This document lists primary analyses in order in the following paper:
Freedman, G. & Dainer-Best, J. (2022). Who is more willing to engage in social rejection? The roles of self-esteem, rejection sensitivity, and negative affect in social rejection decisions.
This paper has been accepted in The Journal of Social Psychology; a pre-print can be found at https://doi.org/10.31234/osf.io/jdx2q
knitr::opts_chunk$set(echo = TRUE,
message = FALSE)
if(! require(pacman)) install.packages("pacman")## Loading required package: pacmanpacman::p_load(lme4, here, gt, arm, corrr, tidyverse)here::i_am("Emotions and Rejection.Rproj")ear1 <- read_csv(here::here("data and codebooks", "study1a_processed.csv"))
ear1 <- ear1 %>% filter(filter == 1)As predicted, self-esteem was negatively correlated with forecasted difficulty of engaging in rejection
cor.test(ear1$OverallDifficulty_Avg, ear1$RosenbergSelfEsteem_Sum)##
## Pearson's product-moment correlation
##
## data: ear1$OverallDifficulty_Avg and ear1$RosenbergSelfEsteem_Sum
## t = -4.9666, df = 209, p-value = 1.414e-06
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## -0.4406391 -0.1985528
## sample estimates:
## cor
## -0.3249079and [self-esteem was] negatively correlated with forecasting negative emotions in response to thinking about hypothetical rejection
cor.test(ear1$OverallNegEmo_Avg, ear1$RosenbergSelfEsteem_Sum)##
## Pearson's product-moment correlation
##
## data: ear1$OverallNegEmo_Avg and ear1$RosenbergSelfEsteem_Sum
## t = -6.2881, df = 209, p-value = 1.848e-09
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## -0.5066335 -0.2788108
## sample estimates:
## cor
## -0.3988589contrary to hypotheses, self-esteem was not associated with likelihood of engaging in rejection in the vignettes
cor.test(ear1$OverallLikelyEnd_Avg, ear1$RosenbergSelfEsteem_Sum)##
## Pearson's product-moment correlation
##
## data: ear1$OverallLikelyEnd_Avg and ear1$RosenbergSelfEsteem_Sum
## t = 0.69969, df = 204, p-value = 0.4849
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## -0.08836291 0.18439735
## sample estimates:
## cor
## 0.04892943Self-esteem was also not associated with percent of relationships participants had ended
cor.test(ear1$PercentRelationshipsYouEnded, ear1$RosenbergSelfEsteem_Sum)##
## Pearson's product-moment correlation
##
## data: ear1$PercentRelationshipsYouEnded and ear1$RosenbergSelfEsteem_Sum
## t = -1.8071, df = 153, p-value = 0.07272
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## -0.29547328 0.01339478
## sample estimates:
## cor
## -0.1445587there was a positive correlation between general distress and forecasted difficulty of engaging in rejection
cor.test(ear1$MASQ_GD, ear1$OverallDifficulty_Avg)##
## Pearson's product-moment correlation
##
## data: ear1$MASQ_GD and ear1$OverallDifficulty_Avg
## t = 4.8644, df = 212, p-value = 2.237e-06
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.1908661 0.4326033
## sample estimates:
## cor
## 0.3168713there was no association between general distress and the overall likelihood of rejecting in the vignettes
cor.test(ear1$MASQ_GD, ear1$OverallLikelyEnd_Avg)##
## Pearson's product-moment correlation
##
## data: ear1$MASQ_GD and ear1$OverallLikelyEnd_Avg
## t = 0.24292, df = 207, p-value = 0.8083
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## -0.1191061 0.1522472
## sample estimates:
## cor
## 0.01688137An exploratory linear mixed effects model showed, however, that there was an interaction between relationship type and general distress in the prediction of likelihood of rejecting
reltype_gd_likelihood <- ear1 %>%
dplyr::select(ID, MASQ_GD,
LikelihoodEndingFriendships_Avg,
LikelihoodEndingRomantic_Avg) %>%
pivot_longer(
cols = c(LikelihoodEndingFriendships_Avg,
LikelihoodEndingRomantic_Avg),
names_to = "RelationshipType",
values_to = "LikelihoodEnding") %>%
mutate(
RelationshipType =
factor(RelationshipType,
levels =
c("LikelihoodEndingFriendships_Avg",
"LikelihoodEndingRomantic_Avg"),
labels = c("Friendship", "Romantic"))) %>%
na.omit()
summary(nlme::lme(LikelihoodEnding ~ RelationshipType * MASQ_GD, random = ~ 1 | ID, data = reltype_gd_likelihood))## Linear mixed-effects model fit by REML
## Data: reltype_gd_likelihood
## AIC BIC logLik
## 820.0477 844.1739 -404.0239
##
## Random effects:
## Formula: ~1 | ID
## (Intercept) Residual
## StdDev: 0.3467155 0.5335001
##
## Fixed effects: LikelihoodEnding ~ RelationshipType * MASQ_GD
## Value Std.Error DF t-value p-value
## (Intercept) 3.1258980 0.14031235 206 22.278138 0.0000
## RelationshipTypeRomantic 0.9021859 0.16638223 206 5.422369 0.0000
## MASQ_GD 0.0107422 0.00456949 206 2.350851 0.0197
## RelationshipTypeRomantic:MASQ_GD -0.0198782 0.00541850 206 -3.668576 0.0003
## Correlation:
## (Intr) RltnTR MASQ_G
## RelationshipTypeRomantic -0.593
## MASQ_GD -0.949 0.563
## RelationshipTypeRomantic:MASQ_GD 0.563 -0.949 -0.593
##
## Standardized Within-Group Residuals:
## Min Q1 Med Q3 Max
## -3.11833516 -0.52844891 0.03927846 0.62744346 1.84277314
##
## Number of Observations: 416
## Number of Groups: 208model_reltype_gd_likelihood <- lme4::lmer(
LikelihoodEnding ~ RelationshipType * MASQ_GD + (1 | ID), data = reltype_gd_likelihood)
#summary(model_reltype_gd_likelihood)
model_reltype_gd_likelihood.sim <- arm::sim(model_reltype_gd_likelihood)
#simulated uncertainty for fixed effects b weights
fixef.model_reltype_gd_likelihood.sim <- lme4::fixef(model_reltype_gd_likelihood.sim)
# colnames(fixef.model_reltype_gd_likelihood.sim)
quantile(fixef.model_reltype_gd_likelihood.sim[, 4], # interaction
probs = c(.025, .975))## 2.5% 97.5%
## -0.02857632 -0.01052478quantile(fixef.model_reltype_gd_likelihood.sim[, 2], # RelType
probs = c(.025, .975))## 2.5% 97.5%
## 0.5852915 1.1560440there was no association between general distress and the percent of romantic relationships ended by the participant
cor.test(ear1$MASQ_GD, ear1$PercentRelationshipsYouEnded)##
## Pearson's product-moment correlation
##
## data: ear1$MASQ_GD and ear1$PercentRelationshipsYouEnded
## t = 1.6013, df = 156, p-value = 0.1113
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## -0.02955896 0.27779272
## sample estimates:
## cor
## 0.1271683linear mixed effects model with predictors of general distress and whether rejection was explicit or passive found no effect of general distress on likelihood of rejection
gdrejtype_data <- ear1 %>%
select(ID, MASQ_GD, OverallLikelyExplicit_Avg,
OverallLikelyGhost_Avg) %>%
pivot_longer(cols = c(OverallLikelyExplicit_Avg,
OverallLikelyGhost_Avg),
names_to = "RejectionType",
values_to = "Likelihood_Avg") %>%
mutate(RejectionType =
factor(RejectionType,
levels =
c("OverallLikelyExplicit_Avg",
"OverallLikelyGhost_Avg"),
labels = c("Explicit", "Ghost"))) %>%
na.omit()
gd_rejtype_likelihood <- nlme::lme(Likelihood_Avg ~ RejectionType * MASQ_GD, random = ~ 1 | ID, data = gdrejtype_data)
summary(gd_rejtype_likelihood)## Linear mixed-effects model fit by REML
## Data: gdrejtype_data
## AIC BIC logLik
## 992.3765 1016.646 -490.1882
##
## Random effects:
## Formula: ~1 | ID
## (Intercept) Residual
## StdDev: 3.629812e-05 0.7456459
##
## Fixed effects: Likelihood_Avg ~ RejectionType * MASQ_GD
## Value Std.Error DF t-value p-value
## (Intercept) 3.629775 0.16234462 211 22.358456 0.0000
## RejectionTypeGhost -0.843486 0.22958996 211 -3.673882 0.0003
## MASQ_GD -0.001333 0.00529821 211 -0.251620 0.8016
## RejectionTypeGhost:MASQ_GD 0.005871 0.00749280 211 0.783555 0.4342
## Correlation:
## (Intr) RjctTG MASQ_G
## RejectionTypeGhost -0.707
## MASQ_GD -0.949 0.671
## RejectionTypeGhost:MASQ_GD 0.671 -0.949 -0.707
##
## Standardized Within-Group Residuals:
## Min Q1 Med Q3 Max
## -2.68166049 -0.62254326 0.02532607 0.69454474 2.40617999
##
## Number of Observations: 426
## Number of Groups: 213gd_rejtype_likelihood <- lme4::lmer(Likelihood_Avg ~ RejectionType * MASQ_GD + (1 | ID), data = gdrejtype_data)
#summary(gd_rejtype_likelihood)
gd_rejtype_likelihood.sim <- arm::sim(gd_rejtype_likelihood)
#simulated uncertainty for fixed effects
fixef.gd_rejtype_likelihood.sim <- lme4::fixef(gd_rejtype_likelihood.sim)
# colnames(fixef.gd_rejtype_likelihood.sim)
quantile(fixef.gd_rejtype_likelihood.sim[, 4], # interaction
probs = c(.025, .975))## 2.5% 97.5%
## -0.008973711 0.018922296general distress was positively correlated with negative emotions in response to thinking about rejecting in the hypothetical scenarios
cor.test(ear1$MASQ_GD, ear1$OverallNegEmo_Avg)##
## Pearson's product-moment correlation
##
## data: ear1$MASQ_GD and ear1$OverallNegEmo_Avg
## t = 6.1861, df = 212, p-value = 3.133e-09
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.2711394 0.4989839
## sample estimates:
## cor
## 0.3910362anxiety was not associated with forecasted difficulty of engaging in rejection
cor.test(ear1$MASQ_AA, ear1$OverallDifficulty_Avg)##
## Pearson's product-moment correlation
##
## data: ear1$MASQ_AA and ear1$OverallDifficulty_Avg
## t = 1.4082, df = 212, p-value = 0.1605
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## -0.03834226 0.22744924
## sample estimates:
## cor
## 0.09626935[anxiety was not associated with] likelihood of rejecting in rejection vignettes
cor.test(ear1$MASQ_AA, ear1$OverallLikelyEnd_Avg)##
## Pearson's product-moment correlation
##
## data: ear1$MASQ_AA and ear1$OverallLikelyEnd_Avg
## t = 0.20995, df = 207, p-value = 0.8339
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## -0.1213642 0.1500083
## sample estimates:
## cor
## 0.01459073[anxiety was not associated with] the percent of romantic relationships ended by the participant
cor.test(ear1$MASQ_AA, ear1$PercentRelationshipsYouEnded)##
## Pearson's product-moment correlation
##
## data: ear1$MASQ_AA and ear1$PercentRelationshipsYouEnded
## t = 1.1816, df = 156, p-value = 0.2392
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## -0.06288014 0.24669727
## sample estimates:
## cor
## 0.0941849The hypothesis that individuals with more anxiety symptoms would be more likely to engage in ghosting than explicit rejection was likewise not supported
aarejtype_data <- ear1 %>%
select(ID, MASQ_AA, OverallLikelyExplicit_Avg,
OverallLikelyGhost_Avg) %>%
pivot_longer(cols = c(OverallLikelyExplicit_Avg,
OverallLikelyGhost_Avg),
names_to = "RejectionType",
values_to = "Likelihood_Avg") %>%
mutate(RejectionType =
factor(RejectionType,
levels =
c("OverallLikelyExplicit_Avg",
"OverallLikelyGhost_Avg"),
labels = c("Explicit", "Ghost"))) %>%
na.omit()
aa_rejtype_likelihood <- nlme::lme(Likelihood_Avg ~ RejectionType * MASQ_AA, random = ~ 1 | ID, data = aarejtype_data)
summary(aa_rejtype_likelihood)## Linear mixed-effects model fit by REML
## Data: aarejtype_data
## AIC BIC logLik
## 991.5845 1015.854 -489.7922
##
## Random effects:
## Formula: ~1 | ID
## (Intercept) Residual
## StdDev: 1.86421e-05 0.7453446
##
## Fixed effects: Likelihood_Avg ~ RejectionType * MASQ_AA
## Value Std.Error DF t-value p-value
## (Intercept) 3.638303 0.13891879 211 26.190146 0.0000
## RejectionTypeGhost -0.849530 0.19646083 211 -4.324170 0.0000
## MASQ_AA -0.002170 0.00592797 211 -0.366139 0.7146
## RejectionTypeGhost:MASQ_AA 0.008113 0.00838342 211 0.967684 0.3343
## Correlation:
## (Intr) RjctTG MASQ_A
## RejectionTypeGhost -0.707
## MASQ_AA -0.930 0.658
## RejectionTypeGhost:MASQ_AA 0.658 -0.930 -0.707
##
## Standardized Within-Group Residuals:
## Min Q1 Med Q3 Max
## -2.64706643 -0.63121991 0.01211365 0.69449085 2.20858081
##
## Number of Observations: 426
## Number of Groups: 213aa_rejtype_likelihood <- lme4::lmer(Likelihood_Avg ~ RejectionType * MASQ_AA + (1 | ID), data = aarejtype_data)
#summary(aa_rejtype_likelihood)
aa_rejtype_likelihood.sim <- arm::sim(aa_rejtype_likelihood)
#simulated uncertainty for fixed effects
fixef.aa_rejtype_likelihood.sim <- lme4::fixef(aa_rejtype_likelihood.sim)
#colnames(fixef.aa_rejtype_likelihood.sim)
quantile(fixef.aa_rejtype_likelihood.sim[, 4], # interaction
probs = c(.025, .975))## 2.5% 97.5%
## -0.007882003 0.026918718anxiety symptoms were positively correlated with negative emotions in response to thinking about rejecting in the hypothetical scenarios
cor.test(ear1$MASQ_AA, ear1$OverallNegEmo_Avg)##
## Pearson's product-moment correlation
##
## data: ear1$MASQ_AA and ear1$OverallNegEmo_Avg
## t = 3.7504, df = 212, p-value = 0.0002278
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.1193097 0.3711357
## sample estimates:
## cor
## 0.2494349There was a positive correlation between anhedonic depressive symptoms and forecasted difficulty of engaging in rejection
cor.test(ear1$MASQ_AD, ear1$OverallDifficulty_Avg)##
## Pearson's product-moment correlation
##
## data: ear1$MASQ_AD and ear1$OverallDifficulty_Avg
## t = 2.4575, df = 212, p-value = 0.01479
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.03305056 0.29398397
## sample estimates:
## cor
## 0.1664294there was no association between anhedonic depressive symptoms and the likelihood of rejecting in the vignettes
cor.test(ear1$MASQ_AD, ear1$OverallLikelyEnd_Avg)##
## Pearson's product-moment correlation
##
## data: ear1$MASQ_AD and ear1$OverallLikelyEnd_Avg
## t = -0.5045, df = 207, p-value = 0.6144
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## -0.1699497 0.1011523
## sample estimates:
## cor
## -0.0350434[there was no association between anhedonic depressive symptoms and] the percent of romantic relationships ended by the participant
cor.test(ear1$MASQ_AD, ear1$PercentRelationshipsYouEnded)##
## Pearson's product-moment correlation
##
## data: ear1$MASQ_AD and ear1$PercentRelationshipsYouEnded
## t = 0.65966, df = 156, p-value = 0.5104
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## -0.1042571 0.2071760
## sample estimates:
## cor
## 0.05274182a linear mixed effects model with predictors of anhedonic depressive symptoms and whether rejection was explicit or passive found no effect of anhedonic depressive symptoms on likelihood of rejection
adrejtype_data <- ear1 %>%
select(ID, MASQ_AD, OverallLikelyExplicit_Avg,
OverallLikelyGhost_Avg) %>%
pivot_longer(cols = c(OverallLikelyExplicit_Avg,
OverallLikelyGhost_Avg),
names_to = "RejectionType",
values_to = "Likelihood_Avg") %>%
mutate(RejectionType =
factor(RejectionType,
levels =
c("OverallLikelyExplicit_Avg",
"OverallLikelyGhost_Avg"),
labels = c("Explicit", "Ghost"))) %>%
na.omit()
ad_rejtype_likelihood <- nlme::lme(Likelihood_Avg ~ RejectionType * MASQ_AD, random = ~ 1 | ID, data = adrejtype_data)
summary(ad_rejtype_likelihood)## Linear mixed-effects model fit by REML
## Data: adrejtype_data
## AIC BIC logLik
## 991.5338 1015.804 -489.7669
##
## Random effects:
## Formula: ~1 | ID
## (Intercept) Residual
## StdDev: 3.825843e-05 0.7454856
##
## Fixed effects: Likelihood_Avg ~ RejectionType * MASQ_AD
## Value Std.Error DF t-value p-value
## (Intercept) 3.776210 0.19502261 211 19.362933 0.0000
## RejectionTypeGhost -0.838771 0.27580361 211 -3.041188 0.0027
## MASQ_AD -0.006150 0.00624936 211 -0.984027 0.3262
## RejectionTypeGhost:MASQ_AD 0.005513 0.00883794 211 0.623798 0.5334
## Correlation:
## (Intr) RjctTG MASQ_A
## RejectionTypeGhost -0.707
## MASQ_AD -0.965 0.682
## RejectionTypeGhost:MASQ_AD 0.682 -0.965 -0.707
##
## Standardized Within-Group Residuals:
## Min Q1 Med Q3 Max
## -2.5775516 -0.5997020 -0.0186325 0.6921174 2.3324050
##
## Number of Observations: 426
## Number of Groups: 213ad_rejtype_likelihood <- lme4::lmer(Likelihood_Avg ~ RejectionType * MASQ_AD + (1 | ID), data = adrejtype_data)
#summary(ad_rejtype_likelihood)
ad_rejtype_likelihood.sim <- arm::sim(ad_rejtype_likelihood)
#simulated uncertainty for fixed effects
fixef.ad_rejtype_likelihood.sim <- lme4::fixef(ad_rejtype_likelihood.sim)
#colnames(fixef.ad_rejtype_likelihood.sim)
quantile(fixef.ad_rejtype_likelihood.sim[, 4], # interaction
probs = c(.025, .975))## 2.5% 97.5%
## -0.01143030 0.02163718anhedonic depressive symptoms were positively correlated with negative emotions in response to thinking about rejecting in the hypothetical scenarios
cor.test(ear1$MASQ_AD, ear1$OverallNegEmo_Avg)##
## Pearson's product-moment correlation
##
## data: ear1$MASQ_AD and ear1$OverallNegEmo_Avg
## t = 4.1061, df = 212, p-value = 5.743e-05
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.1424953 0.3912970
## sample estimates:
## cor
## 0.2714247Rejection sensitivity was significantly positively correlated with forecasted difficulty of engaging in rejection
cor.test(ear1$RSQ_Score_Fixed, ear1$OverallDifficulty_Avg)##
## Pearson's product-moment correlation
##
## data: ear1$RSQ_Score_Fixed and ear1$OverallDifficulty_Avg
## t = 3.5806, df = 210, p-value = 0.0004259
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.1086302 0.3628878
## sample estimates:
## cor
## 0.239868[Rejection sensitivity was significantly positively correlated with] forecasted negative emotions in response to thinking about hypothetical rejection
cor.test(ear1$RSQ_Score_Fixed, ear1$OverallNegEmo_Avg)##
## Pearson's product-moment correlation
##
## data: ear1$RSQ_Score_Fixed and ear1$OverallNegEmo_Avg
## t = 2.9907, df = 210, p-value = 0.003116
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.06925818 0.32793827
## sample estimates:
## cor
## 0.2021209Rejection sensitivity was not associated with likelihood of engaging in rejection
cor.test(ear1$RSQ_Score_Fixed, ear1$OverallLikelyEnd_Avg)##
## Pearson's product-moment correlation
##
## data: ear1$RSQ_Score_Fixed and ear1$OverallLikelyEnd_Avg
## t = -0.23948, df = 205, p-value = 0.811
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## -0.1527457 0.1199195
## sample estimates:
## cor
## -0.01672401Self-esteem was associated with forecasts of overall difficulty across vignettes
mod_rse_diff <- lm(data = ear1, OverallDifficulty_Avg ~ RosenbergSelfEsteem_Sum + MASQ_GD + MASQ_AD + MASQ_AA)
summary(mod_rse_diff)##
## Call:
## lm(formula = OverallDifficulty_Avg ~ RosenbergSelfEsteem_Sum +
## MASQ_GD + MASQ_AD + MASQ_AA, data = ear1)
##
## Residuals:
## Min 1Q Median 3Q Max
## -2.78844 -0.65061 0.08807 0.62537 2.45789
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 5.914008 0.779673 7.585 1.12e-12 ***
## RosenbergSelfEsteem_Sum -0.044596 0.015869 -2.810 0.00543 **
## MASQ_GD 0.030146 0.010910 2.763 0.00624 **
## MASQ_AD -0.009047 0.009761 -0.927 0.35509
## MASQ_AA -0.026932 0.010000 -2.693 0.00766 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.9505 on 206 degrees of freedom
## (3 observations deleted due to missingness)
## Multiple R-squared: 0.149, Adjusted R-squared: 0.1325
## F-statistic: 9.019 on 4 and 206 DF, p-value: 9.881e-07confint(mod_rse_diff)## 2.5 % 97.5 %
## (Intercept) 4.376846439 7.451169025
## RosenbergSelfEsteem_Sum -0.075882873 -0.013309207
## MASQ_GD 0.008637041 0.051655021
## MASQ_AD -0.028291018 0.010197161
## MASQ_AA -0.046647029 -0.007216917[Self-esteem was associated with] overall negative emotions
mod_rse_negemo <- lm(data = ear1, OverallNegEmo_Avg ~ RosenbergSelfEsteem_Sum + MASQ_GD + MASQ_AD + MASQ_AA)
summary(mod_rse_negemo)##
## Call:
## lm(formula = OverallNegEmo_Avg ~ RosenbergSelfEsteem_Sum + MASQ_GD +
## MASQ_AD + MASQ_AA, data = ear1)
##
## Residuals:
## Min 1Q Median 3Q Max
## -2.51575 -0.45227 0.09141 0.41719 2.00174
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 4.650065 0.591122 7.866 2.02e-13 ***
## RosenbergSelfEsteem_Sum -0.028165 0.012031 -2.341 0.0202 *
## MASQ_GD 0.018016 0.008271 2.178 0.0305 *
## MASQ_AD 0.004956 0.007400 0.670 0.5038
## MASQ_AA -0.003067 0.007582 -0.405 0.6862
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.7207 on 206 degrees of freedom
## (3 observations deleted due to missingness)
## Multiple R-squared: 0.183, Adjusted R-squared: 0.1671
## F-statistic: 11.53 on 4 and 206 DF, p-value: 1.808e-08confint(mod_rse_negemo)## 2.5 % 97.5 %
## (Intercept) 3.484638951 5.815490073
## RosenbergSelfEsteem_Sum -0.051885509 -0.004444196
## MASQ_GD 0.001708355 0.034323185
## MASQ_AD -0.009634620 0.019545862
## MASQ_AA -0.018014471 0.011880154Rejection sensitivity was not associated with either outcome
summary(lm(data = ear1, OverallDifficulty_Avg ~ RSQ_Score_Fixed + MASQ_GD + MASQ_AD + MASQ_AA))##
## Call:
## lm(formula = OverallDifficulty_Avg ~ RSQ_Score_Fixed + MASQ_GD +
## MASQ_AD + MASQ_AA, data = ear1)
##
## Residuals:
## Min 1Q Median 3Q Max
## -2.75411 -0.67303 -0.01492 0.57863 2.58810
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 3.693e+00 2.896e-01 12.751 < 2e-16 ***
## RSQ_Score_Fixed 3.212e-02 1.457e-02 2.205 0.028524 *
## MASQ_GD 3.714e-02 1.001e-02 3.711 0.000266 ***
## MASQ_AD -1.020e-06 9.149e-03 0.000 0.999911
## MASQ_AA -1.882e-02 9.990e-03 -1.884 0.060928 .
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.9556 on 207 degrees of freedom
## (2 observations deleted due to missingness)
## Multiple R-squared: 0.1275, Adjusted R-squared: 0.1107
## F-statistic: 7.564 on 4 and 207 DF, p-value: 1.048e-05summary(lm(data = ear1, OverallNegEmo_Avg ~ RSQ_Score_Fixed + MASQ_GD + MASQ_AD + MASQ_AA))##
## Call:
## lm(formula = OverallNegEmo_Avg ~ RSQ_Score_Fixed + MASQ_GD +
## MASQ_AD + MASQ_AA, data = ear1)
##
## Residuals:
## Min 1Q Median 3Q Max
## -2.44977 -0.41719 0.07895 0.50707 1.92410
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 3.317749 0.222198 14.931 < 2e-16 ***
## RSQ_Score_Fixed 0.012969 0.011174 1.161 0.24712
## MASQ_GD 0.023899 0.007678 3.113 0.00212 **
## MASQ_AD 0.009854 0.007018 1.404 0.16182
## MASQ_AA 0.001798 0.007664 0.235 0.81474
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.7331 on 207 degrees of freedom
## (2 observations deleted due to missingness)
## Multiple R-squared: 0.1563, Adjusted R-squared: 0.14
## F-statistic: 9.585 on 4 and 207 DF, p-value: 3.949e-07From Study 1a for the correlation table
ear1 %>%
select(MASQ_AD, MASQ_AA, MASQ_GD,
RosenbergSelfEsteem_Sum,
RSQ_Score_Fixed,
DifficultyEndFriendship_Avg,
NegEmo_Friendship_Avg,
DifficultyEndRomantic_Avg,
NegEmo_Romantic_Avg) %>%
corrr::correlate(use = "pairwise.complete.obs",
method = "pearson") %>%
dplyr::select(term, DifficultyEndFriendship_Avg,
NegEmo_Friendship_Avg,
DifficultyEndRomantic_Avg,
NegEmo_Romantic_Avg) %>%
filter(term %in% c("MASQ_AD", "MASQ_AA", "MASQ_GD",
"RosenbergSelfEsteem_Sum",
"RSQ_Score_Fixed")) %>%
corrr::fashion() %>%
gt::gt()| term | DifficultyEndFriendship_Avg | NegEmo_Friendship_Avg | DifficultyEndRomantic_Avg | NegEmo_Romantic_Avg |
|---|---|---|---|---|
| MASQ_AD | .18 | .26 | .12 | .24 |
| MASQ_AA | .04 | .18 | .13 | .27 |
| MASQ_GD | .22 | .30 | .34 | .41 |
| RosenbergSelfEsteem_Sum | -.28 | -.33 | -.30 | -.39 |
| RSQ_Score_Fixed | .24 | .18 | .19 | .19 |
# and the p-value / stats on them
ear1 %>%
select(DifficultyEndFriendship_Avg,
NegEmo_Friendship_Avg,
DifficultyEndRomantic_Avg,
NegEmo_Romantic_Avg,
MASQ_AD, MASQ_AA, MASQ_GD,
RosenbergSelfEsteem_Sum,
RSQ_Score_Fixed) %>%
pivot_longer(
cols = c(DifficultyEndFriendship_Avg,
NegEmo_Friendship_Avg,
DifficultyEndRomantic_Avg,
NegEmo_Romantic_Avg),
names_to = "outcomes",
values_to = "out_val") %>%
pivot_longer(
cols = c(MASQ_AD, MASQ_AA, MASQ_GD,
RosenbergSelfEsteem_Sum,
RSQ_Score_Fixed),
names_to = "predictors",
values_to = "pred_val"
) %>%
nest(data = -c(outcomes, predictors)) %>%
mutate(correlation = map(data,
~ cor.test(.x$out_val, .x$pred_val)),
tidied = map(correlation, broom::tidy)) %>%
unnest(tidied) %>%
select(outcomes, predictors, df = parameter,
estimate, conf.low, conf.high, p.value) %>%
gt::gt() %>%
gt::cols_merge(columns = c(estimate, conf.low,
conf.high),
pattern = "{1} [{2}, {3}]") %>%
gt::cols_label(
estimate = md("*r* [95% CI]"),
df = md("*df*"),
p.value = md("*p*")
) %>%
gt::sub_small_vals(columns = c(estimate,
conf.low,
conf.high),
threshold = .01) %>%
gt::sub_small_vals(columns = p.value,
threshold = .001,
small_pattern = "<.001") %>%
gt::fmt_number(p.value, decimals = 3) %>%
gt::fmt_number(c(estimate, conf.low, conf.high),
decimals = 2)| outcomes | predictors | df | r [95% CI] | p |
|---|---|---|---|---|
| DifficultyEndFriendship_Avg | MASQ_AD | 212 | 0.18 [0.05, 0.31] | 0.009 |
| DifficultyEndFriendship_Avg | MASQ_AA | 212 | 0.04 [−0.09, 0.17] | 0.557 |
| DifficultyEndFriendship_Avg | MASQ_GD | 212 | 0.22 [0.09, 0.35] | 0.001 |
| DifficultyEndFriendship_Avg | RosenbergSelfEsteem_Sum | 209 | −0.28 [−0.40, −0.15] | <.001 |
| DifficultyEndFriendship_Avg | RSQ_Score_Fixed | 210 | 0.24 [0.10, 0.36] | <.001 |
| NegEmo_Friendship_Avg | MASQ_AD | 212 | 0.26 [0.13, 0.38] | <.001 |
| NegEmo_Friendship_Avg | MASQ_AA | 212 | 0.18 [0.05, 0.31] | 0.008 |
| NegEmo_Friendship_Avg | MASQ_GD | 212 | 0.30 [0.17, 0.41] | <.001 |
| NegEmo_Friendship_Avg | RosenbergSelfEsteem_Sum | 209 | −0.33 [−0.45, −0.20] | <.001 |
| NegEmo_Friendship_Avg | RSQ_Score_Fixed | 210 | 0.18 [0.04, 0.30] | 0.010 |
| DifficultyEndRomantic_Avg | MASQ_AD | 212 | 0.12 [−0.02, 0.25] | 0.087 |
| DifficultyEndRomantic_Avg | MASQ_AA | 212 | 0.13 [0.00, 0.26] | 0.057 |
| DifficultyEndRomantic_Avg | MASQ_GD | 212 | 0.34 [0.22, 0.45] | <.001 |
| DifficultyEndRomantic_Avg | RosenbergSelfEsteem_Sum | 209 | −0.30 [−0.41, −0.17] | <.001 |
| DifficultyEndRomantic_Avg | RSQ_Score_Fixed | 210 | 0.19 [0.06, 0.32] | 0.006 |
| NegEmo_Romantic_Avg | MASQ_AD | 212 | 0.24 [0.11, 0.36] | <.001 |
| NegEmo_Romantic_Avg | MASQ_AA | 212 | 0.27 [0.14, 0.39] | <.001 |
| NegEmo_Romantic_Avg | MASQ_GD | 212 | 0.41 [0.30, 0.52] | <.001 |
| NegEmo_Romantic_Avg | RosenbergSelfEsteem_Sum | 209 | −0.39 [−0.50, −0.27] | <.001 |
| NegEmo_Romantic_Avg | RSQ_Score_Fixed | 210 | 0.19 [0.06, 0.32] | 0.006 |
ear2 <- read_csv(here::here("data and codebooks", "study1b_processed.csv"))
ear2 <- ear2 %>% filter(filter == 1)anxiety symptoms were not significantly associated at the preregistered criterion level with forecasting negative emotions for explicit rejection
cor.test(ear2$MASQ_AA, ear2$Explicit_OverallNegEmo_Avg)##
## Pearson's product-moment correlation
##
## data: ear2$MASQ_AA and ear2$Explicit_OverallNegEmo_Avg
## t = 2.7597, df = 262, p-value = 0.006193
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.04832402 0.28305416
## sample estimates:
## cor
## 0.1680705[anxiety symptoms were not significantly associated at the preregistered criterion level with forecasting negative emotions for] ghosting
cor.test(ear2$MASQ_AA, ear2$Ghost_OverallNegEmo_Avg)##
## Pearson's product-moment correlation
##
## data: ear2$MASQ_AA and ear2$Ghost_OverallNegEmo_Avg
## t = 1.7072, df = 262, p-value = 0.08898
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## -0.0160431 0.2227929
## sample estimates:
## cor
## 0.1048871The hypothesis that individuals with more anxiety symptoms would be more likely to engage in ghosting than explicit rejection was not significant at the preregistered criterion level
There was a main effect of preference to explicitly reject
anx_data <- ear2 %>%
dplyr::select(ID, MASQ_AA,
Explicit_OverallLikelyEnd_Avg,
Ghost_OverallLikelyEnd_Avg) %>%
pivot_longer(cols = c(Explicit_OverallLikelyEnd_Avg,
Ghost_OverallLikelyEnd_Avg),
names_to = "RejectionType",
values_to = "Likelihood_Avg") %>%
mutate(RejectionType = factor(
RejectionType, levels =
c("Explicit_OverallLikelyEnd_Avg",
"Ghost_OverallLikelyEnd_Avg"),
labels = c("Explicit", "Ghost"))) %>%
na.omit()
# then run the model
anx_mod <- nlme::lme(Likelihood_Avg ~ RejectionType * MASQ_AA, random = ~ 1 | ID, data = anx_data)
summary(anx_mod)## Linear mixed-effects model fit by REML
## Data: anx_data
## AIC BIC logLik
## 1324.468 1350.037 -656.2341
##
## Random effects:
## Formula: ~1 | ID
## (Intercept) Residual
## StdDev: 5.022968e-05 0.8223895
##
## Fixed effects: Likelihood_Avg ~ RejectionType * MASQ_AA
## Value Std.Error DF t-value p-value
## (Intercept) 4.419812 0.13464958 262 32.82455 0.0000
## RejectionTypeGhost -2.274611 0.19042326 262 -11.94503 0.0000
## MASQ_AA -0.013334 0.00677509 262 -1.96811 0.0501
## RejectionTypeGhost:MASQ_AA 0.024156 0.00958142 262 2.52115 0.0123
## Correlation:
## (Intr) RjctTG MASQ_A
## RejectionTypeGhost -0.707
## MASQ_AA -0.927 0.655
## RejectionTypeGhost:MASQ_AA 0.655 -0.927 -0.707
##
## Standardized Within-Group Residuals:
## Min Q1 Med Q3 Max
## -2.326303758 -0.692852819 0.005728757 0.638303399 3.313434730
##
## Number of Observations: 528
## Number of Groups: 264anx_mod <- lme4::lmer(Likelihood_Avg ~ RejectionType * MASQ_AA + (1 | ID), data = anx_data)
# summary(anx_mod)
anx_mod.sim <- arm::sim(anx_mod)
#simulated uncertainty for fixed effects
fixef.anx_mod.sim <- fixef(anx_mod.sim)
# colnames(fixef.anx_mod.sim)
quantile(fixef.anx_mod.sim[, 4], # interaction
probs = c(.025, .975))## 2.5% 97.5%
## 0.008859006 0.042524213quantile(fixef.anx_mod.sim[, 2], # RejType
probs = c(.025, .975))## 2.5% 97.5%
## -2.585800 -1.996156general distress, anxiety symptoms, and self-esteem were not significantly correlated with the percent of relationships that participants had ended
cor.test(ear2$PercentRelationshipsEnded, ear2$MASQ_GD)##
## Pearson's product-moment correlation
##
## data: ear2$PercentRelationshipsEnded and ear2$MASQ_GD
## t = -0.23685, df = 195, p-value = 0.813
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## -0.1563838 0.1231288
## sample estimates:
## cor
## -0.0169588cor.test(ear2$PercentRelationshipsEnded, ear2$MASQ_AA)##
## Pearson's product-moment correlation
##
## data: ear2$PercentRelationshipsEnded and ear2$MASQ_AA
## t = -0.65875, df = 195, p-value = 0.5108
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## -0.18569400 0.09328872
## sample estimates:
## cor
## -0.04712152cor.test(ear2$PercentRelationshipsEnded, ear2$RosenbergSelfEsteem_Sum)##
## Pearson's product-moment correlation
##
## data: ear2$PercentRelationshipsEnded and ear2$RosenbergSelfEsteem_Sum
## t = 0.80219, df = 193, p-value = 0.4234
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## -0.08354243 0.19656678
## sample estimates:
## cor
## 0.05764663[general distress, anxiety symptoms, and self-esteem were not significantly correlated with] the reported likelihood of ending the relationships described in the vignettes
cor.test(ear2$OverallLikelyEnd_Avg, ear2$MASQ_GD)##
## Pearson's product-moment correlation
##
## data: ear2$OverallLikelyEnd_Avg and ear2$MASQ_GD
## t = -0.55026, df = 262, p-value = 0.5826
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## -0.15407059 0.08710877
## sample estimates:
## cor
## -0.03397554cor.test(ear2$OverallLikelyEnd_Avg, ear2$MASQ_AA)##
## Pearson's product-moment correlation
##
## data: ear2$OverallLikelyEnd_Avg and ear2$MASQ_AA
## t = -0.41608, df = 262, p-value = 0.6777
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## -0.1459714 0.0953255
## sample estimates:
## cor
## -0.02569724cor.test(ear2$OverallLikelyEnd_Avg, ear2$RosenbergSelfEsteem_Sum)##
## Pearson's product-moment correlation
##
## data: ear2$OverallLikelyEnd_Avg and ear2$RosenbergSelfEsteem_Sum
## t = 1.0623, df = 259, p-value = 0.2891
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## -0.05600575 0.18579632
## sample estimates:
## cor
## 0.06586213[general distress, anxiety symptoms, and self-esteem were not significantly correlated with] the reported likelihood of ending the relationships described in the vignettes via explicit rejection or ghosting
cor.test(ear2$Explicit_OverallLikelyEnd_Avg, ear2$MASQ_GD)##
## Pearson's product-moment correlation
##
## data: ear2$Explicit_OverallLikelyEnd_Avg and ear2$MASQ_GD
## t = -1.2405, df = 262, p-value = 0.2159
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## -0.19534088 0.04472362
## sample estimates:
## cor
## -0.07641598cor.test(ear2$Explicit_OverallLikelyEnd_Avg, ear2$MASQ_AA)##
## Pearson's product-moment correlation
##
## data: ear2$Explicit_OverallLikelyEnd_Avg and ear2$MASQ_AA
## t = -2.5168, df = 262, p-value = 0.01244
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## -0.26937378 -0.03353846
## sample estimates:
## cor
## -0.1536434cor.test(ear2$Explicit_OverallLikelyEnd_Avg, ear2$RosenbergSelfEsteem_Sum)##
## Pearson's product-moment correlation
##
## data: ear2$Explicit_OverallLikelyEnd_Avg and ear2$RosenbergSelfEsteem_Sum
## t = 1.0914, df = 259, p-value = 0.2761
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## -0.0542035 0.1875411
## sample estimates:
## cor
## 0.06766181cor.test(ear2$Ghost_OverallLikelyEnd_Avg, ear2$MASQ_GD)##
## Pearson's product-moment correlation
##
## data: ear2$Ghost_OverallLikelyEnd_Avg and ear2$MASQ_GD
## t = 0.10715, df = 262, p-value = 0.9147
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## -0.1141984 0.1272452
## sample estimates:
## cor
## 0.006619885cor.test(ear2$Ghost_OverallLikelyEnd_Avg, ear2$MASQ_AA)##
## Pearson's product-moment correlation
##
## data: ear2$Ghost_OverallLikelyEnd_Avg and ear2$MASQ_AA
## t = 1.3556, df = 262, p-value = 0.1764
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## -0.03765113 0.20214555
## sample estimates:
## cor
## 0.08345522cor.test(ear2$Ghost_OverallLikelyEnd_Avg, ear2$RosenbergSelfEsteem_Sum)##
## Pearson's product-moment correlation
##
## data: ear2$Ghost_OverallLikelyEnd_Avg and ear2$RosenbergSelfEsteem_Sum
## t = 0.042009, df = 259, p-value = 0.9665
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## -0.1188473 0.1239910
## sample estimates:
## cor
## 0.002610327anhedonic depressive symptoms were not significantly associated with forecasting negative emotions for explicit rejection … or ghosting
cor.test(ear2$MASQ_AD, ear2$Explicit_OverallNegEmo_Avg)##
## Pearson's product-moment correlation
##
## data: ear2$MASQ_AD and ear2$Explicit_OverallNegEmo_Avg
## t = 0.2841, df = 262, p-value = 0.7766
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## -0.1033970 0.1379838
## sample estimates:
## cor
## 0.01754909cor.test(ear2$MASQ_AD, ear2$Ghost_OverallNegEmo_Avg)##
## Pearson's product-moment correlation
##
## data: ear2$MASQ_AD and ear2$Ghost_OverallNegEmo_Avg
## t = 1.3445, df = 262, p-value = 0.18
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## -0.03833408 0.20148948
## sample estimates:
## cor
## 0.08277602Anhedonic depressive symptoms were also not associated with the percent of relationships that participants had ended
cor.test(ear2$MASQ_AD, ear2$PercentRelationshipsEnded)##
## Pearson's product-moment correlation
##
## data: ear2$MASQ_AD and ear2$PercentRelationshipsEnded
## t = -0.30613, df = 195, p-value = 0.7598
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## -0.1612186 0.1182411
## sample estimates:
## cor
## -0.02191686there was no statistical support for the hypothesis that individuals with more anhedonic depressive symptoms were more likely to engage in ghosting than explicit rejection
dep_data <- ear2 %>%
dplyr::select(ID, MASQ_AD,
Explicit_OverallLikelyEnd_Avg,
Ghost_OverallLikelyEnd_Avg) %>%
pivot_longer(cols = c(Explicit_OverallLikelyEnd_Avg,
Ghost_OverallLikelyEnd_Avg),
names_to = "RejectionType",
values_to = "Likelihood_Avg") %>%
mutate(RejectionType = factor(
RejectionType, levels =
c("Explicit_OverallLikelyEnd_Avg",
"Ghost_OverallLikelyEnd_Avg"),
labels = c("Explicit", "Ghost"))) %>%
na.omit()
mod_AD <- nlme::lme(Likelihood_Avg ~ RejectionType * MASQ_AD, random = ~ 1 | ID, data = dep_data)
summary(mod_AD)## Linear mixed-effects model fit by REML
## Data: dep_data
## AIC BIC logLik
## 1330.286 1355.855 -659.1431
##
## Random effects:
## Formula: ~1 | ID
## (Intercept) Residual
## StdDev: 3.811419e-05 0.8264172
##
## Fixed effects: Likelihood_Avg ~ RejectionType * MASQ_AD
## Value Std.Error DF t-value p-value
## (Intercept) 4.072943 0.19848995 262 20.519642 0.0000
## RejectionTypeGhost -1.537680 0.28070718 262 -5.477878 0.0000
## MASQ_AD 0.003019 0.00571819 262 0.527981 0.5980
## RejectionTypeGhost:MASQ_AD -0.008704 0.00808674 262 -1.076365 0.2828
## Correlation:
## (Intr) RjctTG MASQ_A
## RejectionTypeGhost -0.707
## MASQ_AD -0.967 0.683
## RejectionTypeGhost:MASQ_AD 0.683 -0.967 -0.707
##
## Standardized Within-Group Residuals:
## Min Q1 Med Q3 Max
## -2.43438897 -0.68553094 0.03684135 0.59723585 3.18881631
##
## Number of Observations: 528
## Number of Groups: 264mod_AD <- lme4::lmer(Likelihood_Avg ~ RejectionType * MASQ_AD + (1 | ID), data = dep_data)
#summary(mod_AD)
mod_AD.sim <- sim(mod_AD)
#simulated uncertainty for fixed effects
fixef.masqAD.sim <- fixef(mod_AD.sim)
colnames(fixef.masqAD.sim)## [1] "(Intercept)" "RejectionTypeGhost"
## [3] "MASQ_AD" "RejectionTypeGhost:MASQ_AD"quantile(fixef.masqAD.sim[, 4], # interaction
probs = c(.025, .975))## 2.5% 97.5%
## -0.022162548 0.008398672rejection sensitivity was significantly positively associated with perceived difficulty of ending relationships via explicit rejection … and ghosting
cor.test(ear2$RSQ_Score_Fixed, ear2$Explicit_OverallDifficulty_Avg)##
## Pearson's product-moment correlation
##
## data: ear2$RSQ_Score_Fixed and ear2$Explicit_OverallDifficulty_Avg
## t = 4.0963, df = 262, p-value = 5.599e-05
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.1284147 0.3555347
## sample estimates:
## cor
## 0.2453382cor.test(ear2$RSQ_Score_Fixed, ear2$Ghost_OverallDifficulty_Avg)##
## Pearson's product-moment correlation
##
## data: ear2$RSQ_Score_Fixed and ear2$Ghost_OverallDifficulty_Avg
## t = 3.057, df = 262, p-value = 0.002467
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.06634082 0.29959613
## sample estimates:
## cor
## 0.1855815rejection sensitivity was not significantly correlated with overall likelihood of ending the relationships
cor.test(ear2$RSQ_Score_Fixed, ear2$OverallLikelyEnd_Avg)##
## Pearson's product-moment correlation
##
## data: ear2$RSQ_Score_Fixed and ear2$OverallLikelyEnd_Avg
## t = -2.7693, df = 262, p-value = 0.00602
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## -0.28358991 -0.04890515
## sample estimates:
## cor
## -0.1686365Nor was it significantly correlated with negative emotions in response to explicitly rejecting … or ghosting
cor.test(ear2$RSQ_Score_Fixed, ear2$Explicit_OverallNegEmo_Avg)##
## Pearson's product-moment correlation
##
## data: ear2$RSQ_Score_Fixed and ear2$Explicit_OverallNegEmo_Avg
## t = 2.567, df = 262, p-value = 0.01081
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.03659593 0.27221055
## sample estimates:
## cor
## 0.1566309cor.test(ear2$RSQ_Score_Fixed, ear2$Ghost_OverallNegEmo_Avg)##
## Pearson's product-moment correlation
##
## data: ear2$RSQ_Score_Fixed and ear2$Ghost_OverallNegEmo_Avg
## t = 2.7976, df = 262, p-value = 0.005531
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.05062279 0.28517256
## sample estimates:
## cor
## 0.170309Scores on the RSQ were also not associated with the percent of relationships that participants had ended
cor.test(ear2$RSQ_Score_Fixed, ear2$PercentRelationshipsEnded)##
## Pearson's product-moment correlation
##
## data: ear2$RSQ_Score_Fixed and ear2$PercentRelationshipsEnded
## t = -1.3487, df = 195, p-value = 0.179
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## -0.23280232 0.04425516
## sample estimates:
## cor
## -0.09613531Self-esteem was associated with forecasted difficulty of making explicit rejections
rse_diff <- lm(data = ear2, Explicit_OverallDifficulty_Avg ~ MASQ_AA + MASQ_AD + MASQ_GD + RosenbergSelfEsteem_Sum)
summary(rse_diff)##
## Call:
## lm(formula = Explicit_OverallDifficulty_Avg ~ MASQ_AA + MASQ_AD +
## MASQ_GD + RosenbergSelfEsteem_Sum, data = ear2)
##
## Residuals:
## Min 1Q Median 3Q Max
## -4.0136 -0.7293 0.0336 0.7912 2.3350
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 5.57909 0.85466 6.528 3.56e-10 ***
## MASQ_AA -0.01970 0.01191 -1.654 0.0993 .
## MASQ_AD -0.01802 0.01000 -1.802 0.0727 .
## MASQ_GD 0.03001 0.01167 2.571 0.0107 *
## RosenbergSelfEsteem_Sum -0.04308 0.01720 -2.504 0.0129 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.126 on 256 degrees of freedom
## (3 observations deleted due to missingness)
## Multiple R-squared: 0.1235, Adjusted R-squared: 0.1098
## F-statistic: 9.02 on 4 and 256 DF, p-value: 7.87e-07confint(rse_diff)## 2.5 % 97.5 %
## (Intercept) 3.896027942 7.262143647
## MASQ_AA -0.043153188 0.003752947
## MASQ_AD -0.037710646 0.001673081
## MASQ_GD 0.007024778 0.052985245
## RosenbergSelfEsteem_Sum -0.076949222 -0.009204853but [Self-esteem was not associated with] ghosting rejections
rse_diff2 <- lm(data = ear2, Ghost_OverallDifficulty_Avg ~ MASQ_AA + MASQ_AD + MASQ_GD + RosenbergSelfEsteem_Sum)
summary(rse_diff2)##
## Call:
## lm(formula = Ghost_OverallDifficulty_Avg ~ MASQ_AA + MASQ_AD +
## MASQ_GD + RosenbergSelfEsteem_Sum, data = ear2)
##
## Residuals:
## Min 1Q Median 3Q Max
## -3.5243 -0.8534 0.0315 0.7855 2.9048
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 4.445849 0.910655 4.882 1.85e-06 ***
## MASQ_AA -0.002815 0.012690 -0.222 0.8246
## MASQ_AD -0.016983 0.010655 -1.594 0.1122
## MASQ_GD 0.028421 0.012434 2.286 0.0231 *
## RosenbergSelfEsteem_Sum -0.017546 0.018327 -0.957 0.3393
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.2 on 256 degrees of freedom
## (3 observations deleted due to missingness)
## Multiple R-squared: 0.06866, Adjusted R-squared: 0.05411
## F-statistic: 4.718 on 4 and 256 DF, p-value: 0.001088confint(rse_diff2)## 2.5 % 97.5 %
## (Intercept) 2.652520721 6.239177601
## MASQ_AA -0.027804730 0.022174601
## MASQ_AD -0.037965182 0.003998889
## MASQ_GD 0.003935224 0.052906929
## RosenbergSelfEsteem_Sum -0.053637107 0.018545738Rejection sensitivity was associated with the difficulty of making explicit rejections … as well as the overall likelihood of rejecting
arsq_diff <- lm(data = ear2, Explicit_OverallDifficulty_Avg ~ MASQ_AA + MASQ_AD + MASQ_GD + RSQ_Score_Fixed)
summary(arsq_diff)##
## Call:
## lm(formula = Explicit_OverallDifficulty_Avg ~ MASQ_AA + MASQ_AD +
## MASQ_GD + RSQ_Score_Fixed, data = ear2)
##
## Residuals:
## Min 1Q Median 3Q Max
## -3.8290 -0.6956 0.0437 0.8160 2.4208
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 3.475983 0.314003 11.070 < 2e-16 ***
## MASQ_AA -0.018738 0.011824 -1.585 0.114254
## MASQ_AD -0.009138 0.008966 -1.019 0.309022
## MASQ_GD 0.038859 0.009872 3.936 0.000106 ***
## RSQ_Score_Fixed 0.036743 0.015769 2.330 0.020576 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.124 on 259 degrees of freedom
## Multiple R-squared: 0.1168, Adjusted R-squared: 0.1032
## F-statistic: 8.562 on 4 and 259 DF, p-value: 1.669e-06confint(arsq_diff)## 2.5 % 97.5 %
## (Intercept) 2.857658512 4.094307469
## MASQ_AA -0.042021126 0.004545893
## MASQ_AD -0.026793278 0.008516325
## MASQ_GD 0.019419136 0.058298540
## RSQ_Score_Fixed 0.005690004 0.067795418arsq_diff2 <- lm(data = ear2, OverallLikelyEnd_Avg ~ MASQ_AA + MASQ_AD + MASQ_GD + RSQ_Score_Fixed)
summary(arsq_diff2)##
## Call:
## lm(formula = OverallLikelyEnd_Avg ~ MASQ_AA + MASQ_AD + MASQ_GD +
## RSQ_Score_Fixed, data = ear2)
##
## Residuals:
## Min 1Q Median 3Q Max
## -2.43669 -0.37254 0.09387 0.41217 1.31834
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 3.9337751 0.1761361 22.334 < 2e-16 ***
## MASQ_AA 0.0009279 0.0066325 0.140 0.88885
## MASQ_AD 0.0034060 0.0050292 0.677 0.49885
## MASQ_GD 0.0016636 0.0055376 0.300 0.76410
## RSQ_Score_Fixed -0.0253923 0.0088457 -2.871 0.00444 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.6307 on 259 degrees of freedom
## Multiple R-squared: 0.03215, Adjusted R-squared: 0.0172
## F-statistic: 2.151 on 4 and 259 DF, p-value: 0.075confint(arsq_diff2)## 2.5 % 97.5 %
## (Intercept) 3.586933989 4.280616200
## MASQ_AA -0.012132725 0.013988441
## MASQ_AD -0.006497262 0.013309202
## MASQ_GD -0.009240878 0.012568020
## RSQ_Score_Fixed -0.042810882 -0.007973655From Study 1b for the correlation table
ear2 %>%
select(MASQ_AD, MASQ_AA, MASQ_GD,
RosenbergSelfEsteem_Sum,
RSQ_Score_Fixed,
Explicit_OverallDifficulty_Avg,
Explicit_OverallNegEmo_Avg,
Ghost_OverallDifficulty_Avg,
Ghost_OverallNegEmo_Avg) %>%
corrr::correlate(use = "pairwise.complete.obs",
method = "pearson") %>%
dplyr::select(term, Explicit_OverallDifficulty_Avg,
Explicit_OverallNegEmo_Avg,
Ghost_OverallDifficulty_Avg,
Ghost_OverallNegEmo_Avg) %>%
filter(term %in% c("MASQ_AD", "MASQ_AA", "MASQ_GD",
"RosenbergSelfEsteem_Sum",
"RSQ_Score_Fixed")) %>%
corrr::fashion() %>%
gt::gt()| term | Explicit_OverallDifficulty_Avg | Explicit_OverallNegEmo_Avg | Ghost_OverallDifficulty_Avg | Ghost_OverallNegEmo_Avg |
|---|---|---|---|---|
| MASQ_AD | .11 | .02 | .03 | .08 |
| MASQ_AA | .12 | .17 | .15 | .10 |
| MASQ_GD | .30 | .27 | .24 | .24 |
| RosenbergSelfEsteem_Sum | -.30 | -.23 | -.19 | -.21 |
| RSQ_Score_Fixed | .25 | .16 | .19 | .17 |
# and the p-value / stats on them
ear2 %>%
select(Explicit_OverallDifficulty_Avg,
Explicit_OverallNegEmo_Avg,
Ghost_OverallDifficulty_Avg,
Ghost_OverallNegEmo_Avg,
MASQ_AD, MASQ_AA, MASQ_GD,
RosenbergSelfEsteem_Sum,
RSQ_Score_Fixed) %>%
pivot_longer(
cols = c(Explicit_OverallDifficulty_Avg,
Explicit_OverallNegEmo_Avg,
Ghost_OverallDifficulty_Avg,
Ghost_OverallNegEmo_Avg),
names_to = "outcomes",
values_to = "out_val") %>%
pivot_longer(
cols = c(MASQ_AD, MASQ_AA, MASQ_GD,
RosenbergSelfEsteem_Sum,
RSQ_Score_Fixed),
names_to = "predictors",
values_to = "pred_val"
) %>%
nest(data = -c(outcomes, predictors)) %>%
mutate(correlation = map(data,
~ cor.test(.x$out_val, .x$pred_val)),
tidied = map(correlation, broom::tidy)) %>%
unnest(tidied) %>%
select(outcomes, predictors, df = parameter,
estimate, conf.low, conf.high, p.value) %>%
gt::gt() %>%
gt::cols_merge(columns = c(estimate, conf.low,
conf.high),
pattern = "{1} [{2}, {3}]") %>%
gt::cols_label(
estimate = md("*r* [95% CI]"),
df = md("*df*"),
p.value = md("*p*")
) %>%
gt::sub_small_vals(columns = c(estimate,
conf.low,
conf.high),
threshold = .01) %>%
gt::sub_small_vals(columns = p.value,
threshold = .001,
small_pattern = "<.001") %>%
gt::fmt_number(p.value, decimals = 3) %>%
gt::fmt_number(c(estimate, conf.low, conf.high),
decimals = 2)| outcomes | predictors | df | r [95% CI] | p |
|---|---|---|---|---|
| Explicit_OverallDifficulty_Avg | MASQ_AD | 262 | 0.11 [−0.01, 0.23] | 0.072 |
| Explicit_OverallDifficulty_Avg | MASQ_AA | 262 | 0.12 [−0.01, 0.23] | 0.061 |
| Explicit_OverallDifficulty_Avg | MASQ_GD | 262 | 0.30 [0.19, 0.41] | <.001 |
| Explicit_OverallDifficulty_Avg | RosenbergSelfEsteem_Sum | 259 | −0.30 [−0.41, −0.19] | <.001 |
| Explicit_OverallDifficulty_Avg | RSQ_Score_Fixed | 262 | 0.25 [0.13, 0.36] | <.001 |
| Explicit_OverallNegEmo_Avg | MASQ_AD | 262 | 0.02 [−0.10, 0.14] | 0.777 |
| Explicit_OverallNegEmo_Avg | MASQ_AA | 262 | 0.17 [0.05, 0.28] | 0.006 |
| Explicit_OverallNegEmo_Avg | MASQ_GD | 262 | 0.27 [0.16, 0.38] | <.001 |
| Explicit_OverallNegEmo_Avg | RosenbergSelfEsteem_Sum | 259 | −0.23 [−0.34, −0.11] | <.001 |
| Explicit_OverallNegEmo_Avg | RSQ_Score_Fixed | 262 | 0.16 [0.04, 0.27] | 0.011 |
| Ghost_OverallDifficulty_Avg | MASQ_AD | 262 | 0.03 [−0.09, 0.15] | 0.600 |
| Ghost_OverallDifficulty_Avg | MASQ_AA | 262 | 0.15 [0.03, 0.27] | 0.014 |
| Ghost_OverallDifficulty_Avg | MASQ_GD | 262 | 0.24 [0.13, 0.35] | <.001 |
| Ghost_OverallDifficulty_Avg | RosenbergSelfEsteem_Sum | 259 | −0.19 [−0.30, −0.07] | 0.002 |
| Ghost_OverallDifficulty_Avg | RSQ_Score_Fixed | 262 | 0.19 [0.07, 0.30] | 0.002 |
| Ghost_OverallNegEmo_Avg | MASQ_AD | 262 | 0.08 [−0.04, 0.20] | 0.180 |
| Ghost_OverallNegEmo_Avg | MASQ_AA | 262 | 0.10 [−0.02, 0.22] | 0.089 |
| Ghost_OverallNegEmo_Avg | MASQ_GD | 262 | 0.24 [0.12, 0.35] | <.001 |
| Ghost_OverallNegEmo_Avg | RosenbergSelfEsteem_Sum | 259 | −0.21 [−0.33, −0.09] | <.001 |
| Ghost_OverallNegEmo_Avg | RSQ_Score_Fixed | 262 | 0.17 [0.05, 0.29] | 0.006 |
ear3 <- read_csv(here::here("data and codebooks", "study2_processed.csv"))Contrary to predictions, self-esteem was not significantly correlated with difficulty rejecting
cor.test(ear3$rse, ear3$difficulty)##
## Pearson's product-moment correlation
##
## data: ear3$rse and ear3$difficulty
## t = -0.052032, df = 252, p-value = 0.9585
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## -0.1263113 0.1198552
## sample estimates:
## cor
## -0.00327767Self-esteem was, however, negatively correlated with negative emotions experienced while rejecting
cor.test(ear3$rse, ear3$neg_emotions)##
## Pearson's product-moment correlation
##
## data: ear3$rse and ear3$neg_emotions
## t = -3.6047, df = 252, p-value = 0.0003766
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## -0.3353794 -0.1011069
## sample estimates:
## cor
## -0.2214357Contrary to predictions, rejection sensitivity was not significantly correlated with difficulty rejecting
cor.test(ear3$arsq_full, ear3$difficulty)##
## Pearson's product-moment correlation
##
## data: ear3$arsq_full and ear3$difficulty
## t = 0.61522, df = 252, p-value = 0.539
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## -0.08476262 0.16104288
## sample estimates:
## cor
## 0.03872595Rejection sensitivity was, however, positively correlated with negative emotions experienced while rejecting
cor.test(ear3$arsq_full, ear3$neg_emotions)##
## Pearson's product-moment correlation
##
## data: ear3$arsq_full and ear3$neg_emotions
## t = 3.6408, df = 252, p-value = 0.0003298
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.1033008 0.3373457
## sample estimates:
## cor
## 0.2235431A logistic regression did not find a significant effect of adult RSQ score predicting the method participants had used to reject
arsq_type_reg <- ear3 %>%
filter(reject_method %in% c("direct", "indirect")) %>%
mutate(reject_method = factor(reject_method)) %>%
glm(reject_method ~ arsq_full,
family = "binomial",
data = .)
summary(arsq_type_reg)##
## Call:
## glm(formula = reject_method ~ arsq_full, family = "binomial",
## data = .)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -1.3588 -0.9869 -0.9037 1.3301 1.5433
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.03197 0.34489 -2.992 0.00277 **
## arsq_full 0.04466 0.02456 1.818 0.06906 .
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 327.19 on 244 degrees of freedom
## Residual deviance: 323.84 on 243 degrees of freedom
## (4 observations deleted due to missingness)
## AIC: 327.84
##
## Number of Fisher Scoring iterations: 4general distress was positively correlated with difficulty experienced rejecting others
cor.test(ear3$masq_gd, ear3$difficulty)##
## Pearson's product-moment correlation
##
## data: ear3$masq_gd and ear3$difficulty
## t = 3.3463, df = 256, p-value = 0.0009417
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.08471005 0.31887023
## sample estimates:
## cor
## 0.2047173General distress was also positively correlated with negative emotions experienced while rejecting
cor.test(ear3$masq_gd, ear3$neg_emotions)##
## Pearson's product-moment correlation
##
## data: ear3$masq_gd and ear3$neg_emotions
## t = 7.0179, df = 256, p-value = 2.012e-11
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.2939756 0.4993109
## sample estimates:
## cor
## 0.4016796Anhedonic depression was not correlated with difficulty rejecting others
cor.test(ear3$masq_ad, ear3$difficulty)##
## Pearson's product-moment correlation
##
## data: ear3$masq_ad and ear3$difficulty
## t = -0.70908, df = 256, p-value = 0.4789
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## -0.16550402 0.07827447
## sample estimates:
## cor
## -0.04427383[Anhedonic depression was] positively correlated with negative emotions experienced while rejecting
cor.test(ear3$masq_ad, ear3$neg_emotions)##
## Pearson's product-moment correlation
##
## data: ear3$masq_ad and ear3$neg_emotions
## t = 2.1917, df = 256, p-value = 0.02931
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.01381543 0.25363293
## sample estimates:
## cor
## 0.1357115A logistic regression did not find a significant effect of MASQ-AD predicting the method participants had used to reject
ad_type_reg <- ear3 %>%
filter(reject_method %in% c("direct", "indirect")) %>%
mutate(reject_method = factor(reject_method)) %>%
glm(reject_method ~ masq_ad,
family = "binomial",
data = .)
summary(ad_type_reg)##
## Call:
## glm(formula = reject_method ~ masq_ad, family = "binomial", data = .)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -1.0876 -1.0048 -0.9325 1.3439 1.5534
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.15011 0.60956 -1.887 0.0592 .
## masq_ad 0.01870 0.01621 1.154 0.2486
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 332.02 on 248 degrees of freedom
## Residual deviance: 330.67 on 247 degrees of freedom
## AIC: 334.67
##
## Number of Fisher Scoring iterations: 4anxiety symptoms on the MASQ were positively correlated with the negative emotions experienced while rejecting
cor.test(ear3$masq_aa, ear3$neg_emotions)##
## Pearson's product-moment correlation
##
## data: ear3$masq_aa and ear3$neg_emotions
## t = 5.1136, df = 256, p-value = 6.196e-07
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.189347 0.411267
## sample estimates:
## cor
## 0.3044324anxiety was also negatively associated with how difficult participants found it to reject others
cor.test(ear3$masq_aa, ear3$difficulty)##
## Pearson's product-moment correlation
##
## data: ear3$masq_aa and ear3$difficulty
## t = 2.4937, df = 256, p-value = 0.01327
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.03248172 0.27102390
## sample estimates:
## cor
## 0.1539959A logistic regression did not find a significant effect of anxiety symptoms predicting the method participants had used to reject
aa_type_reg <- ear3 %>%
filter(reject_method %in% c("direct", "indirect")) %>%
mutate(reject_method = factor(reject_method)) %>%
glm(reject_method ~ masq_aa,
family = "binomial",
data = .)
summary(aa_type_reg)##
## Call:
## glm(formula = reject_method ~ masq_aa, family = "binomial", data = .)
##
## Deviance Residuals:
## Min 1Q Median 3Q Max
## -1.2381 -0.9826 -0.9164 1.3526 1.4854
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.94761 0.35568 -2.664 0.00772 **
## masq_aa 0.02475 0.01690 1.465 0.14288
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 332.02 on 248 degrees of freedom
## Residual deviance: 329.87 on 247 degrees of freedom
## AIC: 333.87
##
## Number of Fisher Scoring iterations: 4even with covariates self-esteem was associated with negative emotions after rejecting
rse_neg_mod <- lm(data = ear3,
neg_emotions ~ rse + masq_gd + masq_aa + masq_ad)
summary(rse_neg_mod)##
## Call:
## lm(formula = neg_emotions ~ rse + masq_gd + masq_aa + masq_ad,
## data = ear3)
##
## Residuals:
## Min 1Q Median 3Q Max
## -3.5431 -0.9674 0.0546 0.9491 3.5584
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) -0.271943 1.242208 -0.219 0.8269
## rse 0.055539 0.023623 2.351 0.0195 *
## masq_gd 0.083651 0.015582 5.369 1.82e-07 ***
## masq_aa 0.021772 0.014910 1.460 0.1455
## masq_ad 0.008386 0.015499 0.541 0.5890
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.456 on 249 degrees of freedom
## (5 observations deleted due to missingness)
## Multiple R-squared: 0.2047, Adjusted R-squared: 0.192
## F-statistic: 16.03 on 4 and 249 DF, p-value: 1.088e-11confint(rse_neg_mod)## 2.5 % 97.5 %
## (Intercept) -2.718516611 2.17463049
## rse 0.009012054 0.10206505
## masq_gd 0.052962216 0.11433895
## masq_aa -0.007594096 0.05113734
## masq_ad -0.022140660 0.03891218Given that participants indicated both direct rejection and indirect rejection (ghosting), we included an interaction term in the model of method of rejection interacting with self-esteem. We found that the interaction was significant
rse_neg_inter_mod <- lm(data = ear3 %>% filter(reject_method != "different"), neg_emotions ~ rse * reject_method + masq_gd + masq_ad + masq_aa)
summary(rse_neg_inter_mod)##
## Call:
## lm(formula = neg_emotions ~ rse * reject_method + masq_gd + masq_ad +
## masq_aa, data = ear3 %>% filter(reject_method != "different"))
##
## Residuals:
## Min 1Q Median 3Q Max
## -3.2064 -0.9245 0.0652 0.8683 3.6049
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.34718 1.29853 0.267 0.7894
## rse 0.03158 0.02632 1.200 0.2313
## reject_methodindirect -1.59736 0.70314 -2.272 0.0240 *
## masq_gd 0.08038 0.01598 5.031 9.62e-07 ***
## masq_ad 0.01104 0.01595 0.692 0.4894
## masq_aa 0.02292 0.01515 1.513 0.1317
## rse:reject_methodindirect 0.06196 0.02720 2.278 0.0236 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.463 on 237 degrees of freedom
## (5 observations deleted due to missingness)
## Multiple R-squared: 0.2132, Adjusted R-squared: 0.1933
## F-statistic: 10.7 on 6 and 237 DF, p-value: 1.586e-10confint(rse_neg_inter_mod)## 2.5 % 97.5 %
## (Intercept) -2.210946638 2.90531577
## rse -0.020267208 0.08343486
## reject_methodindirect -2.982552918 -0.21216029
## masq_gd 0.048907079 0.11184966
## masq_ad -0.020379259 0.04246849
## masq_aa -0.006931243 0.05277650
## rse:reject_methodindirect 0.008373134 0.11554873individuals who had used indirect methods to reject experienced more negative emotions if they had higher self-esteem … while those who used direct methods did not experience a significant difference
rse_indirect <- lm(data = filter(ear3, reject_method == "indirect"), neg_emotions ~ rse + masq_gd + masq_ad + masq_aa)
summary(rse_indirect)##
## Call:
## lm(formula = neg_emotions ~ rse + masq_gd + masq_ad + masq_aa,
## data = filter(ear3, reject_method == "indirect"))
##
## Residuals:
## Min 1Q Median 3Q Max
## -3.3138 -0.7223 -0.0060 0.9419 3.0422
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) -2.616643 2.365722 -1.106 0.27171
## rse 0.122059 0.045431 2.687 0.00863 **
## masq_gd 0.109083 0.032461 3.360 0.00115 **
## masq_ad 0.009898 0.026466 0.374 0.70931
## masq_aa 0.017360 0.026344 0.659 0.51164
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.46 on 88 degrees of freedom
## (3 observations deleted due to missingness)
## Multiple R-squared: 0.203, Adjusted R-squared: 0.1668
## F-statistic: 5.603 on 4 and 88 DF, p-value: 0.0004585confint(rse_indirect)## 2.5 % 97.5 %
## (Intercept) -7.31801806 2.08473204
## rse 0.03177525 0.21234251
## masq_gd 0.04457305 0.17359293
## masq_ad -0.04269746 0.06249324
## masq_aa -0.03499305 0.06971266rse_direct <- lm(data = filter(ear3, reject_method == "direct"), neg_emotions ~ rse + masq_gd + masq_ad + masq_aa)
summary(rse_direct)##
## Call:
## lm(formula = neg_emotions ~ rse + masq_gd + masq_ad + masq_aa,
## data = filter(ear3, reject_method == "direct"))
##
## Residuals:
## Min 1Q Median 3Q Max
## -3.1923 -0.9806 0.1328 0.8421 3.6212
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.75489 1.53691 0.491 0.62404
## rse 0.02249 0.02926 0.769 0.44326
## masq_gd 0.07049 0.01854 3.802 0.00021 ***
## masq_ad 0.01417 0.02022 0.701 0.48451
## masq_aa 0.02180 0.01894 1.151 0.25170
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.472 on 146 degrees of freedom
## (2 observations deleted due to missingness)
## Multiple R-squared: 0.2258, Adjusted R-squared: 0.2046
## F-statistic: 10.64 on 4 and 146 DF, p-value: 1.349e-07confint(rse_direct)## 2.5 % 97.5 %
## (Intercept) -2.28258567 3.79235749
## rse -0.03533225 0.08032017
## masq_gd 0.03384515 0.10712867
## masq_ad -0.02579285 0.05413815
## masq_aa -0.01563622 0.05922927The association between rejection sensitivity and negative emotions experienced while rejecting was not significant when the MASQ subscales were included in the regression
rsq_negemo_mod <- lm(data = ear3,
neg_emotions ~ arsq_full + masq_gd + masq_aa + masq_ad)
summary(rsq_negemo_mod)##
## Call:
## lm(formula = neg_emotions ~ arsq_full + masq_gd + masq_aa + masq_ad,
## data = ear3)
##
## Residuals:
## Min 1Q Median 3Q Max
## -3.8896 -0.9589 0.1845 0.9599 3.5062
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.259790 0.483233 4.676 4.79e-06 ***
## arsq_full 0.009444 0.020883 0.452 0.6515
## masq_gd 0.056565 0.013236 4.274 2.74e-05 ***
## masq_aa 0.025940 0.015300 1.695 0.0912 .
## masq_ad -0.008051 0.013905 -0.579 0.5631
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.481 on 249 degrees of freedom
## (5 observations deleted due to missingness)
## Multiple R-squared: 0.1858, Adjusted R-squared: 0.1727
## F-statistic: 14.21 on 4 and 249 DF, p-value: 1.857e-10when rejection method was included in the regression with MASQ covariates, there was a significant interaction
rsq_rejmetho_mod <- lm(data = ear3 %>% filter(reject_method != "different"), neg_emotions ~ arsq_full * reject_method + masq_gd + masq_ad + masq_aa)
summary(rsq_rejmetho_mod)##
## Call:
## lm(formula = neg_emotions ~ arsq_full * reject_method + masq_gd +
## masq_ad + masq_aa, data = ear3 %>% filter(reject_method !=
## "different"))
##
## Residuals:
## Min 1Q Median 3Q Max
## -3.6526 -0.9533 0.1933 0.9702 3.6320
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 1.965223 0.519045 3.786 0.000194 ***
## arsq_full 0.042761 0.026266 1.628 0.104857
## reject_methodindirect 0.886240 0.516390 1.716 0.087428 .
## masq_gd 0.053298 0.013495 3.950 0.000103 ***
## masq_ad -0.008768 0.014300 -0.613 0.540362
## masq_aa 0.028357 0.015558 1.823 0.069615 .
## arsq_full:reject_methodindirect -0.076711 0.036571 -2.098 0.037001 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.489 on 237 degrees of freedom
## (5 observations deleted due to missingness)
## Multiple R-squared: 0.1938, Adjusted R-squared: 0.1734
## F-statistic: 9.497 on 6 and 237 DF, p-value: 2.366e-09confint(rsq_rejmetho_mod)## 2.5 % 97.5 %
## (Intercept) 0.942691696 2.987754789
## arsq_full -0.008984166 0.094506876
## reject_methodindirect -0.131059246 1.903540191
## masq_gd 0.026713494 0.079883009
## masq_ad -0.036940517 0.019403775
## masq_aa -0.002292667 0.059005747
## arsq_full:reject_methodindirect -0.148757207 -0.004665102Neither simple effect had a significant regression coefficient when MASQ subscales were included in the regression
summary(lm(data = filter(ear3, reject_method == "direct"), neg_emotions ~ arsq_full + masq_gd + masq_aa + masq_ad))##
## Call:
## lm(formula = neg_emotions ~ arsq_full + masq_gd + masq_aa + masq_ad,
## data = filter(ear3, reject_method == "direct"))
##
## Residuals:
## Min 1Q Median 3Q Max
## -3.6751 -0.9769 0.2159 0.9555 3.6957
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 1.847896 0.632060 2.924 0.00402 **
## arsq_full 0.038513 0.028208 1.365 0.17428
## masq_gd 0.054353 0.016271 3.341 0.00106 **
## masq_aa 0.022176 0.020072 1.105 0.27108
## masq_ad -0.001578 0.019047 -0.083 0.93407
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.493 on 144 degrees of freedom
## (4 observations deleted due to missingness)
## Multiple R-squared: 0.2223, Adjusted R-squared: 0.2007
## F-statistic: 10.29 on 4 and 144 DF, p-value: 2.339e-07summary(lm(data = filter(ear3, reject_method == "indirect"), neg_emotions ~ arsq_full + masq_gd + masq_aa + masq_ad))##
## Call:
## lm(formula = neg_emotions ~ arsq_full + masq_gd + masq_aa + masq_ad,
## data = filter(ear3, reject_method == "indirect"))
##
## Residuals:
## Min 1Q Median 3Q Max
## -3.1265 -0.6867 0.1250 0.9569 3.2197
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 3.00308 0.82269 3.650 0.000439 ***
## arsq_full -0.02992 0.03359 -0.891 0.375473
## masq_gd 0.05028 0.02454 2.049 0.043360 *
## masq_aa 0.03693 0.02531 1.459 0.148070
## masq_ad -0.01661 0.02213 -0.751 0.454864
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.501 on 90 degrees of freedom
## (1 observation deleted due to missingness)
## Multiple R-squared: 0.1497, Adjusted R-squared: 0.1119
## F-statistic: 3.961 on 4 and 90 DF, p-value: 0.005239Without covariates, and mirroring the trend that was apparent in the full model, higher rejection sensitivity was associated with decreased negative emotions for direct rejection methods … but not for indirect methods
summary(lm(data = filter(ear3, reject_method == "direct"), neg_emotions ~ arsq_full))##
## Call:
## lm(formula = neg_emotions ~ arsq_full, data = filter(ear3, reject_method ==
## "direct"))
##
## Residuals:
## Min 1Q Median 3Q Max
## -3.6871 -1.0577 0.1127 1.1992 3.4492
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.91865 0.32889 8.874 2.22e-15 ***
## arsq_full 0.09810 0.02467 3.976 0.00011 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.592 on 147 degrees of freedom
## (4 observations deleted due to missingness)
## Multiple R-squared: 0.0971, Adjusted R-squared: 0.09096
## F-statistic: 15.81 on 1 and 147 DF, p-value: 0.0001095confint(lm(data = filter(ear3, reject_method == "direct"), neg_emotions ~ arsq_full))## 2.5 % 97.5 %
## (Intercept) 2.26867966 3.5686145
## arsq_full 0.04933942 0.1468515summary(lm(data = filter(ear3, reject_method == "indirect"), neg_emotions ~ arsq_full))##
## Call:
## lm(formula = neg_emotions ~ arsq_full, data = filter(ear3, reject_method ==
## "indirect"))
##
## Residuals:
## Min 1Q Median 3Q Max
## -3.444 -1.274 0.272 1.095 2.829
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 3.94879 0.44285 8.917 4e-14 ***
## arsq_full 0.01834 0.03035 0.604 0.547
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.598 on 93 degrees of freedom
## (1 observation deleted due to missingness)
## Multiple R-squared: 0.00391, Adjusted R-squared: -0.0068
## F-statistic: 0.3651 on 1 and 93 DF, p-value: 0.5472confint(lm(data = filter(ear3, reject_method == "indirect"), neg_emotions ~ arsq_full))## 2.5 % 97.5 %
## (Intercept) 3.06938743 4.82820048
## arsq_full -0.04193542 0.07861667From Study 2 for the correlation table
ear3 %>%
select(masq_ad, masq_aa, masq_gd,
rse, arsq_full,
difficulty, neg_emotions) %>%
corrr::correlate(use = "pairwise.complete.obs",
method = "pearson") %>%
dplyr::select(term, difficulty, neg_emotions) %>%
filter(term %in% c("masq_ad", "masq_aa", "masq_gd",
"rse", "arsq_full")) %>%
corrr::fashion() %>%
gt::gt()| term | difficulty | neg_emotions |
|---|---|---|
| masq_ad | -.04 | .14 |
| masq_aa | .15 | .30 |
| masq_gd | .20 | .40 |
| rse | -.00 | -.22 |
| arsq_full | .04 | .22 |
# and the p-value / stats on them
ear3 %>%
select(masq_ad, masq_aa, masq_gd, rse, arsq_full,
difficulty, neg_emotions) %>%
pivot_longer(
cols = c(difficulty, neg_emotions),
names_to = "outcomes",
values_to = "out_val") %>%
pivot_longer(
cols = c(masq_ad, masq_aa, masq_gd, rse,
arsq_full),
names_to = "predictors",
values_to = "pred_val"
) %>%
nest(data = -c(outcomes, predictors)) %>%
mutate(correlation = map(data,
~ cor.test(.x$out_val, .x$pred_val)),
tidied = map(correlation, broom::tidy)) %>%
unnest(tidied) %>%
select(outcomes, predictors, df = parameter,
estimate, conf.low, conf.high, p.value) %>%
gt::gt() %>%
gt::cols_merge(columns = c(estimate, conf.low,
conf.high),
pattern = "{1} [{2}, {3}]") %>%
gt::cols_label(
estimate = md("*r* [95% CI]"),
df = md("*df*"),
p.value = md("*p*")
) %>%
gt::sub_small_vals(columns = c(estimate,
conf.low,
conf.high),
threshold = .01) %>%
gt::sub_small_vals(columns = p.value,
threshold = .001,
small_pattern = "<.001") %>%
gt::fmt_number(p.value, decimals = 3) %>%
gt::fmt_number(c(estimate, conf.low, conf.high),
decimals = 2)| outcomes | predictors | df | r [95% CI] | p |
|---|---|---|---|---|
| difficulty | masq_ad | 256 | −0.04 [−0.17, 0.08] | 0.479 |
| difficulty | masq_aa | 256 | 0.15 [0.03, 0.27] | 0.013 |
| difficulty | masq_gd | 256 | 0.20 [0.08, 0.32] | <.001 |
| difficulty | rse | 252 | 0.00 [−0.13, 0.12] | 0.959 |
| difficulty | arsq_full | 252 | 0.04 [−0.08, 0.16] | 0.539 |
| neg_emotions | masq_ad | 256 | 0.14 [0.01, 0.25] | 0.029 |
| neg_emotions | masq_aa | 256 | 0.30 [0.19, 0.41] | <.001 |
| neg_emotions | masq_gd | 256 | 0.40 [0.29, 0.50] | <.001 |
| neg_emotions | rse | 252 | −0.22 [−0.34, −0.10] | <.001 |
| neg_emotions | arsq_full | 252 | 0.22 [0.10, 0.34] | <.001 |