Initialization

rm(list=ls())
library(relativeVariability)
library(rsurveyutils)
library(here)
library(tidyverse)
library(corrplot)
# library(ggthemr)
# theme_set(theme_bw())
# ggthemr("fresh")

library(lmerTest)
library(sjPlot)

# library(aTSA)
library(tidylog)

Preprocessing

Read data

filepath <- here("data/ESM_W2_WITH_MISSING_V1_no exclusions.csv")
df_raw <- read_csv(filepath)

## Warning: One or more parsing issues, call `problems()` on your data frame for details,
## e.g.:
##   dat <- vroom(...)
##   problems(dat)

Process data

df_emos <- df_raw %>% 
  select(ID, timeStampSent, matches("panas.*scale.*beep"), -matches("new")) %>% 
  # TODO: what is panas scale new
  rename(
    panas_neg = PANAS_negativescale_beep,
    panas_pos = PANAS_positivescale_beep
  )

df_er <- df_raw %>% 
  select(ID, timeStampSent, matches("Emotionregulation_[a-z]+$"),
         -Emotionregulation_selfcompassion1,
         -Emotionregulation_selfcompassion2,
         -Emotionregulation_selfcompassion,
         -Emotionregulation_engagement) %>% 
  rename_all(~str_replace(., "Emotionregulation_", "er_"))

df_indiff <- df_raw %>% 
  select(ID, Sex, Birth_year, Birth_month, Ethnicity, 
         matches("SCARED.*scale"), matches("SMFQ.*scale"),
         # matches("panas"),
         -matches("omit")) %>% 
  rename_all(~str_remove(., "_scale_w2")) %>% 
  distinct() %>%
  left_join(
    df_emos %>% 
      filter(!is.na(panas_neg) & !is.na(panas_pos)) %>% 
      group_by(ID) %>% 
      summarize(
        panas_neg = mean(panas_neg, na.rm=T),
        panas_pos = mean(panas_pos, na.rm=T),
        n_beeps = n()
      )
  ) %>% 
  mutate_at(vars(SCARED, SMFQ, panas_neg, panas_pos), list(z=scale)) 
# confirm 1 row per person
df_indiff %>% 
  count(ID) %>% 
  arrange(desc(n))

## # A tibble: 265 × 2
##    ID        n
##    <chr> <int>
##  1 10015     1
##  2 10016     1
##  3 10017     1
##  4 10018     1
##  5 10020     1
##  6 10021     1
##  7 10022     1
##  8 10026     1
##  9 10027     1
## 10 10028     1
## # ℹ 255 more rows

df_indiff_2 <- df_raw %>% 
  select(ID, Sex, Birth_year, Birth_month, Ethnicity, 
         matches("SCARED.*scale"), matches("SMFQ.*scale"),
         # matches("panas"),
         -matches("omit")) %>% 
  rename_all(~str_remove(., "_scale_w2")) %>% 
  distinct() %>%
  left_join(
    df_emos # %>% 
      # filter(!is.na(panas_neg) & !is.na(panas_pos)) %>% 
  ) %>% 
  mutate_at(vars(SCARED, SMFQ, panas_neg, panas_pos), list(z=scale))

Calculate Metrics

Mean ER

df_mean_er_strat <- df_er %>% 
  pivot_longer(cols=starts_with("er_"), 
               names_to="strat", 
               values_to="rating") %>% 
  drop_na(rating) %>% 
  group_by(ID, strat) %>%
  summarize(
    mean_er = mean(rating)
  ) %>%
  ungroup

Relative Within SD

d <- df_er %>% 
  pivot_longer(cols=starts_with("er_"), 
               names_to="strat", 
               values_to="rating") %>% 
  drop_na(rating)

d %>% 
  group_by(ID, strat) %>% 
  summarize(
    n = n()
  ) %>%
  # arrange(desc(n))
  arrange(n)

## # A tibble: 1,325 × 3
## # Groups:   ID [265]
##    ID    strat                n
##    <chr> <chr>            <int>
##  1 1029  er_distraction       1
##  2 1029  er_reappraisal       1
##  3 1029  er_rumination        1
##  4 1029  er_socialsupport     1
##  5 1029  er_suppresion        1
##  6 8048  er_distraction       1
##  7 8048  er_reappraisal       1
##  8 8048  er_rumination        1
##  9 8048  er_socialsupport     1
## 10 8048  er_suppresion        1
## # ℹ 1,315 more rows

df_rwsd <- d %>% 
  group_by(ID, strat) %>% 
  mutate(n = n()) %>% 
  filter(n > 1) %>% 
  filter(mean(rating) < 7) %>%
  filter(mean(rating) > 1) %>%
  summarize(
    rwsd = relativeSD(rating, 1, 7),
    n = n()
  ) %>% 
  summarize(
    rwsd = mean(rwsd),
    n = mean(n)
  ) %>% 
  ungroup

hist(df_rwsd$rwsd)

# rwsd for each ER strat
df_rwsd_st <- d %>% 
  group_by(ID, strat) %>% 
  mutate(n = n()) %>% 
  filter(n > 1) %>% 
  filter(mean(rating) < 7) %>%
  filter(mean(rating) > 1) %>%
  summarize(
    rwsd = relativeSD(rating, 1, 7),
    n = n()
  )

Relative Between SD

d <- df_er %>% 
  pivot_longer(cols=starts_with("er_"), 
               names_to="strat", 
               values_to="rating") %>% 
  drop_na(rating)

d %>% 
  group_by(ID, timeStampSent) %>% 
  summarize(
    n = n()
  ) %>%
  # arrange(desc(n))
  arrange(n)

## # A tibble: 12,590 × 3
## # Groups:   ID [265]
##    ID    timeStampSent     n
##    <chr>         <dbl> <int>
##  1 7034     1652955304     2
##  2 11187    1653992106     3
##  3 11143    1653768014     4
##  4 11143    1653940813     4
##  5 11143    1654027212     4
##  6 11143    1654157103     4
##  7 11143    1654449309     4
##  8 11170    1653768014     4
##  9 2026     1650967206     4
## 10 2026     1651782618     4
## # ℹ 12,580 more rows

df_rbsd <- d %>% 
  group_by(ID, timeStampSent) %>% 
  mutate(n = n()) %>% 
  filter(n > 1) %>% 
  filter(mean(rating) > 1) %>%
  filter(mean(rating) < 7) %>%
  summarize(
    rbsd = relativeSD(rating, 1, 7),
    n = n()
  ) %>% 
  summarize(
    rbsd = mean(rbsd),
    n = mean(n)
  )

hist(df_rbsd$rbsd)

# rbsd across moments
df_rbsd_st <- d %>% 
  group_by(ID, timeStampSent) %>% 
  mutate(n = n()) %>% 
  filter(n > 1) %>% 
  filter(mean(rating) > 1) %>%
  filter(mean(rating) < 7) %>%
  summarize(
    rbsd = relativeSD(rating, 1, 7),
    n = n()
  )

Successive Difference

d <- df_er %>% 
  pivot_longer(cols=starts_with("er_"), 
               names_to="strat", 
               values_to="rating") %>% 
  drop_na(rating)

d %>% 
  group_by(ID, strat) %>% 
  summarize(
    n = n()
  ) %>%
  # arrange(desc(n))
  arrange(n)

## # A tibble: 1,325 × 3
## # Groups:   ID [265]
##    ID    strat                n
##    <chr> <chr>            <int>
##  1 1029  er_distraction       1
##  2 1029  er_reappraisal       1
##  3 1029  er_rumination        1
##  4 1029  er_socialsupport     1
##  5 1029  er_suppresion        1
##  6 8048  er_distraction       1
##  7 8048  er_reappraisal       1
##  8 8048  er_rumination        1
##  9 8048  er_socialsupport     1
## 10 8048  er_suppresion        1
## # ℹ 1,315 more rows

df_rwrmssd <- d %>% 
  group_by(ID, strat) %>% 
  mutate(n = n()) %>% 
  filter(n > 1) %>% 
  filter(mean(rating) < 7) %>%
  filter(mean(rating) > 1) %>%
  summarize(
    rwrmssd = relativeRMSSD(rating, 1, 7),
    n = n()
  ) %>% 
  summarize(
    rwrmssd = mean(rwrmssd),
    n = mean(n)
  ) %>% 
  ungroup

hist(df_rwrmssd$rwrmssd)

# rwrmssd for each ER strat
df_rwrmssd_st <- d %>% 
  group_by(ID, strat) %>% 
  mutate(n = n()) %>% 
  filter(n > 1) %>% 
  filter(mean(rating) < 7) %>%
  filter(mean(rating) > 1) %>%
  summarize(
    rwrmssd = relativeRMSSD(rating, 1, 7),
    n = n()
  ) %>%
  ungroup

Bray-Curtis

d <- df_er %>% 
  pivot_longer(cols=starts_with("er_"), 
               names_to="strat", 
               values_to="rating") %>% 
  group_by(ID, strat) %>% 
  arrange(timeStampSent) %>% 
  mutate(
    rating_abs_diff = abs(rating - lag(rating)),
    rating_sum = rating + lag(rating),
  ) %>%
  ungroup %>% 
  drop_na(rating_abs_diff)

# momentary level (state)
df_bc_st <- d %>% 
  group_by(ID, timeStampSent) %>%
  summarize(
    bc = sum(rating_abs_diff, na.rm=T)/sum(rating_sum, na.rm=T)
  )

hist(df_bc_st$bc)

# individual level (trait)
df_bc_tr <- d %>%
  group_by(ID) %>%
  summarize(
    bc = sum(rating_abs_diff, na.rm=T)/sum(rating_sum, na.rm=T),
    n = n()
  ) 

hist(df_bc_tr$bc)

# checking for a relationship between number of observations and bc
df_bc_tr %>% 
  ggplot(aes(x = n, y = bc)) +
  geom_smooth(se=F, size=.4, color="red") + 
  geom_smooth(method="lm") + 
  geom_point()

## Warning: Using `size` aesthetic for lines was deprecated in ggplot2 3.4.0.
## ℹ Please use `linewidth` instead.
## This warning is displayed once every 8 hours.
## Call `lifecycle::last_lifecycle_warnings()` to see where this warning was
## generated.

Wentropy (Emotion Regulation Diversity)

# momentary (state)
df_wentropy_st <- df_er %>% 
  pivot_longer(cols=starts_with("er_"), 
               names_to="strat", 
               values_to="rating") %>% 
  drop_na(rating) %>% 
  group_by(ID, timeStampSent) %>% 
  mutate(
    pr = rating/7,
    ln_pr = log(pr),
    pr_ln_pr = pr*ln_pr
    ) %>%
  summarize(
    wentropy = -sum(pr_ln_pr)
  ) %>% 
  ungroup

hist(df_wentropy_st$wentropy)

# individual (trait)
df_wentropy_tr <- df_er %>% 
  pivot_longer(cols=starts_with("er_"), 
               names_to="strat", 
               values_to="rating") %>% 
  drop_na(rating) %>% 
  group_by(ID, strat) %>% 
  summarize(
    pr = sum(rating)/(7*n()),
    ln_pr = log(pr),
    pr_ln_pr = pr*ln_pr
    ) %>%
  summarize(
    wentropy = -sum(pr_ln_pr)
  ) %>% 
  ungroup

hist(df_wentropy_tr$wentropy)

Entropy

# momentary (state)
df_entropy_st <- df_er %>% 
  pivot_longer(cols=starts_with("er_"), 
               names_to="strat", 
               values_to="rating") %>% 
  drop_na(rating) %>% 
  group_by(ID, timeStampSent) %>% 
  mutate(
    pr = rating/sum(rating),
    ln_pr = log(pr),
    pr_ln_pr = pr*ln_pr
    ) %>%
  summarize(
    entropy = -sum(pr_ln_pr)
  ) %>% 
  ungroup

hist(df_entropy_st$entropy, 50)

# individual (trait)
df_entropy_tr <- df_er %>% 
  pivot_longer(cols=starts_with("er_"), 
               names_to="strat", 
               values_to="rating") %>% 
  drop_na(rating) %>% 
  group_by(ID, strat) %>% 
  summarize(
    sum_strategy_rating = sum(rating),
    ) %>%
  mutate(
    pr = sum_strategy_rating/sum(sum_strategy_rating),
    ln_pr = log(pr),
    pr_ln_pr = pr*ln_pr
    ) %>%
  summarize(
    entropy = -sum(pr_ln_pr)
  ) %>% 
  ungroup

hist(df_entropy_tr$entropy, breaks=30)

Trait (between-person analysis)

Cor Between Metrics

d <- df_rwsd %>% 
  select(-n) %>% 
  full_join(df_rbsd %>% select(-n), by="ID") %>%
  full_join(df_rwrmssd %>% select(-n), by="ID") %>%
  full_join(df_bc_tr %>% select(-n), by="ID") %>%
  full_join(df_wentropy_tr, by="ID") %>%
  full_join(df_entropy_tr, by="ID") %>% 
  full_join(df_mean_er_strat, by="ID") %>%
  select(-ID) %>% 
  select(-strat)

## select: dropped one variable (n)
## select: dropped one variable (n)
## full_join: added one column (rbsd)
##            > rows only in x                         0
##            > rows only in df_rbsd %>% select(-n)    1
##            > matched rows                         253
##            >                                     =====
##            > rows total                           254
## select: dropped one variable (n)
## full_join: added one column (rwrmssd)
##            > rows only in x                            1
##            > rows only in df_rwrmssd %>% select(-n)    0
##            > matched rows                            253
##            >                                        =====
##            > rows total                              254
## select: dropped one variable (n)
## full_join: added one column (bc)
##            > rows only in x                          6
##            > rows only in df_bc_tr %>% select(-n)    9
##            > matched rows                          248
##            >                                      =====
##            > rows total                            263
## full_join: added one column (wentropy)
##            > rows only in x                 0
##            > rows only in df_wentropy_tr    2
##            > matched rows                 263
##            >                             =====
##            > rows total                   265
## full_join: added one column (entropy)
##            > rows only in x                0
##            > rows only in df_entropy_tr    0
##            > matched rows                265
##            >                            =====
##            > rows total                  265
## full_join: added 2 columns (strat, mean_er)
##            > rows only in x                     0
##            > rows only in df_mean_er_strat      0
##            > matched rows                   1,325    (includes duplicates)
##            >                               =======
##            > rows total                     1,325
## select: dropped one variable (ID)
## select: dropped one variable (strat)

corrplot::corrplot(cor(d, use="pairwise.complete.obs"))

cor(d, use="pairwise.complete.obs")

##                 rwsd        rbsd     rwrmssd          bc   wentropy     entropy
## rwsd      1.00000000  0.69307528  0.90168342 -0.01388751 -0.1229160 -0.05263748
## rbsd      0.69307528  1.00000000  0.64130309 -0.03940082 -0.2611491 -0.29040009
## rwrmssd   0.90168342  0.64130309  1.00000000 -0.11901781 -0.1643954  0.02033524
## bc       -0.01388751 -0.03940082 -0.11901781  1.00000000  0.5667578  0.01220801
## wentropy -0.12291598 -0.26114908 -0.16439543  0.56675784  1.0000000  0.21150563
## entropy  -0.05263748 -0.29040009  0.02033524  0.01220801  0.2115056  1.00000000
## mean_er  -0.44514455 -0.20574042 -0.55858523  0.21007821 -0.2679604 -0.16430195
##             mean_er
## rwsd     -0.4451445
## rbsd     -0.2057404
## rwrmssd  -0.5585852
## bc        0.2100782
## wentropy -0.2679604
## entropy  -0.1643019
## mean_er   1.0000000

Relative Within SD

df_rwsd_outcomes <- df_rwsd %>% 
  left_join(df_indiff)

## Joining with `by = join_by(ID)`
## left_join: added 13 columns (Sex, Birth_year, Birth_month, Ethnicity, SCARED,
## …)
## > rows only in x 0
## > rows only in df_indiff ( 12)
## > matched rows 253
## > =====
## > rows total 253

Depression

fit <- lm(SMFQ ~ rwsd, data=df_rwsd_outcomes)
tab_model(fit)

	SMFQ
Predictors	Estimates	CI	p
(Intercept)	1.66	1.45 – 1.88	<0.001
rwsd	-0.29	-0.57 – -0.01	0.039
Observations	245
R² / R² adjusted	0.017 / 0.013

plot_model(fit, type="slope")

## `geom_smooth()` using formula = 'y ~ x'
## `geom_smooth()` using formula = 'y ~ x'

Anxiety

fit <- lm(SCARED ~ rwsd, data=df_rwsd_outcomes)
tab_model(fit)

	SCARED
Predictors	Estimates	CI	p
(Intercept)	2.19	1.98 – 2.41	<0.001
rwsd	-0.35	-0.63 – -0.06	0.017
Observations	244
R² / R² adjusted	0.023 / 0.019

plot_model(fit, type="slope")

## `geom_smooth()` using formula = 'y ~ x'
## `geom_smooth()` using formula = 'y ~ x'

Neg Affect

fit <- lm(panas_neg ~ rwsd, data=df_rwsd_outcomes)
tab_model(fit)

	panas_neg
Predictors	Estimates	CI	p
(Intercept)	2.84	2.45 – 3.22	<0.001
rwsd	-1.37	-1.88 – -0.87	<0.001
Observations	253
R² / R² adjusted	0.103 / 0.100

plot_model(fit, type="slope")

## `geom_smooth()` using formula = 'y ~ x'
## `geom_smooth()` using formula = 'y ~ x'

Pos Affect

fit <- lm(panas_pos ~ rwsd, data=df_rwsd_outcomes)
tab_model(fit)

	panas_pos
Predictors	Estimates	CI	p
(Intercept)	3.97	3.40 – 4.55	<0.001
rwsd	1.56	0.81 – 2.31	<0.001
Observations	253
R² / R² adjusted	0.062 / 0.059

plot_model(fit, type="slope")

## `geom_smooth()` using formula = 'y ~ x'
## `geom_smooth()` using formula = 'y ~ x'

Relative Between SD

df_rbsd_outcomes <- df_rbsd %>% 
  left_join(df_indiff)

## Joining with `by = join_by(ID)`
## left_join: added 13 columns (Sex, Birth_year, Birth_month, Ethnicity, SCARED,
## …)
## > rows only in x 0
## > rows only in df_indiff ( 11)
## > matched rows 254
## > =====
## > rows total 254

Depression

fit <- lm(SMFQ ~ rbsd, data=df_rbsd_outcomes)
tab_model(fit)

	SMFQ
Predictors	Estimates	CI	p
(Intercept)	1.55	1.40 – 1.70	<0.001
rbsd	-0.16	-0.38 – 0.05	0.141
Observations	246
R² / R² adjusted	0.009 / 0.005

plot_model(fit, type="slope")

## `geom_smooth()` using formula = 'y ~ x'
## `geom_smooth()` using formula = 'y ~ x'

Anxiety

fit <- lm(SCARED ~ rbsd, data=df_rbsd_outcomes)
tab_model(fit)

	SCARED
Predictors	Estimates	CI	p
(Intercept)	2.03	1.87 – 2.19	<0.001
rbsd	-0.14	-0.36 – 0.08	0.212
Observations	245
R² / R² adjusted	0.006 / 0.002

plot_model(fit, type="slope")

## `geom_smooth()` using formula = 'y ~ x'
## `geom_smooth()` using formula = 'y ~ x'

Neg Affect

fit <- lm(panas_neg ~ rbsd, data=df_rbsd_outcomes)
tab_model(fit)

	panas_neg
Predictors	Estimates	CI	p
(Intercept)	2.23	1.94 – 2.51	<0.001
rbsd	-0.61	-1.01 – -0.21	0.003
Observations	254
R² / R² adjusted	0.034 / 0.030

plot_model(fit, type="slope")

## `geom_smooth()` using formula = 'y ~ x'
## `geom_smooth()` using formula = 'y ~ x'

Pos Affect

fit <- lm(panas_pos ~ rbsd, data=df_rbsd_outcomes)
tab_model(fit)

	panas_pos
Predictors	Estimates	CI	p
(Intercept)	4.69	4.27 – 5.11	<0.001
rbsd	0.66	0.07 – 1.26	0.028
Observations	254
R² / R² adjusted	0.019 / 0.015

plot_model(fit, type="slope")

## `geom_smooth()` using formula = 'y ~ x'
## `geom_smooth()` using formula = 'y ~ x'

Successive Difference

df_rwrmssd_outcomes <- df_rwrmssd %>% 
  left_join(df_indiff)

## Joining with `by = join_by(ID)`
## left_join: added 13 columns (Sex, Birth_year, Birth_month, Ethnicity, SCARED,
## …)
## > rows only in x 0
## > rows only in df_indiff ( 12)
## > matched rows 253
## > =====
## > rows total 253

Depression

fit <- lm(SMFQ ~ rwrmssd, data=df_rwrmssd_outcomes)
tab_model(fit)

	SMFQ
Predictors	Estimates	CI	p
(Intercept)	1.68	1.51 – 1.85	<0.001
rwrmssd	-0.39	-0.65 – -0.13	0.004
Observations	245
R² / R² adjusted	0.034 / 0.030

plot_model(fit, type="slope")

## `geom_smooth()` using formula = 'y ~ x'
## `geom_smooth()` using formula = 'y ~ x'

Anxiety

fit <- lm(SCARED ~ rwrmssd, data=df_rwrmssd_outcomes)
tab_model(fit)

	SCARED
Predictors	Estimates	CI	p
(Intercept)	2.24	2.07 – 2.40	<0.001
rwrmssd	-0.49	-0.76 – -0.23	<0.001
Observations	244
R² / R² adjusted	0.053 / 0.049

plot_model(fit, type="slope")

## `geom_smooth()` using formula = 'y ~ x'
## `geom_smooth()` using formula = 'y ~ x'

Neg Affect

fit <- lm(panas_neg ~ rwrmssd, data=df_rwrmssd_outcomes)
tab_model(fit)

	panas_neg
Predictors	Estimates	CI	p
(Intercept)	2.78	2.49 – 3.08	<0.001
rwrmssd	-1.58	-2.03 – -1.12	<0.001
Observations	253
R² / R² adjusted	0.156 / 0.153

plot_model(fit, type="slope")

## `geom_smooth()` using formula = 'y ~ x'
## `geom_smooth()` using formula = 'y ~ x'

Pos Affect

fit <- lm(panas_pos ~ rwrmssd, data=df_rwrmssd_outcomes)
tab_model(fit)

	panas_pos
Predictors	Estimates	CI	p
(Intercept)	4.23	3.78 – 4.68	<0.001
rwrmssd	1.47	0.77 – 2.18	<0.001
Observations	253
R² / R² adjusted	0.064 / 0.060

plot_model(fit, type="slope")

## `geom_smooth()` using formula = 'y ~ x'
## `geom_smooth()` using formula = 'y ~ x'

Bray-Curtis

df_bc_tr_outcomes <- df_bc_tr %>% 
  left_join(df_indiff)

## Joining with `by = join_by(ID)`
## left_join: added 13 columns (Sex, Birth_year, Birth_month, Ethnicity, SCARED,
## …)
## > rows only in x 0
## > rows only in df_indiff ( 8)
## > matched rows 257
## > =====
## > rows total 257

Depression

fit <- lm(SMFQ ~ bc, data=df_bc_tr_outcomes)
tab_model(fit)

	SMFQ
Predictors	Estimates	CI	p
(Intercept)	1.37	1.27 – 1.47	<0.001
bc	0.45	-0.04 – 0.95	0.074
Observations	249
R² / R² adjusted	0.013 / 0.009

plot_model(fit, type="slope")

## `geom_smooth()` using formula = 'y ~ x'
## `geom_smooth()` using formula = 'y ~ x'

Anxiety

fit <- lm(SCARED ~ bc, data=df_bc_tr_outcomes)
tab_model(fit)

	SCARED
Predictors	Estimates	CI	p
(Intercept)	1.79	1.68 – 1.89	<0.001
bc	0.84	0.33 – 1.34	0.001
Observations	249
R² / R² adjusted	0.041 / 0.038

plot_model(fit, type="slope")

## `geom_smooth()` using formula = 'y ~ x'
## `geom_smooth()` using formula = 'y ~ x'

Neg Affect

fit <- lm(panas_neg ~ bc, data=df_bc_tr_outcomes)
tab_model(fit)

	panas_neg
Predictors	Estimates	CI	p
(Intercept)	1.38	1.20 – 1.56	<0.001
bc	2.51	1.62 – 3.40	<0.001
Observations	257
R² / R² adjusted	0.108 / 0.105

plot_model(fit, type="slope")

## `geom_smooth()` using formula = 'y ~ x'
## `geom_smooth()` using formula = 'y ~ x'

Pos Affect

fit <- lm(panas_pos ~ bc, data=df_bc_tr_outcomes)
tab_model(fit)

	panas_pos
Predictors	Estimates	CI	p
(Intercept)	5.66	5.39 – 5.93	<0.001
bc	-2.97	-4.31 – -1.64	<0.001
Observations	257
R² / R² adjusted	0.070 / 0.066

plot_model(fit, type="slope")

## `geom_smooth()` using formula = 'y ~ x'
## `geom_smooth()` using formula = 'y ~ x'

Wentropy (Emotion Regulation Diversity)

df_wentropy_tr_outcomes <- df_wentropy_tr %>% 
  left_join(df_indiff)

## Joining with `by = join_by(ID)`
## left_join: added 13 columns (Sex, Birth_year, Birth_month, Ethnicity, SCARED,
## …)
## > rows only in x 0
## > rows only in df_indiff ( 0)
## > matched rows 265
## > =====
## > rows total 265

Depression

fit <- lm(SMFQ ~ wentropy, data=df_wentropy_tr_outcomes)
tab_model(fit)

	SMFQ
Predictors	Estimates	CI	p
(Intercept)	1.31	0.95 – 1.67	<0.001
wentropy	0.08	-0.15 – 0.32	0.497
Observations	257
R² / R² adjusted	0.002 / -0.002

plot_model(fit, type="slope")

## `geom_smooth()` using formula = 'y ~ x'
## `geom_smooth()` using formula = 'y ~ x'

Anxiety

fit <- lm(SCARED ~ wentropy, data=df_wentropy_tr_outcomes)
tab_model(fit)

	SCARED
Predictors	Estimates	CI	p
(Intercept)	1.71	1.33 – 2.09	<0.001
wentropy	0.13	-0.11 – 0.38	0.284
Observations	256
R² / R² adjusted	0.005 / 0.001

plot_model(fit, type="slope")

## `geom_smooth()` using formula = 'y ~ x'
## `geom_smooth()` using formula = 'y ~ x'

Neg Affect

fit <- lm(panas_neg ~ wentropy, data=df_wentropy_tr_outcomes)
tab_model(fit)

	panas_neg
Predictors	Estimates	CI	p
(Intercept)	0.89	0.21 – 1.57	0.010
wentropy	0.59	0.15 – 1.03	0.008
Observations	265
R² / R² adjusted	0.026 / 0.022

plot_model(fit, type="slope")

## `geom_smooth()` using formula = 'y ~ x'
## `geom_smooth()` using formula = 'y ~ x'

Pos Affect

fit <- lm(panas_pos ~ wentropy, data=df_wentropy_tr_outcomes)
tab_model(fit)

	panas_pos
Predictors	Estimates	CI	p
(Intercept)	7.08	6.09 – 8.06	<0.001
wentropy	-1.25	-1.88 – -0.61	<0.001
Observations	265
R² / R² adjusted	0.053 / 0.049

plot_model(fit, type="slope")

## `geom_smooth()` using formula = 'y ~ x'
## `geom_smooth()` using formula = 'y ~ x'

Entropy

df_entropy_tr_outcomes <- df_entropy_tr %>% 
  left_join(df_indiff)

## Joining with `by = join_by(ID)`
## left_join: added 13 columns (Sex, Birth_year, Birth_month, Ethnicity, SCARED,
## …)
## > rows only in x 0
## > rows only in df_indiff ( 0)
## > matched rows 265
## > =====
## > rows total 265

Depression

fit <- lm(SMFQ ~ entropy, data=df_entropy_tr_outcomes)
tab_model(fit)

	SMFQ
Predictors	Estimates	CI	p
(Intercept)	3.78	2.13 – 5.42	<0.001
entropy	-1.47	-2.51 – -0.44	0.005
Observations	257
R² / R² adjusted	0.030 / 0.026

plot_model(fit, type="slope")

## `geom_smooth()` using formula = 'y ~ x'
## `geom_smooth()` using formula = 'y ~ x'

Anxiety

fit <- lm(SCARED ~ entropy, data=df_entropy_tr_outcomes)
tab_model(fit)

	SCARED
Predictors	Estimates	CI	p
(Intercept)	4.25	2.54 – 5.97	<0.001
entropy	-1.47	-2.55 – -0.39	0.008
Observations	256
R² / R² adjusted	0.028 / 0.024

plot_model(fit, type="slope")

## `geom_smooth()` using formula = 'y ~ x'
## `geom_smooth()` using formula = 'y ~ x'

Neg Affect

fit <- lm(panas_neg ~ entropy, data=df_entropy_tr_outcomes)
tab_model(fit)

	panas_neg
Predictors	Estimates	CI	p
(Intercept)	3.82	0.69 – 6.96	0.017
entropy	-1.28	-3.25 – 0.70	0.204
Observations	265
R² / R² adjusted	0.006 / 0.002

plot_model(fit, type="slope")

## `geom_smooth()` using formula = 'y ~ x'
## `geom_smooth()` using formula = 'y ~ x'

Pos Affect

fit <- lm(panas_pos ~ entropy, data=df_entropy_tr_outcomes)
tab_model(fit)

	panas_pos
Predictors	Estimates	CI	p
(Intercept)	2.60	-2.02 – 7.23	0.268
entropy	1.62	-1.29 – 4.53	0.274
Observations	265
R² / R² adjusted	0.005 / 0.001

plot_model(fit, type="slope")

## `geom_smooth()` using formula = 'y ~ x'
## `geom_smooth()` using formula = 'y ~ x'

Momentary (within-person analysis)

Cor Between Metrics

df_bc_st_mean <- df_bc_st %>% 
  group_by(ID) %>% 
  summarize(mean_bc = mean(bc, na.rm = TRUE))

## group_by: one grouping variable (ID)
## summarize: now 257 rows and 2 columns, ungrouped

df_wentropy_st_mean <- df_wentropy_st %>% 
  group_by(ID) %>% 
  summarize(mean_wentropy = mean(wentropy, na.rm = TRUE))

## group_by: one grouping variable (ID)
## summarize: now 265 rows and 2 columns, ungrouped

df_entropy_st_mean <- df_entropy_st %>% 
  group_by(ID) %>% 
  summarize(mean_entropy = mean(entropy, na.rm = TRUE))

## group_by: one grouping variable (ID)
## summarize: now 265 rows and 2 columns, ungrouped

d <- df_rwsd %>% 
  select(-n) %>% 
  full_join(df_rbsd %>% select(-n), by="ID") %>%
  full_join(df_rwrmssd %>% select(-n), by="ID") %>%
  full_join(df_bc_st_mean, by="ID") %>%
  full_join(df_wentropy_st_mean, by="ID") %>%
  full_join(df_entropy_st_mean, by="ID") %>% 
  full_join(df_mean_er_strat, by="ID") %>%
  select(-ID) %>% 
  select(-strat)

## select: dropped one variable (n)
## select: dropped one variable (n)
## full_join: added one column (rbsd)
##            > rows only in x                         0
##            > rows only in df_rbsd %>% select(-n)    1
##            > matched rows                         253
##            >                                     =====
##            > rows total                           254
## select: dropped one variable (n)
## full_join: added one column (rwrmssd)
##            > rows only in x                            1
##            > rows only in df_rwrmssd %>% select(-n)    0
##            > matched rows                            253
##            >                                        =====
##            > rows total                              254
## full_join: added one column (mean_bc)
##            > rows only in x                6
##            > rows only in df_bc_st_mean    9
##            > matched rows                248
##            >                            =====
##            > rows total                  263
## full_join: added one column (mean_wentropy)
##            > rows only in x                      0
##            > rows only in df_wentropy_st_mean    2
##            > matched rows                      263
##            >                                  =====
##            > rows total                        265
## full_join: added one column (mean_entropy)
##            > rows only in x                     0
##            > rows only in df_entropy_st_mean    0
##            > matched rows                     265
##            >                                 =====
##            > rows total                       265
## full_join: added 2 columns (strat, mean_er)
##            > rows only in x                     0
##            > rows only in df_mean_er_strat      0
##            > matched rows                   1,325    (includes duplicates)
##            >                               =======
##            > rows total                     1,325
## select: dropped one variable (ID)
## select: dropped one variable (strat)

corrplot::corrplot(cor(d, use="pairwise.complete.obs"))

cor(d, use="pairwise.complete.obs")

##                     rwsd        rbsd     rwrmssd     mean_bc mean_wentropy
## rwsd           1.0000000  0.69307528  0.90168342 -0.14082009   -0.23403019
## rbsd           0.6930753  1.00000000  0.64130309 -0.06915301   -0.36418921
## rwrmssd        0.9016834  0.64130309  1.00000000 -0.27621472   -0.05835954
## mean_bc       -0.1408201 -0.06915301 -0.27621472  1.00000000   -0.16626100
## mean_wentropy -0.2340302 -0.36418921 -0.05835954 -0.16626100    1.00000000
## mean_entropy  -0.0435414 -0.33794978  0.12736986 -0.48980349    0.43857187
## mean_er       -0.4451445 -0.20574042 -0.55858523  0.38774922   -0.57420085
##               mean_entropy    mean_er
## rwsd            -0.0435414 -0.4451445
## rbsd            -0.3379498 -0.2057404
## rwrmssd          0.1273699 -0.5585852
## mean_bc         -0.4898035  0.3877492
## mean_wentropy    0.4385719 -0.5742008
## mean_entropy     1.0000000 -0.3772359
## mean_er         -0.3772359  1.0000000

Relative Within SD

# all ER strategies
df_rwsd_outcomes <- df_rwsd %>% 
  left_join(df_indiff_2) %>% 
  left_join(
    df_mean_er_strat %>%
      group_by(ID) %>%
      summarize(mean_ID_er = mean(mean_er, na.rm = TRUE)) %>%
      select(ID, mean_ID_er) %>% 
      ungroup() 
    )

## Joining with `by = join_by(ID)`
## left_join: added 13 columns (Sex, Birth_year, Birth_month, Ethnicity, SCARED,
## …)
## > rows only in x 0
## > rows only in df_indiff_2 ( 836)
## > matched rows 17,573 (includes duplicates)
## > ========
## > rows total 17,573
## group_by: one grouping variable (ID)
## summarize: now 265 rows and 2 columns, ungrouped
## select: no changes
## ungroup: no grouping variables remain
## Joining with `by = join_by(ID)`
## left_join: added one column (mean_ID_er)
## > rows only in x 0
## > rows only in df_mean_er_strat %>% gr.. ( 12)
## > matched rows 17,573
## > ========
## > rows total 17,573

Neg Affect

fit <- df_rwsd_outcomes %>% 
  lmer(panas_neg_z ~ rwsd + (1|ID), data = .)
tab_model(fit)

	panas_neg_z
Predictors	Estimates	CI	p
(Intercept)	0.95	0.61 – 1.30	<0.001
rwsd	-1.24	-1.69 – -0.79	<0.001
Random Effects
σ²	0.45
τ₀₀ _ID	0.51
ICC	0.53
N _ID	253
Observations	12099
Marginal R² / Conditional R²	0.060 / 0.561

# controlling for mean ER
df_rwsd_outcomes %>% 
  lmer(panas_neg_z ~ rwsd + mean_ID_er + (1|ID), data = .) %>% 
  tab_model()

	panas_neg_z
Predictors	Estimates	CI	p
(Intercept)	0.05	-0.44 – 0.55	0.828
rwsd	-0.63	-1.13 – -0.13	0.013
mean ID er	0.18	0.11 – 0.25	<0.001
Random Effects
σ²	0.45
τ₀₀ _ID	0.47
ICC	0.51
N _ID	253
Observations	12099
Marginal R² / Conditional R²	0.106 / 0.563

relative within SD predicts less neg affect

Pos Affect

fit <- df_rwsd_outcomes %>% 
  lmer(panas_pos_z ~ rwsd + (1|ID), data = .)
tab_model(fit)

	panas_pos_z
Predictors	Estimates	CI	p
(Intercept)	-0.73	-1.08 – -0.39	<0.001
rwsd	0.95	0.49 – 1.40	<0.001
Random Effects
σ²	0.47
τ₀₀ _ID	0.51
ICC	0.52
N _ID	253
Observations	12096
Marginal R² / Conditional R²	0.035 / 0.538

# controlling for mean ER
df_rwsd_outcomes %>% 
  lmer(panas_pos_z ~ rwsd + mean_ID_er + (1|ID), data = .) %>% 
  tab_model()

	panas_pos_z
Predictors	Estimates	CI	p
(Intercept)	-0.80	-1.32 – -0.29	0.002
rwsd	0.99	0.47 – 1.51	<0.001
mean ID er	0.01	-0.06 – 0.09	0.717
Random Effects
σ²	0.47
τ₀₀ _ID	0.51
ICC	0.52
N _ID	253
Observations	12096
Marginal R² / Conditional R²	0.035 / 0.538

relative within SD predicts more pos affect

Relative Between SD

df_rbsd_outcomes <- df_rbsd_st %>% 
  left_join(df_indiff_2) %>% 
  left_join(
    df_mean_er_strat %>%
      group_by(ID) %>%
      summarize(mean_ID_er = mean(mean_er, na.rm = TRUE)) %>%
      select(ID, mean_ID_er) %>% 
      ungroup() 
    ) %>% 
  mutate(across(rbsd, ~scale(., scale = F), .names = "{.col}_pc")) %>% 
  mutate(across(rbsd_pc, ~scale(., scale = T), .names = "{.col}_z"))

## Joining with `by = join_by(ID, timeStampSent)`
## left_join: added 12 columns (Sex, Birth_year, Birth_month, Ethnicity, SCARED,
## …)
## > rows only in x 0
## > rows only in df_indiff_2 (11,761)
## > matched rows 6,648
## > ========
## > rows total 6,648
## group_by: one grouping variable (ID)
## summarize: now 265 rows and 2 columns, ungrouped
## select: no changes
## ungroup: no grouping variables remain
## Joining with `by = join_by(ID)`
## left_join: added one column (mean_ID_er)
## > rows only in x 0
## > rows only in df_mean_er_strat %>% gr.. ( 11)
## > matched rows 6,648
## > =======
## > rows total 6,648
## mutate (grouped): new variable 'rbsd_pc' (double) with 3,128 unique values and
## 0% NA
## mutate (grouped): new variable 'rbsd_pc_z' (double) with 3,107 unique values
## and 1% NA

Neg Affect

fit <- df_rbsd_outcomes %>% 
  lmer(panas_neg_z ~ rbsd_pc_z + (1|ID), data = .)
tab_model(fit)

	panas_neg_z
Predictors	Estimates	CI	p
(Intercept)	0.33	0.23 – 0.44	<0.001
rbsd pc z	-0.04	-0.06 – -0.03	<0.001
Random Effects
σ²	0.54
τ₀₀ _ID	0.60
ICC	0.53
N _ID	231
Observations	6612
Marginal R² / Conditional R²	0.002 / 0.528

# controlling for mean ER
df_rbsd_outcomes %>% 
  lmer(panas_neg_z ~ rbsd_pc_z + mean_ID_er + (1|ID), data = .) %>% 
  tab_model()

	panas_neg_z
Predictors	Estimates	CI	p
(Intercept)	0.18	-0.05 – 0.42	0.130
rbsd pc z	-0.04	-0.06 – -0.03	<0.001
mean ID er	0.05	-0.02 – 0.13	0.167
Random Effects
σ²	0.54
τ₀₀ _ID	0.60
ICC	0.53
N _ID	231
Observations	6612
Marginal R² / Conditional R²	0.005 / 0.529

relative between SD predicts less neg affect (but effect size is really small)

Pos Affect

fit <- df_rbsd_outcomes %>% 
  lmer(panas_pos_z ~ rbsd_pc_z + (1|ID), data = .)
tab_model(fit)

	panas_pos_z
Predictors	Estimates	CI	p
(Intercept)	-0.18	-0.27 – -0.09	<0.001
rbsd pc z	0.04	0.02 – 0.05	<0.001
Random Effects
σ²	0.49
τ₀₀ _ID	0.48
ICC	0.49
N _ID	231
Observations	6610
Marginal R² / Conditional R²	0.001 / 0.495

# controlling for mean ER
df_rbsd_outcomes %>% 
  lmer(panas_pos_z ~ rbsd_pc_z + mean_ID_er + (1|ID), data = .) %>% 
  tab_model()

	panas_pos_z
Predictors	Estimates	CI	p
(Intercept)	-0.28	-0.49 – -0.06	0.011
rbsd pc z	0.04	0.02 – 0.05	<0.001
mean ID er	0.04	-0.03 – 0.10	0.322
Random Effects
σ²	0.49
τ₀₀ _ID	0.48
ICC	0.49
N _ID	231
Observations	6610
Marginal R² / Conditional R²	0.003 / 0.495

relative between SD predicts more pos affect (but effect size is really small)

Successive Difference

df_rwrmssd_outcomes <- df_rwrmssd %>% 
  left_join(df_indiff_2) %>% 
  left_join(
    df_mean_er_strat %>%
      group_by(ID) %>%
      summarize(mean_ID_er = mean(mean_er, na.rm = TRUE)) %>%
      select(ID, mean_ID_er) %>% 
      ungroup() 
    )

## Joining with `by = join_by(ID)`
## left_join: added 13 columns (Sex, Birth_year, Birth_month, Ethnicity, SCARED,
## …)
## > rows only in x 0
## > rows only in df_indiff_2 ( 836)
## > matched rows 17,573 (includes duplicates)
## > ========
## > rows total 17,573
## group_by: one grouping variable (ID)
## summarize: now 265 rows and 2 columns, ungrouped
## select: no changes
## ungroup: no grouping variables remain
## Joining with `by = join_by(ID)`
## left_join: added one column (mean_ID_er)
## > rows only in x 0
## > rows only in df_mean_er_strat %>% gr.. ( 12)
## > matched rows 17,573
## > ========
## > rows total 17,573

Neg Affect

fit <- df_rwrmssd_outcomes %>% 
  lmer(panas_neg_z ~ rwrmssd + (1|ID), data = .)
tab_model(fit)

	panas_neg_z
Predictors	Estimates	CI	p
(Intercept)	0.90	0.64 – 1.17	<0.001
rwrmssd	-1.42	-1.83 – -1.02	<0.001
Random Effects
σ²	0.45
τ₀₀ _ID	0.48
ICC	0.52
N _ID	253
Observations	12099
Marginal R² / Conditional R²	0.092 / 0.562

# controlling for mean ER
df_rwrmssd_outcomes %>% 
  lmer(panas_neg_z ~ rwrmssd + mean_ID_er + (1|ID), data = .) %>% 
  tab_model()

	panas_neg_z
Predictors	Estimates	CI	p
(Intercept)	0.21	-0.26 – 0.68	0.378
rwrmssd	-0.86	-1.37 – -0.35	0.001
mean ID er	0.14	0.06 – 0.22	0.001
Random Effects
σ²	0.45
τ₀₀ _ID	0.46
ICC	0.51
N _ID	253
Observations	12099
Marginal R² / Conditional R²	0.116 / 0.564

successive differences predicts less neg affect

Pos Affect

fit <- df_rwrmssd_outcomes %>% 
  lmer(panas_pos_z ~ rwrmssd + (1|ID), data = .)
tab_model(fit)

	panas_pos_z
Predictors	Estimates	CI	p
(Intercept)	-0.58	-0.85 – -0.31	<0.001
rwrmssd	0.89	0.47 – 1.31	<0.001
Random Effects
σ²	0.47
τ₀₀ _ID	0.51
ICC	0.52
N _ID	253
Observations	12096
Marginal R² / Conditional R²	0.036 / 0.538

# controlling for mean ER
df_rwrmssd_outcomes %>% 
  lmer(panas_pos_z ~ rwrmssd + mean_ID_er + (1|ID), data = .) %>% 
  tab_model()

	panas_pos_z
Predictors	Estimates	CI	p
(Intercept)	-0.81	-1.31 – -0.32	0.001
rwrmssd	1.08	0.54 – 1.62	<0.001
mean ID er	0.05	-0.04 – 0.13	0.268
Random Effects
σ²	0.47
τ₀₀ _ID	0.51
ICC	0.52
N _ID	253
Observations	12096
Marginal R² / Conditional R²	0.038 / 0.539

successive differences predicts more pos affect (large effect size)

Bray-Curtis

df_bc_outcomes <- df_bc_st %>% 
  left_join(df_indiff_2) %>% 
  left_join(
    df_mean_er_strat %>%
      group_by(ID) %>%
      summarize(mean_ID_er = mean(mean_er, na.rm = TRUE)) %>%
      select(ID, mean_ID_er) %>% 
      ungroup() 
    ) %>% 
  mutate(across(bc, ~scale(., scale = F), .names = "{.col}_pc")) %>% 
  mutate(across(bc_pc, ~scale(., scale = T), .names = "{.col}_z"))

## Joining with `by = join_by(ID, timeStampSent)`
## left_join: added 12 columns (Sex, Birth_year, Birth_month, Ethnicity, SCARED,
## …)
## > rows only in x 0
## > rows only in df_indiff_2 (8,410)
## > matched rows 9,999
## > =======
## > rows total 9,999
## group_by: one grouping variable (ID)
## summarize: now 265 rows and 2 columns, ungrouped
## select: no changes
## ungroup: no grouping variables remain
## Joining with `by = join_by(ID)`
## left_join: added one column (mean_ID_er)
## > rows only in x 0
## > rows only in df_mean_er_strat %>% gr.. ( 8)
## > matched rows 9,999
## > =======
## > rows total 9,999
## mutate (grouped): new variable 'bc_pc' (double) with 3,901 unique values and 0%
## NA
## mutate (grouped): new variable 'bc_pc_z' (double) with 3,901 unique values and
## 4% NA

Neg Affect

fit <- df_bc_outcomes %>% 
  lmer(panas_neg_z ~ bc_pc_z + (1|ID), data = .)
tab_model(fit)

	panas_neg_z
Predictors	Estimates	CI	p
(Intercept)	0.04	-0.06 – 0.14	0.421
bc pc z	0.10	0.09 – 0.11	<0.001
Random Effects
σ²	0.44
τ₀₀ _ID	0.59
ICC	0.58
N _ID	244
Observations	9578
Marginal R² / Conditional R²	0.009 / 0.580

# controlling for mean ER
df_bc_outcomes %>% 
  lmer(panas_neg_z ~ bc_pc_z + mean_ID_er + (1|ID), data = .) %>% 
  tab_model()

	panas_neg_z
Predictors	Estimates	CI	p
(Intercept)	-0.51	-0.70 – -0.32	<0.001
bc pc z	0.10	0.09 – 0.11	<0.001
mean ID er	0.22	0.15 – 0.28	<0.001
Random Effects
σ²	0.44
τ₀₀ _ID	0.50
ICC	0.54
N _ID	244
Observations	9578
Marginal R² / Conditional R²	0.101 / 0.583

BC predicts MORE neg affect (effect size does not change when controlling for mean ER; small effect size)

Pos Affect

fit <- df_bc_outcomes %>% 
  lmer(panas_pos_z ~ bc_pc_z + (1|ID), data = .)
tab_model(fit)

	panas_pos_z
Predictors	Estimates	CI	p
(Intercept)	-0.04	-0.14 – 0.05	0.381
bc pc z	-0.05	-0.06 – -0.04	<0.001
Random Effects
σ²	0.46
τ₀₀ _ID	0.56
ICC	0.55
N _ID	244
Observations	9575
Marginal R² / Conditional R²	0.002 / 0.547

# controlling for mean ER
df_bc_outcomes %>% 
  lmer(panas_pos_z ~ bc_pc_z + mean_ID_er + (1|ID), data = .) %>% 
  tab_model()

	panas_pos_z
Predictors	Estimates	CI	p
(Intercept)	0.09	-0.11 – 0.29	0.382
bc pc z	-0.05	-0.06 – -0.04	<0.001
mean ID er	-0.05	-0.12 – 0.02	0.141
Random Effects
σ²	0.46
τ₀₀ _ID	0.55
ICC	0.54
N _ID	244
Observations	9575
Marginal R² / Conditional R²	0.008 / 0.548

BC predicts LESS pos affect (effect size does not change when controlling for mean ER; small effect size)

Wentropy (Emotion Regulation Diversity)

df_wentropy_outcomes <- df_wentropy_st %>% 
  left_join(df_indiff_2) %>% 
  left_join(
    df_mean_er_strat %>%
      group_by(ID) %>%
      summarize(mean_ID_er = mean(mean_er, na.rm = TRUE)) %>%
      select(ID, mean_ID_er) %>% 
      ungroup() 
    ) %>% 
  mutate(across(wentropy, ~scale(., scale = F), .names = "{.col}_pc")) %>% 
  mutate(across(wentropy_pc, ~scale(., scale = T), .names = "{.col}_z"))

## Joining with `by = join_by(ID, timeStampSent)`
## left_join: added 12 columns (Sex, Birth_year, Birth_month, Ethnicity, SCARED,
## …)
## > rows only in x 0
## > rows only in df_indiff_2 ( 5,819)
## > matched rows 12,590
## > ========
## > rows total 12,590
## group_by: one grouping variable (ID)
## summarize: now 265 rows and 2 columns, ungrouped
## select: no changes
## ungroup: no grouping variables remain
## Joining with `by = join_by(ID)`
## left_join: added one column (mean_ID_er)
## > rows only in x 0
## > rows only in df_mean_er_strat %>% gr.. ( 0)
## > matched rows 12,590
## > ========
## > rows total 12,590
## mutate: new variable 'wentropy_pc' (double) with 436 unique values and 0% NA
## mutate: new variable 'wentropy_pc_z' (double) with 436 unique values and 0% NA

Neg Affect

fit <- df_wentropy_outcomes %>% 
  lmer(panas_neg_z ~ wentropy_pc_z + (1|ID), data = .)
tab_model(fit)

	panas_neg_z
Predictors	Estimates	CI	p
(Intercept)	0.02	-0.07 – 0.11	0.724
wentropy pc z	-0.05	-0.07 – -0.03	<0.001
Random Effects
σ²	0.43
τ₀₀ _ID	0.56
ICC	0.56
N _ID	265
Observations	12590
Marginal R² / Conditional R²	0.003 / 0.566

# controlling for mean ER
df_wentropy_outcomes %>% 
  lmer(panas_neg_z ~ wentropy_pc_z + mean_ID_er + (1|ID), data = .) %>% 
  tab_model()

	panas_neg_z
Predictors	Estimates	CI	p
(Intercept)	-0.52	-0.69 – -0.35	<0.001
wentropy pc z	-0.05	-0.06 – -0.03	<0.001
mean ID er	0.22	0.16 – 0.28	<0.001
Random Effects
σ²	0.43
τ₀₀ _ID	0.47
ICC	0.52
N _ID	265
Observations	12590
Marginal R² / Conditional R²	0.107 / 0.574

Wentropy predicts less neg affect (but small effect size)

Pos Affect

fit <- df_wentropy_outcomes %>% 
  lmer(panas_pos_z ~ wentropy_pc_z + (1|ID), data = .)
tab_model(fit)

	panas_pos_z
Predictors	Estimates	CI	p
(Intercept)	-0.00	-0.09 – 0.09	0.929
wentropy pc z	-0.04	-0.06 – -0.02	<0.001
Random Effects
σ²	0.46
τ₀₀ _ID	0.54
ICC	0.54
N _ID	265
Observations	12587
Marginal R² / Conditional R²	0.002 / 0.542

# controlling for mean ER
df_wentropy_outcomes %>% 
  lmer(panas_pos_z ~ wentropy_pc_z + mean_ID_er + (1|ID), data = .) %>% 
  tab_model()

	panas_pos_z
Predictors	Estimates	CI	p
(Intercept)	0.23	0.04 – 0.41	0.016
wentropy pc z	-0.04	-0.06 – -0.02	<0.001
mean ID er	-0.09	-0.16 – -0.03	0.005
Random Effects
σ²	0.46
τ₀₀ _ID	0.53
ICC	0.53
N _ID	265
Observations	12587
Marginal R² / Conditional R²	0.013 / 0.541

Wentropy predicts LESS pos affecr (but small effect size)

Entropy

df_entropy_outcomes <- df_entropy_st %>% 
  left_join(df_indiff_2) %>% 
  left_join(
    df_mean_er_strat %>%
      group_by(ID) %>%
      summarize(mean_ID_er = mean(mean_er, na.rm = TRUE)) %>%
      select(ID, mean_ID_er) %>% 
      ungroup() 
    ) %>% 
  mutate(across(entropy, ~scale(., scale = F), .names = "{.col}_pc")) %>% 
  mutate(across(entropy_pc, ~scale(., scale = T), .names = "{.col}_z"))

## Joining with `by = join_by(ID, timeStampSent)`
## left_join: added 12 columns (Sex, Birth_year, Birth_month, Ethnicity, SCARED,
## …)
## > rows only in x 0
## > rows only in df_indiff_2 ( 5,819)
## > matched rows 12,590
## > ========
## > rows total 12,590
## group_by: one grouping variable (ID)
## summarize: now 265 rows and 2 columns, ungrouped
## select: no changes
## ungroup: no grouping variables remain
## Joining with `by = join_by(ID)`
## left_join: added one column (mean_ID_er)
## > rows only in x 0
## > rows only in df_mean_er_strat %>% gr.. ( 0)
## > matched rows 12,590
## > ========
## > rows total 12,590
## mutate: new variable 'entropy_pc' (double) with 416 unique values and 0% NA
## mutate: new variable 'entropy_pc_z' (double) with 416 unique values and 0% NA

Neg Affect

fit <- df_entropy_outcomes %>% 
  lmer(panas_neg_z ~ entropy_pc_z + (1|ID), data = .)
tab_model(fit)

	panas_neg_z
Predictors	Estimates	CI	p
(Intercept)	0.01	-0.08 – 0.10	0.748
entropy pc z	-0.09	-0.11 – -0.07	<0.001
Random Effects
σ²	0.43
τ₀₀ _ID	0.54
ICC	0.56
N _ID	265
Observations	12590
Marginal R² / Conditional R²	0.008 / 0.562

# controlling for mean ER
df_entropy_outcomes %>% 
  lmer(panas_neg_z ~ entropy_pc_z + mean_ID_er + (1|ID), data = .) %>% 
  tab_model()

	panas_neg_z
Predictors	Estimates	CI	p
(Intercept)	-0.51	-0.68 – -0.34	<0.001
entropy pc z	-0.09	-0.10 – -0.07	<0.001
mean ID er	0.21	0.15 – 0.27	<0.001
Random Effects
σ²	0.43
τ₀₀ _ID	0.46
ICC	0.52
N _ID	265
Observations	12590
Marginal R² / Conditional R²	0.114 / 0.571

Entropy predicts less neg affect (but small effect size)

Pos Affect

fit <- df_entropy_outcomes %>% 
  lmer(panas_pos_z ~ entropy_pc_z + (1|ID), data = .)
tab_model(fit)

	panas_pos_z
Predictors	Estimates	CI	p
(Intercept)	-0.00	-0.09 – 0.08	0.918
entropy pc z	0.05	0.03 – 0.06	<0.001
Random Effects
σ²	0.46
τ₀₀ _ID	0.54
ICC	0.54
N _ID	265
Observations	12587
Marginal R² / Conditional R²	0.002 / 0.541

# controlling for mean ER
df_entropy_outcomes %>% 
  lmer(panas_pos_z ~ entropy_pc_z + mean_ID_er + (1|ID), data = .) %>% 
  tab_model()

	panas_pos_z
Predictors	Estimates	CI	p
(Intercept)	0.17	-0.02 – 0.35	0.077
entropy pc z	0.05	0.03 – 0.06	<0.001
mean ID er	-0.07	-0.13 – -0.00	0.037
Random Effects
σ²	0.46
τ₀₀ _ID	0.53
ICC	0.54
N _ID	265
Observations	12587
Marginal R² / Conditional R²	0.014 / 0.543

Entropy predicts more pos affect (but small effect size)

wave2_analysis

Angela

2025-09-08

Initialization

Preprocessing

Read data

Process data

Calculate Metrics

Mean ER

Relative Within SD

Relative Between SD

Successive Difference

Bray-Curtis

Wentropy (Emotion Regulation Diversity)

Entropy

Trait (between-person analysis)

Cor Between Metrics

Relative Within SD

Depression

Anxiety

Neg Affect

Pos Affect

Relative Between SD

Depression

Anxiety

Neg Affect

Pos Affect

Successive Difference

Depression

Anxiety

Neg Affect

Pos Affect

Bray-Curtis

Depression

Anxiety

Neg Affect

Pos Affect

Wentropy (Emotion Regulation Diversity)

Depression

Anxiety

Neg Affect

Pos Affect

Entropy

Depression

Anxiety

Neg Affect

Pos Affect

Momentary (within-person analysis)

Cor Between Metrics

Relative Within SD

Neg Affect

Pos Affect

Relative Between SD

Neg Affect

Pos Affect

Successive Difference

Neg Affect

Pos Affect

Bray-Curtis

Neg Affect

Pos Affect

Wentropy (Emotion Regulation Diversity)

Neg Affect

Pos Affect

Entropy

Neg Affect

Pos Affect