Meta-analysis: Personality traits and emotional variability

I. Meta-Data

To run this notebook, the following directories and files need to exist:

• Data folder:
  • final_dat.xlsx: meta-analytic dataset
  • metaSEM.RData: Big Five correlation matrices for meta-analytic SEM
  • Rayyan_ProQuest.csv, Rayyan_PsyArXiv.csv, Rayyan_published.csv, Rayyan_rerun.csv: abstract screening decisions
• Scripts folder:
  • 01_helpers.R: helper functions
  • 02_analysis.Rmd: this current RMarkdown file

Descriptive information on all unique samples:

# reads in data
dat <- rio::import("../Data/final_dat.xlsx", na = c("NA")) %>%
  select(-c(Search, Authors, Title, Year, 
            Personality_Scale, Emotion_Scale,
            Notes)) %>%
  mutate(Location = if_else(grepl("US", Location), "US", Location),
         ID = case_when(
           !is.na(Study_Code) ~ paste0(Paper_Code, "_", Study_Code),
           TRUE ~ Paper_Code)) %>%
  select(ID, everything())

# removes 2 duplicated datasets
dat <- dat %>%
  filter(!ID %in% c("Kuppens_2010_study1", "Koval_2013"))

# reads in abstract screening
rayyan_rerun <- rio::import("../Data/Rayyan_rerun.csv") %>% select(notes)
rayyan_pub <- rio::import("../Data/Rayyan_published.csv") %>% select(notes)
rayyan_pq <- rio::import("../Data/Rayyan_ProQuest.csv") %>% select(notes)
rayyan_arx <- rio::import("../Data/Rayyan_PsyArXiv.csv") %>% select(notes)
rayyan <- rbind(rayyan_rerun, rayyan_pub, rayyan_pq, rayyan_arx) %>% na.omit()

# clean up sample characteristics
dat <- dat %>%
  mutate_at(vars(Age_Mean, Age_SD, White_Perc, Female_Perc),
            ~round(., 2))

# show all unique samples info
dat %>%
  select(ID:Time_No) %>%
  unique() %>%
  kable() %>% kable_styling() %>%
  scroll_box(width="800px",
             height="500px")

	ID	Paper_Code	Study_Code	Study_Quality	Study_Quality_Reli	Sample_Size	Age_Mean	Age_SD	Female_Perc	White_Perc	Location	Duration	Frequency	Time_No
1	Alshamsi_2015	Alshamsi_2015	NA	good	fair	52	36.00	NA	9.00	NA	Italy	30 days	3 times a day	90
21	Armeli_2007	Armeli_2007	NA	good	good	98	43.50	8.69	50.00	95.00	NA	21 days	3 times a day	63
23	Aurora_2021	Aurora_2021	NA	good	good	195	20.05	2.38	79.00	81.00	US	10 days	5 times a day	50
28	Balducci_2020	Balducci_2020	NA	fair	good	213	45.50	10.40	42.50	NA	Italy	10 days	1 time a day	10
30	Bartley_2011	Bartley_2011	NA	good	good	366	20.14	2.10	68.50	37.60	US	5 days	1 time a day	5
40	Bringmann_2016_study1	Bringmann_2016	study1	fair	good	95	19.00	1.00	62.00	NA	Belgium	7 days	10 times a day	70
50	Bringmann_2016_study2	Bringmann_2016	study2	fair	good	79	24.00	8.00	63.00	NA	Belgium	14 days	10 times a day	140
60	Buhler_2020_wave2	Buhler_2020	wave2	good	good	2626	32.81	13.85	NA	NA	Austria, Germany, Switzerland	14 days	1 time a day	14
70	Burns_2015	Burns_2015	NA	good	good	45	28.09	10.80	68.00	NA	Australia	14 days	1 time a day	14
74	Clegg_2021	Clegg_2021	NA	good	good	267	20.26	0.70	67.00	60.00	Canada	20 days	1 time a day	20
84	Conner_2012_condition1	Conner_2012	condition1	good	good	54	19.90	2.40	59.00	81.00	New Zealand	13 days	1 time a day	13
89	Conner_2012_condition2	Conner_2012	condition2	good	good	54	19.90	2.40	59.00	81.00	New Zealand	13 days	3 times a day	39
94	Conner_2012_condition3	Conner_2012	condition3	good	good	54	19.90	2.40	59.00	81.00	New Zealand	13 days	6 times a day	78
99	Conner_2015	Conner_2015	NA	good	good	658	19.80	1.70	70.21	79.20	New Zealand	13 days	1 time a day	13
109	DMello_2021	DMello_2021	NA	good	good	598	34.36	9.39	42.00	NA	US	56 days	1 time a day	56
119	Dauvier_2019	Dauvier_2019	NA	fair	fair	191	38.50	17.40	64.00	NA	France	14 days	5 times a day	70
123	Denissen_2008	Denissen_2008	NA	good	good	1233	27.67	9.77	88.60	NA	Germany	30 days	1 time a day	30
133	Dunkley_2014_wave1	Dunkley_2014	wave1	good	good	196	40.94	12.25	66.33	78.00	Canada	14 days	1 time a day	14
148	Eid_1999	Eid_1999	NA	fair	good	180	NA	NA	55.00	NA	US	51 days	1 time a day	51
183	Galla_2015	Galla_2015	NA	good	fair	129	14.70	0.35	59.00	85.00	US	14 days	1 time a day	14
185	Gartland_2014	Gartland_2014	NA	good	fair	103	35.00	NA	70.87	90.30	United Kingdom	14 days	1 time a day	14
195	Geukes_2017_wave1	Geukes_2017	wave1	fair	fair	131	21.01	3.65	81.00	NA	Germany	21 days	event-contingent	112
205	Hisler_2020_sample1	Hisler_2020	sample1	good	good	211	18.78	1.05	58.00	NA	US	30 days	1 time a day	30
207	Howland_2017_wave1	Howland_2017	wave1	good	good	575	18.76	1.09	52.00	86.00	US	30 days	1 time a day	30
208	Jimenez_2022_study2	Jimenez_2022	study2	good	fair	279	18.94	1.21	57.20	59.50	US	7 days	5 times a day	35
218	Jones_2022	Jones_2022	NA	good	good	121	42.10	12.01	60.00	89.00	US	7 days	8 times a day	56
238	Kalokerinos_2017	Kalokerinos_2017	NA	fair	fair	114	35.23	11.87	50.00	NA	NA	7 days	1 time a day	7
248	Komulainen_2014	Komulainen_2014	NA	good	good	104	23.00	3.69	82.69	NA	Finland	7 days	10 times a day	70
258	Kritzler_2020	Kritzler_2020	NA	good	fair	206	25.17	8.03	79.61	NA	Germany	7 days	5 times a day	35
268	Kroencke_2020	Kroencke_2020	NA	good	good	1609	33.70	12.70	78.00	NA	Germany	14 days	6 times a day	84
273	Kuijpers_2022_study1	Kuijpers_2022	study1	fair	good	92	30.00	11.84	62.00	NA	Belgium	20 days	5 times a day	100
274	Kuijpers_2022_study2	Kuijpers_2022	study2	good	good	80	32.00	12.50	57.00	NA	Belgium	10 days	3 times a day	30
275	Kukk_2022	Kukk_2022	NA	good	good	96	21.50	6.70	100.00	100.00	Estonia	3 days	7 times a day	21
285	Kuppens_2007_study1	Kuppens_2007	study1	good	good	58	22.00	NA	68.97	NA	Belgium	7 days	9 times a day	63
290	Laferton_2020	Laferton_2020	NA	good	good	98	25.46	6.65	83.70	NA	Austria, Germany, Switzerland	10 days	1 time a day	10
292	Leki_2017_study3	Leki_2017	study3	good	good	52	19.54	NA	65.38	NA	US	21 days	1 time a day	21
298	Mey_2020	Mey_2020	NA	good	good	70	23.93	3.15	59.00	NA	Germany	28 days	5 times a day	140
299	Mill_2016	Mill_2016	NA	fair	fair	110	44.75	3.25	63.64	NA	Estonia	14 days	7 times a day	98
314	Ottenstein_2020	Ottenstein_2020	NA	good	good	72	22.85	2.40	68.00	NA	Germany	21 days	3 times a day	63
320	Pavani_2017	Pavani_2017	NA	good	good	78	44.55	18.01	62.00	NA	France	14 days	5 times a day	70
324	Pelt_2020	Pelt_2020	NA	good	fair	1223	29.47	10.49	86.26	NA	Germany	30 days	1 time a day	30
334	Roberts_1997	Roberts_1997	NA	good	good	92	18.70	1.30	100.00	NA	US	7 days	1 time a day	7
338	Slavish_2018	Slavish_2018	NA	good	good	242	46.80	10.90	66.50	27.30	US	14 days	2 times a day	28
348	Steffens_2017	Steffens_2017	NA	fair	good	32	28.80	5.60	53.13	NA	Canada	7 days	11 times a day	77
353	Sun_2017_study3	Sun_2017	study3	good	good	62	21.40	3.55	62.90	NA	Australia	7 days	6 times a day	42
354	Watson_1992_study2	Watson_1992	study2	fair	fair	127	NA	NA	NA	NA	US	45 days	1 time a day	45
364	Watson_unpub_study1	Watson_unpub	study1	NA	NA	366	NA	NA	71.31	NA	US	55 days	1 time a day	55
374	Watson_unpub_study2	Watson_unpub	study2	NA	NA	295	NA	NA	69.49	NA	US	55 days	1 time a day	55
384	Weltz_2016	Weltz_2016	NA	good	good	1634	19.23	1.41	53.70	79.60	US	30 day	1 time a day	30
385	Wenzel_2015_study2	Wenzel_2015	study2	fair	good	108	25.20	6.60	80.56	NA	Germany	6 days	1 time a day	6
390	Willroth_2020_sampleUSCA	Willroth_2020	sampleUSCA	good	good	130	47.00	17.00	97.48	56.00	US	16 days	1 time a day	16
392	Willroth_2020_sampleUSUG	Willroth_2020	sampleUSUG	good	fair	184	19.00	2.00	72.11	26.00	US	21 days	1 time a day	21
394	Wilson_2017_sample1	Wilson_2017	sample1	good	fair	124	20.10	2.30	67.00	43.50	US	6 days	6 times a day	36
404	Wilson_2017_sample2	Wilson_2017	sample2	good	fair	415	19.30	2.00	68.00	50.00	US	14 days	4 times a day	56
412	Wilt_2017_sample1	Wilt_2017	sample1	good	good	40	23.50	5.61	72.50	NA	US	14 days	6 times a day	84
416	Wilt_2017_sample2	Wilt_2017	sample2	good	good	40	20.60	2.22	82.50	NA	US	14 days	6 times a day	84
420	Wilt_2012_sample1	Wilt_2012	sample1	fair	good	44	NA	NA	NA	NA	US	13 days	5 times a day	65
421	Wilt_2012_sample2	Wilt_2012	sample2	fair	good	62	27.90	NA	NA	NA	US	10 days	5 times a day	50
422	Wilt_2012_sample3	Wilt_2012	sample3	fair	good	48	NA	NA	NA	NA	US	10 weeks	1 time a week	10
423	Wilt_2012_sample4	Wilt_2012	sample4	fair	good	97	NA	NA	NA	NA	US	10 weeks	1 time a week	10
424	Wilt_2019	Wilt_2019	NA	good	good	78	26.60	7.90	80.77	66.67	US	7 days	4 times a day	28
434	Windsor_2021	Windsor_2021	NA	good	good	73	88.71	2.99	67.00	NA	Australia	7 days	5 times a day	35
436	Zhang_2019	Zhang_2019	NA	good	good	139	19.50	1.07	73.00	NA	China	14 days	1 time a day	14
446	Long_2003	Long_2003	NA	good	good	163	20.00	NA	80.98	NA	US	7 days	1 time a day	7
456	Mackinnon_2021_sample1	Mackinnon_2021	sample1	good	good	263	21.37	1.89	79.80	78.30	Canada	20 days	1 time a day	20
466	Ryvkina_2023_S1W1	Ryvkina_2023	S1W1	good	good	313	23.00	6.80	78.20	NA	Germany	14 days	6 times a day	84
468	Ryvkina_2023_S2W2	Ryvkina_2023	S2W2	good	good	914	41.00	12.40	80.60	NA	Germany	14 days	6 times a day	84
478	Shui_2023	Shui_2023	NA	good	good	80	19.10	NA	NA	NA	China	14 days	1 time a day	14
488	Smith-Pickering_2022	Smith-Pickering_2022	NA	good	good	290	31.85	1.98	60.00	51.20	NA	13 days	1 time a day	13
498	Smith-DeNunzio_2022	Smith-DeNunzio_2022	NA	good	good	114	36.20	9.30	48.00	86.00	NA	5 days	1 time a day	5
501	Sweeny_2020_study1	Sweeny_2020	study1	good	fair	120	NA	NA	68.00	17.00	US	4 days	1 time a day	4
505	Sweeny_2020_study2	Sweeny_2020	study2	good	fair	203	NA	NA	61.00	67.00	US	4 months	2 times a month	8
509	Sweeny_unpub	Sweeny_unpub	NA	NA	NA	102	27.61	7.16	53.00	63.00	NA	8 days	1 time a day	8
519	Kalokerinos_2019_study2	Kalokerinos_2019	study2	NA	NA	101	18.64	1.45	NA	NA	Belgium	9 days	10 times a day	90
529	Grommisch_2020	Grommisch_2020	NA	NA	NA	179	27.02	8.98	65.00	NA	Australia	21 days	9 times a day	189
539	Erbas_2018_wave1	Erbas_2018	wave1	NA	NA	200	18.32	0.97	55.00	NA	Belgium	7 days	10 times a day	70
549	Haines_2016	Haines_2016	NA	NA	NA	78	23.26	3.54	61.00	NA	Australia	4 days	10 times a day	40
559	Sels_2017	Sels_2017	NA	NA	NA	100	27.75	10.60	50.00	NA	Belgium	7 days	10 times a day	70
569	Brans_2013_study2	Brans_2013	study2	NA	NA	95	19.06	1.28	62.11	NA	Belgium	7 days	10 times a day	70
579	Pasyugina_2015	Pasyugina_2015	NA	NA	NA	101	21.40	2.15	73.70	NA	Belgium	9 days	10 times a day	90
589	Medland_2020	Medland_2020	NA	NA	NA	132	21.14	3.51	66.70	18.20	Australia	7 days	8 times a day	56
594	Kuppens_2010_study2	Kuppens_2010	study2	NA	NA	60	23.00	NA	66.67	NA	Belgium	4 days	50 times a day	200
599	Koval_2019_study1	Koval_2019	study1	NA	NA	81	22.33	5.47	100.00	46.90	Australia	7 days	10 times a day	70
609	Koval_2019_study2	Koval_2019	study2	NA	NA	87	23.52	4.11	100.00	32.20	Australia	7 days	10 times a day	70
619	Koval_2019_study3	Koval_2019	study3	NA	NA	100	26.46	6.12	100.00	65.00	US	7 days	10 times a day	70
629	Dejonckheere_2019	Dejonckheere_2019	NA	NA	NA	100	24.12	6.87	77.00	NA	Belgium	14 days	7 times a day	98
639	Kalokerinos_unpub	Kalokerinos_unpub	NA	NA	NA	NA	NA	NA	NA	NA	United Kingdom	7 days	1 time a day	7
649	Greenaway_unpub	Greenaway_unpub	NA	NA	NA	NA	NA	NA	NA	NA	United Kingdom	7 days	1 time a day	7

II. Descriptives

1. General Information

There are a total of 658 effects from 88 studies/samples from 73 papers, with a total of \(N =\) 20813 people.

# numeric descriptives
dat %>% select(ID, Sample_Size, Age_Mean, Female_Perc, White_Perc, Time_No) %>%
  unique() %>%
  select(-ID) %>%
  descr(stats = "common", order = "p")

Descriptive Statistics
dat
N: 88

	Sample_Size	Age_Mean	Female_Perc	White_Perc	Time_No
Mean	247.36	27.29	67.96	64.30	48.53
Std.Dev	402.37	10.88	15.16	23.42	39.58
Min	32.00	14.70	9.00	17.00	4.00
Median	109.00	23.50	66.85	66.84	39.50
Max	2626.00	88.71	100.00	100.00	200.00
N.Valid	86.00	77.00	78.00	32.00	88.00
Pct.Valid	97.73	87.50	88.64	36.36	100.00

# frequencies of samples
freq(dat %>% select(ID, Location) %>% unique() %>% pull(Location), order = "freq")

Frequencies

	Freq	% Valid	% Valid Cum.	% Total	% Total Cum.
US	31	37.35	37.35	35.23	35.23
Belgium	12	14.46	51.81	13.64	48.86
Germany	10	12.05	63.86	11.36	60.23
Australia	8	9.64	73.49	9.09	69.32
Canada	4	4.82	78.31	4.55	73.86
New Zealand	4	4.82	83.13	4.55	78.41
United Kingdom	3	3.61	86.75	3.41	81.82
Austria, Germany, Switzerland	2	2.41	89.16	2.27	84.09
China	2	2.41	91.57	2.27	86.36
Estonia	2	2.41	93.98	2.27	88.64
France	2	2.41	96.39	2.27	90.91
Italy	2	2.41	98.80	2.27	93.18
Finland	1	1.20	100.00	1.14	94.32
	5			5.68	100.00
Total	88	100.00	100.00	100.00	100.00

freq(dat %>% select(ID, Frequency) %>% unique() %>% pull(Frequency), order = "freq")

Frequencies

	Freq	% Valid	% Valid Cum.	% Total	% Total Cum.
1 time a day	37	42.05	42.05	42.05	42.05
10 times a day	12	13.64	55.68	13.64	55.68
5 times a day	10	11.36	67.05	11.36	67.05
6 times a day	8	9.09	76.14	9.09	76.14
3 times a day	5	5.68	81.82	5.68	81.82
7 times a day	3	3.41	85.23	3.41	85.23
1 time a week	2	2.27	87.50	2.27	87.50
4 times a day	2	2.27	89.77	2.27	89.77
8 times a day	2	2.27	92.05	2.27	92.05
9 times a day	2	2.27	94.32	2.27	94.32
11 times a day	1	1.14	95.45	1.14	95.45
2 times a day	1	1.14	96.59	1.14	96.59
2 times a month	1	1.14	97.73	1.14	97.73
50 times a day	1	1.14	98.86	1.14	98.86
event-contingent	1	1.14	100.00	1.14	100.00
	0			0.00	100.00
Total	88	100.00	100.00	100.00	100.00

freq(dat %>% select(ID, Personality) %>% unique() %>% pull(Personality), order = "freq")

Frequencies

	Freq	% Valid	% Valid Cum.	% Total	% Total Cum.
N	79	24.09	24.09	24.09	24.09
E	72	21.95	46.04	21.95	46.04
C	63	19.21	65.24	19.21	65.24
A	58	17.68	82.93	17.68	82.93
O	56	17.07	100.00	17.07	100.00
	0			0.00	100.00
Total	328	100.00	100.00	100.00	100.00

freq(dat %>% select(ID, Emotion_Cat) %>% unique() %>% pull(Emotion_Cat), order = "freq")

Frequencies

	Freq	% Valid	% Valid Cum.	% Total	% Total Cum.
pa	76	48.72	48.72	48.72	48.72
na	73	46.79	95.51	46.79	95.51
neu	7	4.49	100.00	4.49	100.00
	0			0.00	100.00
Total	156	100.00	100.00	100.00	100.00

freq(dat %>% select(ID, Raw) %>% unique() %>% pull(Raw), order = "freq")

Frequencies

	Freq	% Valid	% Valid Cum.	% Total	% Total Cum.
yes	80	90.91	90.91	90.91	90.91
no	8	9.09	100.00	9.09	100.00
	0			0.00	100.00
Total	88	100.00	100.00	100.00	100.00

freq(dat %>% select(ID, Study_Quality) %>% unique() %>% pull(Study_Quality), order = "freq")

Frequencies

	Freq	% Valid	% Valid Cum.	% Total	% Total Cum.
good	54	77.14	77.14	61.36	61.36
fair	16	22.86	100.00	18.18	79.55
	18			20.45	100.00
Total	88	100.00	100.00	100.00	100.00

Kappa’s agreement on study quality ratings and proportion of agreement:

# kappa's agreement on study quality ratings
rating_dat <- dat %>% 
         select(ID, Study_Quality, Study_Quality_Reli) %>% 
         unique() %>% 
         select(Study_Quality, Study_Quality_Reli)
kappa2(ratings = rating_dat)

##  Cohen's Kappa for 2 Raters (Weights: unweighted)
## 
##  Subjects = 70 
##    Raters = 2 
##     Kappa = 0.109 
## 
##         z = 0.91 
##   p-value = 0.363

# proportion of agreement
sum(rating_dat$Study_Quality == rating_dat$Study_Quality_Reli, na.rm=T)/
  nrow(na.omit(rating_dat)) * 100

## [1] 68.57143

Kappa’s agreement on abstract screening and proportion of agreement:

# clean rayyan
rayyan$notes <- sub(".*?\\{", "", rayyan$notes)
rayyan$notes <- sub("\\}.*", "", rayyan$notes)
rayyan <- separate(rayyan, notes, into = paste0("decision_", 1:2), sep = ",")
rayyan$decision_1 <- sub(".*?>", "", rayyan$decision_1)
rayyan$decision_2 <- sub(".*?>", "", rayyan$decision_2)
rayyan$decision_1 <- gsub('"', '', rayyan$decision_1)
rayyan$decision_2 <- gsub('"', '', rayyan$decision_2)
rayyan$decision_1 <- sub("Maybe", "Included", rayyan$decision_1)
rayyan$decision_2 <- sub("Maybe", "Included", rayyan$decision_2)

# kappa's agreement
kappa2(ratings = rayyan)

##  Cohen's Kappa for 2 Raters (Weights: unweighted)
## 
##  Subjects = 4848 
##    Raters = 2 
##     Kappa = 0.383 
## 
##         z = 27.2 
##   p-value = 0

# proportion of agreement
sum(rayyan$decision_1 == rayyan$decision_2, na.rm=T)/
  nrow(na.omit(rayyan)) * 100

## [1] 80.44554

# compute sampling variances `vi` for Cor_SD
dat <- escalc(measure = "COR",
              ri = Cor_SD,
              ni = Cor_N, 
              data = dat,
              slab = ID) %>% select(-yi) %>%
  rename(vi_SD = vi)

# compute sampling variances `vi` for Cor_RVI
dat <- escalc(measure = "COR",
              ri = Cor_RVI,
              ni = Cor_N, 
              data = dat,
              slab = ID) %>% select(-yi) %>%
  rename(vi_RVI = vi)

# compute sampling variances `vi` for Cor_Mean
dat <- escalc(measure = "COR",
              ri = Cor_Mean,
              ni = Cor_N, 
              data = dat,
              slab = ID) %>% select(-yi) %>%
  rename(vi_Mean = vi)

# compute sampling variances `vi` for Cor_BCLSM
dat <- escalc(measure = "COR",
              ri = Cor_BCLSM,
              ni = BCLSM_N, 
              data = dat,
              slab = ID) %>% select(-yi) %>%
  rename(vi_BCLSM = vi)

# initialize table for all results
results_df <- data.frame(
  trait = rep(c("A", "C", "E", "N", "O"), each = 3),
  emo = c("all", "pa", "na"),
  cor_SD = NA,
  p_SD = NA,
  CI_SD = NA,
  k_SD = NA,
  Q_SD = NA,
  Qp_SD = NA,
  PI_SD = NA,
  I2_SD = NA,
  tau2_SD = NA,
  cor_RVI = NA,
  p_RVI = NA,
  CI_RVI = NA,
  k_RVI = NA,
  Q_RVI = NA,
  Qp_RVI = NA,
  PI_RVI = NA,
  I2_RVI = NA,
  tau2_RVI = NA,
  cor_BCLSM = NA,
  p_BCLSM = NA,
  CI_BCLSM = NA,
  k_BCLSM = NA,
  Q_BCLSM = NA,
  Qp_BCLSM = NA,
  PI_BCLSM = NA,
  I2_BCLSM = NA,
  tau2_BCLSM = NA
)

2. Compare SD and RVI

These un-preregistered exploratory analyses aim to further clarify the relationship between the two metrics: SD and RVI, especially in relation to the mean and scale boundaries.

# create an index of how far away from the boundaries the mean is
dat <- dat %>%
  rowwise() %>%
  mutate(
    # distance from the closer boundary (min or max)
    mean_dist = min(Mean_Affect - Min, Max - Mean_Affect)
  )
# divide by whole range
dat$mean_dist <- dat$mean_dist/(dat$Max - dat$Min)
dat$mean_relative <- dat$Mean_Affect/dat$Max

# create unique data frame of only emotions
dat_emo <- dat %>% 
  select(ID, Emotion_Cat, Mean_SD, Mean_RVI, SD_RVI, mean_dist, mean_relative) %>%
  unique()

# descriptives of correlations between
#   (1) mean and sd
#   (2) mean and rvi
#   (3) sd and rvi
# and distance from boundaries
dat_emo %>%
  group_by(Emotion_Cat) %>%
  select(mean_dist, mean_relative, Mean_SD, Mean_RVI, SD_RVI) %>%
  descr(stats = "common")

Descriptive Statistics
dat_emo
Group: Emotion_Cat = na
N: 77

	mean_dist	mean_relative	Mean_RVI	Mean_SD	SD_RVI
Mean	0.18	0.29	-0.16	0.56	0.45
Std.Dev	0.08	0.11	0.18	0.16	0.27
Min	0.05	0.09	-0.58	-0.13	-0.27
Median	0.18	0.32	-0.15	0.57	0.50
Max	0.47	0.63	0.25	0.86	0.98
N.Valid	69.00	69.00	69.00	69.00	69.00
Pct.Valid	89.61	89.61	89.61	89.61	89.61

Group: Emotion_Cat = neu
N: 7

	mean_dist	mean_relative	Mean_RVI	Mean_SD	SD_RVI
Mean	0.32	0.58	-0.01	-0.07	0.78
Std.Dev	0.09	0.18	0.30	0.37	0.37
Min	0.12	0.29	-0.52	-0.43	-0.04
Median	0.35	0.65	-0.03	-0.18	0.92
Max	0.39	0.74	0.42	0.68	0.99
N.Valid	7.00	7.00	7.00	7.00	7.00
Pct.Valid	100.00	100.00	100.00	100.00	100.00

Group: Emotion_Cat = pa
N: 78

	mean_dist	mean_relative	Mean_RVI	Mean_SD	SD_RVI
Mean	0.40	0.56	-0.05	-0.04	0.85
Std.Dev	0.08	0.10	0.19	0.25	0.20
Min	0.07	0.20	-0.62	-0.60	-0.04
Median	0.41	0.57	-0.07	-0.06	0.91
Max	0.50	0.75	0.41	0.78	0.99
N.Valid	71.00	71.00	71.00	71.00	71.00
Pct.Valid	91.03	91.03	91.03	91.03	91.03

# scatterplot between distance from boundaries and correlation between SD and RVI
ggplot(data = na.omit(dat_emo),
       aes(x = mean_dist, y = SD_RVI, 
           color = Emotion_Cat,
           shape = Emotion_Cat)) +
  geom_point(size = 2) +
  theme_classic() +
  labs(
    x = "Distance from scale boundaries in proportion",
    y = "Correlation between SD and RVI",
    title = paste(
      "Scatterplot between distance from boundaries and SD-RVI correlations\n",
      "r =", round(cor(dat_emo$mean_dist, dat_emo$SD_RVI, use = "pair"), 2),
      "p < .001"),
    color = "Emotion Type",
    shape = "Emotion Type"
  )

# scatterplot between distance from boundaries and mean-SD correlations
ggplot(data = na.omit(dat_emo),
       aes(x = mean_dist, y = Mean_SD, 
           color = Emotion_Cat,
           shape = Emotion_Cat)) +
  geom_point(size = 2) +
  theme_classic() +
  labs(
    x = "Distance from scale boundaries in proportion",
    y = "Correlation between mean and SD",
    title = paste(
      "Scatterplot between distance from boundaries and mean-SD correlations\n",
      "r =", round(cor(dat_emo$mean_dist, dat_emo$Mean_SD, use = "pair"), 2),
      "p < .001"),
    color = "Emotion Type",
    shape = "Emotion Type"
  )

# scatterplot between distance from boundaries and mean-RVI correlations
ggplot(data = na.omit(dat_emo),
       aes(x = mean_dist, y = Mean_RVI, 
           color = Emotion_Cat,
           shape = Emotion_Cat)) +
  geom_point(size = 2) +
  theme_classic() +
  labs(
    x = "Distance from scale boundaries in proportion",
    y = "Correlation between mean and RVI",
    title = paste(
      "Scatterplot between distance from boundaries and mean-RVI correlations\n",
      "r =", round(cor(dat_emo$mean_dist, dat_emo$Mean_RVI, use = "pair"), 2),
      "p < .001"),
    color = "Emotion Type",
    shape = "Emotion Type"
  )

# histogram of relative means
ggplot(data = na.omit(dat_emo[dat_emo$Emotion_Cat == "na",]),
       aes(x = mean_relative)) + 
  geom_histogram(binwidth = 0.03,
                 color = "#E69F00",
                 fill = "#710c0c") +
  theme_classic() +
  labs(
    title = "Histogram of Mean-Level on Negative Affect",
    x = "Relative Mean Level"
  ) + 
  xlim(0.1, 1.0)

## Warning: Removed 1 rows containing non-finite values (`stat_bin()`).

## Warning: Removed 2 rows containing missing values (`geom_bar()`).

ggplot(data = na.omit(dat_emo[dat_emo$Emotion_Cat == "pa",]),
       aes(x = mean_relative)) + 
  geom_histogram(binwidth = 0.03,
                 color = "#E69F00",
                 fill = "#710c0c") +
  theme_classic() +
  labs(
    title = "Histogram of Mean-Level on Positive Affect",
    x = "Relative Mean Level"
  ) +
  xlim(0.1, 1.0)

## Warning: Removed 2 rows containing missing values (`geom_bar()`).

III. Results

1. Neuroticism

results_df <- fit_meta(trait = "N", .data = dat, output = results_df)

## 
## Multivariate Meta-Analysis Model (k = 150; method: REML)
## 
## Variance Components:
## 
## outer factor: ID                  (nlvls = 74)
## inner factor: factor(Emotion_Cat) (nlvls = 2)
## 
##             estim    sqrt  fixed 
## tau^2      0.0044  0.0662     no 
## rho        0.4273             no 
## 
## Test for Residual Heterogeneity:
## QE(df = 148) = 310.5485, p-val < .0001
## 
## Number of estimates:   150
## Number of clusters:    74
## Estimates per cluster: 1-7 (mean: 2.03, median: 2)
## 
## Test of Moderators (coefficient 2):¹
## F(df1 = 1, df2 = 53.61) = 230.9184, p-val < .0001
## 
## Model Results:
## 
##                        estimate      se¹      tval¹     df¹    pval¹    ci.lb¹ 
## intrcpt                  0.3033  0.0117    25.9549    59.4   <.0001    0.2799  
## factor(Emotion_Cat)pa   -0.2248  0.0148   -15.1960   53.61   <.0001   -0.2544  
##                          ci.ub¹      
## intrcpt                 0.3267   *** 
## factor(Emotion_Cat)pa  -0.1951   *** 
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## 1) results based on cluster-robust inference (var-cov estimator: CR2,
##    approx t/F-tests and confidence intervals, df: Satterthwaite approx)

## 
## Multivariate Meta-Analysis Model (k = 122; method: REML)
## 
## Variance Components:
## 
## outer factor: ID                    (nlvls = 61)
## inner factor: factor(Study_Quality) (nlvls = 2)
## 
##             estim    sqrt  fixed 
## tau^2      0.0058  0.0765     no 
## rho        0.5000             no 
## 
## Test for Residual Heterogeneity:
## QE(df = 120) = 829.9909, p-val < .0001
## 
## Test of Moderators (coefficient 2):
## QM(df = 1) = 0.0469, p-val = 0.8286
## 
## Model Results:
## 
##                            estimate      se     zval    pval    ci.lb   ci.ub 
## intrcpt                      0.2183  0.0300   7.2872  <.0001   0.1596  0.2770 
## factor(Study_Quality)good   -0.0071  0.0329  -0.2166  0.8286  -0.0715  0.0573 
##                                
## intrcpt                    *** 
## factor(Study_Quality)good      
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

## 
## Multivariate Meta-Analysis Model (k = 133; method: REML)
## 
## Variance Components:
## 
## outer factor: ID                  (nlvls = 68)
## inner factor: factor(Emotion_Cat) (nlvls = 2)
## 
##             estim    sqrt  fixed 
## tau^2      0.0086  0.0926     no 
## rho        0.5067             no 
## 
## Test for Residual Heterogeneity:
## QE(df = 131) = 374.5068, p-val < .0001
## 
## Number of estimates:   133
## Number of clusters:    68
## Estimates per cluster: 1-4 (mean: 1.96, median: 2)
## 
## Test of Moderators (coefficient 2):¹
## F(df1 = 1, df2 = 52.71) = 47.9640, p-val < .0001
## 
## Model Results:
## 
##                        estimate      se¹     tval¹     df¹    pval¹    ci.lb¹ 
## intrcpt                 -0.0626  0.0165   -3.7876   58.54   0.0004   -0.0957  
## factor(Emotion_Cat)pa    0.1253  0.0181    6.9256   52.71   <.0001    0.0890  
##                          ci.ub¹      
## intrcpt                -0.0295   *** 
## factor(Emotion_Cat)pa   0.1616   *** 
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## 1) results based on cluster-robust inference (var-cov estimator: CR2,
##    approx t/F-tests and confidence intervals, df: Satterthwaite approx)

## 
## Multivariate Meta-Analysis Model (k = 109; method: REML)
## 
## Variance Components:
## 
## outer factor: ID                    (nlvls = 57)
## inner factor: factor(Study_Quality) (nlvls = 2)
## 
##             estim    sqrt  fixed 
## tau^2      0.0081  0.0902     no 
## rho        0.5000             no 
## 
## Test for Residual Heterogeneity:
## QE(df = 107) = 500.8071, p-val < .0001
## 
## Number of estimates:   109
## Number of clusters:    57
## Estimates per cluster: 1-4 (mean: 1.91, median: 2)
## 
## Test of Moderators (coefficient 2):¹
## F(df1 = 1, df2 = 9.88) = 1.0270, p-val = 0.3350
## 
## Model Results:
## 
##                            estimate      se¹     tval¹    df¹    pval¹ 
## intrcpt                      0.0293  0.0334    0.8764   7.32   0.4087  
## factor(Study_Quality)good   -0.0375  0.0370   -1.0134   9.88   0.3350  
##                              ci.lb¹   ci.ub¹    
## intrcpt                    -0.0490   0.1076     
## factor(Study_Quality)good  -0.1202   0.0451     
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## 1) results based on cluster-robust inference (var-cov estimator: CR2,
##    approx t/F-tests and confidence intervals, df: Satterthwaite approx)

## 
## Multivariate Meta-Analysis Model (k = 132; method: REML)
## 
## Variance Components:
## 
## outer factor: ID                  (nlvls = 67)
## inner factor: factor(Emotion_Cat) (nlvls = 2)
## 
##             estim    sqrt  fixed 
## tau^2      0.0078  0.0881     no 
## rho        0.7246             no 
## 
## Test for Residual Heterogeneity:
## QE(df = 130) = 322.5981, p-val < .0001
## 
## Number of estimates:   132
## Number of clusters:    67
## Estimates per cluster: 1-4 (mean: 1.97, median: 2)
## 
## Test of Moderators (coefficient 2):¹
## F(df1 = 1, df2 = 45.27) = 35.5427, p-val < .0001
## 
## Model Results:
## 
##                        estimate      se¹     tval¹     df¹    pval¹    ci.lb¹ 
## intrcpt                  0.1599  0.0158   10.1124   57.26   <.0001    0.1283  
## factor(Emotion_Cat)pa   -0.0926  0.0155   -5.9618   45.27   <.0001   -0.1238  
##                          ci.ub¹      
## intrcpt                 0.1916   *** 
## factor(Emotion_Cat)pa  -0.0613   *** 
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## 1) results based on cluster-robust inference (var-cov estimator: CR2,
##    approx t/F-tests and confidence intervals, df: Satterthwaite approx)

## 
## Multivariate Meta-Analysis Model (k = 108; method: REML)
## 
## Variance Components:
## 
## outer factor: ID                    (nlvls = 56)
## inner factor: factor(Study_Quality) (nlvls = 2)
## 
##             estim    sqrt  fixed 
## tau^2      0.0072  0.0849     no 
## rho        0.5000             no 
## 
## Test for Residual Heterogeneity:
## QE(df = 106) = 370.6036, p-val < .0001
## 
## Number of estimates:   108
## Number of clusters:    56
## Estimates per cluster: 1-4 (mean: 1.93, median: 2)
## 
## Test of Moderators (coefficient 2):¹
## F(df1 = 1, df2 = 9.93) = 0.0942, p-val = 0.7652
## 
## Model Results:
## 
##                            estimate      se¹     tval¹    df¹    pval¹ 
## intrcpt                      0.1300  0.0430    3.0223   7.32   0.0183  
## factor(Study_Quality)good   -0.0139  0.0454   -0.3070   9.93   0.7652  
##                              ci.lb¹   ci.ub¹    
## intrcpt                     0.0292   0.2307   * 
## factor(Study_Quality)good  -0.1153   0.0874     
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## 1) results based on cluster-robust inference (var-cov estimator: CR2,
##    approx t/F-tests and confidence intervals, df: Satterthwaite approx)

2. Extraversion

results_df <- fit_meta(trait = "E", .data = dat, output = results_df)

## 
## Multivariate Meta-Analysis Model (k = 130; method: REML)
## 
## Variance Components:
## 
## outer factor: ID                  (nlvls = 66)
## inner factor: factor(Emotion_Cat) (nlvls = 2)
## 
##             estim    sqrt  fixed 
## tau^2      0.0065  0.0809     no 
## rho        0.7993             no 
## 
## Test for Residual Heterogeneity:
## QE(df = 128) = 273.5444, p-val < .0001
## 
## Number of estimates:   130
## Number of clusters:    66
## Estimates per cluster: 1-7 (mean: 1.97, median: 2)
## 
## Test of Moderators (coefficient 2):¹
## F(df1 = 1, df2 = 36.35) = 69.1654, p-val < .0001
## 
## Model Results:
## 
##                        estimate      se¹     tval¹     df¹    pval¹    ci.lb¹ 
## intrcpt                 -0.0507  0.0136   -3.7374   50.15   0.0005   -0.0779  
## factor(Emotion_Cat)pa    0.1240  0.0149    8.3166   36.35   <.0001    0.0938  
##                          ci.ub¹      
## intrcpt                -0.0235   *** 
## factor(Emotion_Cat)pa   0.1542   *** 
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## 1) results based on cluster-robust inference (var-cov estimator: CR2,
##    approx t/F-tests and confidence intervals, df: Satterthwaite approx)

## 
## Multivariate Meta-Analysis Model (k = 103; method: REML)
## 
## Variance Components:
## 
## outer factor: ID                    (nlvls = 54)
## inner factor: factor(Study_Quality) (nlvls = 2)
## 
##             estim    sqrt  fixed 
## tau^2      0.0055  0.0740     no 
## rho        0.5000             no 
## 
## Test for Residual Heterogeneity:
## QE(df = 101) = 365.6576, p-val < .0001
## 
## Test of Moderators (coefficient 2):
## QM(df = 1) = 1.9100, p-val = 0.1670
## 
## Model Results:
## 
##                            estimate      se     zval    pval    ci.lb   ci.ub 
## intrcpt                      0.0402  0.0285   1.4113  0.1582  -0.0156  0.0961 
## factor(Study_Quality)good   -0.0448  0.0324  -1.3820  0.1670  -0.1084  0.0187 
##                              
## intrcpt                      
## factor(Study_Quality)good    
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

## 
## Multivariate Meta-Analysis Model (k = 113; method: REML)
## 
## Variance Components:
## 
## outer factor: ID                  (nlvls = 60)
## inner factor: factor(Emotion_Cat) (nlvls = 2)
## 
##             estim    sqrt  fixed 
## tau^2      0.0039  0.0623     no 
## rho        1.0000             no 
## 
## Test for Residual Heterogeneity:
## QE(df = 111) = 190.1526, p-val < .0001
## 
## Number of estimates:   113
## Number of clusters:    60
## Estimates per cluster: 1-4 (mean: 1.88, median: 2)
## 
## Test of Moderators (coefficient 2):¹
## F(df1 = 1, df2 = 15.86) = 7.5824, p-val = 0.0142
## 
## Model Results:
## 
##                        estimate      se¹     tval¹     df¹    pval¹    ci.lb¹ 
## intrcpt                  0.0790  0.0127    6.2388   39.33   <.0001    0.0534  
## factor(Emotion_Cat)pa   -0.0330  0.0120   -2.7536   15.86   0.0142   -0.0585  
##                          ci.ub¹      
## intrcpt                 0.1046   *** 
## factor(Emotion_Cat)pa  -0.0076     * 
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## 1) results based on cluster-robust inference (var-cov estimator: CR2,
##    approx t/F-tests and confidence intervals, df: Satterthwaite approx)

## 
## Multivariate Meta-Analysis Model (k = 90; method: REML)
## 
## Variance Components:
## 
## outer factor: ID                    (nlvls = 50)
## inner factor: factor(Study_Quality) (nlvls = 2)
## 
##             estim    sqrt  fixed 
## tau^2      0.0030  0.0551     no 
## rho        0.5000             no 
## 
## Test for Residual Heterogeneity:
## QE(df = 88) = 156.9802, p-val < .0001
## 
## Number of estimates:   90
## Number of clusters:    50
## Estimates per cluster: 1-4 (mean: 1.80, median: 2)
## 
## Test of Moderators (coefficient 2):¹
## F(df1 = 1, df2 = 13.8) = 1.7301, p-val = 0.2098
## 
## Model Results:
## 
##                            estimate      se¹     tval¹    df¹    pval¹ 
## intrcpt                      0.0839  0.0218    3.8542   9.47   0.0035  
## factor(Study_Quality)good   -0.0339  0.0257   -1.3153   13.8   0.2098  
##                              ci.lb¹   ci.ub¹     
## intrcpt                     0.0350   0.1327   ** 
## factor(Study_Quality)good  -0.0892   0.0214      
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## 1) results based on cluster-robust inference (var-cov estimator: CR2,
##    approx t/F-tests and confidence intervals, df: Satterthwaite approx)

## 
## Multivariate Meta-Analysis Model (k = 113; method: REML)
## 
## Variance Components:
## 
## outer factor: ID                  (nlvls = 60)
## inner factor: factor(Emotion_Cat) (nlvls = 2)
## 
##             estim    sqrt  fixed 
## tau^2      0.0056  0.0746     no 
## rho        0.9977             no 
## 
## Test for Residual Heterogeneity:
## QE(df = 111) = 203.0535, p-val < .0001
## 
## Number of estimates:   113
## Number of clusters:    60
## Estimates per cluster: 1-4 (mean: 1.88, median: 2)
## 
## Test of Moderators (coefficient 2):¹
## F(df1 = 1, df2 = 15.18) = 40.5171, p-val < .0001
## 
## Model Results:
## 
##                        estimate      se¹     tval¹     df¹    pval¹    ci.lb¹ 
## intrcpt                 -0.0040  0.0139   -0.2896   42.92   0.7735   -0.0320  
## factor(Emotion_Cat)pa    0.0797  0.0125    6.3653   15.18   <.0001    0.0530  
##                         ci.ub¹      
## intrcpt                0.0239       
## factor(Emotion_Cat)pa  0.1064   *** 
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## 1) results based on cluster-robust inference (var-cov estimator: CR2,
##    approx t/F-tests and confidence intervals, df: Satterthwaite approx)

## 
## Multivariate Meta-Analysis Model (k = 90; method: REML)
## 
## Variance Components:
## 
## outer factor: ID                    (nlvls = 50)
## inner factor: factor(Study_Quality) (nlvls = 2)
## 
##             estim    sqrt  fixed 
## tau^2      0.0043  0.0656     no 
## rho        0.5000             no 
## 
## Test for Residual Heterogeneity:
## QE(df = 88) = 196.6602, p-val < .0001
## 
## Number of estimates:   90
## Number of clusters:    50
## Estimates per cluster: 1-4 (mean: 1.80, median: 2)
## 
## Test of Moderators (coefficient 2):¹
## F(df1 = 1, df2 = 14.83) = 2.3764, p-val = 0.1442
## 
## Model Results:
## 
##                            estimate      se¹     tval¹     df¹    pval¹ 
## intrcpt                      0.0722  0.0243    2.9652    9.89   0.0143  
## factor(Study_Quality)good   -0.0443  0.0287   -1.5416   14.83   0.1442  
##                              ci.lb¹   ci.ub¹    
## intrcpt                     0.0179   0.1265   * 
## factor(Study_Quality)good  -0.1056   0.0170     
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## 1) results based on cluster-robust inference (var-cov estimator: CR2,
##    approx t/F-tests and confidence intervals, df: Satterthwaite approx)

3. Agreeableness

results_df <- fit_meta(trait = "A", .data = dat, output = results_df)

## 
## Multivariate Meta-Analysis Model (k = 112; method: REML)
## 
## Variance Components:
## 
## outer factor: ID                  (nlvls = 53)
## inner factor: factor(Emotion_Cat) (nlvls = 2)
## 
##             estim    sqrt  fixed 
## tau^2      0.0088  0.0939     no 
## rho        0.8579             no 
## 
## Test for Residual Heterogeneity:
## QE(df = 110) = 269.6759, p-val < .0001
## 
## Number of estimates:   112
## Number of clusters:    53
## Estimates per cluster: 1-7 (mean: 2.11, median: 2)
## 
## Test of Moderators (coefficient 2):¹
## F(df1 = 1, df2 = 31.84) = 44.3148, p-val < .0001
## 
## Model Results:
## 
##                        estimate      se¹     tval¹     df¹    pval¹    ci.lb¹ 
## intrcpt                 -0.1045  0.0163   -6.4158   45.96   <.0001   -0.1373  
## factor(Emotion_Cat)pa    0.1022  0.0154    6.6569   31.84   <.0001    0.0709  
##                          ci.ub¹      
## intrcpt                -0.0717   *** 
## factor(Emotion_Cat)pa   0.1335   *** 
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## 1) results based on cluster-robust inference (var-cov estimator: CR2,
##    approx t/F-tests and confidence intervals, df: Satterthwaite approx)

## 
## Multivariate Meta-Analysis Model (k = 84; method: REML)
## 
## Variance Components:
## 
## outer factor: ID                    (nlvls = 40)
## inner factor: factor(Study_Quality) (nlvls = 2)
## 
##             estim    sqrt  fixed 
## tau^2      0.0106  0.1032     no 
## rho        0.5000             no 
## 
## Test for Residual Heterogeneity:
## QE(df = 82) = 310.0257, p-val < .0001
## 
## Test of Moderators (coefficient 2):
## QM(df = 1) = 0.0338, p-val = 0.8540
## 
## Model Results:
## 
##                            estimate      se     zval    pval    ci.lb   ci.ub 
## intrcpt                     -0.0713  0.0421  -1.6916  0.0907  -0.1539  0.0113 
## factor(Study_Quality)good    0.0087  0.0472   0.1840  0.8540  -0.0838  0.1012 
##                              
## intrcpt                    . 
## factor(Study_Quality)good    
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

## 
## Multivariate Meta-Analysis Model (k = 97; method: REML)
## 
## Variance Components:
## 
## outer factor: ID                  (nlvls = 48)
## inner factor: factor(Emotion_Cat) (nlvls = 2)
## 
##             estim    sqrt  fixed 
## tau^2      0.0091  0.0952     no 
## rho        0.8895             no 
## 
## Test for Residual Heterogeneity:
## QE(df = 95) = 232.6265, p-val < .0001
## 
## Number of estimates:   97
## Number of clusters:    48
## Estimates per cluster: 1-4 (mean: 2.02, median: 2)
## 
## Test of Moderators (coefficient 2):¹
## F(df1 = 1, df2 = 25.68) = 9.4137, p-val = 0.0050
## 
## Model Results:
## 
##                        estimate      se¹     tval¹     df¹    pval¹    ci.lb¹ 
## intrcpt                  0.0506  0.0168    3.0109   41.15   0.0044    0.0167  
## factor(Emotion_Cat)pa   -0.0492  0.0160   -3.0682   25.68   0.0050   -0.0821  
##                          ci.ub¹     
## intrcpt                 0.0845   ** 
## factor(Emotion_Cat)pa  -0.0162   ** 
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## 1) results based on cluster-robust inference (var-cov estimator: CR2,
##    approx t/F-tests and confidence intervals, df: Satterthwaite approx)

## 
## Multivariate Meta-Analysis Model (k = 73; method: REML)
## 
## Variance Components:
## 
## outer factor: ID                    (nlvls = 37)
## inner factor: factor(Study_Quality) (nlvls = 2)
## 
##             estim    sqrt  fixed 
## tau^2      0.0092  0.0961     no 
## rho        0.5000             no 
## 
## Test for Residual Heterogeneity:
## QE(df = 71) = 212.5704, p-val < .0001
## 
## Number of estimates:   73
## Number of clusters:    37
## Estimates per cluster: 1-4 (mean: 1.97, median: 2)
## 
## Test of Moderators (coefficient 2):¹
## F(df1 = 1, df2 = 7.71) = 0.3587, p-val = 0.5664
## 
## Model Results:
## 
##                            estimate      se¹     tval¹    df¹    pval¹ 
## intrcpt                     -0.0184  0.0721   -0.2546    5.5   0.8082  
## factor(Study_Quality)good    0.0445  0.0744    0.5989   7.71   0.5664  
##                              ci.lb¹   ci.ub¹    
## intrcpt                    -0.1988   0.1621     
## factor(Study_Quality)good  -0.1281   0.2172     
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## 1) results based on cluster-robust inference (var-cov estimator: CR2,
##    approx t/F-tests and confidence intervals, df: Satterthwaite approx)

## 
## Multivariate Meta-Analysis Model (k = 97; method: REML)
## 
## Variance Components:
## 
## outer factor: ID                  (nlvls = 48)
## inner factor: factor(Emotion_Cat) (nlvls = 2)
## 
##             estim    sqrt  fixed 
## tau^2      0.0096  0.0982     no 
## rho        0.8740             no 
## 
## Test for Residual Heterogeneity:
## QE(df = 95) = 248.2551, p-val < .0001
## 
## Number of estimates:   97
## Number of clusters:    48
## Estimates per cluster: 1-4 (mean: 2.02, median: 2)
## 
## Test of Moderators (coefficient 2):¹
## F(df1 = 1, df2 = 26.78) = 4.9074, p-val = 0.0354
## 
## Model Results:
## 
##                        estimate      se¹     tval¹     df¹    pval¹    ci.lb¹ 
## intrcpt                 -0.0352  0.0184   -1.9177    41.4   0.0621   -0.0723  
## factor(Emotion_Cat)pa    0.0366  0.0165    2.2153   26.78   0.0354    0.0027  
##                         ci.ub¹    
## intrcpt                0.0019   . 
## factor(Emotion_Cat)pa  0.0706   * 
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## 1) results based on cluster-robust inference (var-cov estimator: CR2,
##    approx t/F-tests and confidence intervals, df: Satterthwaite approx)

## 
## Multivariate Meta-Analysis Model (k = 73; method: REML)
## 
## Variance Components:
## 
## outer factor: ID                    (nlvls = 37)
## inner factor: factor(Study_Quality) (nlvls = 2)
## 
##             estim    sqrt  fixed 
## tau^2      0.0090  0.0948     no 
## rho        0.5000             no 
## 
## Test for Residual Heterogeneity:
## QE(df = 71) = 214.3815, p-val < .0001
## 
## Number of estimates:   73
## Number of clusters:    37
## Estimates per cluster: 1-4 (mean: 1.97, median: 2)
## 
## Test of Moderators (coefficient 2):¹
## F(df1 = 1, df2 = 7.73) = 0.9816, p-val = 0.3518
## 
## Model Results:
## 
##                            estimate      se¹     tval¹    df¹    pval¹ 
## intrcpt                     -0.0811  0.0715   -1.1352    5.5   0.3033  
## factor(Study_Quality)good    0.0730  0.0737    0.9908   7.73   0.3518  
##                              ci.lb¹   ci.ub¹    
## intrcpt                    -0.2599   0.0976     
## factor(Study_Quality)good  -0.0980   0.2440     
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## 1) results based on cluster-robust inference (var-cov estimator: CR2,
##    approx t/F-tests and confidence intervals, df: Satterthwaite approx)

4. Conscientiousness

results_df <- fit_meta(trait = "C", .data = dat, output = results_df)

## 
## Multivariate Meta-Analysis Model (k = 128; method: REML)
## 
## Variance Components:
## 
## outer factor: ID                  (nlvls = 58)
## inner factor: factor(Emotion_Cat) (nlvls = 2)
## 
##             estim    sqrt  fixed 
## tau^2      0.0029  0.0537     no 
## rho        1.0000             no 
## 
## Test for Residual Heterogeneity:
## QE(df = 126) = 181.8460, p-val = 0.0008
## 
## Number of estimates:   128
## Number of clusters:    58
## Estimates per cluster: 1-10 (mean: 2.21, median: 2)
## 
## Test of Moderators (coefficient 2):¹
## F(df1 = 1, df2 = 20.49) = 95.8076, p-val < .0001
## 
## Model Results:
## 
##                        estimate      se¹      tval¹     df¹    pval¹    ci.lb¹ 
## intrcpt                 -0.1211  0.0110   -10.9755   41.94   <.0001   -0.1434  
## factor(Emotion_Cat)pa    0.1074  0.0110     9.7881   20.49   <.0001    0.0845  
##                          ci.ub¹      
## intrcpt                -0.0988   *** 
## factor(Emotion_Cat)pa   0.1302   *** 
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## 1) results based on cluster-robust inference (var-cov estimator: CR2,
##    approx t/F-tests and confidence intervals, df: Satterthwaite approx)

## 
## Multivariate Meta-Analysis Model (k = 100; method: REML)
## 
## Variance Components:
## 
## outer factor: ID                    (nlvls = 45)
## inner factor: factor(Study_Quality) (nlvls = 2)
## 
##             estim    sqrt  fixed 
## tau^2      0.0017  0.0413     no 
## rho        0.5000             no 
## 
## Test for Residual Heterogeneity:
## QE(df = 98) = 203.0795, p-val < .0001
## 
## Test of Moderators (coefficient 2):
## QM(df = 1) = 0.8303, p-val = 0.3622
## 
## Model Results:
## 
##                            estimate      se     zval    pval    ci.lb    ci.ub 
## intrcpt                     -0.0905  0.0257  -3.5162  0.0004  -0.1410  -0.0401 
## factor(Study_Quality)good    0.0254  0.0279   0.9112  0.3622  -0.0292   0.0800 
##                                
## intrcpt                    *** 
## factor(Study_Quality)good      
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

## 
## Multivariate Meta-Analysis Model (k = 105; method: REML)
## 
## Variance Components:
## 
## outer factor: ID                  (nlvls = 53)
## inner factor: factor(Emotion_Cat) (nlvls = 2)
## 
##             estim    sqrt  fixed 
## tau^2      0.0098  0.0992     no 
## rho        0.3678             no 
## 
## Test for Residual Heterogeneity:
## QE(df = 103) = 274.8285, p-val < .0001
## 
## Number of estimates:   105
## Number of clusters:    53
## Estimates per cluster: 1-4 (mean: 1.98, median: 2)
## 
## Test of Moderators (coefficient 2):¹
## F(df1 = 1, df2 = 43.52) = 5.6366, p-val = 0.0221
## 
## Model Results:
## 
##                        estimate      se¹     tval¹     df¹    pval¹    ci.lb¹ 
## intrcpt                  0.0510  0.0188    2.7212   44.76   0.0092    0.0133  
## factor(Emotion_Cat)pa   -0.0534  0.0225   -2.3742   43.52   0.0221   -0.0987  
##                          ci.ub¹     
## intrcpt                 0.0888   ** 
## factor(Emotion_Cat)pa  -0.0081    * 
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## 1) results based on cluster-robust inference (var-cov estimator: CR2,
##    approx t/F-tests and confidence intervals, df: Satterthwaite approx)

## 
## Multivariate Meta-Analysis Model (k = 81; method: REML)
## 
## Variance Components:
## 
## outer factor: ID                    (nlvls = 42)
## inner factor: factor(Study_Quality) (nlvls = 2)
## 
##             estim    sqrt  fixed 
## tau^2      0.0079  0.0887     no 
## rho        0.5000             no 
## 
## Test for Residual Heterogeneity:
## QE(df = 79) = 265.1099, p-val < .0001
## 
## Number of estimates:   81
## Number of clusters:    42
## Estimates per cluster: 1-4 (mean: 1.93, median: 2)
## 
## Test of Moderators (coefficient 2):¹
## F(df1 = 1, df2 = 7.23) = 2.6208, p-val = 0.1482
## 
## Model Results:
## 
##                            estimate      se¹     tval¹    df¹    pval¹ 
## intrcpt                     -0.0459  0.0472   -0.9727   5.41   0.3722  
## factor(Study_Quality)good    0.0817  0.0505    1.6189   7.23   0.1482  
##                              ci.lb¹   ci.ub¹    
## intrcpt                    -0.1645   0.0727     
## factor(Study_Quality)good  -0.0369   0.2002     
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## 1) results based on cluster-robust inference (var-cov estimator: CR2,
##    approx t/F-tests and confidence intervals, df: Satterthwaite approx)

## 
## Multivariate Meta-Analysis Model (k = 103; method: REML)
## 
## Variance Components:
## 
## outer factor: ID                  (nlvls = 52)
## inner factor: factor(Emotion_Cat) (nlvls = 2)
## 
##             estim    sqrt  fixed 
## tau^2      0.0073  0.0857     no 
## rho        1.0000             no 
## 
## Test for Residual Heterogeneity:
## QE(df = 101) = 218.5766, p-val < .0001
## 
## Number of estimates:   103
## Number of clusters:    52
## Estimates per cluster: 1-4 (mean: 1.98, median: 2)
## 
## Test of Moderators (coefficient 2):¹
## F(df1 = 1, df2 = 15.03) = 8.7916, p-val = 0.0096
## 
## Model Results:
## 
##                        estimate      se¹     tval¹     df¹    pval¹    ci.lb¹ 
## intrcpt                 -0.0428  0.0156   -2.7362   43.45   0.0090   -0.0743  
## factor(Emotion_Cat)pa    0.0337  0.0114    2.9651   15.03   0.0096    0.0095  
##                          ci.ub¹     
## intrcpt                -0.0113   ** 
## factor(Emotion_Cat)pa   0.0579   ** 
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## 1) results based on cluster-robust inference (var-cov estimator: CR2,
##    approx t/F-tests and confidence intervals, df: Satterthwaite approx)

## 
## Multivariate Meta-Analysis Model (k = 79; method: REML)
## 
## Variance Components:
## 
## outer factor: ID                    (nlvls = 41)
## inner factor: factor(Study_Quality) (nlvls = 2)
## 
##             estim    sqrt  fixed 
## tau^2      0.0060  0.0773     no 
## rho        0.5000             no 
## 
## Test for Residual Heterogeneity:
## QE(df = 77) = 169.9816, p-val < .0001
## 
## Number of estimates:   79
## Number of clusters:    41
## Estimates per cluster: 1-4 (mean: 1.93, median: 2)
## 
## Test of Moderators (coefficient 2):¹
## F(df1 = 1, df2 = 7.07) = 1.9392, p-val = 0.2060
## 
## Model Results:
## 
##                            estimate      se¹     tval¹    df¹    pval¹ 
## intrcpt                     -0.0977  0.0587   -1.6639   5.31   0.1536  
## factor(Study_Quality)good    0.0845  0.0607    1.3925   7.07   0.2060  
##                              ci.lb¹   ci.ub¹    
## intrcpt                    -0.2460   0.0506     
## factor(Study_Quality)good  -0.0587   0.2278     
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## 1) results based on cluster-robust inference (var-cov estimator: CR2,
##    approx t/F-tests and confidence intervals, df: Satterthwaite approx)

5. Openness

results_df <- fit_meta(trait = "O", .data = dat, output = results_df)

## 
## Multivariate Meta-Analysis Model (k = 107; method: REML)
## 
## Variance Components:
## 
## outer factor: ID                  (nlvls = 51)
## inner factor: factor(Emotion_Cat) (nlvls = 2)
## 
##             estim    sqrt  fixed 
## tau^2      0.0098  0.0992     no 
## rho        0.9363             no 
## 
## Test for Residual Heterogeneity:
## QE(df = 105) = 259.2885, p-val < .0001
## 
## Number of estimates:   107
## Number of clusters:    51
## Estimates per cluster: 1-7 (mean: 2.10, median: 2)
## 
## Test of Moderators (coefficient 2):¹
## F(df1 = 1, df2 = 25.3) = 24.3791, p-val < .0001
## 
## Model Results:
## 
##                        estimate      se¹     tval¹     df¹    pval¹    ci.lb¹ 
## intrcpt                 -0.0031  0.0164   -0.1870   44.66   0.8525   -0.0362  
## factor(Emotion_Cat)pa    0.0707  0.0143    4.9375    25.3   <.0001    0.0412  
##                         ci.ub¹      
## intrcpt                0.0300       
## factor(Emotion_Cat)pa  0.1001   *** 
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## 1) results based on cluster-robust inference (var-cov estimator: CR2,
##    approx t/F-tests and confidence intervals, df: Satterthwaite approx)

## 
## Multivariate Meta-Analysis Model (k = 79; method: REML)
## 
## Variance Components:
## 
## outer factor: ID                    (nlvls = 38)
## inner factor: factor(Study_Quality) (nlvls = 2)
## 
##             estim    sqrt  fixed 
## tau^2      0.0093  0.0963     no 
## rho        0.5000             no 
## 
## Test for Residual Heterogeneity:
## QE(df = 77) = 245.8930, p-val < .0001
## 
## Test of Moderators (coefficient 2):
## QM(df = 1) = 0.0165, p-val = 0.8979
## 
## Model Results:
## 
##                            estimate      se    zval    pval    ci.lb   ci.ub    
## intrcpt                      0.0292  0.0403  0.7229  0.4697  -0.0499  0.1082    
## factor(Study_Quality)good    0.0058  0.0454  0.1283  0.8979  -0.0832  0.0948    
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

## 
## Multivariate Meta-Analysis Model (k = 92; method: REML)
## 
## Variance Components:
## 
## outer factor: ID                  (nlvls = 46)
## inner factor: factor(Emotion_Cat) (nlvls = 2)
## 
##             estim    sqrt  fixed 
## tau^2      0.0070  0.0836     no 
## rho        0.4541             no 
## 
## Test for Residual Heterogeneity:
## QE(df = 90) = 202.3818, p-val < .0001
## 
## Number of estimates:   92
## Number of clusters:    46
## Estimates per cluster: 1-4 (mean: 2.00, median: 2)
## 
## Test of Moderators (coefficient 2):¹
## F(df1 = 1, df2 = 34.78) = 0.2959, p-val = 0.5900
## 
## Model Results:
## 
##                        estimate      se¹    tval¹     df¹    pval¹    ci.lb¹ 
## intrcpt                  0.0450  0.0167   2.6886   37.61   0.0106    0.0111  
## factor(Emotion_Cat)pa    0.0115  0.0212   0.5439   34.78   0.5900   -0.0315  
##                         ci.ub¹    
## intrcpt                0.0788   * 
## factor(Emotion_Cat)pa  0.0546     
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## 1) results based on cluster-robust inference (var-cov estimator: CR2,
##    approx t/F-tests and confidence intervals, df: Satterthwaite approx)

## 
## Multivariate Meta-Analysis Model (k = 68; method: REML)
## 
## Variance Components:
## 
## outer factor: ID                    (nlvls = 35)
## inner factor: factor(Study_Quality) (nlvls = 2)
## 
##             estim    sqrt  fixed 
## tau^2      0.0047  0.0684     no 
## rho        0.5000             no 
## 
## Test for Residual Heterogeneity:
## QE(df = 66) = 170.1374, p-val < .0001
## 
## Number of estimates:   68
## Number of clusters:    35
## Estimates per cluster: 1-4 (mean: 1.94, median: 2)
## 
## Test of Moderators (coefficient 2):¹
## F(df1 = 1, df2 = 7.18) = 0.0090, p-val = 0.9271
## 
## Model Results:
## 
##                            estimate      se¹    tval¹    df¹    pval¹    ci.lb¹ 
## intrcpt                      0.0588  0.0445   1.3208   5.22   0.2415   -0.0542  
## factor(Study_Quality)good    0.0045  0.0474   0.0948   7.18   0.9271   -0.1070  
##                             ci.ub¹    
## intrcpt                    0.1718     
## factor(Study_Quality)good  0.1159     
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## 1) results based on cluster-robust inference (var-cov estimator: CR2,
##    approx t/F-tests and confidence intervals, df: Satterthwaite approx)

## 
## Multivariate Meta-Analysis Model (k = 92; method: REML)
## 
## Variance Components:
## 
## outer factor: ID                  (nlvls = 46)
## inner factor: factor(Emotion_Cat) (nlvls = 2)
## 
##             estim    sqrt  fixed 
## tau^2      0.0067  0.0817     no 
## rho        0.9587             no 
## 
## Test for Residual Heterogeneity:
## QE(df = 90) = 181.2950, p-val < .0001
## 
## Number of estimates:   92
## Number of clusters:    46
## Estimates per cluster: 1-4 (mean: 2.00, median: 2)
## 
## Test of Moderators (coefficient 2):¹
## F(df1 = 1, df2 = 16.75) = 17.2601, p-val = 0.0007
## 
## Model Results:
## 
##                        estimate      se¹    tval¹     df¹    pval¹    ci.lb¹ 
## intrcpt                  0.0137  0.0150   0.9119   37.35   0.3676   -0.0167  
## factor(Emotion_Cat)pa    0.0598  0.0144   4.1545   16.75   0.0007    0.0294  
##                         ci.ub¹      
## intrcpt                0.0441       
## factor(Emotion_Cat)pa  0.0903   *** 
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## 1) results based on cluster-robust inference (var-cov estimator: CR2,
##    approx t/F-tests and confidence intervals, df: Satterthwaite approx)

## 
## Multivariate Meta-Analysis Model (k = 68; method: REML)
## 
## Variance Components:
## 
## outer factor: ID                    (nlvls = 35)
## inner factor: factor(Study_Quality) (nlvls = 2)
## 
##             estim    sqrt  fixed 
## tau^2      0.0062  0.0786     no 
## rho        0.5000             no 
## 
## Test for Residual Heterogeneity:
## QE(df = 66) = 167.7987, p-val < .0001
## 
## Number of estimates:   68
## Number of clusters:    35
## Estimates per cluster: 1-4 (mean: 1.94, median: 2)
## 
## Test of Moderators (coefficient 2):¹
## F(df1 = 1, df2 = 7.49) = 0.0038, p-val = 0.9525
## 
## Model Results:
## 
##                            estimate      se¹     tval¹    df¹    pval¹ 
## intrcpt                      0.0666  0.0552    1.2058   5.34   0.2786  
## factor(Study_Quality)good   -0.0036  0.0578   -0.0616   7.49   0.9525  
##                              ci.lb¹   ci.ub¹    
## intrcpt                    -0.0726   0.2057     
## factor(Study_Quality)good  -0.1385   0.1314     
## 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## 1) results based on cluster-robust inference (var-cov estimator: CR2,
##    approx t/F-tests and confidence intervals, df: Satterthwaite approx)

Summary

Here are the full results for all emotions, positive, and negative emotions using within-person SD and RVI

results_df %>% 
  knitr::kable() %>% 
  kable_styling(bootstrap_options = c("striped")) %>%
  column_spec(c(5, 9, 14, 18), width_min = "1.25in") %>%
  scroll_box(width = "900px")

trait	emo	cor_SD	p_SD	CI_SD	k_SD	Q_SD	Qp_SD	PI_SD	I2_SD	tau2_SD	cor_RVI	p_RVI	CI_RVI	k_RVI	Q_RVI	Qp_RVI	PI_RVI	I2_RVI	tau2_RVI	cor_BCLSM	p_BCLSM	CI_BCLSM	k_BCLSM	Q_BCLSM	Qp_BCLSM	PI_BCLSM	I2_BCLSM	tau2_BCLSM
A	all	-0.06	0.00028	[-0.08 - -0.03]	118	360.97	0	[-0.24 - 0.13]	71.26	0.01	0.02	0.12407	[-0.01 - 0.06]	103	264.74	0	[-0.16 - 0.21]	69.81	0.01	-0.02	0.21665	[-0.05 - 0.01]	103	271.44	0	[-0.21 - 0.17]	68.88	0.01
A	pa	-0.01	0.7135	[-0.04 - 0.03]	55	126.82	0	[-0.19 - 0.18]	64.2	0.01	0.01	0.60933	[-0.03 - 0.05]	48	126.12	0	[-0.2 - 0.22]	71.74	0.01	0	0.94373	[-0.04 - 0.04]	48	117.87	0	[-0.2 - 0.19]	68.34	0.01
A	na	-0.1	0	[-0.13 - -0.07]	57	142.86	0	[-0.26 - 0.07]	64.77	0.01	0.05	0.00373	[0.02 - 0.08]	49	106.51	0	[-0.11 - 0.21]	64.52	0.01	-0.03	0.1036	[-0.06 - 0.01]	49	130.38	0	[-0.21 - 0.15]	68.36	0.01
C	all	-0.07	0	[-0.09 - -0.05]	134	267.41	0	[-0.17 - 0.03]	51.19	0	0.02	0.21709	[-0.01 - 0.05]	111	324.93	0	[-0.16 - 0.2]	73.43	0.01	-0.03	0.04071	[-0.06 - 0]	109	243.67	0	[-0.21 - 0.15]	64.71	0.01
C	pa	-0.02	0.12861	[-0.04 - 0.01]	63	96.22	0.00349	[-0.11 - 0.08]	29.37	0	0	0.98598	[-0.04 - 0.03]	53	119.01	0	[-0.19 - 0.19]	65.68	0.01	-0.01	0.48149	[-0.04 - 0.02]	51	109.01	0	[-0.18 - 0.16]	62.92	0.01
C	na	-0.12	0	[-0.14 - -0.1]	65	85.63	0.03688	[-0.2 - -0.03]	28.47	0	0.05	0.0109	[0.01 - 0.09]	52	155.82	0	[-0.17 - 0.27]	75.48	0.01	-0.04	0.01286	[-0.07 - -0.01]	52	109.57	0	[-0.2 - 0.12]	61.98	0.01
E	all	0.01	0.31601	[-0.01 - 0.04]	137	440.6	0	[-0.15 - 0.17]	69.02	0.01	0.06	0	[0.04 - 0.09]	120	198.72	0.00001	[-0.06 - 0.19]	43.6	0	0.04	0.00455	[0.01 - 0.06]	120	257.9	0	[-0.12 - 0.19]	55.19	0.01
E	pa	0.07	0.00005	[0.04 - 0.1]	67	156.16	0	[-0.1 - 0.24]	61.45	0.01	0.05	0.00072	[0.02 - 0.08]	59	102.82	0.00026	[-0.08 - 0.18]	45.12	0	0.07	0.00002	[0.04 - 0.1]	59	108.91	0.00006	[-0.07 - 0.2]	47.67	0
E	na	-0.05	0.0005	[-0.07 - -0.02]	63	117.38	0.00003	[-0.18 - 0.08]	53.25	0	0.07	0	[0.05 - 0.1]	54	87.33	0.00208	[-0.04 - 0.19]	43.84	0	0	0.9991	[-0.03 - 0.03]	54	94.15	0.00043	[-0.13 - 0.13]	50.09	0
N	all	0.21	0	[0.19 - 0.23]	156	921.74	0	[0.07 - 0.35]	81.63	0.02	0	0.95411	[-0.02 - 0.03]	139	545.07	0	[-0.17 - 0.17]	75.03	0.01	0.12	0	[0.1 - 0.15]	138	430.96	0	[-0.05 - 0.29]	70.48	0.01
N	pa	0.08	0	[0.06 - 0.1]	72	126.79	0.00005	[-0.04 - 0.2]	47.4	0	0.06	0.00004	[0.03 - 0.09]	64	118.38	0.00003	[-0.08 - 0.2]	51.22	0	0.07	0	[0.04 - 0.1]	64	118.05	0.00003	[-0.06 - 0.21]	50.61	0
N	na	0.3	0	[0.28 - 0.32]	78	183.76	0	[0.16 - 0.44]	62	0.01	-0.06	0.00074	[-0.09 - -0.03]	69	256.13	0	[-0.27 - 0.16]	75.66	0.01	0.16	0	[0.13 - 0.19]	68	204.55	0	[-0.05 - 0.36]	74.15	0.01
O	all	0.03	0.04588	[0 - 0.06]	113	316.95	0	[-0.16 - 0.22]	67.63	0.01	0.05	0.00049	[0.02 - 0.08]	98	216.75	0	[-0.1 - 0.2]	60.61	0.01	0.04	0.00519	[0.01 - 0.07]	98	221.82	0	[-0.12 - 0.21]	58.65	0.01
O	pa	0.06	0.00184	[0.02 - 0.1]	53	146.81	0	[-0.14 - 0.27]	70.29	0.01	0.06	0.00213	[0.02 - 0.09]	46	106.16	0	[-0.12 - 0.23]	62.39	0.01	0.07	0.00053	[0.03 - 0.1]	46	107.77	0	[-0.11 - 0.24]	62.67	0.01
O	na	0.01	0.55243	[-0.02 - 0.04]	54	112.48	0	[-0.15 - 0.17]	60.18	0.01	0.05	0.00717	[0.01 - 0.08]	46	96.22	0.00001	[-0.11 - 0.2]	60.28	0.01	0.02	0.07638	[0 - 0.05]	46	73.52	0.00462	[-0.09 - 0.14]	42.86	0

rio::export(results_df, "final_results.RDS")

IV. Robustness and Supplementary Materials

1. Compare SD and RVI

To more accurately directly compare approaches using SD and RVI, we are refitting all models for within-person SD using the subset of the data for which raw data were available to calculate RVI.

# duplicate results_df structure
robust_df <- data.frame(
  trait = rep(c("A", "C", "E", "N", "O")),
  cor_SD = NA,
  p_SD = NA,
  CI_SD = NA,
  k_SD = NA,
  Q_SD = NA,
  Qp_SD = NA
)

for(trait in c("N", "E", "A", "C", "O")){
  meta_obj <- robust(rma.mv(yi = Cor_SD,
                            V = vi_SD,
                            data = dat[dat$Personality == trait & 
                                       !is.na(dat$Cor_RVI), ],
                            method = "REML",
                            level = 95, 
                            digits = 4,
                            slab = ID,
                            random = ~1 | ID),
                     clubSandwich = T,
                     cluster = ID)
  robust_df[robust_df$trait == trait, c("cor_SD", "p_SD", "CI_SD", "k_SD",
                                        "Q_SD", "Qp_SD")] <- c(
             round(meta_obj$b, 2), round(meta_obj$pval, 5), 
             paste0("[", round(meta_obj$ci.lb, 2), 
                    " - ", round(meta_obj$ci.ub, 2), "]"), 
             meta_obj$k,
             round(meta_obj$QE, 2), round(meta_obj$QEp, 5))
}

robust_df %>% knitr::kable() %>% kable_styling(bootstrap_options = c("striped"))

trait	cor_SD	p_SD	CI_SD	k_SD	Q_SD	Qp_SD
A	-0.06	0.00065	[-0.09 - -0.03]	103	315.52	0
C	-0.07	0	[-0.09 - -0.05]	111	230.87	0
E	0.01	0.41467	[-0.02 - 0.04]	120	382.75	0
N	0.2	0	[0.18 - 0.23]	139	775.56	0
O	0.02	0.1753	[-0.01 - 0.05]	98	273.63	0

2. Publication Bias

2a. With Emotion SD

for(trait in c("N", "E", "A", "C", "O")) {
  
  # get model object
  mod_obj <- get(paste0(trait, "_sd"))
  
  # funnel plot
  funnel(mod_obj,    
       xlim=c(-1,1), 
       xlab= paste("Correlations:", trait, "- emotion SD"),
       steps = 4, 
       digits = c(1, 2),
       back = "white")
  
  # test for funnel plot asymmetry
  print(ranktest(mod_obj))
}

## 
## Rank Correlation Test for Funnel Plot Asymmetry
## 
## Kendall's tau = -0.1639, p = 0.0023

## Warning in cor.test.default(yi.star, vi, method = "kendall", exact = exact):
## Cannot compute exact p-value with ties

## 
## Rank Correlation Test for Funnel Plot Asymmetry
## 
## Kendall's tau = 0.0944, p = 0.1019

## Warning in cor.test.default(yi.star, vi, method = "kendall", exact = exact):
## Cannot compute exact p-value with ties

## 
## Rank Correlation Test for Funnel Plot Asymmetry
## 
## Kendall's tau = -0.0380, p = 0.5423

## Warning in cor.test.default(yi.star, vi, method = "kendall", exact = exact):
## Cannot compute exact p-value with ties

## 
## Rank Correlation Test for Funnel Plot Asymmetry
## 
## Kendall's tau = 0.0588, p = 0.3135

## Warning in cor.test.default(yi.star, vi, method = "kendall", exact = exact):
## Cannot compute exact p-value with ties

## 
## Rank Correlation Test for Funnel Plot Asymmetry
## 
## Kendall's tau = -0.0754, p = 0.2366

2b. With Emotion RVI

for(trait in c("N", "E", "A", "C", "O")) {
  
  # get model object
  mod_obj <- get(paste0(trait, "_rvi"))
  
  # funnel plot
  funnel(mod_obj,    
       xlim=c(-1,1), 
       xlab= paste("Correlations:", trait, "- emotion RVI"),
       steps = 4, 
       digits = c(1, 2),
       back = "white")
  
  # test for funnel plot asymmetry
  print(ranktest(mod_obj))
}

## 
## Rank Correlation Test for Funnel Plot Asymmetry
## 
## Kendall's tau = -0.0495, p = 0.3888

## 
## Rank Correlation Test for Funnel Plot Asymmetry
## 
## Kendall's tau = 0.0754, p = 0.2237

## 
## Rank Correlation Test for Funnel Plot Asymmetry
## 
## Kendall's tau = -0.0695, p = 0.3004

## 
## Rank Correlation Test for Funnel Plot Asymmetry
## 
## Kendall's tau = 0.0143, p = 0.8269

## 
## Rank Correlation Test for Funnel Plot Asymmetry
## 
## Kendall's tau = -0.0848, p = 0.2179

2c. With Emotion BCLSM

for(trait in c("N", "E", "A", "C", "O")) {
  
  # get model object
  mod_obj <- get(paste0(trait, "_bclsm"))
  
  # funnel plot
  funnel(mod_obj,    
       xlim=c(-1,1), 
       xlab= paste("Correlations:", trait, "- emotion BCLSM"),
       steps = 4, 
       digits = c(1, 2),
       back = "white")
  
  # test for funnel plot asymmetry
  print(ranktest(mod_obj))
}

## Warning in cor.test.default(yi.star, vi, method = "kendall", exact = exact):
## Cannot compute exact p-value with ties

## 
## Rank Correlation Test for Funnel Plot Asymmetry
## 
## Kendall's tau = -0.1560, p = 0.0067

## Warning in cor.test.default(yi.star, vi, method = "kendall", exact = exact):
## Cannot compute exact p-value with ties

## 
## Rank Correlation Test for Funnel Plot Asymmetry
## 
## Kendall's tau = 0.1073, p = 0.0823

## Warning in cor.test.default(yi.star, vi, method = "kendall", exact = exact):
## Cannot compute exact p-value with ties

## 
## Rank Correlation Test for Funnel Plot Asymmetry
## 
## Kendall's tau = -0.0543, p = 0.4167

## Warning in cor.test.default(yi.star, vi, method = "kendall", exact = exact):
## Cannot compute exact p-value with ties

## 
## Rank Correlation Test for Funnel Plot Asymmetry
## 
## Kendall's tau = 0.0602, p = 0.3539

## Warning in cor.test.default(yi.star, vi, method = "kendall", exact = exact):
## Cannot compute exact p-value with ties

## 
## Rank Correlation Test for Funnel Plot Asymmetry
## 
## Kendall's tau = -0.0794, p = 0.2472

3. Robust Variance Estimation with {robumeta}

# duplicate results_df structure
robumeta_df <- data.frame(
  trait = rep(c("A", "C", "E", "N", "O"), each = 3),
  emo = c("all", "pa", "na"),
  cor_SD = NA,
  p_SD = NA,
  CI_SD = NA,
  k_SD = NA,
  cor_RVI = NA,
  p_RVI = NA,
  CI_RVI = NA,
  k_RVI = NA,
  cor_BCLSM = NA,
  p_BCLSM = NA,
  CI_BCLSM = NA,
  k_BCLSM = NA
)
for(trait in c("N", "E", "A", "C", "O")) {
  robumeta_df <- fit_robumeta(trait = trait, .data = dat,
                              output = robumeta_df)
}

robumeta_df %>% 
  knitr::kable() %>% 
  kable_styling(bootstrap_options = c("striped"))

trait	emo	cor_SD	p_SD	CI_SD	k_SD	cor_RVI	p_RVI	CI_RVI	k_RVI	cor_BCLSM	p_BCLSM	CI_BCLSM	k_BCLSM
A	all	-0.05	0.00014	[-0.08 - -0.03]	58	0.02	0.13531	[-0.01 - 0.05]	53	-0.02	0.19379	[-0.05 - 0.01]	53
A	pa	0	0.87177	[-0.04 - 0.03]	51	0.01	0.61568	[-0.03 - 0.05]	46	0	0.97278	[-0.04 - 0.04]	46
A	na	-0.1	0	[-0.13 - -0.07]	50	0.05	0.00302	[0.02 - 0.08]	45	-0.03	0.12066	[-0.06 - 0.01]	45
C	all	-0.07	0	[-0.09 - -0.05]	63	0.02	0.27244	[-0.01 - 0.04]	58	-0.03	0.01655	[-0.06 - -0.01]	57
C	pa	-0.01	0.20345	[-0.04 - 0.01]	55	0	0.92785	[-0.04 - 0.03]	50	-0.01	0.51245	[-0.04 - 0.02]	49
C	na	-0.12	0	[-0.14 - -0.09]	55	0.05	0.00889	[0.01 - 0.09]	49	-0.04	0.01029	[-0.07 - -0.01]	49
E	all	0.01	0.30154	[-0.01 - 0.04]	72	0.07	0	[0.04 - 0.09]	66	0.04	0.00474	[0.01 - 0.06]	66
E	pa	0.07	0.00003	[0.04 - 0.1]	63	0.05	0.00033	[0.03 - 0.08]	57	0.07	0.00001	[0.04 - 0.1]	57
E	na	-0.05	0.00078	[-0.07 - -0.02]	57	0.07	0	[0.05 - 0.1]	51	0	0.97006	[-0.03 - 0.03]	51
N	all	0.19	0	[0.18 - 0.21]	79	0	0.71545	[-0.02 - 0.03]	73	0.12	0	[0.1 - 0.15]	72
N	pa	0.08	0	[0.06 - 0.1]	68	0.06	0.00002	[0.04 - 0.09]	62	0.07	0	[0.04 - 0.1]	62
N	na	0.3	0	[0.28 - 0.32]	71	-0.06	0.00081	[-0.09 - -0.03]	65	0.16	0	[0.13 - 0.19]	64
O	all	0.03	0.01999	[0.01 - 0.06]	56	0.05	0.00033	[0.02 - 0.08]	51	0.04	0.00276	[0.02 - 0.07]	51
O	pa	0.06	0.00101	[0.03 - 0.1]	49	0.06	0.00221	[0.02 - 0.1]	44	0.07	0.00044	[0.03 - 0.1]	44
O	na	0.01	0.4969	[-0.02 - 0.04]	48	0.05	0.00729	[0.01 - 0.08]	43	0.02	0.09496	[0 - 0.05]	43

4. Personality and Mean Affects

Although this meta-analysis did not focus on the relationship between personality traits and mean level of affect, this association can be examined using the raw data that we obtained and are presented here for further context in understanding our patterns of results.

# initialize table for all results
mean_df <- data.frame(
  trait = rep(c("A", "C", "E", "N", "O"), each = 3),
  emo = c("all", "pa", "na"),
  cor_Mean = NA,
  p_Mean = NA,
  CI_Mean = NA,
  k_Mean = NA,
  Q_Mean = NA,
  Qp_Mean = NA,
  PI_Mean = NA,
  I2_Mean = NA,
  tau2_Mean = NA
)

Here are the full results for all emotions, positive, and negative emotions using within-person Means

mean_df %>% 
  knitr::kable() %>% 
  kable_styling(bootstrap_options = c("striped")) %>%
  column_spec(c(5, 9), width_min = "1.25in") %>%
  scroll_box(width = "900px")

trait	emo	cor_Mean	p_Mean	CI_Mean	k_Mean	Q_Mean	Qp_Mean	PI_Mean	I2_Mean	tau2_Mean
A	all	0.01	0.60184	[-0.02 - 0.03]	103	1586.27	0	[-0.1 - 0.11]	93.55	0.05
A	pa	0.2	0	[0.15 - 0.24]	48	231.76	0	[-0.05 - 0.44]	83.1	0.02
A	na	-0.19	0	[-0.23 - -0.14]	49	226.57	0	[-0.41 - 0.04]	84.79	0.02
C	all	0.01	0.23951	[-0.01 - 0.04]	109	2078.06	0	[-0.11 - 0.14]	94.42	0.06
C	pa	0.2	0	[0.16 - 0.25]	51	298.95	0	[-0.06 - 0.47]	89.09	0.03
C	na	-0.18	0	[-0.22 - -0.14]	52	338.82	0	[-0.44 - 0.08]	89.01	0.03
E	all	0.13	0	[0.09 - 0.17]	120	2397.52	0	[-0.17 - 0.43]	93.52	0.05
E	pa	0.28	0	[0.23 - 0.33]	59	446.25	0	[-0.02 - 0.58]	87.46	0.03
E	na	-0.11	0	[-0.15 - -0.08]	54	154.86	0	[-0.31 - 0.09]	73.82	0.01
N	all	0.06	0.00574	[0.02 - 0.1]	138	6898.32	0	[-0.28 - 0.4]	97.77	0.13
N	pa	-0.3	0	[-0.33 - -0.26]	64	299.68	0	[-0.54 - -0.06]	85.34	0.02
N	na	0.38	0	[0.35 - 0.42]	68	724.31	0	[0.15 - 0.62]	92.16	0.03
O	all	0.06	0.00387	[0.02 - 0.09]	98	927.34	0	[-0.17 - 0.28]	88.9	0.03
O	pa	0.15	0	[0.09 - 0.2]	46	399.55	0	[-0.18 - 0.48]	89.18	0.03
O	na	-0.04	0.04149	[-0.08 - 0]	46	151.91	0	[-0.25 - 0.17]	78.89	0.01

5. Meta-Traits SEM Model

load("../Data/metaSEM.RData")

5a. Positive Affect

#### SD ####
pa_sd_fixed <- tssem1(pa_sd_data, pa_sd_n, method="FEM")

sd_model <- "## Factor loadings
             Stability =~ A + C + N
             Plasticity =~ E + O
             ## Factor predictions
             EmoSD ~ Stability
             EmoSD ~ Plasticity
             ## Factor correlation
             Stability ~~ Plasticity"

sd_RAM <- lavaan2RAM(sd_model, obs.variables=c("A","C","E","N","O","EmoSD"),
                     A.notation="on", S.notation="with")
pa_sd_fixed2 <- tssem2(pa_sd_fixed, RAM=sd_RAM, intervals="z")
summary(pa_sd_fixed2)

## 
## Call:
## wls(Cov = coef.tssem1FEM(tssem1.obj), aCov = vcov.tssem1FEM(tssem1.obj), 
##     n = sum(tssem1.obj$n), RAM = RAM, Amatrix = Amatrix, Smatrix = Smatrix, 
##     Fmatrix = Fmatrix, diag.constraints = diag.constraints, cor.analysis = tssem1.obj$cor.analysis, 
##     intervals.type = intervals.type, mx.algebras = mx.algebras, 
##     mxModel.Args = mxModel.Args, subset.variables = subset.variables, 
##     model.name = model.name, suppressWarnings = suppressWarnings, 
##     silent = silent, run = run)
## 
## 95% confidence intervals: z statistic approximation
## Coefficients:
##                          Estimate Std.Error    lbound    ubound  z value
## AonStability             0.473171  0.011321  0.450981  0.495360  41.7946
## ConStability             0.431907  0.011229  0.409898  0.453915  38.4633
## EonPlasticity            0.769474  0.020475  0.729344  0.809604  37.5814
## EmoSDonPlasticity        0.263130  0.023993  0.216104  0.310156  10.9668
## EmoSDonStability        -0.221771  0.023282 -0.267402 -0.176140  -9.5256
## NonStability            -0.636689  0.011688 -0.659597 -0.613780 -54.4733
## OonPlasticity            0.356359  0.012635  0.331596  0.381122  28.2052
## StabilitywithPlasticity  0.621138  0.019179  0.583547  0.658729  32.3857
##                                      Pr(>|z|)    
## AonStability            < 0.00000000000000022 ***
## ConStability            < 0.00000000000000022 ***
## EonPlasticity           < 0.00000000000000022 ***
## EmoSDonPlasticity       < 0.00000000000000022 ***
## EmoSDonStability        < 0.00000000000000022 ***
## NonStability            < 0.00000000000000022 ***
## OonPlasticity           < 0.00000000000000022 ***
## StabilitywithPlasticity < 0.00000000000000022 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Goodness-of-fit indices:
##                                                 Value
## Sample size                                11353.0000
## Chi-square of target model                   136.8936
## DF of target model                             7.0000
## p value of target model                        0.0000
## Number of constraints imposed on "Smatrix"     0.0000
## DF manually adjusted                           0.0000
## Chi-square of independence model            3936.1566
## DF of independence model                      15.0000
## RMSEA                                          0.0404
## RMSEA lower 95% CI                             0.0347
## RMSEA upper 95% CI                             0.0465
## SRMR                                           0.0238
## TLI                                            0.9290
## CFI                                            0.9669
## AIC                                          122.8936
## BIC                                           71.5330
## OpenMx status1: 0 ("0" or "1": The optimization is considered fine.
## Other values indicate problems.)

plot(pa_sd_fixed2, color="#E69F00", what = "std")

#### RVI ####
pa_rvi_fixed <- tssem1(pa_rvi_data, pa_rvi_n, method="FEM")

rvi_model <- "## Factor loadings
              Stability =~ A + C + N
              Plasticity =~ E + O
              ## Factor predictions
              EmoRVI ~ Stability
              EmoRVI ~ Plasticity
              ## Factor correlation
              Stability ~~ Plasticity"

rvi_RAM <- lavaan2RAM(rvi_model, obs.variables=c("A","C","E","N","O","EmoRVI"),
                      A.notation="on", S.notation="with")
pa_rvi_fixed2 <- tssem2(pa_rvi_fixed, RAM=rvi_RAM, intervals="z")
summary(pa_rvi_fixed2)

## 
## Call:
## wls(Cov = coef.tssem1FEM(tssem1.obj), aCov = vcov.tssem1FEM(tssem1.obj), 
##     n = sum(tssem1.obj$n), RAM = RAM, Amatrix = Amatrix, Smatrix = Smatrix, 
##     Fmatrix = Fmatrix, diag.constraints = diag.constraints, cor.analysis = tssem1.obj$cor.analysis, 
##     intervals.type = intervals.type, mx.algebras = mx.algebras, 
##     mxModel.Args = mxModel.Args, subset.variables = subset.variables, 
##     model.name = model.name, suppressWarnings = suppressWarnings, 
##     silent = silent, run = run)
## 
## 95% confidence intervals: z statistic approximation
## Coefficients:
##                          Estimate Std.Error    lbound    ubound  z value
## AonStability             0.473235  0.011332  0.451024  0.495446  41.7605
## ConStability             0.431988  0.011240  0.409957  0.454019  38.4319
## EonPlasticity            0.777162  0.021523  0.734978  0.819346  36.1084
## EmoRVIonPlasticity       0.194151  0.022072  0.150890  0.237412   8.7961
## EmoRVIonStability       -0.191677  0.021685 -0.234179 -0.149175  -8.8392
## NonStability            -0.635725  0.011694 -0.658645 -0.612805 -54.3635
## OonPlasticity            0.352891  0.012912  0.327584  0.378198  27.3306
## StabilitywithPlasticity  0.616449  0.019575  0.578082  0.654816  31.4909
##                                      Pr(>|z|)    
## AonStability            < 0.00000000000000022 ***
## ConStability            < 0.00000000000000022 ***
## EonPlasticity           < 0.00000000000000022 ***
## EmoRVIonPlasticity      < 0.00000000000000022 ***
## EmoRVIonStability       < 0.00000000000000022 ***
## NonStability            < 0.00000000000000022 ***
## OonPlasticity           < 0.00000000000000022 ***
## StabilitywithPlasticity < 0.00000000000000022 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Goodness-of-fit indices:
##                                                 Value
## Sample size                                11353.0000
## Chi-square of target model                   136.1270
## DF of target model                             7.0000
## p value of target model                        0.0000
## Number of constraints imposed on "Smatrix"     0.0000
## DF manually adjusted                           0.0000
## Chi-square of independence model            3831.0264
## DF of independence model                      15.0000
## RMSEA                                          0.0403
## RMSEA lower 95% CI                             0.0346
## RMSEA upper 95% CI                             0.0464
## SRMR                                           0.0239
## TLI                                            0.9275
## CFI                                            0.9662
## AIC                                          122.1270
## BIC                                           70.7663
## OpenMx status1: 0 ("0" or "1": The optimization is considered fine.
## Other values indicate problems.)

plot(pa_rvi_fixed2, color="#E69F00", what = "std")

#### BCLSM ####
pa_bclsm_fixed <- tssem1(pa_bclsm_data, pa_bclsm_n, method="FEM")

bclsm_model <- "## Factor loadings
                Stability =~ A + C + N
                Plasticity =~ E + O
                ## Factor predictions
                EmoBCLSM ~ Stability
                EmoBCLSM ~ Plasticity
                ## Factor correlation
                Stability ~~ Plasticity"

bclsm_RAM <- lavaan2RAM(bclsm_model, obs.variables=c("A","C","E","N","O","EmoBCLSM"),
                        A.notation="on", S.notation="with")
pa_bclsm_fixed2 <- tssem2(pa_bclsm_fixed, RAM=bclsm_RAM, intervals="z")
summary(pa_bclsm_fixed2)

## 
## Call:
## wls(Cov = coef.tssem1FEM(tssem1.obj), aCov = vcov.tssem1FEM(tssem1.obj), 
##     n = sum(tssem1.obj$n), RAM = RAM, Amatrix = Amatrix, Smatrix = Smatrix, 
##     Fmatrix = Fmatrix, diag.constraints = diag.constraints, cor.analysis = tssem1.obj$cor.analysis, 
##     intervals.type = intervals.type, mx.algebras = mx.algebras, 
##     mxModel.Args = mxModel.Args, subset.variables = subset.variables, 
##     model.name = model.name, suppressWarnings = suppressWarnings, 
##     silent = silent, run = run)
## 
## 95% confidence intervals: z statistic approximation
## Coefficients:
##                          Estimate Std.Error    lbound    ubound  z value
## AonStability             0.473685  0.011307  0.451523  0.495847  41.8916
## ConStability             0.432258  0.011220  0.410267  0.454249  38.5252
## EonPlasticity            0.765933  0.020329  0.726089  0.805777  37.6768
## EmoBCLSMonPlasticity     0.262166  0.024226  0.214684  0.309647  10.8218
## EmoBCLSMonStability     -0.231069  0.023420 -0.276971 -0.185167  -9.8664
## NonStability            -0.635099  0.011657 -0.657946 -0.612253 -54.4835
## OonPlasticity            0.358656  0.012671  0.333820  0.383491  28.3043
## StabilitywithPlasticity  0.623754  0.019146  0.586229  0.661279  32.5793
##                                      Pr(>|z|)    
## AonStability            < 0.00000000000000022 ***
## ConStability            < 0.00000000000000022 ***
## EonPlasticity           < 0.00000000000000022 ***
## EmoBCLSMonPlasticity    < 0.00000000000000022 ***
## EmoBCLSMonStability     < 0.00000000000000022 ***
## NonStability            < 0.00000000000000022 ***
## OonPlasticity           < 0.00000000000000022 ***
## StabilitywithPlasticity < 0.00000000000000022 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Goodness-of-fit indices:
##                                                 Value
## Sample size                                11353.0000
## Chi-square of target model                   138.5594
## DF of target model                             7.0000
## p value of target model                        0.0000
## Number of constraints imposed on "Smatrix"     0.0000
## DF manually adjusted                           0.0000
## Chi-square of independence model            3921.4603
## DF of independence model                      15.0000
## RMSEA                                          0.0407
## RMSEA lower 95% CI                             0.0349
## RMSEA upper 95% CI                             0.0467
## SRMR                                           0.0241
## TLI                                            0.9278
## CFI                                            0.9663
## AIC                                          124.5594
## BIC                                           73.1987
## OpenMx status1: 0 ("0" or "1": The optimization is considered fine.
## Other values indicate problems.)

plot(pa_bclsm_fixed2, color="#E69F00", what = "std")

5b. Negative Affect

#### SD ####
na_sd_fixed <- tssem1(na_sd_data, na_sd_n, method="FEM")

na_sd_fixed2 <- tssem2(na_sd_fixed, RAM=sd_RAM, intervals="z")
summary(na_sd_fixed2)

## 
## Call:
## wls(Cov = coef.tssem1FEM(tssem1.obj), aCov = vcov.tssem1FEM(tssem1.obj), 
##     n = sum(tssem1.obj$n), RAM = RAM, Amatrix = Amatrix, Smatrix = Smatrix, 
##     Fmatrix = Fmatrix, diag.constraints = diag.constraints, cor.analysis = tssem1.obj$cor.analysis, 
##     intervals.type = intervals.type, mx.algebras = mx.algebras, 
##     mxModel.Args = mxModel.Args, subset.variables = subset.variables, 
##     model.name = model.name, suppressWarnings = suppressWarnings, 
##     silent = silent, run = run)
## 
## 95% confidence intervals: z statistic approximation
## Coefficients:
##                          Estimate Std.Error    lbound    ubound z value
## AonStability            -0.430858  0.010118 -0.450690 -0.411026 -42.581
## ConStability            -0.406421  0.010124 -0.426264 -0.386578 -40.143
## EonPlasticity            0.790943  0.021938  0.747946  0.833941  36.053
## EmoSDonPlasticity        0.214120  0.021334  0.172307  0.255934  10.037
## EmoSDonStability         0.499876  0.020134  0.460413  0.539338  24.827
## NonStability             0.703086  0.010275  0.682948  0.723224  68.430
## OonPlasticity            0.333950  0.012184  0.310071  0.357829  27.410
## StabilitywithPlasticity -0.590124  0.018905 -0.627177 -0.553070 -31.215
##                                      Pr(>|z|)    
## AonStability            < 0.00000000000000022 ***
## ConStability            < 0.00000000000000022 ***
## EonPlasticity           < 0.00000000000000022 ***
## EmoSDonPlasticity       < 0.00000000000000022 ***
## EmoSDonStability        < 0.00000000000000022 ***
## NonStability            < 0.00000000000000022 ***
## OonPlasticity           < 0.00000000000000022 ***
## StabilitywithPlasticity < 0.00000000000000022 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Goodness-of-fit indices:
##                                                 Value
## Sample size                                12846.0000
## Chi-square of target model                   254.3705
## DF of target model                             7.0000
## p value of target model                        0.0000
## Number of constraints imposed on "Smatrix"     0.0000
## DF manually adjusted                           0.0000
## Chi-square of independence model            5392.6121
## DF of independence model                      15.0000
## RMSEA                                          0.0525
## RMSEA lower 95% CI                             0.0470
## RMSEA upper 95% CI                             0.0581
## SRMR                                           0.0324
## TLI                                            0.9014
## CFI                                            0.9540
## AIC                                          240.3705
## BIC                                          188.1450
## OpenMx status1: 0 ("0" or "1": The optimization is considered fine.
## Other values indicate problems.)

plot(na_sd_fixed2, color="#710c0c", what = "std")

#### RVI ####
na_rvi_fixed <- tssem1(na_rvi_data, na_rvi_n, method="FEM")

na_rvi_fixed2 <- tssem2(na_rvi_fixed, RAM=rvi_RAM, intervals="z")
summary(na_rvi_fixed2)

## 
## Call:
## wls(Cov = coef.tssem1FEM(tssem1.obj), aCov = vcov.tssem1FEM(tssem1.obj), 
##     n = sum(tssem1.obj$n), RAM = RAM, Amatrix = Amatrix, Smatrix = Smatrix, 
##     Fmatrix = Fmatrix, diag.constraints = diag.constraints, cor.analysis = tssem1.obj$cor.analysis, 
##     intervals.type = intervals.type, mx.algebras = mx.algebras, 
##     mxModel.Args = mxModel.Args, subset.variables = subset.variables, 
##     model.name = model.name, suppressWarnings = suppressWarnings, 
##     silent = silent, run = run)
## 
## 95% confidence intervals: z statistic approximation
## Coefficients:
##                          Estimate Std.Error    lbound    ubound  z value
## AonStability             0.468984  0.010697  0.448018  0.489949  43.8432
## ConStability             0.437138  0.010630  0.416303  0.457972  41.1224
## EonPlasticity            0.778905  0.020919  0.737904  0.819905  37.2343
## EmoRVIonPlasticity       0.060348  0.018090  0.024891  0.095804   3.3359
## EmoRVIonStability        0.068833  0.018933  0.031726  0.105941   3.6357
## NonStability            -0.626232  0.011002 -0.647795 -0.604668 -56.9190
## OonPlasticity            0.344360  0.012103  0.320640  0.368081  28.4534
## StabilitywithPlasticity  0.621126  0.019003  0.583880  0.658372  32.6849
##                                      Pr(>|z|)    
## AonStability            < 0.00000000000000022 ***
## ConStability            < 0.00000000000000022 ***
## EonPlasticity           < 0.00000000000000022 ***
## EmoRVIonPlasticity                  0.0008503 ***
## EmoRVIonStability                   0.0002773 ***
## NonStability            < 0.00000000000000022 ***
## OonPlasticity           < 0.00000000000000022 ***
## StabilitywithPlasticity < 0.00000000000000022 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Goodness-of-fit indices:
##                                                 Value
## Sample size                                12846.0000
## Chi-square of target model                   101.0201
## DF of target model                             7.0000
## p value of target model                        0.0000
## Number of constraints imposed on "Smatrix"     0.0000
## DF manually adjusted                           0.0000
## Chi-square of independence model            4047.3962
## DF of independence model                      15.0000
## RMSEA                                          0.0323
## RMSEA lower 95% CI                             0.0269
## RMSEA upper 95% CI                             0.0381
## SRMR                                           0.0171
## TLI                                            0.9500
## CFI                                            0.9767
## AIC                                           87.0201
## BIC                                           34.7946
## OpenMx status1: 0 ("0" or "1": The optimization is considered fine.
## Other values indicate problems.)

plot(na_rvi_fixed2, color="#710c0c", what = "std")

#### BCLSM ####
na_bclsm_fixed <- tssem1(na_bclsm_data, na_bclsm_n, method="FEM")

na_bclsm_fixed2 <- tssem2(na_bclsm_fixed, RAM=bclsm_RAM, intervals="z")
summary(na_bclsm_fixed2)

## 
## Call:
## wls(Cov = coef.tssem1FEM(tssem1.obj), aCov = vcov.tssem1FEM(tssem1.obj), 
##     n = sum(tssem1.obj$n), RAM = RAM, Amatrix = Amatrix, Smatrix = Smatrix, 
##     Fmatrix = Fmatrix, diag.constraints = diag.constraints, cor.analysis = tssem1.obj$cor.analysis, 
##     intervals.type = intervals.type, mx.algebras = mx.algebras, 
##     mxModel.Args = mxModel.Args, subset.variables = subset.variables, 
##     model.name = model.name, suppressWarnings = suppressWarnings, 
##     silent = silent, run = run)
## 
## 95% confidence intervals: z statistic approximation
## Coefficients:
##                          Estimate Std.Error    lbound    ubound  z value
## AonStability            -0.449572  0.010501 -0.470154 -0.428990 -42.8118
## ConStability            -0.419175  0.010472 -0.439700 -0.398650 -40.0276
## EonPlasticity            0.783714  0.021244  0.742076  0.825351  36.8907
## EmoBCLSMonPlasticity     0.189593  0.020655  0.149110  0.230076   9.1791
## EmoBCLSMonStability      0.329433  0.020125  0.289990  0.368877  16.3696
## NonStability             0.670217  0.010905  0.648843  0.691590  61.4586
## OonPlasticity            0.339828  0.012116  0.316081  0.363576  28.0470
## StabilitywithPlasticity -0.606806  0.018907 -0.643864 -0.569748 -32.0935
##                                      Pr(>|z|)    
## AonStability            < 0.00000000000000022 ***
## ConStability            < 0.00000000000000022 ***
## EonPlasticity           < 0.00000000000000022 ***
## EmoBCLSMonPlasticity    < 0.00000000000000022 ***
## EmoBCLSMonStability     < 0.00000000000000022 ***
## NonStability            < 0.00000000000000022 ***
## OonPlasticity           < 0.00000000000000022 ***
## StabilitywithPlasticity < 0.00000000000000022 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Goodness-of-fit indices:
##                                                 Value
## Sample size                                12846.0000
## Chi-square of target model                   210.2549
## DF of target model                             7.0000
## p value of target model                        0.0000
## Number of constraints imposed on "Smatrix"     0.0000
## DF manually adjusted                           0.0000
## Chi-square of independence model            4643.0578
## DF of independence model                      15.0000
## RMSEA                                          0.0475
## RMSEA lower 95% CI                             0.0421
## RMSEA upper 95% CI                             0.0532
## SRMR                                           0.0288
## TLI                                            0.9059
## CFI                                            0.9561
## AIC                                          196.2549
## BIC                                          144.0294
## OpenMx status1: 0 ("0" or "1": The optimization is considered fine.
## Other values indicate problems.)

plot(na_bclsm_fixed2, color="#710c0c", what = "std")