNNon -0.265
## RepNOut -0.260 0.497
## scale(pltc) 0.122 -0.233 -0.228
## RpNNn:scl() -0.130 0.491 0.244 -0.474
## RpNOt:scl() -0.128 0.244 0.491 -0.465 0.497No influence of politic interacting with main effect of reputation
Reputation
# ADifference from ingroup
m <- lmer( scale(toInG) ~ RepN * scale(reput)+ ( scale(reput) | subID)+ (1|issues), data = longDf, control=lmerControl(optimizer="bobyqa",
optCtrl=list(maxfun=2e5)))
summary(m)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: scale(toInG) ~ RepN * scale(reput) + (scale(reput) | subID) +
## (1 | issues)
## Data: longDf
## Control: lmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 2e+05))
##
## REML criterion at convergence: 101139.8
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -3.0242 -0.7614 0.1752 0.7821 2.5945
##
## Random effects:
## Groups Name Variance Std.Dev. Corr
## subID (Intercept) 0.132357 0.36381
## scale(reput) 0.007459 0.08636 0.63
## issues (Intercept) 0.051248 0.22638
## Residual 0.808220 0.89901
## Number of obs: 37917, groups: subID, 384; issues, 100
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 4.567e-02 4.003e-02 4.262e+02 1.141 0.255
## RepNNon -2.482e-02 4.642e-02 3.811e+02 -0.535 0.593
## RepNOut -1.180e-01 4.725e-02 3.810e+02 -2.497 0.013 *
## scale(reput) 1.351e-02 2.518e-02 1.364e+02 0.537 0.592
## RepNNon:scale(reput) 4.821e-03 1.550e-02 3.764e+02 0.311 0.756
## RepNOut:scale(reput) -5.058e-04 1.579e-02 3.774e+02 -0.032 0.974
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) RepNNn RepNOt scl(r) RNN:()
## RepNNon -0.586
## RepNOut -0.576 0.497
## scale(rept) 0.154 -0.133 -0.130
## RpNNn:scl() -0.250 0.426 0.212 -0.312
## RpNOt:scl() -0.245 0.211 0.426 -0.306 0.497# Difference from outgroup
m <- lmer( scale(eval) ~ RepN * scale(reput) +( scale(reput) | subID) + ( 1 | issues), data = longDf, control=lmerControl(optimizer="bobyqa",
optCtrl=list(maxfun=2e5)))
summary(m)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: scale(eval) ~ RepN * scale(reput) + (scale(reput) | subID) +
## (1 | issues)
## Data: longDf
## Control: lmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 2e+05))
##
## REML criterion at convergence: 94083.1
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -3.3524 -0.6820 0.1458 0.7029 3.8134
##
## Random effects:
## Groups Name Variance Std.Dev. Corr
## subID (Intercept) 0.03940 0.1985
## scale(reput) 0.05829 0.2414 0.51
## issues (Intercept) 0.24204 0.4920
## Residual 0.66344 0.8145
## Number of obs: 37917, groups: subID, 384; issues, 100
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 0.01802 0.05271 126.55868 0.342 0.7330
## RepNNon -0.02050 0.02659 380.21845 -0.771 0.4411
## RepNOut -0.03805 0.02707 380.08665 -1.406 0.1607
## scale(reput) -0.01321 0.05408 139.51488 -0.244 0.8073
## RepNNon:scale(reput) -0.01479 0.03158 380.91489 -0.469 0.6397
## RepNOut:scale(reput) -0.06250 0.03214 380.99587 -1.944 0.0526 .
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) RepNNn RepNOt scl(r) RNN:()
## RepNNon -0.255
## RepNOut -0.251 0.497
## scale(rept) 0.067 -0.132 -0.130
## RpNNn:scl() -0.114 0.448 0.222 -0.295
## RpNOt:scl() -0.112 0.222 0.448 -0.290 0.497No effect of reput interacting with Rep
Honesty interacting with rep IV
# Approval evaluations
m <- lmer( scale(toInG) ~ RepN * scale(honest) + ( scale(honest) | subID)+ (1|issues), data = longDf, control=lmerControl(optimizer="bobyqa",
optCtrl=list(maxfun=2e5)))
summary(m)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: scale(toInG) ~ RepN * scale(honest) + (scale(honest) | subID) +
## (1 | issues)
## Data: longDf
## Control: lmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 2e+05))
##
## REML criterion at convergence: 99804.1
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -3.4940 -0.7303 0.1817 0.7487 2.7291
##
## Random effects:
## Groups Name Variance Std.Dev. Corr
## subID (Intercept) 0.13265 0.3642
## scale(honest) 0.04232 0.2057 0.67
## issues (Intercept) 0.04944 0.2223
## Residual 0.77228 0.8788
## Number of obs: 37917, groups: subID, 384; issues, 100
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 0.04579 0.03979 431.13826 1.151 0.2505
## RepNNon -0.02528 0.04640 381.10840 -0.545 0.5862
## RepNOut -0.11803 0.04723 381.04645 -2.499 0.0129 *
## scale(honest) -0.02166 0.02972 251.55519 -0.729 0.4669
## RepNNon:scale(honest) -0.00914 0.02773 381.14950 -0.330 0.7418
## RepNOut:scale(honest) -0.06357 0.02822 380.79846 -2.253 0.0248 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) RepNNn RepNOt scl(h) RNN:()
## RepNNon -0.590
## RepNOut -0.579 0.497
## scale(hnst) 0.328 -0.281 -0.276
## RpNNn:scl() -0.352 0.596 0.296 -0.472
## RpNOt:scl() -0.346 0.296 0.597 -0.464 0.497plot_model(m, type = "pred", terms = c("honest", "RepN"))
# Difference from ingroup
m <- lmer( scale(eval) ~ RepN * scale(honest) + ( scale(honest) | subID) + ( 1 | issues), data = longDf, control=lmerControl(optimizer="bobyqa",
optCtrl=list(maxfun=2e5)))
summary(m)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: scale(eval) ~ RepN * scale(honest) + (scale(honest) | subID) +
## (1 | issues)
## Data: longDf
## Control: lmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 2e+05))
##
## REML criterion at convergence: 96290.2
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -3.2232 -0.7012 0.1541 0.7181 3.1324
##
## Random effects:
## Groups Name Variance Std.Dev. Corr
## subID (Intercept) 0.038732 0.19680
## scale(honest) 0.007684 0.08766 -0.40
## issues (Intercept) 0.147450 0.38399
## Residual 0.714691 0.84539
## Number of obs: 37917, groups: subID, 384; issues, 100
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 0.018139 0.042790 145.191834 0.424 0.672
## RepNNon -0.020542 0.026549 380.058991 -0.774 0.440
## RepNOut -0.038456 0.027021 379.903996 -1.423 0.156
## scale(honest) 0.300206 0.039878 110.958230 7.528 1.46e-11 ***
## RepNNon:scale(honest) 0.010302 0.015127 381.207945 0.681 0.496
## RepNOut:scale(honest) 0.009613 0.015387 380.258990 0.625 0.533
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) RepNNn RepNOt scl(h) RNN:()
## RepNNon -0.314
## RepNOut -0.308 0.497
## scale(hnst) -0.031 0.050 0.049
## RpNNn:scl() 0.082 -0.263 -0.130 -0.192
## RpNOt:scl() 0.081 -0.131 -0.263 -0.189 0.497Participants falsify more for issues that people perceive others would lie about when they think they are interacting with outgroup *In line with prediction
Changeability
# Approval evaluations
m <- lmer( scale(toInG) ~ RepN * scale(change) + ( scale(change) | subID)+ (1|issues), data = longDf, control=lmerControl(optimizer="bobyqa",
optCtrl=list(maxfun=2e5)))
summary(m)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: scale(toInG) ~ RepN * scale(change) + (scale(change) | subID) +
## (1 | issues)
## Data: longDf
## Control: lmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 2e+05))
##
## REML criterion at convergence: 99847.9
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -3.1625 -0.7267 0.1548 0.7632 3.0391
##
## Random effects:
## Groups Name Variance Std.Dev. Corr
## subID (Intercept) 0.13198 0.3633
## scale(change) 0.03906 0.1976 0.78
## issues (Intercept) 0.05105 0.2259
## Residual 0.77533 0.8805
## Number of obs: 37917, groups: subID, 384; issues, 100
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 0.04544 0.03993 426.65121 1.138 0.2558
## RepNNon -0.02506 0.04630 381.13339 -0.541 0.5887
## RepNOut -0.11757 0.04713 381.06484 -2.495 0.0130 *
## scale(change) 0.06349 0.02957 236.98264 2.147 0.0328 *
## RepNNon:scale(change) -0.03614 0.02682 380.49970 -1.348 0.1786
## RepNOut:scale(change) -0.08243 0.02729 380.14423 -3.020 0.0027 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) RepNNn RepNOt scl(c) RNN:()
## RepNNon -0.586
## RepNOut -0.576 0.497
## scale(chng) 0.366 -0.316 -0.310
## RpNNn:scl() -0.404 0.689 0.342 -0.459
## RpNOt:scl() -0.397 0.342 0.689 -0.451 0.497plot_model(m, type = "pred", terms = c("change", "RepN"))
# Difference from ingroup
m <- lmer( scale(eval) ~ RepN * scale(change) + ( scale(change) | subID) + ( 1 | issues), data = longDf, control=lmerControl(optimizer="bobyqa",
optCtrl=list(maxfun=2e5)))
summary(m)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: scale(eval) ~ RepN * scale(change) + (scale(change) | subID) +
## (1 | issues)
## Data: longDf
## Control: lmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 2e+05))
##
## REML criterion at convergence: 96225.6
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -3.2642 -0.7066 0.1483 0.7078 3.4877
##
## Random effects:
## Groups Name Variance Std.Dev. Corr
## subID (Intercept) 0.03860 0.1965
## scale(change) 0.01054 0.1027 0.42
## issues (Intercept) 0.20137 0.4487
## Residual 0.71177 0.8437
## Number of obs: 37917, groups: subID, 384; issues, 100
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 0.017966 0.048672 132.097784 0.369 0.713
## RepNNon -0.020704 0.026503 380.195264 -0.781 0.435
## RepNOut -0.038267 0.026975 380.036388 -1.419 0.157
## scale(change) 0.202417 0.046383 109.673338 4.364 2.9e-05 ***
## RepNNon:scale(change) 0.011211 0.016501 381.136358 0.679 0.497
## RepNOut:scale(change) -0.008847 0.016787 380.497986 -0.527 0.598
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) RepNNn RepNOt scl(c) RNN:()
## RepNNon -0.275
## RepNOut -0.271 0.497
## scale(chng) 0.029 -0.053 -0.052
## RpNNn:scl() -0.081 0.295 0.146 -0.180
## RpNOt:scl() -0.080 0.146 0.295 -0.177 0.497Participants falsify more for issues that people perceive that others are more likely to lie about when they think they are interacting with outgroup (relative to ingroup) *We predicted opposite effect
Hypothesis 2: Between-subjects information main effect
# info main effect full
m <- lmer( scale(toInG) ~Info +( 1| subID) + (1|issues), data = longDf, control=lmerControl(optimizer="bobyqa",
optCtrl=list(maxfun=2e5)))
summary(m)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: scale(toInG) ~ Info + (1 | subID) + (1 | issues)
## Data: longDf
## Control: lmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 2e+05))
##
## REML criterion at convergence: 101293.2
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.9806 -0.7632 0.1772 0.7859 2.6137
##
## Random effects:
## Groups Name Variance Std.Dev.
## subID (Intercept) 0.13261 0.3642
## issues (Intercept) 0.05094 0.2257
## Residual 0.81563 0.9031
## Number of obs: 37917, groups: subID, 384; issues, 100
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) -7.146e-04 2.960e-02 2.400e+02 -0.024 0.9808
## Info1 4.110e-02 1.915e-02 3.821e+02 2.146 0.0325 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr)
## Info1 0.000plot_model(m, type = "pred", terms = c("Info"))
m <- lmer( scale(eval) ~Info +( 1| subID)+ (1|issues) , data = longDf, control=lmerControl(optimizer="bobyqa",
optCtrl=list(maxfun=2e5)))
summary(m)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: scale(eval) ~ Info + (1 | subID) + (1 | issues)
## Data: longDf
## Control: lmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 2e+05))
##
## REML criterion at convergence: 96472.5
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -3.2384 -0.7108 0.1590 0.7224 3.1868
##
## Random effects:
## Groups Name Variance Std.Dev.
## subID (Intercept) 0.03868 0.1967
## issues (Intercept) 0.24104 0.4910
## Residual 0.72241 0.8499
## Number of obs: 37917, groups: subID, 384; issues, 100
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) -0.001276 0.050301 107.316302 -0.025 0.980
## Info1 -0.006441 0.010945 381.027499 -0.588 0.557
##
## Correlation of Fixed Effects:
## (Intr)
## Info1 0.000Participants falsify more when they are shown consensus compared to when they are not shown any info *not sure we predicted this.
Further analyses of hypothesis 2; incorporating issue-specific features
No models with Info and issue-specific features are significant
Hypothesis 3: Strength of affiliation moderating IV #1 & #2
Need to run still
Hypothesis 4: Interaction of between-subjects information and reputation effects
For everyone
m <- lmer( scale(toInG) ~ RepN + Info + ( 1| subID)+ (1|issues), data = longDf, control=lmerControl(optimizer="bobyqa",
optCtrl=list(maxfun=2e5)))
summary(m)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: scale(toInG) ~ RepN + Info + (1 | subID) + (1 | issues)
## Data: longDf
## Control: lmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 2e+05))
##
## REML criterion at convergence: 101295.8
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.9828 -0.7637 0.1777 0.7851 2.6175
##
## Random effects:
## Groups Name Variance Std.Dev.
## subID (Intercept) 0.13106 0.3620
## issues (Intercept) 0.05094 0.2257
## Residual 0.81563 0.9031
## Number of obs: 37917, groups: subID, 384; issues, 100
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 0.04252 0.03990 427.40646 1.066 0.2872
## RepNNon -0.02192 0.04624 380.10891 -0.474 0.6358
## RepNOut -0.11145 0.04716 380.03101 -2.363 0.0186 *
## Info1 0.03791 0.01910 380.05935 1.985 0.0479 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) RepNNn RepNOt
## RepNNon -0.587
## RepNOut -0.577 0.498
## Info1 -0.041 0.035 0.071plot_model(m, type = "pred", terms = c("RepN", "Info"))
m <- lmer( scale(eval) ~ RepN + Info +( 1| subID) + (1|issues), data = longDf, control=lmerControl(optimizer="bobyqa",
optCtrl=list(maxfun=2e5)))
summary(m)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: scale(eval) ~ RepN + Info + (1 | subID) + (1 | issues)
## Data: longDf
## Control: lmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 2e+05))
##
## REML criterion at convergence: 96481.4
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -3.2345 -0.7105 0.1599 0.7220 3.1872
##
## Random effects:
## Groups Name Variance Std.Dev.
## subID (Intercept) 0.03865 0.1966
## issues (Intercept) 0.24104 0.4910
## Residual 0.72241 0.8499
## Number of obs: 37917, groups: subID, 384; issues, 100
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 0.018842 0.052610 127.773155 0.358 0.721
## RepNNon -0.021380 0.026566 379.122361 -0.805 0.421
## RepNOut -0.039857 0.027089 378.943618 -1.471 0.142
## Info1 -0.007583 0.010970 379.002616 -0.691 0.490
##
## Correlation of Fixed Effects:
## (Intr) RepNNn RepNOt
## RepNNon -0.256
## RepNOut -0.251 0.498
## Info1 -0.018 0.036 0.071m <- lmer( scale(toInG) ~ RepN * Info + ( 1| subID)+ (1|issues), data = longDf, control=lmerControl(optimizer="bobyqa",
optCtrl=list(maxfun=2e5)))
summary(m)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: scale(toInG) ~ RepN * Info + (1 | subID) + (1 | issues)
## Data: longDf
## Control: lmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 2e+05))
##
## REML criterion at convergence: 101304.6
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.9827 -0.7635 0.1776 0.7850 2.6183
##
## Random effects:
## Groups Name Variance Std.Dev.
## subID (Intercept) 0.13176 0.3630
## issues (Intercept) 0.05094 0.2257
## Residual 0.81563 0.9031
## Number of obs: 37917, groups: subID, 384; issues, 100
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 0.042848 0.040037 427.574258 1.070 0.2851
## RepNNon -0.022247 0.046416 378.115145 -0.479 0.6320
## RepNOut -0.110975 0.047339 378.030208 -2.344 0.0196 *
## Info1 0.034079 0.033069 378.121933 1.031 0.3034
## RepNNon:Info1 -0.000735 0.046416 378.115859 -0.016 0.9874
## RepNOut:Info1 0.012777 0.047339 378.029904 0.270 0.7874
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) RepNNn RepNOt Info1 RNN:I1
## RepNNon -0.588
## RepNOut -0.577 0.498
## Info1 -0.071 0.061 0.060
## RepNNn:Inf1 0.050 -0.043 -0.043 -0.712
## RepNOt:Inf1 0.049 -0.043 0.004 -0.699 0.498m <- lmer( scale(eval) ~ RepN * Info +( 1| subID) + (1|issues), data = longDf, control=lmerControl(optimizer="bobyqa",
optCtrl=list(maxfun=2e5)))
summary(m)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: scale(eval) ~ RepN * Info + (1 | subID) + (1 | issues)
## Data: longDf
## Control: lmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 2e+05))
##
## REML criterion at convergence: 96490.9
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -3.2363 -0.7108 0.1600 0.7222 3.1909
##
## Random effects:
## Groups Name Variance Std.Dev.
## subID (Intercept) 0.0387 0.1967
## issues (Intercept) 0.2410 0.4910
## Residual 0.7224 0.8499
## Number of obs: 37917, groups: subID, 384; issues, 100
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 0.017858 0.052630 127.954653 0.339 0.735
## RepNNon -0.020405 0.026611 377.049170 -0.767 0.444
## RepNOut -0.038116 0.027138 376.850196 -1.405 0.161
## Info1 0.003872 0.018959 377.085603 0.204 0.838
## RepNNon:Info1 -0.030450 0.026611 377.051146 -1.144 0.253
## RepNOut:Info1 -0.002933 0.027138 376.849377 -0.108 0.914
##
## Correlation of Fixed Effects:
## (Intr) RepNNn RepNOt Info1 RNN:I1
## RepNNon -0.257
## RepNOut -0.252 0.498
## Info1 -0.031 0.061 0.060
## RepNNn:Inf1 0.022 -0.043 -0.043 -0.712
## RepNOt:Inf1 0.022 -0.043 0.004 -0.699 0.498We get 2 main effects but no sig interaction with info + reput
Incorporating issue-specific features
Political
No 3 way with rep * info * politic
Reputation
No 3 way info * rep * reput
Honesty
No sig 3 way info * rep * honest
Changeability
No sig 3 way info * rep * change
Willingness to Self-Censor
For Reputation
m <- lmer( scale(toInG) ~ RepN * scale(WSC) + ( 1 | subID), data = longDf, control=lmerControl(optimizer="bobyqa",
optCtrl=list(maxfun=2e5)))
summary(m)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: scale(toInG) ~ RepN * scale(WSC) + (1 | subID)
## Data: longDf
## Control: lmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 2e+05))
##
## REML criterion at convergence: 88233.4
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.7463 -0.7851 0.2057 0.7806 2.0471
##
## Random effects:
## Groups Name Variance Std.Dev.
## subID (Intercept) 0.1276 0.3571
## Residual 0.8722 0.9339
## Number of obs: 32325, groups: subID, 327
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 0.047652 0.034594 321.080774 1.377 0.1693
## RepNNon -0.025040 0.049333 321.023985 -0.508 0.6121
## RepNOut -0.128801 0.050421 321.050468 -2.555 0.0111 *
## scale(WSC) -0.082331 0.037365 321.371326 -2.203 0.0283 *
## RepNNon:scale(WSC) 0.002895 0.049763 321.180135 0.058 0.9536
## RepNOut:scale(WSC) 0.064408 0.052345 321.182805 1.230 0.2194
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) RepNNn RepNOt s(WSC) RNN:(W
## RepNNon -0.701
## RepNOut -0.686 0.481
## scale(WSC) 0.017 -0.012 -0.012
## RpNNn:(WSC) -0.013 -0.030 0.009 -0.751
## RpNOt:(WSC) -0.012 0.008 0.049 -0.714 0.536For Info
m <- lmer( scale(toInG) ~ Info * scale(WSC) + ( 1 | subID), data = longDf, control=lmerControl(optimizer="bobyqa",
optCtrl=list(maxfun=2e5)))
summary(m)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: scale(toInG) ~ Info * scale(WSC) + (1 | subID)
## Data: longDf
## Control: lmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 2e+05))
##
## REML criterion at convergence: 88233.3
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.7407 -0.7851 0.2033 0.7873 2.0530
##
## Random effects:
## Groups Name Variance Std.Dev.
## subID (Intercept) 0.1289 0.3590
## Residual 0.8722 0.9339
## Number of obs: 32325, groups: subID, 327
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) -0.00423 0.02057 323.01317 -0.206 0.83720
## Info1 0.03843 0.02057 323.01317 1.868 0.06269 .
## scale(WSC) -0.06144 0.02057 323.07004 -2.986 0.00304 **
## Info1:scale(WSC) 0.01953 0.02057 323.07004 0.949 0.34319
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) Info1 s(WSC)
## Info1 -0.009
## scale(WSC) 0.001 -0.069
## Inf1:s(WSC) -0.069 0.001 -0.007Info and Rep
m <- lmer( scale(toInG) ~ Info * scale(WSC) * Rep + ( 1 | subID), data = longDf, control=lmerControl(optimizer="bobyqa",
optCtrl=list(maxfun=2e5)))
summary(m)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: scale(toInG) ~ Info * scale(WSC) * Rep + (1 | subID)
## Data: longDf
## Control: lmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 2e+05))
##
## REML criterion at convergence: 88267.3
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.7386 -0.7832 0.2032 0.7858 2.0532
##
## Random effects:
## Groups Name Variance Std.Dev.
## subID (Intercept) 0.1285 0.3585
## Residual 0.8722 0.9339
## Number of obs: 32325, groups: subID, 327
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) -7.578e-03 2.086e-02 3.150e+02 -0.363 0.71665
## Info1 3.100e-02 2.086e-02 3.150e+02 1.486 0.13829
## scale(WSC) -6.384e-02 2.105e-02 3.151e+02 -3.032 0.00263 **
## Rep1 1.904e-02 2.824e-02 3.150e+02 0.674 0.50050
## Rep2 2.953e-02 2.918e-02 3.150e+02 1.012 0.31233
## Info1:scale(WSC) 2.319e-02 2.105e-02 3.151e+02 1.101 0.27158
## Info1:Rep1 5.939e-02 2.824e-02 3.150e+02 2.103 0.03622 *
## Info1:Rep2 -3.017e-03 2.918e-02 3.150e+02 -0.103 0.91774
## scale(WSC):Rep1 2.317e-02 2.886e-02 3.151e+02 0.803 0.42274
## scale(WSC):Rep2 -1.675e-02 2.841e-02 3.150e+02 -0.590 0.55582
## Info1:scale(WSC):Rep1 -7.138e-04 2.886e-02 3.151e+02 -0.025 0.98028
## Info1:scale(WSC):Rep2 -1.415e-02 2.841e-02 3.150e+02 -0.498 0.61879
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) Info1 sc(WSC) Rep1 Rep2 In1:(WSC) In1:R1 In1:R2
## Info1 -0.012
## scale(WSC) -0.002 -0.060
## Rep1 -0.124 -0.008 0.057
## Rep2 -0.031 0.009 -0.051 -0.417
## Inf1:s(WSC) -0.060 -0.002 -0.027 -0.025 0.009
## Info1:Rep1 -0.008 -0.124 -0.025 -0.019 0.006 0.057
## Info1:Rep2 0.009 -0.031 0.009 0.006 -0.006 -0.051 -0.417
## scl(WSC):R1 0.056 -0.025 -0.088 0.039 -0.002 0.015 -0.083 0.043
## scl(WSC):R2 -0.053 0.009 -0.133 -0.002 -0.041 0.037 0.045 -0.057
## I1:(WSC):R1 -0.025 0.056 0.015 -0.083 0.043 -0.088 0.039 -0.002
## I1:(WSC):R2 0.009 -0.053 0.037 0.045 -0.057 -0.133 -0.002 -0.041
## s(WSC):R1 s(WSC):R2 I1:(WSC):R1
## Info1
## scale(WSC)
## Rep1
## Rep2
## Inf1:s(WSC)
## Info1:Rep1
## Info1:Rep2
## scl(WSC):R1
## scl(WSC):R2 -0.378
## I1:(WSC):R1 -0.018 -0.023
## I1:(WSC):R2 -0.023 -0.003 -0.378Approval Motivation
For Reputation Manip
m <- lmer( scale(toInG) ~ RepN * scale(MLAM) + ( 1 | subID), data = longDf, control=lmerControl(optimizer="bobyqa",
optCtrl=list(maxfun=2e5)))
summary(m)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: scale(toInG) ~ RepN * scale(MLAM) + (1 | subID)
## Data: longDf
## Control: lmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 2e+05))
##
## REML criterion at convergence: 90265.6
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.7630 -0.7773 0.2072 0.8010 2.0563
##
## Random effects:
## Groups Name Variance Std.Dev.
## subID (Intercept) 0.1322 0.3636
## Residual 0.8657 0.9304
## Number of obs: 33156, groups: subID, 336
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 0.05723 0.03581 330.17848 1.598 0.11092
## RepNNon -0.03385 0.04987 330.18599 -0.679 0.49775
## RepNOut -0.13531 0.05074 330.07579 -2.666 0.00804 **
## scale(MLAM) -0.05014 0.03629 330.02943 -1.382 0.16796
## RepNNon:scale(MLAM) -0.01250 0.05225 330.48573 -0.239 0.81114
## RepNOut:scale(MLAM) 0.03152 0.04908 329.95179 0.642 0.52118
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) RepNNn RepNOt s(MLAM RNN:(M
## RepNNon -0.718
## RepNOut -0.706 0.507
## scale(MLAM) -0.021 0.015 0.015
## RpNN:(MLAM) 0.014 -0.002 -0.010 -0.694
## RpNO:(MLAM) 0.015 -0.011 -0.009 -0.739 0.513For Info Manip
m <- lmer( scale(toInG) ~ Info * scale(MLAM) + ( 1 | subID), data = longDf, control=lmerControl(optimizer="bobyqa",
optCtrl=list(maxfun=2e5)))
summary(m)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: scale(toInG) ~ Info * scale(MLAM) + (1 | subID)
## Data: longDf
## Control: lmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 2e+05))
##
## REML criterion at convergence: 90261.6
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.7722 -0.7806 0.2109 0.8004 2.0578
##
## Random effects:
## Groups Name Variance Std.Dev.
## subID (Intercept) 0.1319 0.3631
## Residual 0.8657 0.9304
## Number of obs: 33156, groups: subID, 336
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 0.002686 0.020523 332.072584 0.131 0.8960
## Info1 0.054252 0.020523 332.072584 2.643 0.0086 **
## scale(MLAM) -0.046325 0.020617 332.156212 -2.247 0.0253 *
## Info1:scale(MLAM) 0.011704 0.020617 332.156212 0.568 0.5706
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) Info1 s(MLAM
## Info1 0.035
## scale(MLAM) 0.005 -0.070
## Inf1:(MLAM) -0.070 0.005 -0.101Info and Rep Manip
m <- lmer( scale(toInG) ~ Info * scale(MLAM) * RepN + ( 1 | subID), data = longDf, control=lmerControl(optimizer="bobyqa",
optCtrl=list(maxfun=2e5)))
summary(m)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: scale(toInG) ~ Info * scale(MLAM) * RepN + (1 | subID)
## Data: longDf
## Control: lmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 2e+05))
##
## REML criterion at convergence: 90285.6
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.7664 -0.7796 0.2079 0.8008 2.0651
##
## Random effects:
## Groups Name Variance Std.Dev.
## subID (Intercept) 0.1309 0.3618
## Residual 0.8657 0.9304
## Number of obs: 33156, groups: subID, 336
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 0.057661 0.035662 324.147689 1.617 0.10688
## Info1 0.055306 0.035662 324.147689 1.551 0.12192
## scale(MLAM) -0.045051 0.036786 324.023263 -1.225 0.22159
## RepNNon -0.033586 0.049669 324.156827 -0.676 0.49939
## RepNOut -0.139769 0.051166 324.044493 -2.732 0.00665 **
## Info1:scale(MLAM) -0.020490 0.036786 324.023263 -0.557 0.57791
## Info1:RepNNon -0.016028 0.049669 324.156827 -0.323 0.74712
## Info1:RepNOut 0.008615 0.051166 324.044493 0.168 0.86639
## scale(MLAM):RepNNon -0.019392 0.052501 324.415740 -0.369 0.71210
## scale(MLAM):RepNOut 0.011401 0.049922 323.961736 0.228 0.81950
## Info1:scale(MLAM):RepNNon 0.057698 0.052501 324.415740 1.099 0.27259
## Info1:scale(MLAM):RepNOut 0.055675 0.049922 323.961736 1.115 0.26558
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) Info1 sc(MLAM) RepNNn RepNOt In1:(MLAM) I1:RNN I1:RNO
## Info1 0.017
## scale(MLAM) -0.025 0.027
## RepNNon -0.718 -0.012 0.018
## RepNOut -0.697 -0.012 0.017 0.500
## Inf1:(MLAM) 0.027 -0.025 -0.188 -0.019 -0.019
## Info1:RpNNn -0.012 -0.718 -0.019 0.021 0.008 0.018
## Info1:RpNOt -0.012 -0.697 -0.019 0.008 0.040 0.017 0.500
## s(MLAM):RNN 0.018 -0.019 -0.701 -0.004 -0.012 0.132 0.009 0.013
## s(MLAM):RNO 0.018 -0.020 -0.737 -0.013 -0.009 0.138 0.014 -0.088
## I1:(MLAM):RNN -0.019 0.018 0.132 0.009 0.013 -0.701 -0.004 -0.012
## I1:(MLAM):RNO -0.020 0.018 0.138 0.014 -0.088 -0.737 -0.013 -0.009
## s(MLAM):RNN s(MLAM):RNO I1:(MLAM):RNN
## Info1
## scale(MLAM)
## RepNNon
## RepNOut
## Inf1:(MLAM)
## Info1:RpNNn
## Info1:RpNOt
## s(MLAM):RNN
## s(MLAM):RNO 0.516
## I1:(MLAM):RNN -0.112 -0.097
## I1:(MLAM):RNO -0.097 -0.138 0.516For Reputation Normative Data
m <- lmer( toInG ~ reputF * MLAM + ( reputF | subID), data = longDf, control=lmerControl(optimizer="bobyqa",
optCtrl=list(maxfun=2e5)))
summary(m)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: toInG ~ reputF * MLAM + (reputF | subID)
## Data: longDf
## Control: lmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 2e+05))
##
## REML criterion at convergence: 139090.9
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.8969 -0.7701 0.2082 0.7922 2.1346
##
## Random effects:
## Groups Name Variance Std.Dev. Corr
## subID (Intercept) 2.7894 1.6701
## reputF 0.1835 0.4283 -0.90
## Residual 3.7588 1.9388
## Number of obs: 33156, groups: subID, 336
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 7.07406 0.55371 338.10470 12.776 < 2e-16 ***
## reputF -0.51794 0.13946 327.48846 -3.714 0.000240 ***
## MLAM -0.89577 0.20716 338.14269 -4.324 2.02e-05 ***
## reputF:MLAM 0.19688 0.05222 328.68336 3.770 0.000193 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) reputF MLAM
## reputF -0.960
## MLAM -0.962 0.924
## reputF:MLAM 0.923 -0.962 -0.960ggpredict(m, c("MLAM","reputF")) %>% plot()
- highly significant interaction of approval motivation with reputation normative data
Self They
For Reputation
m <- lmer( scale(toInG) ~ RepN * scale(SGO.st) + ( 1 | subID), data = longDf, control=lmerControl(optimizer="bobyqa",
optCtrl=list(maxfun=2e5)))
summary(m)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: scale(toInG) ~ RepN * scale(SGO.st) + (1 | subID)
## Data: longDf
## Control: lmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 2e+05))
##
## REML criterion at convergence: 87641.1
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.7475 -0.7803 0.2002 0.8034 2.0395
##
## Random effects:
## Groups Name Variance Std.Dev.
## subID (Intercept) 0.1247 0.3531
## Residual 0.8754 0.9356
## Number of obs: 32068, groups: subID, 325
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 0.047699 0.035172 319.121075 1.356 0.1760
## RepNNon -0.048822 0.049296 319.102414 -0.990 0.3227
## RepNOut -0.115522 0.050102 319.069250 -2.306 0.0218 *
## scale(SGO.st) -0.084686 0.040127 318.971077 -2.110 0.0356 *
## RepNNon:scale(SGO.st) -0.002576 0.053677 318.923674 -0.048 0.9617
## RepNOut:scale(SGO.st) 0.037497 0.050858 319.032600 0.737 0.4615
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) RepNNn RepNOt s(SGO. RNN:(S
## RepNNon -0.713
## RepNOut -0.702 0.501
## scl(SGO.st) 0.016 -0.011 -0.011
## RpNN:(SGO.) -0.012 0.019 0.008 -0.748
## RpNO:(SGO.) -0.013 0.009 -0.005 -0.789 0.590For Info
m <- lmer( scale(toInG) ~ Info * scale(SGO.st) + ( 1 | subID), data = longDf, control=lmerControl(optimizer="bobyqa",
optCtrl=list(maxfun=2e5)))
summary(m)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: scale(toInG) ~ Info * scale(SGO.st) + (1 | subID)
## Data: longDf
## Control: lmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 2e+05))
##
## REML criterion at convergence: 87638.6
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.7571 -0.7778 0.2001 0.8035 2.0381
##
## Random effects:
## Groups Name Variance Std.Dev.
## subID (Intercept) 0.1250 0.3535
## Residual 0.8754 0.9356
## Number of obs: 32068, groups: subID, 325
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) -0.004747 0.020311 321.081399 -0.234 0.815351
## Info1 0.037185 0.020311 321.081399 1.831 0.068058 .
## scale(SGO.st) -0.071905 0.020303 321.009834 -3.542 0.000457 ***
## Info1:scale(SGO.st) -0.006944 0.020303 321.009834 -0.342 0.732558
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) Info1 s(SGO.
## Info1 0.028
## scl(SGO.st) -0.001 -0.029
## Inf1:(SGO.) -0.029 -0.001 0.008Info and RepN
m <- lmer( scale(toInG) ~ Info * scale(SGO.st) * RepN + ( 1 | subID), data = longDf, control=lmerControl(optimizer="bobyqa",
optCtrl=list(maxfun=2e5)))
summary(m)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: scale(toInG) ~ Info * scale(SGO.st) * RepN + (1 | subID)
## Data: longDf
## Control: lmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 2e+05))
##
## REML criterion at convergence: 87661.1
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.7493 -0.7778 0.2013 0.8055 2.0406
##
## Random effects:
## Groups Name Variance Std.Dev.
## subID (Intercept) 0.1234 0.3513
## Residual 0.8754 0.9356
## Number of obs: 32068, groups: subID, 325
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 0.037858 0.035494 313.111287 1.067 0.28697
## Info1 0.032293 0.035494 313.111287 0.910 0.36362
## scale(SGO.st) -0.112610 0.043119 313.182138 -2.612 0.00945
## RepNNon -0.035381 0.049443 313.087113 -0.716 0.47478
## RepNOut -0.103567 0.050404 313.037290 -2.055 0.04073
## Info1:scale(SGO.st) 0.066247 0.043119 313.182138 1.536 0.12546
## Info1:RepNNon 0.021121 0.049443 313.087113 0.427 0.66953
## Info1:RepNOut -0.007043 0.050404 313.037290 -0.140 0.88896
## scale(SGO.st):RepNNon 0.018784 0.055915 313.043674 0.336 0.73714
## scale(SGO.st):RepNOut 0.065990 0.053488 313.125954 1.234 0.21822
## Info1:scale(SGO.st):RepNNon -0.127393 0.055915 313.043674 -2.278 0.02338
## Info1:scale(SGO.st):RepNOut -0.072127 0.053488 313.125954 -1.348 0.17848
##
## (Intercept)
## Info1
## scale(SGO.st) **
## RepNNon
## RepNOut *
## Info1:scale(SGO.st)
## Info1:RepNNon
## Info1:RepNOut
## scale(SGO.st):RepNNon
## scale(SGO.st):RepNOut
## Info1:scale(SGO.st):RepNNon *
## Info1:scale(SGO.st):RepNOut
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) Info1 sc(SGO.) RepNNn RepNOt In1:(SGO.) I1:RNN I1:RNO
## Info1 -0.052
## scl(SGO.st) 0.076 -0.160
## RepNNon -0.718 0.037 -0.055
## RepNOut -0.704 0.036 -0.054 0.506
## Inf1:(SGO.) -0.160 0.076 -0.352 0.115 0.113
## Info1:RpNNn 0.037 -0.718 0.115 -0.019 -0.026 -0.055
## Info1:RpNOt 0.036 -0.704 0.113 -0.026 0.025 -0.054 0.506
## s(SGO.):RNN -0.059 0.123 -0.771 0.050 0.041 0.272 -0.109 -0.087
## s(SGO.):RNO -0.062 0.129 -0.806 0.044 0.038 0.284 -0.093 -0.062
## I1:(SGO.):RNN 0.123 -0.059 0.272 -0.109 -0.087 -0.771 0.050 0.041
## I1:(SGO.):RNO 0.129 -0.062 0.284 -0.093 -0.062 -0.806 0.044 0.038
## s(SGO.):RNN s(SGO.):RNO I1:(SGO.):RNN
## Info1
## scl(SGO.st)
## RepNNon
## RepNOut
## Inf1:(SGO.)
## Info1:RpNNn
## Info1:RpNOt
## s(SGO.):RNN
## s(SGO.):RNO 0.622
## I1:(SGO.):RNN -0.183 -0.219
## I1:(SGO.):RNO -0.219 -0.169 0.622- significant three way interaction between self-they overlap, info, and non partisan
For Reputation Normative Data
m <- lmer( toInG ~ reputF * SGO.st + ( reputF | subID), data = longDf, control=lmerControl(optimizer="bobyqa",
optCtrl=list(maxfun=2e5)))
summary(m)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: toInG ~ reputF * SGO.st + (reputF | subID)
## Data: longDf
## Control: lmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 2e+05))
##
## REML criterion at convergence: 134880.8
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.8592 -0.7826 0.1968 0.8016 2.1303
##
## Random effects:
## Groups Name Variance Std.Dev. Corr
## subID (Intercept) 2.7541 1.6595
## reputF 0.1814 0.4259 -0.90
## Residual 3.8028 1.9501
## Number of obs: 32068, groups: subID, 325
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 5.55342 0.29309 329.15296 18.948 < 2e-16 ***
## reputF -0.15312 0.07378 318.34963 -2.075 0.03877 *
## SGO.st -0.28709 0.09560 324.74311 -3.003 0.00288 **
## reputF:SGO.st 0.04952 0.02415 317.94518 2.050 0.04114 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) reputF SGO.st
## reputF -0.962
## SGO.st -0.850 0.819
## rptF:SGO.st 0.816 -0.850 -0.962ggpredict(m, c("SGO.st","reputF")) %>% plot()
- sig interaction with reputation normative data
Self We
For Reputation
m <- lmer( scale(toInG) ~ RepN * scale(SGO.sw) + ( 1 | subID), data = longDf, control=lmerControl(optimizer="bobyqa",
optCtrl=list(maxfun=2e5)))
summary(m)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: scale(toInG) ~ RepN * scale(SGO.sw) + (1 | subID)
## Data: longDf
## Control: lmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 2e+05))
##
## REML criterion at convergence: 86826.8
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.7460 -0.7821 0.1961 0.8031 2.0434
##
## Random effects:
## Groups Name Variance Std.Dev.
## subID (Intercept) 0.1308 0.3616
## Residual 0.8750 0.9354
## Number of obs: 31770, groups: subID, 322
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 0.043265 0.036494 316.108004 1.186 0.2367
## RepNNon -0.034926 0.050838 316.073197 -0.687 0.4926
## RepNOut -0.114906 0.051611 316.050703 -2.226 0.0267 *
## scale(SGO.sw) 0.001873 0.042359 316.033905 0.044 0.9648
## RepNNon:scale(SGO.sw) -0.003947 0.054826 315.934819 -0.072 0.9426
## RepNOut:scale(SGO.sw) 0.046378 0.053702 315.984167 0.864 0.3884
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) RepNNn RepNOt s(SGO. RNN:(S
## RepNNon -0.718
## RepNOut -0.707 0.508
## scl(SGO.sw) -0.037 0.026 0.026
## RpNN:(SGO.) 0.028 0.010 -0.020 -0.773
## RpNO:(SGO.) 0.029 -0.021 -0.037 -0.789 0.609For Info
m <- lmer( scale(toInG) ~ Info * scale(SGO.sw) + ( 1 | subID), data = longDf, control=lmerControl(optimizer="bobyqa",
optCtrl=list(maxfun=2e5)))
summary(m)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: scale(toInG) ~ Info * scale(SGO.sw) + (1 | subID)
## Data: longDf
## Control: lmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 2e+05))
##
## REML criterion at convergence: 86819.9
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.7548 -0.7814 0.1951 0.8060 2.0408
##
## Random effects:
## Groups Name Variance Std.Dev.
## subID (Intercept) 0.1291 0.3593
## Residual 0.8750 0.9354
## Number of obs: 31770, groups: subID, 322
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) -0.001717 0.020754 318.045001 -0.083 0.9341
## Info1 0.036201 0.020754 318.045001 1.744 0.0821 .
## scale(SGO.sw) 0.018973 0.020797 317.902134 0.912 0.3623
## Info1:scale(SGO.sw) -0.047422 0.020797 317.902134 -2.280 0.0233 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) Info1 s(SGO.
## Info1 0.018
## scl(SGO.sw) 0.003 -0.068
## Inf1:(SGO.) -0.068 0.003 -0.069- significant interaction between self-we overlap and info
Info and RepN
m <- lmer( scale(toInG) ~ Info * scale(SGO.sw) * RepN + ( 1 | subID), data = longDf, control=lmerControl(optimizer="bobyqa",
optCtrl=list(maxfun=2e5)))
summary(m)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: scale(toInG) ~ Info * scale(SGO.sw) * RepN + (1 | subID)
## Data: longDf
## Control: lmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 2e+05))
##
## REML criterion at convergence: 86845.3
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.7481 -0.7832 0.1961 0.8052 2.0443
##
## Random effects:
## Groups Name Variance Std.Dev.
## subID (Intercept) 0.1289 0.3591
## Residual 0.8750 0.9354
## Number of obs: 31770, groups: subID, 322
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 4.225e-02 3.631e-02 3.101e+02 1.164 0.2454
## Info1 2.007e-02 3.631e-02 3.101e+02 0.553 0.5808
## scale(SGO.sw) -7.211e-05 4.444e-02 3.101e+02 -0.002 0.9987
## RepNNon -3.427e-02 5.054e-02 3.100e+02 -0.678 0.4982
## RepNOut -1.066e-01 5.187e-02 3.100e+02 -2.056 0.0406 *
## Info1:scale(SGO.sw) 3.482e-03 4.444e-02 3.101e+02 0.078 0.9376
## Info1:RepNNon 2.719e-02 5.054e-02 3.100e+02 0.538 0.5910
## Info1:RepNOut 2.217e-03 5.187e-02 3.100e+02 0.043 0.9659
## scale(SGO.sw):RepNNon -6.168e-04 5.631e-02 3.100e+02 -0.011 0.9913
## scale(SGO.sw):RepNOut 4.383e-02 5.555e-02 3.100e+02 0.789 0.4307
## Info1:scale(SGO.sw):RepNNon -9.263e-02 5.631e-02 3.100e+02 -1.645 0.1010
## Info1:scale(SGO.sw):RepNOut -2.955e-02 5.555e-02 3.100e+02 -0.532 0.5952
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) Info1 sc(SGO.) RepNNn RepNOt In1:(SGO.) I1:RNN I1:RNO
## Info1 -0.046
## scl(SGO.sw) -0.025 -0.026
## RepNNon -0.718 0.033 0.018
## RepNOut -0.700 0.032 0.017 0.503
## Inf1:(SGO.) -0.026 -0.025 -0.320 0.019 0.018
## Info1:RpNNn 0.033 -0.718 0.019 -0.023 -0.023 0.018
## Info1:RpNOt 0.032 -0.700 0.018 -0.023 0.040 0.017 0.503
## s(SGO.):RNN 0.020 0.020 -0.789 0.015 -0.014 0.253 -0.013 -0.014
## s(SGO.):RNO 0.020 0.021 -0.800 -0.014 -0.042 0.256 -0.015 -0.088
## I1:(SGO.):RNN 0.020 0.020 0.253 -0.013 -0.014 -0.789 0.015 -0.014
## I1:(SGO.):RNO 0.021 0.020 0.256 -0.015 -0.088 -0.800 -0.014 -0.042
## s(SGO.):RNN s(SGO.):RNO I1:(SGO.):RNN
## Info1
## scl(SGO.sw)
## RepNNon
## RepNOut
## Inf1:(SGO.)
## Info1:RpNNn
## Info1:RpNOt
## s(SGO.):RNN
## s(SGO.):RNO 0.631
## I1:(SGO.):RNN -0.204 -0.202
## I1:(SGO.):RNO -0.202 -0.186 0.631We they
For Reputation
m <- lmer( scale(toInG) ~ RepN * scale(SGO.wt) + ( 1 | subID), data = longDf, control=lmerControl(optimizer="bobyqa",
optCtrl=list(maxfun=2e5)))
summary(m)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: scale(toInG) ~ RepN * scale(SGO.wt) + (1 | subID)
## Data: longDf
## Control: lmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 2e+05))
##
## REML criterion at convergence: 87371.8
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.7466 -0.7843 0.2011 0.8031 2.0402
##
## Random effects:
## Groups Name Variance Std.Dev.
## subID (Intercept) 0.1282 0.3581
## Residual 0.8751 0.9355
## Number of obs: 31970, groups: subID, 324
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 0.05034 0.03532 318.15532 1.425 0.155
## RepNNon -0.03957 0.04986 318.11288 -0.794 0.428
## RepNOut -0.10938 0.05079 318.08269 -2.153 0.032 *
## scale(SGO.wt) 0.01787 0.03777 318.12641 0.473 0.637
## RepNNon:scale(SGO.wt) -0.05088 0.05256 318.36175 -0.968 0.334
## RepNOut:scale(SGO.wt) -0.02279 0.05026 318.08457 -0.453 0.651
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) RepNNn RepNOt s(SGO. RNN:(S
## RepNNon -0.708
## RepNOut -0.695 0.493
## scl(SGO.wt) 0.022 -0.016 -0.015
## RpNN:(SGO.) -0.016 -0.015 0.011 -0.719
## RpNO:(SGO.) -0.017 0.012 0.026 -0.751 0.540For Info
m <- lmer( scale(toInG) ~ Info * scale(SGO.wt) + ( 1 | subID), data = longDf, control=lmerControl(optimizer="bobyqa",
optCtrl=list(maxfun=2e5)))
summary(m)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: scale(toInG) ~ Info * scale(SGO.wt) + (1 | subID)
## Data: longDf
## Control: lmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 2e+05))
##
## REML criterion at convergence: 87368.3
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.7534 -0.7816 0.1992 0.8059 2.0389
##
## Random effects:
## Groups Name Variance Std.Dev.
## subID (Intercept) 0.1281 0.3579
## Residual 0.8751 0.9355
## Number of obs: 31970, groups: subID, 324
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 2.933e-03 2.064e-02 3.201e+02 0.142 0.8871
## Info1 4.093e-02 2.064e-02 3.201e+02 1.983 0.0482 *
## scale(SGO.wt) -9.730e-03 2.063e-02 3.202e+02 -0.472 0.6375
## Info1:scale(SGO.wt) -8.421e-04 2.063e-02 3.202e+02 -0.041 0.9675
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) Info1 s(SGO.
## Info1 0.037
## scl(SGO.wt) -0.004 -0.076
## Inf1:(SGO.) -0.076 -0.004 0.021Info and RepN
m <- lmer( scale(toInG) ~ Info * scale(SGO.wt) * RepN + ( 1 | subID), data = longDf, control=lmerControl(optimizer="bobyqa",
optCtrl=list(maxfun=2e5)))
summary(m)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: scale(toInG) ~ Info * scale(SGO.wt) * RepN + (1 | subID)
## Data: longDf
## Control: lmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 2e+05))
##
## REML criterion at convergence: 87393.5
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.7495 -0.7855 0.2006 0.8046 2.0424
##
## Random effects:
## Groups Name Variance Std.Dev.
## subID (Intercept) 0.1277 0.3574
## Residual 0.8751 0.9355
## Number of obs: 31970, groups: subID, 324
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 0.04577 0.03568 312.11010 1.283 0.2006
## Info1 0.02047 0.03568 312.11010 0.574 0.5665
## scale(SGO.wt) 0.01055 0.03858 312.16210 0.274 0.7846
## RepNNon -0.02616 0.05028 312.04884 -0.520 0.6033
## RepNOut -0.10053 0.05122 312.03999 -1.962 0.0506 .
## Info1:scale(SGO.wt) 0.02757 0.03858 312.16210 0.715 0.4753
## Info1:RepNNon 0.04720 0.05028 312.04884 0.939 0.3486
## Info1:RepNOut 0.01063 0.05122 312.03999 0.208 0.8357
## scale(SGO.wt):RepNNon -0.05450 0.05332 312.36098 -1.022 0.3075
## scale(SGO.wt):RepNOut -0.01324 0.05106 312.07993 -0.259 0.7956
## Info1:scale(SGO.wt):RepNNon -0.08327 0.05332 312.36098 -1.562 0.1194
## Info1:scale(SGO.wt):RepNOut -0.01739 0.05106 312.07993 -0.341 0.7336
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) Info1 sc(SGO.) RepNNn RepNOt In1:(SGO.) I1:RNN I1:RNO
## Info1 -0.047
## scl(SGO.wt) 0.051 -0.149
## RepNNon -0.710 0.034 -0.036
## RepNOut -0.697 0.033 -0.035 0.494
## Inf1:(SGO.) -0.149 0.051 -0.159 0.106 0.104
## Info1:RpNNn 0.034 -0.710 0.106 -0.003 -0.023 -0.036
## Info1:RpNOt 0.033 -0.697 0.104 -0.023 0.043 -0.035 0.494
## s(SGO.):RNN -0.037 0.108 -0.724 -0.005 0.025 0.115 -0.137 -0.075
## s(SGO.):RNO -0.038 0.113 -0.756 0.027 0.045 0.120 -0.080 -0.062
## I1:(SGO.):RNN 0.108 -0.037 0.115 -0.137 -0.075 -0.724 -0.005 0.025
## I1:(SGO.):RNO 0.113 -0.038 0.120 -0.080 -0.062 -0.756 0.027 0.045
## s(SGO.):RNN s(SGO.):RNO I1:(SGO.):RNN
## Info1
## scl(SGO.wt)
## RepNNon
## RepNOut
## Inf1:(SGO.)
## Info1:RpNNn
## Info1:RpNOt
## s(SGO.):RNN
## s(SGO.):RNO 0.547
## I1:(SGO.):RNN -0.059 -0.087
## I1:(SGO.):RNO -0.087 -0.031 0.547For Reputation Normative Data
m <- lmer( toInG ~ reputF * SGO.wt + ( reputF | subID), data = longDf, control=lmerControl(optimizer="bobyqa",
optCtrl=list(maxfun=2e5)))
summary(m)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: toInG ~ reputF * SGO.wt + (reputF | subID)
## Data: longDf
## Control: lmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 2e+05))
##
## REML criterion at convergence: 134468.2
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.8879 -0.7823 0.1944 0.7989 2.1212
##
## Random effects:
## Groups Name Variance Std.Dev. Corr
## subID (Intercept) 2.8569 1.6902
## reputF 0.1805 0.4248 -0.90
## Residual 3.8018 1.9498
## Number of obs: 31970, groups: subID, 324
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 5.40053 0.30031 328.39982 17.983 <2e-16 ***
## reputF -0.16306 0.07514 318.57977 -2.170 0.0307 *
## SGO.wt -0.22650 0.09465 328.24475 -2.393 0.0173 *
## reputF:SGO.wt 0.05452 0.02367 318.35309 2.303 0.0219 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) reputF SGO.wt
## reputF -0.962
## SGO.wt -0.855 0.822
## rptF:SGO.wt 0.823 -0.856 -0.962ggpredict(m, c("SGO.wt","reputF")) %>% plot()
- significant interaction of we they overlap with reputational normative
Affective Polarization
For Reputation
m <- lmer( scale(toInG) ~ RepN * scale(affPol) + ( 1 | subID), data = longDf, control=lmerControl(optimizer="bobyqa",
optCtrl=list(maxfun=2e5)))
summary(m)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: scale(toInG) ~ RepN * scale(affPol) + (1 | subID)
## Data: longDf
## Control: lmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 2e+05))
##
## REML criterion at convergence: 102925.9
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.7503 -0.7824 0.2072 0.7863 2.0539
##
## Random effects:
## Groups Name Variance Std.Dev.
## subID (Intercept) 0.1047 0.3236
## Residual 0.8670 0.9311
## Number of obs: 37817, groups: subID, 383
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) -4.898e-04 3.069e-02 3.770e+02 -0.016 0.9873
## RepNNon 2.499e-02 4.245e-02 3.770e+02 0.589 0.5564
## RepNOut -3.832e-02 4.360e-02 3.769e+02 -0.879 0.3800
## scale(affPol) 2.138e-01 3.564e-02 3.769e+02 5.999 4.65e-09 ***
## RepNNon:scale(affPol) -3.468e-02 4.924e-02 3.773e+02 -0.704 0.4817
## RepNOut:scale(affPol) -7.887e-02 4.347e-02 3.768e+02 -1.814 0.0704 .
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) RepNNn RepNOt scl(P) RNN:(P
## RepNNon -0.723
## RepNOut -0.704 0.509
## scale(ffPl) -0.256 0.185 0.180
## RpNNn:sc(P) 0.185 -0.126 -0.130 -0.724
## RpNOt:sc(P) 0.210 -0.152 -0.077 -0.820 0.593- marginal interaction with outgroup reputation manipulation
For Info
m <- lmer( scale(toInG) ~ Info * scale(affPol) + ( 1 | subID), data = longDf, control=lmerControl(optimizer="bobyqa",
optCtrl=list(maxfun=2e5)))
summary(m)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: scale(toInG) ~ Info * scale(affPol) + (1 | subID)
## Data: longDf
## Control: lmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 2e+05))
##
## REML criterion at convergence: 102921.3
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.7569 -0.7841 0.2120 0.7896 2.0570
##
## Random effects:
## Groups Name Variance Std.Dev.
## subID (Intercept) 0.1045 0.3232
## Residual 0.8670 0.9311
## Number of obs: 37817, groups: subID, 383
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 1.293e-03 1.720e-02 3.790e+02 0.075 0.9401
## Info1 3.529e-02 1.720e-02 3.790e+02 2.051 0.0409 *
## scale(affPol) 1.678e-01 1.722e-02 3.790e+02 9.744 <2e-16 ***
## Info1:scale(affPol) 5.603e-03 1.722e-02 3.790e+02 0.325 0.7451
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) Info1 scl(P)
## Info1 -0.003
## scale(ffPl) -0.001 -0.027
## Inf1:scl(P) -0.027 -0.001 0.050Info and RepN
m <- lmer( scale(toInG) ~ Info * scale(affPol) * RepN + ( 1 | subID), data = longDf, control=lmerControl(optimizer="bobyqa",
optCtrl=list(maxfun=2e5)))
summary(m)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: scale(toInG) ~ Info * scale(affPol) * RepN + (1 | subID)
## Data: longDf
## Control: lmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 2e+05))
##
## REML criterion at convergence: 102951.8
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.7539 -0.7828 0.2121 0.7891 2.0575
##
## Random effects:
## Groups Name Variance Std.Dev.
## subID (Intercept) 0.1051 0.3241
## Residual 0.8670 0.9311
## Number of obs: 37817, groups: subID, 383
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) -0.00323 0.03084 371.00025 -0.105 0.9166
## Info1 0.01430 0.03084 371.00025 0.464 0.6430
## scale(affPol) 0.20929 0.03592 370.92154 5.827 1.23e-08 ***
## RepNNon 0.02768 0.04262 370.98819 0.649 0.5164
## RepNOut -0.03257 0.04380 370.90741 -0.744 0.4576
## Info1:scale(affPol) 0.03074 0.03592 370.92154 0.856 0.3927
## Info1:RepNNon 0.02792 0.04262 370.98819 0.655 0.5129
## Info1:RepNOut 0.02366 0.04380 370.90741 0.540 0.5895
## scale(affPol):RepNNon -0.02747 0.04952 371.26399 -0.555 0.5794
## scale(affPol):RepNOut -0.07590 0.04388 370.80285 -1.730 0.0845 .
## Info1:scale(affPol):RepNNon -0.03284 0.04952 371.26399 -0.663 0.5077
## Info1:scale(affPol):RepNOut -0.03562 0.04388 370.80285 -0.812 0.4174
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) Info1 scl(P) RepNNn RepNOt In1:(P) I1:RNN I1:RNO s(P):RNN
## Info1 -0.049
## scale(ffPl) -0.244 -0.052
## RepNNon -0.724 0.035 0.177
## RepNOut -0.704 0.034 0.172 0.509
## Inf1:scl(P) -0.052 -0.244 -0.085 0.038 0.037
## Info1:RpNNn 0.035 -0.724 0.038 -0.024 -0.025 0.177
## Info1:RpNOt 0.034 -0.704 0.037 -0.025 0.013 0.172 0.509
## scl(fP):RNN 0.177 0.038 -0.725 -0.119 -0.125 0.062 0.000 -0.027
## scl(fP):RNO 0.200 0.043 -0.819 -0.145 -0.072 0.070 -0.031 -0.030 0.594
## In1:(P):RNN 0.038 0.177 0.062 0.000 -0.027 -0.725 -0.119 -0.125 -0.034
## In1:(P):RNO 0.043 0.200 0.070 -0.031 -0.030 -0.819 -0.145 -0.072 -0.051
## s(P):RNO I1:(P):RNN
## Info1
## scale(ffPl)
## RepNNon
## RepNOut
## Inf1:scl(P)
## Info1:RpNNn
## Info1:RpNOt
## scl(fP):RNN
## scl(fP):RNO
## In1:(P):RNN -0.051
## In1:(P):RNO -0.010 0.594Interacting with information difference
Super fucking cool model
infoDf <- subset(longDf, Info == "Info")
m <- lmer( eval ~ inMinOut * affPol + ( inMinOut | subID) + (1 | issues), data = infoDf, control=lmerControl(optimizer="bobyqa",
optCtrl=list(maxfun=2e5)))
summary(m)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: eval ~ inMinOut * affPol + (inMinOut | subID) + (1 | issues)
## Data: infoDf
## Control: lmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 2e+05))
##
## REML criterion at convergence: 70701.1
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -4.1416 -0.6208 0.0933 0.6521 4.4418
##
## Random effects:
## Groups Name Variance Std.Dev. Corr
## subID (Intercept) 0.1521 0.390
## inMinOut 1.6769 1.295 -0.50
## issues (Intercept) 1.0021 1.001
## Residual 2.2677 1.506
## Number of obs: 18989, groups: subID, 192; issues, 100
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 4.777600 0.115503 160.759702 41.364 < 2e-16 ***
## inMinOut 1.039574 0.186215 190.479907 5.583 8.07e-08 ***
## affPol -0.001462 0.000851 189.114270 -1.718 0.0874 .
## inMinOut:affPol 0.023172 0.002778 190.830484 8.342 1.44e-14 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) inMnOt affPol
## inMinOut -0.204
## affPol -0.420 0.358
## inMnOt:ffPl 0.170 -0.850 -0.424ggpredict(m, c("inMinOut", "affPol")) %>% plot()
Public Self-Consciousness
For Reputation
m <- lmer( scale(toInG) ~ RepN * scale(SCS.pub) + ( 1 | subID), data = longDf, control=lmerControl(optimizer="bobyqa",
optCtrl=list(maxfun=2e5)))
summary(m)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: scale(toInG) ~ RepN * scale(SCS.pub) + (1 | subID)
## Data: longDf
## Control: lmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 2e+05))
##
## REML criterion at convergence: 88185.2
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.7704 -0.7866 0.1980 0.7763 2.0522
##
## Random effects:
## Groups Name Variance Std.Dev.
## subID (Intercept) 0.1359 0.3686
## Residual 0.8651 0.9301
## Number of obs: 32397, groups: subID, 328
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 0.04780 0.03664 322.10006 1.305 0.1930
## RepNNon -0.03436 0.05086 322.12587 -0.676 0.4998
## RepNOut -0.11620 0.05242 322.02680 -2.217 0.0273 *
## scale(SCS.pub) -0.03998 0.03573 322.05714 -1.119 0.2639
## RepNNon:scale(SCS.pub) 0.05468 0.05280 322.16710 1.035 0.3012
## RepNOut:scale(SCS.pub) 0.07375 0.05001 321.95110 1.474 0.1413
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) RepNNn RepNOt s(SCS. RNN:(S
## RepNNon -0.720
## RepNOut -0.699 0.504
## scl(SCS.pb) -0.049 0.036 0.034
## RpNN:(SCS.) 0.033 0.016 -0.023 -0.677
## RpNO:(SCS.) 0.035 -0.025 -0.037 -0.714 0.483For Info
m <- lmer( scale(toInG) ~ Info * scale(SCS.pub) + ( 1 | subID), data = longDf, control=lmerControl(optimizer="bobyqa",
optCtrl=list(maxfun=2e5)))
summary(m)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: scale(toInG) ~ Info * scale(SCS.pub) + (1 | subID)
## Data: longDf
## Control: lmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 2e+05))
##
## REML criterion at convergence: 88182
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.7753 -0.7872 0.1959 0.7809 2.0560
##
## Random effects:
## Groups Name Variance Std.Dev.
## subID (Intercept) 0.1359 0.3686
## Residual 0.8651 0.9301
## Number of obs: 32397, groups: subID, 328
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) -0.002668 0.021003 324.047285 -0.127 0.8990
## Info1 0.046716 0.021003 324.047285 2.224 0.0268 *
## scale(SCS.pub) 0.001529 0.021019 324.014229 0.073 0.9421
## Info1:scale(SCS.pub) 0.011828 0.021019 324.014229 0.563 0.5740
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) Info1 s(SCS.
## Info1 -0.018
## scl(SCS.pb) 0.001 -0.009
## Inf1:(SCS.) -0.009 0.001 -0.043Info and RepN
m <- lmer( scale(toInG) ~ Info * scale(SCS.pub) * RepN + ( 1 | subID), data = longDf, control=lmerControl(optimizer="bobyqa",
optCtrl=list(maxfun=2e5)))
summary(m)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: scale(toInG) ~ Info * scale(SCS.pub) * RepN + (1 | subID)
## Data: longDf
## Control: lmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 2e+05))
##
## REML criterion at convergence: 88208.4
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.7765 -0.7880 0.1960 0.7754 2.0590
##
## Random effects:
## Groups Name Variance Std.Dev.
## subID (Intercept) 0.1361 0.3689
## Residual 0.8651 0.9301
## Number of obs: 32397, groups: subID, 328
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 4.260e-02 3.691e-02 3.161e+02 1.154 0.2493
## Info1 4.816e-02 3.691e-02 3.161e+02 1.305 0.1929
## scale(SCS.pub) -4.252e-02 3.597e-02 3.161e+02 -1.182 0.2381
## RepNNon -3.048e-02 5.111e-02 3.162e+02 -0.596 0.5514
## RepNOut -1.090e-01 5.282e-02 3.160e+02 -2.063 0.0399
## Info1:scale(SCS.pub) 2.639e-02 3.597e-02 3.161e+02 0.734 0.4637
## Info1:RepNNon -1.739e-02 5.111e-02 3.162e+02 -0.340 0.7339
## Info1:RepNOut -9.657e-04 5.282e-02 3.160e+02 -0.018 0.9854
## scale(SCS.pub):RepNNon 5.838e-02 5.306e-02 3.162e+02 1.100 0.2720
## scale(SCS.pub):RepNOut 7.150e-02 5.031e-02 3.160e+02 1.421 0.1563
## Info1:scale(SCS.pub):RepNNon -3.976e-02 5.306e-02 3.162e+02 -0.749 0.4542
## Info1:scale(SCS.pub):RepNOut -7.674e-05 5.031e-02 3.160e+02 -0.002 0.9988
##
## (Intercept)
## Info1
## scale(SCS.pub)
## RepNNon
## RepNOut *
## Info1:scale(SCS.pub)
## Info1:RepNNon
## Info1:RepNOut
## scale(SCS.pub):RepNNon
## scale(SCS.pub):RepNOut
## Info1:scale(SCS.pub):RepNNon
## Info1:scale(SCS.pub):RepNOut
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) Info1 sc(SCS.) RepNNn RepNOt In1:(SCS.) I1:RNN I1:RNO
## Info1 -0.112
## scl(SCS.pb) -0.051 0.012
## RepNNon -0.722 0.081 0.037
## RepNOut -0.699 0.078 0.035 0.505
## Inf1:(SCS.) 0.012 -0.051 -0.108 -0.009 -0.008
## Info1:RpNNn 0.081 -0.722 -0.009 -0.066 -0.057 0.037
## Info1:RpNOt 0.078 -0.699 -0.008 -0.057 -0.008 0.035 0.505
## s(SCS.):RNN 0.034 -0.008 -0.678 0.016 -0.024 0.073 0.033 0.006
## s(SCS.):RNO 0.036 -0.009 -0.715 -0.026 -0.040 0.077 0.006 -0.034
## I1:(SCS.):RNN -0.008 0.034 0.073 0.033 0.006 -0.678 0.016 -0.024
## I1:(SCS.):RNO -0.009 0.036 0.077 0.006 -0.034 -0.715 -0.026 -0.040
## s(SCS.):RNN s(SCS.):RNO I1:(SCS.):RNN
## Info1
## scl(SCS.pb)
## RepNNon
## RepNOut
## Inf1:(SCS.)
## Info1:RpNNn
## Info1:RpNOt
## s(SCS.):RNN
## s(SCS.):RNO 0.485
## I1:(SCS.):RNN -0.026 -0.052
## I1:(SCS.):RNO -0.052 -0.076 0.485Identity Importance
For Reputation
m <- lmer( scale(toInG) ~ RepN * scale(CSEid) + ( 1 | subID), data = longDf, control=lmerControl(optimizer="bobyqa",
optCtrl=list(maxfun=2e5)))
summary(m)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: scale(toInG) ~ RepN * scale(CSEid) + (1 | subID)
## Data: longDf
## Control: lmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 2e+05))
##
## REML criterion at convergence: 89539.8
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.7568 -0.7920 0.2099 0.7965 2.0560
##
## Random effects:
## Groups Name Variance Std.Dev.
## subID (Intercept) 0.1233 0.3512
## Residual 0.8676 0.9314
## Number of obs: 32871, groups: subID, 333
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 0.035992 0.033615 327.090798 1.071 0.2851
## RepNNon -0.015844 0.048209 327.108227 -0.329 0.7426
## RepNOut -0.111926 0.048976 327.020406 -2.285 0.0229 *
## scale(CSEid) 0.084108 0.033700 326.945564 2.496 0.0131 *
## RepNNon:scale(CSEid) 0.011151 0.047592 327.024797 0.234 0.8149
## RepNOut:scale(CSEid) -0.003218 0.049781 326.985811 -0.065 0.9485
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) RepNNn RepNOt s(CSE) RNN:(C
## RepNNon -0.697
## RepNOut -0.686 0.479
## scale(CSEd) -0.025 0.018 0.017
## RpNNn:(CSE) 0.018 -0.041 -0.012 -0.708
## RpNOt:(CSE) 0.017 -0.012 0.037 -0.677 0.479For Info
m <- lmer( scale(toInG) ~ Info * scale(CSEid) + ( 1 | subID), data = longDf, control=lmerControl(optimizer="bobyqa",
optCtrl=list(maxfun=2e5)))
summary(m)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: scale(toInG) ~ Info * scale(CSEid) + (1 | subID)
## Data: longDf
## Control: lmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 2e+05))
##
## REML criterion at convergence: 89537.5
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.7629 -0.7926 0.2143 0.7964 2.0539
##
## Random effects:
## Groups Name Variance Std.Dev.
## subID (Intercept) 0.1238 0.3518
## Residual 0.8676 0.9314
## Number of obs: 32871, groups: subID, 333
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) -0.004466 0.020004 329.089229 -0.223 0.8235
## Info1 0.034108 0.020004 329.089229 1.705 0.0891 .
## scale(CSEid) 0.087547 0.020038 329.033606 4.369 1.68e-05 ***
## Info1:scale(CSEid) 0.001077 0.020038 329.033606 0.054 0.9572
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) Info1 s(CSE)
## Info1 -0.003
## scale(CSEd) -0.004 -0.071
## Inf1:s(CSE) -0.071 -0.004 0.060Info and RepN
m <- lmer( scale(toInG) ~ Info * scale(CSEid) * RepN + ( 1 | subID), data = longDf, control=lmerControl(optimizer="bobyqa",
optCtrl=list(maxfun=2e5)))
summary(m)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: scale(toInG) ~ Info * scale(CSEid) * RepN + (1 | subID)
## Data: longDf
## Control: lmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 2e+05))
##
## REML criterion at convergence: 89564.4
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.7591 -0.7928 0.2080 0.7922 2.0583
##
## Random effects:
## Groups Name Variance Std.Dev.
## subID (Intercept) 0.1240 0.3522
## Residual 0.8676 0.9314
## Number of obs: 32871, groups: subID, 333
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 0.033481 0.033919 321.123824 0.987 0.3243
## Info1 0.017953 0.033919 321.123824 0.529 0.5970
## scale(CSEid) 0.073463 0.035333 320.990447 2.079 0.0384 *
## RepNNon -0.013456 0.048524 321.114091 -0.277 0.7817
## RepNOut -0.107189 0.050203 320.996956 -2.135 0.0335 *
## Info1:scale(CSEid) 0.035272 0.035333 320.990447 0.998 0.3189
## Info1:RepNNon 0.016619 0.048524 321.114091 0.342 0.7322
## Info1:RepNOut 0.029629 0.050203 320.996956 0.590 0.5555
## scale(CSEid):RepNNon 0.022175 0.048912 321.023445 0.453 0.6506
## scale(CSEid):RepNOut -0.003853 0.055253 320.888266 -0.070 0.9444
## Info1:scale(CSEid):RepNNon -0.054320 0.048912 321.023445 -1.111 0.2676
## Info1:scale(CSEid):RepNOut -0.031997 0.055253 320.888266 -0.579 0.5629
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) Info1 sc(CSE) RepNNn RepNOt In1:(CSE) I1:RNN I1:RNO
## Info1 -0.111
## scale(CSEd) -0.018 -0.013
## RepNNon -0.699 0.077 0.013
## RepNOut -0.676 0.075 0.012 0.472
## Inf1:s(CSE) -0.013 -0.018 -0.292 0.009 0.009
## Info1:RpNNn 0.077 -0.699 0.009 -0.043 -0.052 0.013
## Info1:RpNOt 0.075 -0.676 0.009 -0.052 -0.010 0.012 0.472
## sc(CSE):RNN 0.013 0.010 -0.722 -0.034 -0.009 0.211 0.017 -0.007
## sc(CSE):RNO 0.012 0.009 -0.639 -0.008 -0.037 0.186 -0.006 -0.150
## I1:(CSE):RNN 0.010 0.013 0.211 0.017 -0.007 -0.722 -0.034 -0.009
## I1:(CSE):RNO 0.009 0.012 0.186 -0.006 -0.150 -0.639 -0.008 -0.037
## s(CSE):RNN s(CSE):RNO I1:(CSE):RNN
## Info1
## scale(CSEd)
## RepNNon
## RepNOut
## Inf1:s(CSE)
## Info1:RpNNn
## Info1:RpNOt
## sc(CSE):RNN
## sc(CSE):RNO 0.462
## I1:(CSE):RNN -0.119 -0.135
## I1:(CSE):RNO -0.135 0.144 0.462Public Collective Self-Esteem
For Reputation
m <- lmer( scale(toInG) ~ RepN * scale(CSEpub) + ( 1 | subID), data = longDf, control=lmerControl(optimizer="bobyqa",
optCtrl=list(maxfun=2e5)))
summary(m)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: scale(toInG) ~ RepN * scale(CSEpub) + (1 | subID)
## Data: longDf
## Control: lmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 2e+05))
##
## REML criterion at convergence: 89542
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.7553 -0.7887 0.2082 0.7967 2.0555
##
## Random effects:
## Groups Name Variance Std.Dev.
## subID (Intercept) 0.1242 0.3524
## Residual 0.8676 0.9314
## Number of obs: 32871, groups: subID, 333
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 0.04572 0.03381 327.08869 1.352 0.17733
## RepNNon -0.02616 0.04847 327.12599 -0.540 0.58975
## RepNOut -0.12829 0.04909 327.06440 -2.613 0.00939 **
## scale(CSEpub) 0.11332 0.03803 327.09196 2.980 0.00310 **
## RepNNon:scale(CSEpub) -0.03740 0.05111 327.13582 -0.732 0.46482
## RepNOut:scale(CSEpub) -0.05688 0.05005 327.08159 -1.137 0.25656
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) RepNNn RepNOt s(CSE) RNN:(C
## RepNNon -0.698
## RepNOut -0.689 0.481
## scal(CSEpb) 0.075 -0.053 -0.052
## RpNNn:(CSE) -0.056 0.002 0.039 -0.744
## RpNOt:(CSE) -0.057 0.040 0.043 -0.760 0.565For Info
m <- lmer( scale(toInG) ~ Info * scale(CSEpub) + ( 1 | subID), data = longDf, control=lmerControl(optimizer="bobyqa",
optCtrl=list(maxfun=2e5)))
summary(m)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: scale(toInG) ~ Info * scale(CSEpub) + (1 | subID)
## Data: longDf
## Control: lmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 2e+05))
##
## REML criterion at convergence: 89539.9
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.7608 -0.7877 0.2109 0.7950 2.0547
##
## Random effects:
## Groups Name Variance Std.Dev.
## subID (Intercept) 0.1247 0.3532
## Residual 0.8676 0.9314
## Number of obs: 32871, groups: subID, 333
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) -0.003795 0.020054 329.104868 -0.189 0.8500
## Info1 0.044711 0.020054 329.104868 2.230 0.0265 *
## scale(CSEpub) 0.081607 0.020078 329.096193 4.065 6.03e-05 ***
## Info1:scale(CSEpub) 0.011721 0.020078 329.096193 0.584 0.5598
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) Info1 s(CSE)
## Info1 -0.003
## scal(CSEpb) 0.002 0.054
## Inf1:s(CSE) 0.054 0.002 0.048Info and RepN
m <- lmer( scale(toInG) ~ Info * scale(CSEpub) * RepN + ( 1 | subID), data = longDf, control=lmerControl(optimizer="bobyqa",
optCtrl=list(maxfun=2e5)))
summary(m)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: scale(toInG) ~ Info * scale(CSEpub) * RepN + (1 | subID)
## Data: longDf
## Control: lmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 2e+05))
##
## REML criterion at convergence: 89564.4
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.7592 -0.7896 0.2088 0.7984 2.0635
##
## Random effects:
## Groups Name Variance Std.Dev.
## subID (Intercept) 0.1239 0.3520
## Residual 0.8676 0.9314
## Number of obs: 32871, groups: subID, 333
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 0.04180 0.03415 321.08707 1.224 0.22175
## Info1 0.01451 0.03415 321.08707 0.425 0.67106
## scale(CSEpub) 0.10952 0.03826 321.09604 2.863 0.00448 **
## RepNNon -0.01944 0.04882 321.08512 -0.398 0.69083
## RepNOut -0.12076 0.04955 321.02951 -2.437 0.01535 *
## Info1:scale(CSEpub) 0.03135 0.03826 321.09604 0.819 0.41313
## Info1:RepNNon 0.02191 0.04882 321.08512 0.449 0.65394
## Info1:RepNOut 0.05742 0.04955 321.02951 1.159 0.24739
## scale(CSEpub):RepNNon -0.02776 0.05148 321.08983 -0.539 0.59014
## scale(CSEpub):RepNOut -0.04786 0.05042 321.05261 -0.949 0.34313
## Info1:scale(CSEpub):RepNNon -0.01201 0.05148 321.08983 -0.233 0.81567
## Info1:scale(CSEpub):RepNOut -0.05294 0.05042 321.05261 -1.050 0.29449
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) Info1 sc(CSE) RepNNn RepNOt In1:(CSE) I1:RNN I1:RNO
## Info1 -0.125
## scal(CSEpb) 0.091 -0.090
## RepNNon -0.699 0.087 -0.064
## RepNOut -0.689 0.086 -0.063 0.482
## Inf1:s(CSE) -0.090 0.091 -0.086 0.063 0.062
## Info1:RpNNn 0.087 -0.699 0.063 -0.055 -0.060 -0.064
## Info1:RpNOt 0.086 -0.689 0.062 -0.060 -0.012 -0.063 0.482
## sc(CSE):RNN -0.068 0.067 -0.743 0.016 0.047 0.064 0.005 -0.046
## sc(CSE):RNO -0.069 0.068 -0.759 0.049 0.061 0.065 -0.048 0.006
## I1:(CSE):RNN 0.067 -0.068 0.064 0.005 -0.046 -0.743 0.016 0.047
## I1:(CSE):RNO 0.068 -0.069 0.065 -0.048 0.006 -0.759 0.049 0.061
## s(CSE):RNN s(CSE):RNO I1:(CSE):RNN
## Info1
## scal(CSEpb)
## RepNNon
## RepNOut
## Inf1:s(CSE)
## Info1:RpNNn
## Info1:RpNOt
## sc(CSE):RNN
## sc(CSE):RNO 0.564
## I1:(CSE):RNN -0.011 -0.049
## I1:(CSE):RNO -0.049 -0.009 0.564Private Colective Self-Esteem
For Reputation
m <- lmer( scale(toInG) ~ RepN * scale(CSEpriv) + ( 1 | subID), data = longDf, control=lmerControl(optimizer="bobyqa",
optCtrl=list(maxfun=2e5)))
summary(m)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: scale(toInG) ~ RepN * scale(CSEpriv) + (1 | subID)
## Data: longDf
## Control: lmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 2e+05))
##
## REML criterion at convergence: 89525.8
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.7486 -0.7887 0.2110 0.7993 2.0516
##
## Random effects:
## Groups Name Variance Std.Dev.
## subID (Intercept) 0.1178 0.3432
## Residual 0.8676 0.9314
## Number of obs: 32871, groups: subID, 333
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 0.034104 0.032911 327.061856 1.036 0.300851
## RepNNon -0.024353 0.047345 327.101727 -0.514 0.607333
## RepNOut -0.102159 0.048159 327.008177 -2.121 0.034650 *
## scale(CSEpriv) 0.127490 0.035394 326.759635 3.602 0.000365 ***
## RepNNon:scale(CSEpriv) -0.001406 0.048883 327.016716 -0.029 0.977074
## RepNOut:scale(CSEpriv) -0.037972 0.048434 326.839306 -0.784 0.433612
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) RepNNn RepNOt s(CSE) RNN:(C
## RepNNon -0.695
## RepNOut -0.683 0.475
## scl(CSEprv) -0.034 0.023 0.023
## RpNNn:(CSE) 0.024 -0.079 -0.017 -0.724
## RpNOt:(CSE) 0.025 -0.017 0.062 -0.731 0.529For Info
m <- lmer( scale(toInG) ~ Info * scale(CSEpriv) + ( 1 | subID), data = longDf, control=lmerControl(optimizer="bobyqa",
optCtrl=list(maxfun=2e5)))
summary(m)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: scale(toInG) ~ Info * scale(CSEpriv) + (1 | subID)
## Data: longDf
## Control: lmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 2e+05))
##
## REML criterion at convergence: 89522.5
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.7543 -0.7856 0.2109 0.8008 2.0509
##
## Random effects:
## Groups Name Variance Std.Dev.
## subID (Intercept) 0.1178 0.3433
## Residual 0.8676 0.9314
## Number of obs: 32871, groups: subID, 333
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) -0.004662 0.019527 329.078024 -0.239 0.8115
## Info1 0.034366 0.019527 329.078024 1.760 0.0794 .
## scale(CSEpriv) 0.116822 0.019697 329.025512 5.931 7.62e-09 ***
## Info1:scale(CSEpriv) 0.005413 0.019697 329.025512 0.275 0.7836
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) Info1 s(CSE)
## Info1 -0.003
## scl(CSEprv) -0.007 -0.051
## Inf1:s(CSE) -0.051 -0.007 0.131Info and RepN
m <- lmer( scale(toInG) ~ Info * scale(CSEpriv) * RepN + ( 1 | subID), data = longDf, control=lmerControl(optimizer="bobyqa",
optCtrl=list(maxfun=2e5)))
summary(m)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: scale(toInG) ~ Info * scale(CSEpriv) * RepN + (1 | subID)
## Data: longDf
## Control: lmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 2e+05))
##
## REML criterion at convergence: 89548.8
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.7556 -0.7911 0.2111 0.7971 2.0563
##
## Random effects:
## Groups Name Variance Std.Dev.
## subID (Intercept) 0.1177 0.3430
## Residual 0.8676 0.9314
## Number of obs: 32871, groups: subID, 333
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 0.028930 0.033246 321.137484 0.870 0.38486
## Info1 0.007900 0.033246 321.137484 0.238 0.81233
## scale(CSEpriv) 0.123761 0.035610 320.790985 3.475 0.00058
## RepNNon -0.018635 0.047600 321.117302 -0.391 0.69569
## RepNOut -0.091760 0.048458 321.026427 -1.894 0.05918
## Info1:scale(CSEpriv) 0.047105 0.035610 320.790985 1.323 0.18685
## Info1:RepNNon 0.027366 0.047600 321.117302 0.575 0.56575
## Info1:RepNOut 0.044672 0.048458 321.026427 0.922 0.35730
## scale(CSEpriv):RepNNon 0.004005 0.049059 321.010653 0.082 0.93498
## scale(CSEpriv):RepNOut -0.051709 0.050586 320.796209 -1.022 0.30746
## Info1:scale(CSEpriv):RepNNon -0.051005 0.049059 321.010653 -1.040 0.29928
## Info1:scale(CSEpriv):RepNOut -0.079531 0.050586 320.796209 -1.572 0.11689
##
## (Intercept)
## Info1
## scale(CSEpriv) ***
## RepNNon
## RepNOut .
## Info1:scale(CSEpriv)
## Info1:RepNNon
## Info1:RepNOut
## scale(CSEpriv):RepNNon
## scale(CSEpriv):RepNOut
## Info1:scale(CSEpriv):RepNNon
## Info1:scale(CSEpriv):RepNOut
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) Info1 sc(CSE) RepNNn RepNOt In1:(CSE) I1:RNN I1:RNO
## Info1 -0.107
## scl(CSEprv) -0.017 -0.096
## RepNNon -0.698 0.074 0.012
## RepNOut -0.686 0.073 0.012 0.479
## Inf1:s(CSE) -0.096 -0.017 -0.060 0.067 0.066
## Info1:RpNNn 0.074 -0.698 0.067 -0.044 -0.051 0.012
## Info1:RpNOt 0.073 -0.686 0.066 -0.051 -0.016 0.012 0.479
## sc(CSE):RNN 0.012 0.070 -0.726 -0.069 -0.008 0.044 -0.025 -0.048
## sc(CSE):RNO 0.012 0.068 -0.704 -0.008 0.056 0.042 -0.047 -0.067
## I1:(CSE):RNN 0.070 0.012 0.044 -0.025 -0.048 -0.726 -0.069 -0.008
## I1:(CSE):RNO 0.068 0.012 0.042 -0.047 -0.067 -0.704 -0.008 0.056
## s(CSE):RNN s(CSE):RNO I1:(CSE):RNN
## Info1
## scl(CSEprv)
## RepNNon
## RepNOut
## Inf1:s(CSE)
## Info1:RpNNn
## Info1:RpNOt
## sc(CSE):RNN
## sc(CSE):RNO 0.511
## I1:(CSE):RNN -0.026 -0.031
## I1:(CSE):RNO -0.031 0.163 0.511For Reputation Normative Data
m <- lmer( toInG ~ reputF * CSEpriv + ( reputF | subID), data = longDf, control=lmerControl(optimizer="bobyqa",
optCtrl=list(maxfun=2e5)))
summary(m)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: toInG ~ reputF * CSEpriv + (reputF | subID)
## Data: longDf
## Control: lmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 2e+05))
##
## REML criterion at convergence: 137923.2
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.9460 -0.7880 0.2050 0.7949 2.1439
##
## Random effects:
## Groups Name Variance Std.Dev. Corr
## subID (Intercept) 2.4270 1.5579
## reputF 0.1811 0.4255 -0.91
## Residual 3.7678 1.9411
## Number of obs: 32871, groups: subID, 333
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 0.7495 0.7641 326.5690 0.981 0.327372
## reputF 0.6844 0.1983 325.1779 3.452 0.000631 ***
## CSEpriv 0.7304 0.1361 327.6552 5.366 1.52e-07 ***
## reputF:CSEpriv -0.1273 0.0353 325.3285 -3.607 0.000358 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) reputF CSEprv
## reputF -0.964
## CSEpriv -0.981 0.945
## rptF:CSEprv 0.946 -0.981 -0.964ggpredict(m, c("CSEpriv","reputF")) %>% plot()
- highly significant interaction of private collective self-esteem with reputational cost
Extraversion
For Reputation
m <- lmer( scale(toInG) ~ RepN * scale(Extra) + ( 1 | subID), data = longDf, control=lmerControl(optimizer="bobyqa",
optCtrl=list(maxfun=2e5)))
summary(m)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: scale(toInG) ~ RepN * scale(Extra) + (1 | subID)
## Data: longDf
## Control: lmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 2e+05))
##
## REML criterion at convergence: 90267.1
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.7595 -0.7828 0.2025 0.7846 1.9712
##
## Random effects:
## Groups Name Variance Std.Dev.
## subID (Intercept) 0.1284 0.3583
## Residual 0.8709 0.9332
## Number of obs: 33088, groups: subID, 335
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 0.029025 0.035021 329.033623 0.829 0.4078
## RepNNon -0.006413 0.049517 329.107109 -0.130 0.8970
## RepNOut -0.097802 0.049630 329.009353 -1.971 0.0496 *
## scale(Extra) -0.023731 0.036364 329.197235 -0.653 0.5145
## RepNNon:scale(Extra) 0.005607 0.049604 329.030228 0.113 0.9101
## RepNOut:scale(Extra) -0.028812 0.050638 329.176693 -0.569 0.5698
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) RepNNn RepNOt scl(E) RNN:(E
## RepNNon -0.707
## RepNOut -0.706 0.499
## scale(Extr) 0.033 -0.023 -0.023
## RpNNn:sc(E) -0.024 0.013 0.017 -0.733
## RpNOt:sc(E) -0.024 0.017 0.005 -0.718 0.526For Info
m <- lmer( scale(toInG) ~ Info * scale(Extra) + ( 1 | subID), data = longDf, control=lmerControl(optimizer="bobyqa",
optCtrl=list(maxfun=2e5)))
summary(m)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: scale(toInG) ~ Info * scale(Extra) + (1 | subID)
## Data: longDf
## Control: lmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 2e+05))
##
## REML criterion at convergence: 90263
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.7665 -0.7828 0.2048 0.7899 1.9652
##
## Random effects:
## Groups Name Variance Std.Dev.
## subID (Intercept) 0.1281 0.3579
## Residual 0.8709 0.9332
## Number of obs: 33088, groups: subID, 335
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) -0.007194 0.020277 331.079640 -0.355 0.723
## Info1 0.028326 0.020277 331.079640 1.397 0.163
## scale(Extra) -0.028831 0.020283 331.073041 -1.421 0.156
## Info1:scale(Extra) -0.030543 0.020283 331.073041 -1.506 0.133
##
## Correlation of Fixed Effects:
## (Intr) Info1 scl(E)
## Info1 0.027
## scale(Extr) -0.001 0.073
## Inf1:scl(E) 0.073 -0.001 -0.036Info and Rep
m <- lmer( scale(toInG) ~ Info * scale(Extra) * RepN + ( 1 | subID), data = longDf, control=lmerControl(optimizer="bobyqa",
optCtrl=list(maxfun=2e5)))
summary(m)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: scale(toInG) ~ Info * scale(Extra) * RepN + (1 | subID)
## Data: longDf
## Control: lmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 2e+05))
##
## REML criterion at convergence: 90291.3
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.7641 -0.7837 0.2032 0.7884 1.9686
##
## Random effects:
## Groups Name Variance Std.Dev.
## subID (Intercept) 0.1289 0.3591
## Residual 0.8709 0.9332
## Number of obs: 33088, groups: subID, 335
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 0.018556 0.035888 323.155089 0.517 0.605
## Info1 0.022817 0.035888 323.155089 0.636 0.525
## scale(Extra) -0.001338 0.039800 323.357224 -0.034 0.973
## RepNNon 0.002462 0.050276 323.174563 0.049 0.961
## RepNOut -0.081598 0.050539 323.052275 -1.615 0.107
## Info1:scale(Extra) -0.050875 0.039800 323.357224 -1.278 0.202
## Info1:RepNNon -0.005109 0.050276 323.174563 -0.102 0.919
## Info1:RepNOut 0.017789 0.050539 323.052275 0.352 0.725
## scale(Extra):RepNNon -0.020623 0.052488 323.131341 -0.393 0.695
## scale(Extra):RepNOut -0.053070 0.053300 323.253920 -0.996 0.320
## Info1:scale(Extra):RepNNon 0.014796 0.052488 323.131341 0.282 0.778
## Info1:scale(Extra):RepNOut 0.042629 0.053300 323.253920 0.800 0.424
##
## Correlation of Fixed Effects:
## (Intr) Info1 scl(E) RepNNn RepNOt In1:(E) I1:RNN I1:RNO s(E):RNN
## Info1 -0.111
## scale(Extr) -0.054 0.183
## RepNNon -0.714 0.079 0.039
## RepNOut -0.710 0.079 0.038 0.507
## Inf1:scl(E) 0.183 -0.054 -0.368 -0.131 -0.130
## Info1:RpNNn 0.079 -0.714 -0.131 -0.030 -0.056 0.039
## Info1:RpNOt 0.079 -0.710 -0.130 -0.056 0.012 0.038 0.507
## scl(Ex):RNN 0.041 -0.139 -0.758 -0.028 -0.029 0.279 0.130 0.099
## scl(Ex):RNO 0.040 -0.137 -0.747 -0.029 -0.043 0.274 0.098 0.082 0.566
## In1:(E):RNN -0.139 0.041 0.279 0.130 0.099 -0.758 -0.028 -0.029 -0.152
## In1:(E):RNO -0.137 0.040 0.274 0.098 0.082 -0.747 -0.029 -0.043 -0.208
## s(E):RNO I1:(E):RNN
## Info1
## scale(Extr)
## RepNNon
## RepNOut
## Inf1:scl(E)
## Info1:RpNNn
## Info1:RpNOt
## scl(Ex):RNN
## scl(Ex):RNO
## In1:(E):RNN -0.208
## In1:(E):RNO -0.168 0.566Neuroticism
For Reputation
m <- lmer( scale(toInG) ~ RepN * scale(Neur) + ( 1 | subID), data = longDf, control=lmerControl(optimizer="bobyqa",
optCtrl=list(maxfun=2e5)))
summary(m)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: scale(toInG) ~ RepN * scale(Neur) + (1 | subID)
## Data: longDf
## Control: lmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 2e+05))
##
## REML criterion at convergence: 90260
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.7625 -0.7835 0.2011 0.7881 1.9714
##
## Random effects:
## Groups Name Variance Std.Dev.
## subID (Intercept) 0.1254 0.3542
## Residual 0.8709 0.9332
## Number of obs: 33088, groups: subID, 335
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 0.030170 0.034627 329.095440 0.871 0.3842
## RepNNon -0.009291 0.048978 329.147596 -0.190 0.8497
## RepNOut -0.098966 0.049080 329.057753 -2.016 0.0446 *
## scale(Neur) 0.035504 0.035816 329.160387 0.991 0.3223
## RepNNon:scale(Neur) 0.028272 0.049581 329.063019 0.570 0.5689
## RepNOut:scale(Neur) 0.046263 0.049402 329.210277 0.936 0.3497
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) RepNNn RepNOt scl(N) RNN:(N
## RepNNon -0.707
## RepNOut -0.706 0.499
## scale(Neur) 0.011 -0.008 -0.008
## RpNNn:sc(N) -0.008 -0.006 0.006 -0.722
## RpNOt:sc(N) -0.008 0.006 0.013 -0.725 0.524For Info
m <- lmer( scale(toInG) ~ Info * scale(Neur) + ( 1 | subID), data = longDf, control=lmerControl(optimizer="bobyqa",
optCtrl=list(maxfun=2e5)))
summary(m)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: scale(toInG) ~ Info * scale(Neur) + (1 | subID)
## Data: longDf
## Control: lmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 2e+05))
##
## REML criterion at convergence: 90257
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.7693 -0.7816 0.1990 0.7898 1.9654
##
## Random effects:
## Groups Name Variance Std.Dev.
## subID (Intercept) 0.1256 0.3544
## Residual 0.8709 0.9332
## Number of obs: 33088, groups: subID, 335
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) -0.006602 0.020080 331.098717 -0.329 0.74254
## Info1 0.026801 0.020080 331.098717 1.335 0.18289
## scale(Neur) 0.058813 0.020091 331.131809 2.927 0.00366 **
## Info1:scale(Neur) 0.025887 0.020091 331.131809 1.288 0.19848
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) Info1 scl(N)
## Info1 0.027
## scale(Neur) 0.001 -0.062
## Inf1:scl(N) -0.062 0.001 -0.042Info and Rep
m <- lmer( scale(toInG) ~ Info * scale(Neur) * Rep + ( 1 | subID), data = longDf, control=lmerControl(optimizer="bobyqa",
optCtrl=list(maxfun=2e5)))
summary(m)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: scale(toInG) ~ Info * scale(Neur) * Rep + (1 | subID)
## Data: longDf
## Control: lmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 2e+05))
##
## REML criterion at convergence: 90292
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.7637 -0.7783 0.1986 0.7900 1.9636
##
## Random effects:
## Groups Name Variance Std.Dev.
## subID (Intercept) 0.1255 0.3542
## Residual 0.8709 0.9332
## Number of obs: 33088, groups: subID, 335
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) -0.006632 0.020325 323.120522 -0.326 0.74441
## Info1 0.021008 0.020325 323.120522 1.034 0.30209
## scale(Neur) 0.057808 0.020212 323.141151 2.860 0.00451 **
## Rep1 -0.005063 0.027725 323.053183 -0.183 0.85523
## Rep2 0.026894 0.028547 323.164535 0.942 0.34684
## Info1:scale(Neur) 0.030552 0.020212 323.141151 1.512 0.13163
## Info1:Rep1 0.068537 0.027725 323.053183 2.472 0.01395 *
## Info1:Rep2 -0.005518 0.028547 323.164535 -0.193 0.84686
## scale(Neur):Rep1 -0.014115 0.028291 323.105236 -0.499 0.61816
## scale(Neur):Rep2 0.005832 0.028304 323.050057 0.206 0.83690
## Info1:scale(Neur):Rep1 -0.002319 0.028291 323.105236 -0.082 0.93471
## Info1:scale(Neur):Rep2 0.006570 0.028304 323.050057 0.232 0.81659
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) Info1 scl(N) Rep1 Rep2 In1:(N) In1:R1 In1:R2 s(N):R1
## Info1 0.013
## scale(Neur) -0.012 -0.056
## Rep1 -0.102 0.011 0.080
## Rep2 -0.019 0.030 -0.010 -0.435
## Inf1:scl(N) -0.056 -0.012 -0.047 -0.033 0.012
## Info1:Rep1 0.011 -0.102 -0.033 0.021 -0.036 0.080
## Info1:Rep2 0.030 -0.019 0.012 -0.036 0.034 -0.010 -0.435
## scl(Nr):Rp1 0.078 -0.032 -0.029 0.044 -0.042 -0.038 -0.082 0.043
## scl(Nr):Rp2 -0.010 0.012 -0.028 -0.043 -0.019 0.043 0.044 -0.048 -0.469
## Inf1:(N):R1 -0.032 0.078 -0.038 -0.082 0.043 -0.029 0.044 -0.042 -0.075
## Inf1:(N):R2 0.012 -0.010 0.043 0.044 -0.048 -0.028 -0.043 -0.019 0.020
## s(N):R2 I1:(N):R1
## Info1
## scale(Neur)
## Rep1
## Rep2
## Inf1:scl(N)
## Info1:Rep1
## Info1:Rep2
## scl(Nr):Rp1
## scl(Nr):Rp2
## Inf1:(N):R1 0.020
## Inf1:(N):R2 -0.017 -0.469For Reputation Normative Data
m <- lmer( toInG ~ reputF * Neur + ( reputF | subID), data = longDf, control=lmerControl(optimizer="bobyqa",
optCtrl=list(maxfun=2e5)))
summary(m)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: toInG ~ reputF * Neur + (reputF | subID)
## Data: longDf
## Control: lmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 2e+05))
##
## REML criterion at convergence: 138996.6
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.9209 -0.7774 0.1993 0.7894 2.0928
##
## Random effects:
## Groups Name Variance Std.Dev. Corr
## subID (Intercept) 2.9374 1.7139
## reputF 0.1905 0.4365 -0.91
## Residual 3.7828 1.9449
## Number of obs: 33088, groups: subID, 335
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 5.23729 0.44678 335.47309 11.722 <2e-16 ***
## reputF -0.22303 0.11236 328.19686 -1.985 0.0480 *
## Neur -0.20810 0.16215 342.33099 -1.283 0.2002
## reputF:Neur 0.08671 0.04062 329.82000 2.135 0.0335 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) reputF Neur
## reputF -0.963
## Neur -0.939 0.903
## reputF:Neur 0.907 -0.939 -0.964ggpredict(m, c("Neur","reputF")) %>% plot()
- significant interaction of neuroticism with reputational cost
Peripheral
longDf$VoteN <- as.factor(longDf$VoteN)
contrasts(longDf$VoteN) <- contr.sum(3)
m <- lmer( toInG ~ VoteN * reputF + ( reputF | subID) + (1 | issues), data = longDf, control=lmerControl(optimizer="bobyqa",
optCtrl=list(maxfun=2e5)))
summary(m)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: toInG ~ VoteN * reputF + (reputF | subID) + (1 | issues)
## Data: longDf
## Control: lmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 2e+05))
##
## REML criterion at convergence: 157067.7
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.9757 -0.7618 0.1776 0.7810 2.6261
##
## Random effects:
## Groups Name Variance Std.Dev. Corr
## subID (Intercept) 3.4029 1.8447
## reputF 0.1607 0.4009 -0.94
## issues (Intercept) 0.2282 0.4777
## Residual 3.5454 1.8829
## Number of obs: 37917, groups: subID, 384; issues, 100
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 4.76424 0.22349 406.12698 21.318 < 2e-16 ***
## VoteN1 0.84112 0.30138 441.01073 2.791 0.00548 **
## VoteN2 -0.26960 0.40134 377.03261 -0.672 0.50215
## reputF -0.01399 0.05146 381.34488 -0.272 0.78585
## VoteN1:reputF -0.09480 0.07205 417.35338 -1.316 0.18900
## VoteN2:reputF 0.04787 0.09458 381.59885 0.506 0.61307
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) VoteN1 VoteN2 reputF VtN1:F
## VoteN1 -0.474
## VoteN2 0.729 -0.633
## reputF -0.949 0.494 -0.722
## VoteN1:rptF 0.458 -0.981 0.601 -0.507
## VoteN2:rptF -0.708 0.605 -0.972 0.741 -0.609plot_model(m, type = "pred", terms = c("VoteN"))
ggpredict(m, c("VoteN","reputF")) %>% plot()
- significant trump main effect (they move towards outgroup) and they exhibit the opposite response towards reputational costs. Others move towards ingroup on issues perceived as costly for supporting, whereas trump voters move towards outgroup on issues perceived as costly for supporting.
m <- lmer( toInG ~ VoteN * honest + ( honest | subID) + (1 | issues), data = longDf, control=lmerControl(optimizer="bobyqa",
optCtrl=list(maxfun=2e5)))
summary(m)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: toInG ~ VoteN * honest + (honest | subID) + (1 | issues)
## Data: longDf
## Control: lmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 2e+05))
##
## REML criterion at convergence: 155363.6
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -3.4424 -0.7315 0.1795 0.7474 2.7147
##
## Random effects:
## Groups Name Variance Std.Dev. Corr
## subID (Intercept) 7.2205 2.6871
## honest 0.4376 0.6615 -0.98
## issues (Intercept) 0.2160 0.4648
## Residual 3.3752 1.8372
## Number of obs: 37917, groups: subID, 384; issues, 100
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 5.61570 0.65545 131.54122 8.568 2.47e-14 ***
## VoteN1 -3.71553 0.31304 380.33233 -11.869 < 2e-16 ***
## VoteN2 -1.33531 0.50395 381.09506 -2.650 0.00839 **
## honest -0.20486 0.14700 137.23550 -1.394 0.16569
## VoteN1:honest 0.92641 0.07381 381.09186 12.551 < 2e-16 ***
## VoteN2:honest 0.28009 0.11882 381.73092 2.357 0.01892 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) VoteN1 VoteN2 honest VtN1:h
## VoteN1 -0.257
## VoteN2 0.312 -0.797
## honest -0.994 0.266 -0.323
## VoteN1:hnst 0.253 -0.985 0.785 -0.270
## VoteN2:hnst -0.308 0.785 -0.985 0.328 -0.797ggpredict(m, c("VoteN","honest")) %>% plot()
- trump voters will move towards outgroup on issues that are perceived as more honest and towards ingroup on issues perceived as less honest. Biden supporters will move towards ingroup on issues perceived as most honest and towards outgroup on issues perceived as less honest.
m <- lmer( toInG ~ VoteN * change + ( change | subID) + (1 | issues), data = longDf, control=lmerControl(optimizer="bobyqa",
optCtrl=list(maxfun=2e5)))
summary(m)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: toInG ~ VoteN * change + (change | subID) + (1 | issues)
## Data: longDf
## Control: lmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 2e+05))
##
## REML criterion at convergence: 155539.6
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -3.2034 -0.7295 0.1515 0.7624 3.0085
##
## Random effects:
## Groups Name Variance Std.Dev. Corr
## subID (Intercept) 1.5678 1.2521
## change 0.2413 0.4913 -0.93
## issues (Intercept) 0.2230 0.4723
## Residual 3.3885 1.8408
## Number of obs: 37917, groups: subID, 384; issues, 100
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 4.41211 0.29940 132.44350 14.736 <2e-16 ***
## VoteN1 -1.24539 0.14383 377.81752 -8.659 <2e-16 ***
## VoteN2 -0.26319 0.23159 378.84539 -1.136 0.256
## change 0.08587 0.09290 150.88445 0.924 0.357
## VoteN1:change 0.51478 0.05120 379.31620 10.054 <2e-16 ***
## VoteN2:change 0.05690 0.08241 379.88561 0.690 0.490
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) VoteN1 VoteN2 change VtN1:c
## VoteN1 -0.259
## VoteN2 0.314 -0.798
## change -0.973 0.279 -0.339
## VoteN1:chng 0.243 -0.939 0.749 -0.297
## VoteN2:chng -0.295 0.749 -0.939 0.360 -0.797ggpredict(m, c("VoteN","change")) %>% plot()
- Trump voters move towards outgroup on issues perceived as more changeable and towards ingroup on issues perceived as less changeable. Biden voters exhibit opposite pattern.
m <- lmer( toInG ~ VoteN * affPol + ( 1 | subID) + (1 | issues), data = longDf, control=lmerControl(optimizer="bobyqa",
optCtrl=list(maxfun=2e5)))
summary(m)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: toInG ~ VoteN * affPol + (1 | subID) + (1 | issues)
## Data: longDf
## Control: lmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 2e+05))
##
## REML criterion at convergence: 156600.1
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.9940 -0.7642 0.1832 0.7835 2.5849
##
## Random effects:
## Groups Name Variance Std.Dev.
## subID (Intercept) 0.2867 0.5354
## issues (Intercept) 0.2233 0.4726
## Residual 3.5659 1.8884
## Number of obs: 37817, groups: subID, 383; issues, 100
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 4.222e+00 8.287e-02 4.226e+02 50.948 < 2e-16 ***
## VoteN1 1.367e-01 8.234e-02 3.771e+02 1.660 0.09766 .
## VoteN2 8.438e-02 1.198e-01 3.777e+02 0.704 0.48172
## affPol 9.363e-03 1.293e-03 3.768e+02 7.244 2.48e-12 ***
## VoteN1:affPol 4.317e-03 1.463e-03 3.767e+02 2.952 0.00336 **
## VoteN2:affPol -6.123e-04 2.404e-03 3.768e+02 -0.255 0.79908
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) VoteN1 VoteN2 affPol VtN1:P
## VoteN1 -0.365
## VoteN2 0.513 -0.733
## affPol -0.618 0.367 -0.519
## VoteN1:ffPl 0.322 -0.775 0.614 -0.636
## VoteN2:ffPl -0.404 0.544 -0.739 0.784 -0.826ggpredict(m, c("VoteN","affPol")) %>% plot()
- More affectively polarized Biden and Other votes will move more towards ingroup more. Trump voters more tempered.
infoDf <- subset(longDf, Info == "Info")
m <- lmer( toInG ~ VoteN * inMinOut + ( inMinOut | subID) + (1 | issues), data = longDf, control=lmerControl(optimizer="bobyqa",
optCtrl=list(maxfun=2e5)))
summary(m)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: toInG ~ VoteN * inMinOut + (inMinOut | subID) + (1 | issues)
## Data: longDf
## Control: lmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 2e+05))
##
## REML criterion at convergence: 154001.3
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -4.0022 -0.7284 0.1364 0.7558 2.8240
##
## Random effects:
## Groups Name Variance Std.Dev. Corr
## subID (Intercept) 0.3737 0.6113
## inMinOut 0.4430 0.6656 -0.63
## issues (Intercept) 0.2485 0.4985
## Residual 3.2587 1.8052
## Number of obs: 37917, groups: subID, 384; issues, 100
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 4.72759 0.06999 282.93121 67.550 < 2e-16 ***
## VoteN1 0.35342 0.05583 381.46723 6.330 6.88e-10 ***
## VoteN2 -0.09436 0.08983 381.45123 -1.050 0.29419
## inMinOut 1.35077 0.06405 363.87023 21.088 < 2e-16 ***
## VoteN1:inMinOut -0.23845 0.08609 440.53941 -2.770 0.00585 **
## VoteN2:inMinOut -0.17179 0.11776 365.02397 -1.459 0.14546
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) VoteN1 VoteN2 inMnOt VN1:MO
## VoteN1 -0.429
## VoteN2 0.520 -0.797
## inMinOut -0.336 0.287 -0.361
## VotN1:nMnOt 0.139 -0.418 0.338 -0.522
## VotN2:nMnOt -0.256 0.389 -0.481 0.744 -0.644ggpredict(m, c("VoteN","inMinOut")) %>% plot()
- Biden voters incorporate Democrat favoring info and reject Republican favoring info most whereas Trump supports more equally weight Democrat and Republican favoring info.
MODELS THAT WORK
m <- lmer( scale(eval) ~ RepN * Info * scale(MLAM) + scale(bootEvalIn) + ( scale(bootEvalIn) + scale(reput) | subID), data = longDf, control=lmerControl(optimizer="bobyqa",
optCtrl=list(maxfun=2e5)))
summary(m)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: scale(eval) ~ RepN * Info * scale(MLAM) + scale(bootEvalIn) +
## (scale(bootEvalIn) + scale(reput) | subID)
## Data: longDf
## Control: lmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 2e+05))
##
## REML criterion at convergence: 71750
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -4.0475 -0.5662 0.0733 0.6213 3.7942
##
## Random effects:
## Groups Name Variance Std.Dev. Corr
## subID (Intercept) 0.03186 0.1785
## scale(bootEvalIn) 0.05320 0.2307 -0.33
## scale(reput) 0.01541 0.1241 0.09 0.31
## Residual 0.48057 0.6932
## Number of obs: 33156, groups: subID, 336
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 0.010636 0.017605 337.455295 0.604 0.5462
## RepNNon -0.019371 0.024071 321.526016 -0.805 0.4216
## RepNOut -0.014946 0.024808 322.095333 -0.602 0.5473
## Info1 0.021300 0.017277 321.118176 1.233 0.2185
## scale(MLAM) 0.016058 0.017842 322.488276 0.900 0.3688
## scale(bootEvalIn) 0.647243 0.012732 333.376886 50.836 <2e-16 ***
## RepNNon:Info1 -0.047001 0.024071 321.525790 -1.953 0.0517 .
## RepNOut:Info1 -0.041188 0.024808 322.092052 -1.660 0.0978 .
## RepNNon:scale(MLAM) -0.006983 0.025459 322.542430 -0.274 0.7841
## RepNOut:scale(MLAM) 0.003273 0.024214 322.486456 0.135 0.8926
## Info1:scale(MLAM) 0.002131 0.017842 322.489449 0.119 0.9050
## RepNNon:Info1:scale(MLAM) 0.034944 0.025459 322.544416 1.373 0.1708
## RepNOut:Info1:scale(MLAM) -0.008353 0.024214 322.486532 -0.345 0.7304
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1##
## Correlation matrix not shown by default, as p = 13 > 12.
## Use print(x, correlation=TRUE) or
## vcov(x) if you need itm <- lmer( scale(eval) ~ RepN * scale(SGO.sw) + scale(bootEvalIn) + ( scale(bootEvalIn) + scale(reput) | subID), data = longDf, control=lmerControl(optimizer="bobyqa",
optCtrl=list(maxfun=2e5)))
summary(m)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula:
## scale(eval) ~ RepN * scale(SGO.sw) + scale(bootEvalIn) + (scale(bootEvalIn) +
## scale(reput) | subID)
## Data: longDf
## Control: lmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 2e+05))
##
## REML criterion at convergence: 68748.8
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -4.0513 -0.5678 0.0753 0.6187 3.7868
##
## Random effects:
## Groups Name Variance Std.Dev. Corr
## subID (Intercept) 0.03084 0.1756
## scale(bootEvalIn) 0.04775 0.2185 -0.28
## scale(reput) 0.01536 0.1239 0.10 0.28
## Residual 0.48135 0.6938
## Number of obs: 31770, groups: subID, 322
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 6.837e-05 1.794e-02 3.251e+02 0.004 0.9970
## RepNNon -4.440e-04 2.467e-02 3.137e+02 -0.018 0.9857
## RepNOut 4.593e-03 2.506e-02 3.143e+02 0.183 0.8547
## scale(SGO.sw) 4.289e-02 2.057e-02 3.142e+02 2.086 0.0378 *
## scale(bootEvalIn) 6.513e-01 1.247e-02 3.185e+02 52.227 <2e-16 ***
## RepNNon:scale(SGO.sw) -3.099e-02 2.662e-02 3.140e+02 -1.164 0.2452
## RepNOut:scale(SGO.sw) -5.208e-02 2.609e-02 3.149e+02 -1.996 0.0468 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) RepNNn RepNOt s(SGO. sc(EI) RNN:(S
## RepNNon -0.708
## RepNOut -0.698 0.507
## scl(SGO.sw) -0.037 0.027 0.027
## scl(btEvlI) -0.161 0.001 0.002 -0.001
## RpNN:(SGO.) 0.029 0.009 -0.021 -0.773 0.001
## RpNO:(SGO.) 0.029 -0.022 -0.038 -0.788 0.001 0.609m <- lmer( scale(toInG) ~ RepN * scale(politic) * scale(affPol) + ( scale(politic) | subID) + (1 | issues), data = repDf, control=lmerControl(optimizer="bobyqa",
optCtrl=list(maxfun=2e5)))
summary(m)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula:
## scale(toInG) ~ RepN * scale(politic) * scale(affPol) + (scale(politic) |
## subID) + (1 | issues)
## Data: repDf
## Control: lmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 2e+05))
##
## REML criterion at convergence: 44163.8
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -4.2980 -0.6722 0.0478 0.6731 3.3918
##
## Random effects:
## Groups Name Variance Std.Dev. Corr
## subID (Intercept) 0.04914 0.2217
## scale(politic) 0.02067 0.1438 0.46
## issues (Intercept) 0.30995 0.5567
## Residual 0.54229 0.7364
## Number of obs: 19284, groups: subID, 195; issues, 100
##
## Fixed effects:
## Estimate Std. Error df t value
## (Intercept) -0.034922 0.064045 160.045966 -0.545
## RepNNon 0.058288 0.043240 189.015037 1.348
## RepNOut 0.039312 0.041709 188.994535 0.943
## scale(politic) -0.257300 0.059824 126.766130 -4.301
## scale(affPol) 0.188818 0.034738 189.040776 5.435
## RepNNon:scale(politic) 0.013590 0.029903 189.081331 0.454
## RepNOut:scale(politic) 0.024448 0.028843 189.025313 0.848
## RepNNon:scale(affPol) 0.012443 0.049318 189.077633 0.252
## RepNOut:scale(affPol) -0.059063 0.041752 188.948514 -1.415
## scale(politic):scale(affPol) 0.037345 0.024019 188.964761 1.555
## RepNNon:scale(politic):scale(affPol) 0.005315 0.034107 189.170200 0.156
## RepNOut:scale(politic):scale(affPol) -0.028815 0.028870 188.921518 -0.998
## Pr(>|t|)
## (Intercept) 0.586
## RepNNon 0.179
## RepNOut 0.347
## scale(politic) 3.36e-05 ***
## scale(affPol) 1.67e-07 ***
## RepNNon:scale(politic) 0.650
## RepNOut:scale(politic) 0.398
## RepNNon:scale(affPol) 0.801
## RepNOut:scale(affPol) 0.159
## scale(politic):scale(affPol) 0.122
## RepNNon:scale(politic):scale(affPol) 0.876
## RepNOut:scale(politic):scale(affPol) 0.320
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) RepNNn RepNOt scl(p) scl(P) RpNN:() RpNO:() RNN:(P RNO:(P
## RepNNon -0.362
## RepNOut -0.375 0.556
## scale(pltc) 0.070 -0.104 -0.107
## scale(ffPl) -0.103 0.153 0.158 -0.029
## RpNNn:scl() -0.140 0.386 0.215 -0.268 0.059
## RpNOt:scl() -0.145 0.215 0.387 -0.278 0.061 0.556
## RpNNn:sc(P) 0.073 -0.098 -0.111 0.021 -0.704 -0.038 -0.043
## RpNOt:sc(P) 0.086 -0.127 -0.091 0.025 -0.832 -0.049 -0.035 0.586
## scl(pl):(P) -0.040 0.059 0.061 -0.076 0.386 0.152 0.158 -0.272 -0.322
## RpNN:():(P) 0.028 -0.038 -0.043 0.054 -0.272 -0.097 -0.111 0.386 0.226
## RpNO:():(P) 0.033 -0.049 -0.035 0.063 -0.321 -0.127 -0.091 0.226 0.386
## s():(P RNN:():
## RepNNon
## RepNOut
## scale(pltc)
## scale(ffPl)
## RpNNn:scl()
## RpNOt:scl()
## RpNNn:sc(P)
## RpNOt:sc(P)
## scl(pl):(P)
## RpNN:():(P) -0.704
## RpNO:():(P) -0.832 0.586m <- lmer( scale(toInG) ~ RepN * scale(politic) * scale(affPol) + ( scale(politic) | subID) + ( 1 | issues), data = demDf, control=lmerControl(optimizer="bobyqa",
optCtrl=list(maxfun=2e5)))
summary(m)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula:
## scale(toInG) ~ RepN * scale(politic) * scale(affPol) + (scale(politic) |
## subID) + (1 | issues)
## Data: demDf
## Control: lmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 2e+05))
##
## REML criterion at convergence: 41829.1
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -4.3065 -0.4977 0.1294 0.5884 3.9200
##
## Random effects:
## Groups Name Variance Std.Dev. Corr
## subID (Intercept) 0.06255 0.2501
## scale(politic) 0.02853 0.1689 -0.50
## issues (Intercept) 0.31430 0.5606
## Residual 0.52142 0.7221
## Number of obs: 18533, groups: subID, 188; issues, 100
##
## Fixed effects:
## Estimate Std. Error df t value
## (Intercept) -1.434e-02 6.456e-02 1.603e+02 -0.222
## RepNNon 3.743e-04 4.484e-02 1.818e+02 0.008
## RepNOut 2.876e-02 5.045e-02 1.818e+02 0.570
## scale(politic) 2.571e-01 6.045e-02 1.283e+02 4.253
## scale(affPol) 1.311e-01 3.936e-02 1.817e+02 3.330
## RepNNon:scale(politic) 1.271e-02 3.165e-02 1.800e+02 0.401
## RepNOut:scale(politic) 3.062e-02 3.561e-02 1.799e+02 0.860
## RepNNon:scale(affPol) -8.381e-02 5.201e-02 1.822e+02 -1.611
## RepNOut:scale(affPol) -2.997e-02 4.927e-02 1.817e+02 -0.608
## scale(politic):scale(affPol) -7.876e-03 2.778e-02 1.797e+02 -0.284
## RepNNon:scale(politic):scale(affPol) 2.002e-02 3.673e-02 1.807e+02 0.545
## RepNOut:scale(politic):scale(affPol) -1.987e-02 3.477e-02 1.798e+02 -0.572
## Pr(>|t|)
## (Intercept) 0.82456
## RepNNon 0.99335
## RepNOut 0.56935
## scale(politic) 4.03e-05 ***
## scale(affPol) 0.00105 **
## RepNNon:scale(politic) 0.68860
## RepNOut:scale(politic) 0.39096
## RepNNon:scale(affPol) 0.10882
## RepNOut:scale(affPol) 0.54378
## scale(politic):scale(affPol) 0.77709
## RepNNon:scale(politic):scale(affPol) 0.58629
## RepNOut:scale(politic):scale(affPol) 0.56831
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) RepNNn RepNOt scl(p) scl(P) RpNN:() RpNO:() RNN:(P RNO:(P
## RepNNon -0.354
## RepNOut -0.315 0.453
## scale(pltc) -0.082 0.119 0.105
## scale(ffPl) -0.141 0.204 0.181 0.047
## RpNNn:scl() 0.157 -0.444 -0.201 -0.267 -0.090
## RpNOt:scl() 0.140 -0.201 -0.444 -0.237 -0.080 0.453
## RpNNn:sc(P) 0.107 -0.134 -0.137 -0.036 -0.757 0.059 0.061
## RpNOt:sc(P) 0.113 -0.163 -0.041 -0.038 -0.799 0.072 0.018 0.605
## scl(pl):(P) 0.063 -0.090 -0.080 -0.107 -0.444 0.204 0.181 0.336 0.355
## RpNN:():(P) -0.047 0.059 0.061 0.081 0.336 -0.134 -0.137 -0.443 -0.268
## RpNO:():(P) -0.050 0.072 0.018 0.085 0.355 -0.163 -0.041 -0.268 -0.444
## s():(P RNN:():
## RepNNon
## RepNOut
## scale(pltc)
## scale(ffPl)
## RpNNn:scl()
## RpNOt:scl()
## RpNNn:sc(P)
## RpNOt:sc(P)
## scl(pl):(P)
## RpNN:():(P) -0.756
## RpNO:():(P) -0.799 0.604m <- lmer( scale(toInG) ~ RepN * scale(affPol) * partyN + ( 1 | subID) + ( 1 | issues), data = longDf, control=lmerControl(optimizer="bobyqa",
optCtrl=list(maxfun=2e5)))
summary(m)## Linear mixed model fit by REML. t-tests use Satterthwaite's method [
## lmerModLmerTest]
## Formula: scale(toInG) ~ RepN * scale(affPol) * partyN + (1 | subID) +
## (1 | issues)
## Data: longDf
## Control: lmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 2e+05))
##
## REML criterion at convergence: 100739.2
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.9782 -0.7623 0.1846 0.7852 2.5920
##
## Random effects:
## Groups Name Variance Std.Dev.
## subID (Intercept) 0.04831 0.2198
## issues (Intercept) 0.05110 0.2260
## Residual 0.81595 0.9033
## Number of obs: 37817, groups: subID, 383; issues, 100
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) -1.729e-02 3.195e-02 2.922e+02 -0.541 0.589
## RepNNon 3.926e-02 3.090e-02 3.710e+02 1.271 0.205
## RepNOut 3.119e-02 3.179e-02 3.707e+02 0.981 0.327
## scale(affPol) 1.588e-01 2.751e-02 3.705e+02 5.773 1.65e-08
## partyN1 2.499e-01 2.258e-02 3.708e+02 11.067 < 2e-16
## RepNNon:scale(affPol) -4.364e-02 3.724e-02 3.718e+02 -1.172 0.242
## RepNOut:scale(affPol) -4.318e-02 3.401e-02 3.705e+02 -1.269 0.205
## RepNNon:partyN1 -2.115e-02 3.090e-02 3.710e+02 -0.684 0.494
## RepNOut:partyN1 9.378e-04 3.179e-02 3.707e+02 0.029 0.976
## scale(affPol):partyN1 -6.100e-03 2.751e-02 3.705e+02 -0.222 0.825
## RepNNon:scale(affPol):partyN1 -5.567e-02 3.724e-02 3.718e+02 -1.495 0.136
## RepNOut:scale(affPol):partyN1 7.106e-03 3.401e-02 3.705e+02 0.209 0.835
##
## (Intercept)
## RepNNon
## RepNOut
## scale(affPol) ***
## partyN1 ***
## RepNNon:scale(affPol)
## RepNOut:scale(affPol)
## RepNNon:partyN1
## RepNOut:partyN1
## scale(affPol):partyN1
## RepNNon:scale(affPol):partyN1
## RepNOut:scale(affPol):partyN1
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) RepNNn RepNOt scl(P) prtyN1 RpNN:(P) RpNO:(P) RNN:N1 RNO:N1
## RepNNon -0.516
## RepNOut -0.502 0.519
## scale(ffPl) -0.222 0.230 0.224
## partyN1 0.017 -0.017 -0.017 -0.273
## RpNNn:sc(P) 0.164 -0.178 -0.165 -0.739 0.202
## RpNOt:sc(P) 0.180 -0.186 -0.133 -0.809 0.221 0.598
## RpNNn:prtN1 -0.012 -0.008 0.012 0.200 -0.731 -0.233 -0.162
## RpNOt:prtN1 -0.012 0.012 0.114 0.194 -0.710 -0.143 -0.171 0.519
## scl(ffP):N1 -0.193 0.200 0.194 0.370 -0.315 -0.273 -0.299 0.230 0.224
## RpNN:(P):N1 0.143 -0.233 -0.143 -0.273 0.233 0.307 0.221 -0.178 -0.165
## RpNO:(P):N1 0.156 -0.162 -0.171 -0.299 0.255 0.221 0.404 -0.186 -0.133
## s(P):N RNN:(P):
## RepNNon
## RepNOut
## scale(ffPl)
## partyN1
## RpNNn:sc(P)
## RpNOt:sc(P)
## RpNNn:prtN1
## RpNOt:prtN1
## scl(ffP):N1
## RpNN:(P):N1 -0.739
## RpNO:(P):N1 -0.809 0.598