1A: Loading data, setting up
library(prcr)
library(tidyverse)
ctk <- read_csv("~/downloads/ctk.csv")
esm_df <- haven::read_sav('/Users/joshuarosenberg/Dropbox/1_Research/SciMo/JRST\ MEP\ Paper\ Final\ Docs/esm_data.sav')
to_remove <- esm_df$stud_ID == "AXR050594" & esm_df$month == 10 & esm_df$day == 20 & esm_df$year == 8 & esm_df$signal == 2 # removing 10/20/8, ID = "AXR050594", second beeper
esm_df <- esm_df[!to_remove, ]
df <- esm_df %>%
mutate(behavioral_scale = ifelse(!is.na(conc) & !is.na(hardwk), (conc + hardwk) / 2,
ifelse(is.na(conc), hardwk, conc)),
cognitive_scale = ifelse(!is.na(imp_y) & !is.na(imp_fut), (imp_y + imp_fut) / 2,
ifelse(is.na(imp_y), imp_fut, imp_y)),
affective_scale = ifelse(!is.na(enjoy) & !is.na(interest), (enjoy + interest) / 2,
ifelse(is.na(enjoy), interest, enjoy)))
df <- df[ctk$ctk, ]
p6 <- create_profiles(df,
behavioral_scale,
cognitive_scale,
affective_scale,
n_profiles = 6,
to_center = T,
to_scale = T)
knitr::opts_chunk$set(cache=TRUE)
cv_output <- cross_validate(df,
behavioral_scale,
cognitive_scale,
affective_scale,
n_profiles = 6,
to_center = T,
to_scale = T,
k = 30)
knitr::kable(cv_output)
| 1 |
0.44 |
0.53 |
| 2 |
0.75 |
0.79 |
| 3 |
0.71 |
0.76 |
| 4 |
0.90 |
0.92 |
| 5 |
0.57 |
0.64 |
| 6 |
0.55 |
0.62 |
| 7 |
0.63 |
0.69 |
| 8 |
0.74 |
0.78 |
| 9 |
0.49 |
0.57 |
| 10 |
0.71 |
0.76 |
| 11 |
0.92 |
0.93 |
| 12 |
0.56 |
0.63 |
| 13 |
0.85 |
0.88 |
| 14 |
0.65 |
0.71 |
| 15 |
0.54 |
0.62 |
| 16 |
0.66 |
0.72 |
| 17 |
0.70 |
0.75 |
| 18 |
0.61 |
0.67 |
| 19 |
0.58 |
0.65 |
| 20 |
0.68 |
0.73 |
| 21 |
0.59 |
0.66 |
| 22 |
0.89 |
0.91 |
| 23 |
0.67 |
0.72 |
| 24 |
0.62 |
0.69 |
| 25 |
0.68 |
0.74 |
| 26 |
0.92 |
0.93 |
| 27 |
0.63 |
0.69 |
| 28 |
0.68 |
0.73 |
| 29 |
0.79 |
0.83 |
| 30 |
0.60 |
0.67 |
1B: Processing data
library(tidyverse)
library(lme4)
to_m <- p6$.data
dcd <- p6$data_with_dummy_codes
dff <- bind_cols(to_m, dcd)
dff$response_date <- paste0(dff$month, "-", dff$day)
dff$beep_ID <- paste0(dff$response_date, dff$signal, dff$teacher_ID)
df_ss <- select(dff, contains("cluster"), -cluster, teacher_ID, stud_ID, beep_ID)
2A: Null / Variance Components Models
knitr::opts_chunk$set(cache=TRUE)
m1 <- glmer(cluster_1 ~ 1 +
(1|teacher_ID) + (1|stud_ID) + (1|beep_ID),
family = binomial(link = "logit"),
control=lme4::glmerControl(optimizer="bobyqa",
optCtrl=list(maxfun=2e5)),
data = df_ss)
m2 <- glmer(cluster_2 ~ 1 +
(1|teacher_ID) + (1|stud_ID) + (1|beep_ID),
family = binomial(link = "logit"),
control=lme4::glmerControl(optimizer="bobyqa",
optCtrl=list(maxfun=2e5)),
data = df_ss)
m3 <- glmer(cluster_3 ~ 1 +
(1|teacher_ID) + (1|stud_ID) + (1|beep_ID),
family = binomial(link = "logit"),
control=lme4::glmerControl(optimizer="bobyqa",
optCtrl=list(maxfun=2e5)),
data = df_ss)
m4 <- glmer(cluster_4 ~ 1 +
(1|teacher_ID) + (1|stud_ID) + (1|beep_ID),
family = binomial(link = "logit"),
control=lme4::glmerControl(optimizer="bobyqa",
optCtrl=list(maxfun=2e5)),
data = df_ss)
m5 <- glmer(cluster_5 ~ 1 +
(1|teacher_ID) + (1|stud_ID) + (1|beep_ID),
family = binomial(link = "logit"),
control=lme4::glmerControl(optimizer="bobyqa",
optCtrl=list(maxfun=2e5)),
data = df_ss)
m6 <- glmer(cluster_6 ~ 1 +
(1|teacher_ID) + (1|stud_ID) + (1|beep_ID),
family = binomial(link = "logit"),
control=lme4::glmerControl(optimizer="bobyqa",
optCtrl=list(maxfun=2e5)),
data = df_ss)
xm1 <- sjstats::icc(m1)
xm2 <- sjstats::icc(m2)
xm3 <- sjstats::icc(m3)
xm4 <- sjstats::icc(m4)
xm5 <- sjstats::icc(m5)
xm6 <- sjstats::icc(m6)
2B: ICCs for each null model with plot of profiles
plot(p6)
df_with_ICCs <- data.frame(profile_1 = xm1,
profile_2 = xm2,
profile_3 = xm3,
profile_4 = xm4,
profile_5 = xm5,
profile_6 = xm6)
row.names(df_with_ICCs) <- c("Momentary RE", "Student RE", "Teacher RE")
knitr::kable(df_with_ICCs, digits = 4)
| Momentary RE |
0.0000000 |
0.12122591 |
0.08742863 |
0.006808985 |
0.05680992 |
0.04423836 |
| Student RE |
0.3951807 |
0.27105319 |
0.46877232 |
0.303127910 |
0.40472867 |
0.17004291 |
| Teacher RE |
0.0000000 |
0.04996376 |
0.00000000 |
0.005911475 |
0.01024088 |
0.01520584 |
3A: Preparation for analysis with MLMs
student_df <- haven::read_sav("~/documents/myscimo/scimo_student_survey.sav")
student_df <- rename(student_df, stud_ID = Stud_ID)
student_df <- select(student_df, -teacher_ID)
teacher_df <- haven::read_sav("~/documents/myscimo/scimo_teacher_survey.sav")
student_df$stud_ID <- as.vector(student_df$stud_ID)
df_ss$stud_ID <- as.vector(df_ss$stud_ID)
df_joined <- left_join(df_ss, student_df, by = "stud_ID")
# df_joined <- rename(df_joined, teacher_ID = teacher_ID1)
df_joined$teacher_ID <- as.character(df_joined$teacher_ID)
teacher_df$teacher_ID <- as.character(teacher_df$teacher_ID)
df_joined <- left_join(df_joined, teacher_df, by = "teacher_ID")
df$teacher_ID <- as.character(df$teacher_ID)
df_j <- bind_cols(df_joined, df)
3B: Constructing variables
# IPs
df_j$instructional_practice <- as.factor(df_j$instructional_practice)
df_j$instructional_practice <- car::recode(df_j$instructional_practice,
"'1' = 'Lecture';
c('2', '3') = 'Individual Work';
c('4', '5') = 'Group Work';
c('6', '8', '7') = 'Quiz and Test';
c('9') = 'Discussion';
c('11', '10') = 'Presentation';
c('12', '13') = 'Video';
c('15', '14', '16') = 'Laboratory';
c('17') = 'Non-instructional';
c('18') = NA")
df_j$instructional_practice <- as.character(df_j$instructional_practice)
df_j$instructional_practice <- ifelse((df_j$instructional_practice == "Discussion" | df_j$instructional_practice == "Non-instructional" |
df_j$instructional_practice == "Presentation" | df_j$instructional_practice == "Video" | df_j$instructional_practice == "Group Work"), "Other", df_j$instructional_practice)
# For classroom activity: use df_j$instructional_practice
# Choices
df_j$ch_who[is.na(df_j$ch_who)] <- 0
df_j$ch_howdo[is.na(df_j$ch_howdo)] <- 0
df_j$ch_mat[is.na(df_j$ch_mat)] <- 0
df_j$ch_time[is.na(df_j$ch_time)] <- 0
df_j$ch_doing[is.na(df_j$ch_doing)] <- 0
df_j$ch_topic[is.na(df_j$ch_topic)] <- 0
df_j$ch_defin[is.na(df_j$ch_defin)] <- 0
df_j$ch_other[is.na(df_j$ch_other)] <- 0
df_j$ch_none[is.na(df_j$ch_none)] <- 0
df_j$ch_none <- ifelse(((df_j$ch_who == 0 & df_j$ch_howdo == 0 & df_j$ch_mat == 0 & df_j$ch_time == 0 & df_j$ch_doing == 0 &
df_j$ch_defin == 0 & df_j$ch_topic == 0 & df_j$ch_other == 0) & df_j$ch_none == 0), 1, df_j$ch_none)
df_j$ch_none <- ifelse(((df_j$ch_who == 1 | df_j$ch_howdo == 1 | df_j$ch_mat == 1 | df_j$ch_time == 1 | df_j$ch_doing |
df_j$ch_defin == 1 | df_j$ch_topic == 1 | df_j$ch_other == 1) & df_j$ch_none == 1), 0, df_j$ch_none)
df_j$ch_doing_defining_topic <- ifelse(df_j$ch_defin == 1 | df_j$ch_topic == 1 | df_j$ch_doing, 1, 0)
df_j$ch_defining_topic <- ifelse(df_j$ch_defin == 1 | df_j$ch_topic == 1, 1, 0)
df_j$framing <- ifelse(df_j$ch_defin == 1 | df_j$ch_topic == 1 | df_j$ch_doing, 1, 0)
df_j$any_choice <- ifelse(df_j$framing == 1 | df_j$ch_mat == 1 | df_j$ch_howdo == 1 | df_j$ch_time == 1 |
df_j$ch_who == 1 | df_j$ch_other == 1, 1, 0) # this is the same as ch_none above
# For any vs. none use: df_j$any_choice
# For Framing use: df_j$framing
# For who use: df_j$ch_who
# need race, gender, grade, interest at T1, competence at T1, prior grades
df_j$race <- as.factor(as.vector(df_j$race))
df_j$race <- forcats::fct_recode(df_j$race,
Black = "3",
White = "4",
Hispanic = "2",
Other = "1",
Other = "5",
Other = "6")
df_j$race[df_j$race == "NaN"] <- NA
df_j$gender <- as.factor(df_j$gender)
df_j$gender <- forcats::fct_recode(df_j$gender,
Male = "1",
Female = "2")
df_j$grade <- as.factor(df_j$grade)
df_j$grade <- forcats::fct_recode(df_j$grade,
ninth = "1",
tenth = "2",
eleventh = "3",
twelth = "4")
df_j <- filter(df_j, gengrades != 9 & gengrades != 98)
df_j <- filter(df_j, instructional_practice != "NA")
df_j$gengrades <- as.integer(as.vector(df_j$gengrades))
df_j$interestsci <- as.vector(df_j$interestsci)
df_j$expectwell <- as.vector(df_j$expectwell)
# df_j$CumGPA_2010 <- as.numeric(as.vector(df_j$CumGPA_2010))
df_j$instructional_practice[is.na(df_j$instructional_practice)] <- NA
df_j$instructional_practice <- as.factor(df_j$instructional_practice)
df_j$instructional_practice <- forcats::fct_relevel(df_j$instructional_practice,
"Other")
df_j$race <- forcats::fct_relevel(df_j$race, "Black")
4A: Analysis with MLMs
knitr::opts_chunk$set(cache=TRUE)
m1 <- glmer(cluster_1 ~ 1 +
gender + race + grade + gengrades + interestsci + expectwell + instructional_practice +
(1|teacher_ID) + (1|stud_ID) + (1|beep_ID),
family = binomial(link = "logit"),
control=lme4::glmerControl(optimizer="bobyqa",
optCtrl=list(maxfun=2e5)),
data = df_j)
m1_out <- broom::tidy(m1)
sjPlot::sjt.glmer(m1, show.re.var = F, show.ci = F, show.se = T)
|
|
|
cluster_1
|
|
|
|
Odds Ratio
|
std. Error
|
p
|
|
Fixed Parts
|
|
(Intercept)
|
|
0.18
|
0.11
|
.004
|
|
gender (Female)
|
|
0.81
|
0.18
|
.346
|
|
race (Other)
|
|
2.52
|
1.23
|
.059
|
|
race (Hispanic)
|
|
1.55
|
0.57
|
.236
|
|
race (White)
|
|
0.59
|
0.23
|
.178
|
|
grade (tenth)
|
|
1.39
|
0.43
|
.281
|
|
grade (eleventh)
|
|
1.02
|
0.29
|
.948
|
|
grade (twelth)
|
|
7.23
|
4.95
|
.004
|
|
gengrades
|
|
0.94
|
0.07
|
.473
|
|
interestsci
|
|
1.14
|
0.16
|
.362
|
|
expectwell
|
|
0.65
|
0.11
|
.011
|
|
instructional_practice (Individual Work)
|
|
0.92
|
0.16
|
.634
|
|
instructional_practice (Laboratory)
|
|
0.84
|
0.14
|
.275
|
|
instructional_practice (Lecture)
|
|
1.22
|
0.20
|
.242
|
|
instructional_practice (NaN)
|
|
0.87
|
0.48
|
.793
|
|
instructional_practice (Quiz and Test)
|
|
1.14
|
0.19
|
.429
|
|
Random Parts
|
|
Nbeep_ID
|
|
244
|
|
Nstud_ID
|
|
203
|
|
Nteacher_ID
|
|
13
|
|
ICCbeep_ID
|
|
0.000
|
|
ICCstud_ID
|
|
0.331
|
|
ICCteacher_ID
|
|
0.001
|
|
Observations
|
|
3613
|
|
Deviance
|
|
2519.743
|
m2 <- glmer(cluster_2 ~ 1 +
gender + race + grade + gengrades + interestsci + expectwell + instructional_practice +
(1|teacher_ID) + (1|stud_ID) + (1|beep_ID),
family = binomial(link = "logit"),
control=lme4::glmerControl(optimizer="bobyqa",
optCtrl=list(maxfun=2e5)),
data = df_j)
m2_out <- broom::tidy(m2)
sjPlot::sjt.glmer(m2, show.re.var = F, show.ci = F, show.se = T)
|
|
|
cluster_2
|
|
|
|
Odds Ratio
|
std. Error
|
p
|
|
Fixed Parts
|
|
(Intercept)
|
|
0.07
|
0.04
|
<.001
|
|
gender (Female)
|
|
1.34
|
0.30
|
.189
|
|
race (Other)
|
|
0.53
|
0.29
|
.251
|
|
race (Hispanic)
|
|
1.07
|
0.40
|
.866
|
|
race (White)
|
|
0.85
|
0.33
|
.674
|
|
grade (tenth)
|
|
1.07
|
0.33
|
.833
|
|
grade (eleventh)
|
|
0.77
|
0.23
|
.388
|
|
grade (twelth)
|
|
0.57
|
0.44
|
.466
|
|
gengrades
|
|
1.13
|
0.09
|
.136
|
|
interestsci
|
|
0.76
|
0.11
|
.053
|
|
expectwell
|
|
0.94
|
0.16
|
.721
|
|
instructional_practice (Individual Work)
|
|
1.65
|
0.37
|
.025
|
|
instructional_practice (Laboratory)
|
|
0.57
|
0.14
|
.023
|
|
instructional_practice (Lecture)
|
|
1.44
|
0.33
|
.115
|
|
instructional_practice (NaN)
|
|
0.82
|
0.72
|
.825
|
|
instructional_practice (Quiz and Test)
|
|
5.27
|
1.14
|
<.001
|
|
Random Parts
|
|
Nbeep_ID
|
|
244
|
|
Nstud_ID
|
|
203
|
|
Nteacher_ID
|
|
13
|
|
ICCbeep_ID
|
|
0.071
|
|
ICCstud_ID
|
|
0.291
|
|
ICCteacher_ID
|
|
0.000
|
|
Observations
|
|
3613
|
|
Deviance
|
|
1842.132
|
m3 <- glmer(cluster_3 ~ 1 +
gender + race + grade + gengrades + interestsci + expectwell + instructional_practice +
(1|teacher_ID) + (1|stud_ID) + (1|beep_ID),
family = binomial(link = "logit"),
control=lme4::glmerControl(optimizer="bobyqa",
optCtrl=list(maxfun=2e5)),
data = df_j)
m3_out <- broom::tidy(m3)
sjPlot::sjt.glmer(m3, show.re.var = F, show.ci = F, show.se = T)
|
|
|
cluster_3
|
|
|
|
Odds Ratio
|
std. Error
|
p
|
|
Fixed Parts
|
|
(Intercept)
|
|
0.49
|
0.40
|
.378
|
|
gender (Female)
|
|
0.48
|
0.15
|
.016
|
|
race (Other)
|
|
1.60
|
1.07
|
.486
|
|
race (Hispanic)
|
|
0.67
|
0.34
|
.437
|
|
race (White)
|
|
0.47
|
0.25
|
.156
|
|
grade (tenth)
|
|
0.74
|
0.33
|
.499
|
|
grade (eleventh)
|
|
1.67
|
0.66
|
.196
|
|
grade (twelth)
|
|
0.25
|
0.29
|
.235
|
|
gengrades
|
|
0.67
|
0.08
|
<.001
|
|
interestsci
|
|
0.93
|
0.18
|
.709
|
|
expectwell
|
|
0.68
|
0.17
|
.115
|
|
instructional_practice (Individual Work)
|
|
0.56
|
0.16
|
.038
|
|
instructional_practice (Laboratory)
|
|
1.47
|
0.36
|
.119
|
|
instructional_practice (Lecture)
|
|
0.94
|
0.24
|
.818
|
|
instructional_practice (NaN)
|
|
0.43
|
0.42
|
.386
|
|
instructional_practice (Quiz and Test)
|
|
0.87
|
0.24
|
.619
|
|
Random Parts
|
|
Nbeep_ID
|
|
244
|
|
Nstud_ID
|
|
203
|
|
Nteacher_ID
|
|
13
|
|
ICCbeep_ID
|
|
0.082
|
|
ICCstud_ID
|
|
0.421
|
|
ICCteacher_ID
|
|
0.000
|
|
Observations
|
|
3613
|
|
Deviance
|
|
1433.118
|
m4 <- glmer(cluster_4 ~ 1 +
gender + race + grade + gengrades + interestsci + expectwell + instructional_practice +
(1|teacher_ID) + (1|stud_ID) + (1|beep_ID),
family = binomial(link = "logit"),
control=lme4::glmerControl(optimizer="bobyqa",
optCtrl=list(maxfun=2e5)),
data = df_j)
m4_out <- broom::tidy(m4)
sjPlot::sjt.glmer(m4, show.re.var = F, show.ci = F, show.se = T)
|
|
|
cluster_4
|
|
|
|
Odds Ratio
|
std. Error
|
p
|
|
Fixed Parts
|
|
(Intercept)
|
|
0.07
|
0.04
|
<.001
|
|
gender (Female)
|
|
1.05
|
0.21
|
.830
|
|
race (Other)
|
|
0.34
|
0.17
|
.030
|
|
race (Hispanic)
|
|
0.63
|
0.22
|
.175
|
|
race (White)
|
|
1.36
|
0.47
|
.374
|
|
grade (tenth)
|
|
1.22
|
0.34
|
.483
|
|
grade (eleventh)
|
|
1.15
|
0.29
|
.583
|
|
grade (twelth)
|
|
0.38
|
0.28
|
.192
|
|
gengrades
|
|
1.00
|
0.07
|
.995
|
|
interestsci
|
|
1.29
|
0.17
|
.055
|
|
expectwell
|
|
1.18
|
0.18
|
.252
|
|
instructional_practice (Individual Work)
|
|
1.31
|
0.21
|
.089
|
|
instructional_practice (Laboratory)
|
|
0.80
|
0.13
|
.169
|
|
instructional_practice (Lecture)
|
|
1.24
|
0.20
|
.184
|
|
instructional_practice (NaN)
|
|
0.82
|
0.38
|
.674
|
|
instructional_practice (Quiz and Test)
|
|
0.92
|
0.16
|
.654
|
|
Random Parts
|
|
Nbeep_ID
|
|
244
|
|
Nstud_ID
|
|
203
|
|
Nteacher_ID
|
|
13
|
|
ICCbeep_ID
|
|
0.006
|
|
ICCstud_ID
|
|
0.291
|
|
ICCteacher_ID
|
|
0.000
|
|
Observations
|
|
3613
|
|
Deviance
|
|
2640.616
|
m5 <- glmer(cluster_5 ~ 1 +
gender + race + grade + gengrades + interestsci + expectwell + instructional_practice +
(1|teacher_ID) + (1|stud_ID) + (1|beep_ID),
family = binomial(link = "logit"),
control=lme4::glmerControl(optimizer="bobyqa",
optCtrl=list(maxfun=2e5)),
data = df_j)
m5_out <- broom::tidy(m5)
sjPlot::sjt.glmer(m5, show.re.var = F, show.ci = F, show.se = T)
|
|
|
cluster_5
|
|
|
|
Odds Ratio
|
std. Error
|
p
|
|
Fixed Parts
|
|
(Intercept)
|
|
0.05
|
0.04
|
<.001
|
|
gender (Female)
|
|
1.74
|
0.43
|
.025
|
|
race (Other)
|
|
1.04
|
0.59
|
.946
|
|
race (Hispanic)
|
|
1.06
|
0.44
|
.888
|
|
race (White)
|
|
1.38
|
0.58
|
.450
|
|
grade (tenth)
|
|
0.45
|
0.16
|
.030
|
|
grade (eleventh)
|
|
0.53
|
0.18
|
.057
|
|
grade (twelth)
|
|
0.15
|
0.13
|
.033
|
|
gengrades
|
|
1.32
|
0.11
|
.001
|
|
interestsci
|
|
0.94
|
0.15
|
.701
|
|
expectwell
|
|
1.27
|
0.22
|
.173
|
|
instructional_practice (Individual Work)
|
|
1.34
|
0.26
|
.126
|
|
instructional_practice (Laboratory)
|
|
0.96
|
0.18
|
.833
|
|
instructional_practice (Lecture)
|
|
0.90
|
0.18
|
.593
|
|
instructional_practice (NaN)
|
|
2.25
|
1.28
|
.153
|
|
instructional_practice (Quiz and Test)
|
|
0.52
|
0.11
|
.002
|
|
Random Parts
|
|
Nbeep_ID
|
|
244
|
|
Nstud_ID
|
|
203
|
|
Nteacher_ID
|
|
13
|
|
ICCbeep_ID
|
|
0.055
|
|
ICCstud_ID
|
|
0.372
|
|
ICCteacher_ID
|
|
0.006
|
|
Observations
|
|
3613
|
|
Deviance
|
|
2491.594
|
m6 <- glmer(cluster_6 ~ 1 +
gender + race + grade + gengrades + interestsci + expectwell + instructional_practice +
(1|teacher_ID) + (1|stud_ID) + (1|beep_ID),
family = binomial(link = "logit"),
control=lme4::glmerControl(optimizer="bobyqa",
optCtrl=list(maxfun=2e5)),
data = df_j)
m6_out <- broom::tidy(m6)
sjPlot::sjt.glmer(m6, show.re.var = F, show.ci = F, show.se = T)
|
|
|
cluster_6
|
|
|
|
Odds Ratio
|
std. Error
|
p
|
|
Fixed Parts
|
|
(Intercept)
|
|
0.17
|
0.07
|
<.001
|
|
gender (Female)
|
|
0.99
|
0.15
|
.968
|
|
race (Other)
|
|
1.26
|
0.46
|
.526
|
|
race (Hispanic)
|
|
1.32
|
0.35
|
.294
|
|
race (White)
|
|
1.39
|
0.38
|
.225
|
|
grade (tenth)
|
|
1.46
|
0.32
|
.086
|
|
grade (eleventh)
|
|
1.46
|
0.30
|
.065
|
|
grade (twelth)
|
|
1.52
|
0.75
|
.399
|
|
gengrades
|
|
0.90
|
0.05
|
.051
|
|
interestsci
|
|
0.95
|
0.09
|
.604
|
|
expectwell
|
|
1.17
|
0.13
|
.155
|
|
instructional_practice (Individual Work)
|
|
0.59
|
0.10
|
.001
|
|
instructional_practice (Laboratory)
|
|
1.43
|
0.22
|
.021
|
|
instructional_practice (Lecture)
|
|
0.66
|
0.11
|
.013
|
|
instructional_practice (NaN)
|
|
0.86
|
0.40
|
.748
|
|
instructional_practice (Quiz and Test)
|
|
0.35
|
0.07
|
<.001
|
|
Random Parts
|
|
Nbeep_ID
|
|
244
|
|
Nstud_ID
|
|
203
|
|
Nteacher_ID
|
|
13
|
|
ICCbeep_ID
|
|
0.029
|
|
ICCstud_ID
|
|
0.166
|
|
ICCteacher_ID
|
|
0.000
|
|
Observations
|
|
3613
|
|
Deviance
|
|
2846.657
|
4B: ICCs
xm1 <- sjstats::icc(m1)
xm2 <- sjstats::icc(m2)
xm3 <- sjstats::icc(m3)
xm4 <- sjstats::icc(m4)
xm5 <- sjstats::icc(m5)
xm6 <- sjstats::icc(m6)
df1_with_ICCs <- data.frame(profile_1 = xm1,
profile_2 = xm2,
profile_3 = xm3,
profile_4 = xm4,
profile_5 = xm5,
profile_6 = xm6)
row.names(df_with_ICCs) <- c("Momentary RE", "Student RE", "Teacher RE")
df_with_ICCs
## profile_1 profile_2 profile_3 profile_4 profile_5
## Momentary RE 0.0000000 0.12122591 0.08742863 0.006808985 0.05680992
## Student RE 0.3951807 0.27105319 0.46877232 0.303127910 0.40472867
## Teacher RE 0.0000000 0.04996376 0.00000000 0.005911475 0.01024088
## profile_6
## Momentary RE 0.04423836
## Student RE 0.17004291
## Teacher RE 0.01520584