knitr::clean_cache()
## NULL
# data <- params$data
targets::tar_load(esm_long)
data <- esm_long
library(brms)
## Loading required package: Rcpp
## Loading 'brms' package (version 2.14.4). Useful instructions
## can be found by typing help('brms'). A more detailed introduction
## to the package is available through vignette('brms_overview').
##
## Attaching package: 'brms'
## The following object is masked from 'package:stats':
##
## ar
library(bayesplot)
## This is bayesplot version 1.8.0
## - Online documentation and vignettes at mc-stan.org/bayesplot
## - bayesplot theme set to bayesplot::theme_default()
## * Does _not_ affect other ggplot2 plots
## * See ?bayesplot_theme_set for details on theme setting
data$int_val_change <- data$int_val_post - data$int_val_pre
3. Just post - how do students’ momentary experiences relate to changes in their interest?
base model
bf_1 <- bf(frustrated ~ 1 +
course +
(1|s|number) +
(1|p|assign))
bf_2 <- bf(confident ~ 1 +
course +
(1|s|number) +
(1|p|assign))
bf_3 <- bf(interested ~ 1 +
course +
(1|s|number) +
(1|p|assign))
bf_int <- bf(int_val_post ~ course)
m3a_pre <- brm(bf_1 + bf_2 + bf_3 +
bf_int,
data = data,
chains = 4, cores = 4, iter = 1000,
control = list(max_treedepth = 20))
## Setting 'rescor' to TRUE by default for this model
## Warning: In the future, 'rescor' will be set to FALSE by default for all models.
## It is thus recommended to explicitely set 'rescor' via 'set_recor' instead of
## using the default.
## Warning: Rows containing NAs were excluded from the model.
## Compiling Stan program...
## Start sampling
## Warning: Bulk Effective Samples Size (ESS) is too low, indicating posterior means and medians may be unreliable.
## Running the chains for more iterations may help. See
## http://mc-stan.org/misc/warnings.html#bulk-ess
## Warning: Tail Effective Samples Size (ESS) is too low, indicating posterior variances and tail quantiles may be unreliable.
## Running the chains for more iterations may help. See
## http://mc-stan.org/misc/warnings.html#tail-ess
m3a_pre
## Family: MV(gaussian, gaussian, gaussian, gaussian)
## Links: mu = identity; sigma = identity
## mu = identity; sigma = identity
## mu = identity; sigma = identity
## mu = identity; sigma = identity
## Formula: frustrated ~ 1 + course + (1 | s | number) + (1 | p | assign)
## confident ~ 1 + course + (1 | s | number) + (1 | p | assign)
## interested ~ 1 + course + (1 | s | number) + (1 | p | assign)
## int_val_post ~ course
## Data: data (Number of observations: 776)
## Samples: 4 chains, each with iter = 1000; warmup = 500; thin = 1;
## total post-warmup samples = 2000
##
## Group-Level Effects:
## ~assign (Number of levels: 25)
## Estimate Est.Error l-95% CI
## sd(frustrated_Intercept) 0.33 0.06 0.23
## sd(confident_Intercept) 0.19 0.04 0.12
## sd(interested_Intercept) 0.15 0.03 0.10
## cor(frustrated_Intercept,confident_Intercept) -0.88 0.10 -0.99
## cor(frustrated_Intercept,interested_Intercept) -0.88 0.10 -0.99
## cor(confident_Intercept,interested_Intercept) 0.93 0.07 0.73
## u-95% CI Rhat Bulk_ESS Tail_ESS
## sd(frustrated_Intercept) 0.47 1.01 568 1000
## sd(confident_Intercept) 0.28 1.00 676 1106
## sd(interested_Intercept) 0.22 1.00 677 971
## cor(frustrated_Intercept,confident_Intercept) -0.61 1.01 559 1028
## cor(frustrated_Intercept,interested_Intercept) -0.61 1.00 904 1172
## cor(confident_Intercept,interested_Intercept) 1.00 1.00 1421 1734
##
## ~number (Number of levels: 65)
## Estimate Est.Error l-95% CI
## sd(frustrated_Intercept) 1.01 0.10 0.83
## sd(confident_Intercept) 0.73 0.07 0.60
## sd(interested_Intercept) 0.57 0.06 0.47
## cor(frustrated_Intercept,confident_Intercept) -0.56 0.09 -0.73
## cor(frustrated_Intercept,interested_Intercept) -0.11 0.13 -0.34
## cor(confident_Intercept,interested_Intercept) 0.48 0.11 0.24
## u-95% CI Rhat Bulk_ESS Tail_ESS
## sd(frustrated_Intercept) 1.22 1.01 253 328
## sd(confident_Intercept) 0.90 1.00 356 632
## sd(interested_Intercept) 0.69 1.00 358 593
## cor(frustrated_Intercept,confident_Intercept) -0.36 1.01 312 555
## cor(frustrated_Intercept,interested_Intercept) 0.15 1.02 362 508
## cor(confident_Intercept,interested_Intercept) 0.66 1.01 376 753
##
## Population-Level Effects:
## Estimate Est.Error l-95% CI u-95% CI Rhat Bulk_ESS
## frustrated_Intercept 2.56 0.16 2.26 2.87 1.02 170
## confident_Intercept 4.15 0.11 3.94 4.37 1.03 243
## interested_Intercept 4.35 0.08 4.17 4.51 1.02 329
## intvalpost_Intercept 1.18 0.04 1.10 1.26 1.00 2800
## frustrated_courseCOSC111 0.41 0.43 -0.44 1.28 1.02 348
## confident_courseCOSC111 -0.55 0.32 -1.23 0.09 1.02 324
## interested_courseCOSC111 -0.56 0.26 -1.08 -0.06 1.01 457
## intvalpost_courseCOSC111 -0.73 0.14 -1.00 -0.46 1.00 2707
## Tail_ESS
## frustrated_Intercept 325
## confident_Intercept 572
## interested_Intercept 551
## intvalpost_Intercept 1524
## frustrated_courseCOSC111 626
## confident_courseCOSC111 738
## interested_courseCOSC111 765
## intvalpost_courseCOSC111 1576
##
## Family Specific Parameters:
## Estimate Est.Error l-95% CI u-95% CI Rhat Bulk_ESS Tail_ESS
## sigma_frustrated 0.81 0.05 0.73 0.91 1.01 286 506
## sigma_confident 0.73 0.05 0.64 0.84 1.02 242 530
## sigma_interested 0.69 0.05 0.59 0.80 1.02 366 774
## sigma_intvalpost 1.11 0.03 1.05 1.16 1.00 2524 1440
##
## Residual Correlations:
## Estimate Est.Error l-95% CI u-95% CI Rhat
## rescor(frustrated,confident) -0.40 0.07 -0.54 -0.25 1.01
## rescor(frustrated,interested) -0.33 0.09 -0.49 -0.13 1.01
## rescor(confident,interested) 0.59 0.06 0.47 0.70 1.02
## rescor(frustrated,intvalpost) -0.37 0.13 -0.59 -0.09 1.01
## rescor(confident,intvalpost) 0.54 0.09 0.33 0.69 1.02
## rescor(interested,intvalpost) 0.71 0.05 0.59 0.79 1.02
## Bulk_ESS Tail_ESS
## rescor(frustrated,confident) 265 419
## rescor(frustrated,interested) 262 424
## rescor(confident,interested) 265 345
## rescor(frustrated,intvalpost) 224 424
## rescor(confident,intvalpost) 244 508
## rescor(interested,intvalpost) 360 764
##
## Samples were drawn using sampling(NUTS). For each parameter, Bulk_ESS
## and Tail_ESS are effective sample size measures, and Rhat is the potential
## scale reduction factor on split chains (at convergence, Rhat = 1).
adding gender and pre self-efficacy
bf_1 <- bf(frustrated ~ 1 +
gender +
course +
self_eff +
(1|s|number) +
(1|p|assign))
bf_2 <- bf(confident ~ 1 +
gender +
course +
self_eff +
(1|s|number) +
(1|p|assign))
bf_3 <- bf(interested ~ 1 +
gender +
course +
self_eff +
(1|s|number) +
(1|p|assign))
bf_int <- bf(int_val_post ~ course + gender + self_eff)
m3e_pre <- brm(bf_1 + bf_2 + bf_3 +
bf_int,
data = data,
chains = 4, cores = 4, iter = 1000,
control = list(max_treedepth = 20))
## Setting 'rescor' to TRUE by default for this model
## Warning: In the future, 'rescor' will be set to FALSE by default for all models.
## It is thus recommended to explicitely set 'rescor' via 'set_recor' instead of
## using the default.
## Warning: Rows containing NAs were excluded from the model.
## Compiling Stan program...
## recompiling to avoid crashing R session
## Start sampling
## Warning: Bulk Effective Samples Size (ESS) is too low, indicating posterior means and medians may be unreliable.
## Running the chains for more iterations may help. See
## http://mc-stan.org/misc/warnings.html#bulk-ess
## Warning: Tail Effective Samples Size (ESS) is too low, indicating posterior variances and tail quantiles may be unreliable.
## Running the chains for more iterations may help. See
## http://mc-stan.org/misc/warnings.html#tail-ess
m3e_pre
## Family: MV(gaussian, gaussian, gaussian, gaussian)
## Links: mu = identity; sigma = identity
## mu = identity; sigma = identity
## mu = identity; sigma = identity
## mu = identity; sigma = identity
## Formula: frustrated ~ 1 + gender + course + self_eff + (1 | s | number) + (1 | p | assign)
## confident ~ 1 + gender + course + self_eff + (1 | s | number) + (1 | p | assign)
## interested ~ 1 + gender + course + self_eff + (1 | s | number) + (1 | p | assign)
## int_val_post ~ course + gender + self_eff
## Data: data (Number of observations: 770)
## Samples: 4 chains, each with iter = 1000; warmup = 500; thin = 1;
## total post-warmup samples = 2000
##
## Group-Level Effects:
## ~assign (Number of levels: 25)
## Estimate Est.Error l-95% CI
## sd(frustrated_Intercept) 0.33 0.06 0.23
## sd(confident_Intercept) 0.19 0.04 0.12
## sd(interested_Intercept) 0.15 0.03 0.10
## cor(frustrated_Intercept,confident_Intercept) -0.86 0.10 -0.99
## cor(frustrated_Intercept,interested_Intercept) -0.88 0.10 -0.99
## cor(confident_Intercept,interested_Intercept) 0.93 0.07 0.75
## u-95% CI Rhat Bulk_ESS Tail_ESS
## sd(frustrated_Intercept) 0.47 1.00 518 993
## sd(confident_Intercept) 0.28 1.00 667 1176
## sd(interested_Intercept) 0.22 1.00 666 1232
## cor(frustrated_Intercept,confident_Intercept) -0.59 1.01 510 1120
## cor(frustrated_Intercept,interested_Intercept) -0.62 1.00 696 920
## cor(confident_Intercept,interested_Intercept) 1.00 1.00 1130 1498
##
## ~number (Number of levels: 64)
## Estimate Est.Error l-95% CI
## sd(frustrated_Intercept) 0.96 0.10 0.79
## sd(confident_Intercept) 0.63 0.07 0.52
## sd(interested_Intercept) 0.57 0.06 0.46
## cor(frustrated_Intercept,confident_Intercept) -0.45 0.11 -0.63
## cor(frustrated_Intercept,interested_Intercept) -0.09 0.13 -0.33
## cor(confident_Intercept,interested_Intercept) 0.53 0.11 0.30
## u-95% CI Rhat Bulk_ESS Tail_ESS
## sd(frustrated_Intercept) 1.17 1.00 352 626
## sd(confident_Intercept) 0.77 1.00 477 908
## sd(interested_Intercept) 0.70 1.01 517 859
## cor(frustrated_Intercept,confident_Intercept) -0.21 1.00 398 620
## cor(frustrated_Intercept,interested_Intercept) 0.18 1.01 357 553
## cor(confident_Intercept,interested_Intercept) 0.72 1.01 377 772
##
## Population-Level Effects:
## Estimate Est.Error l-95% CI u-95% CI Rhat Bulk_ESS
## frustrated_Intercept 3.11 0.25 2.63 3.59 1.01 312
## confident_Intercept 3.57 0.16 3.27 3.89 1.01 462
## interested_Intercept 4.05 0.15 3.76 4.33 1.00 447
## intvalpost_Intercept 0.79 0.07 0.65 0.93 1.00 2649
## frustrated_genderMale -0.17 0.31 -0.76 0.44 1.01 318
## frustrated_courseCOSC111 0.34 0.44 -0.47 1.24 1.00 545
## frustrated_self_eff -0.56 0.16 -0.87 -0.26 1.01 326
## confident_genderMale 0.25 0.21 -0.16 0.66 1.00 338
## confident_courseCOSC111 -0.45 0.30 -1.03 0.17 1.00 428
## confident_self_eff 0.53 0.11 0.32 0.73 1.00 332
## interested_genderMale 0.25 0.19 -0.10 0.63 1.01 389
## interested_courseCOSC111 -0.40 0.27 -0.94 0.12 1.00 479
## interested_self_eff 0.16 0.10 -0.03 0.34 1.01 394
## intvalpost_courseCOSC111 -0.48 0.14 -0.75 -0.21 1.00 2254
## intvalpost_genderMale 0.10 0.09 -0.08 0.29 1.00 2066
## intvalpost_self_eff 0.43 0.05 0.34 0.53 1.00 2282
## Tail_ESS
## frustrated_Intercept 539
## confident_Intercept 697
## interested_Intercept 760
## intvalpost_Intercept 1472
## frustrated_genderMale 505
## frustrated_courseCOSC111 813
## frustrated_self_eff 515
## confident_genderMale 390
## confident_courseCOSC111 724
## confident_self_eff 582
## interested_genderMale 459
## interested_courseCOSC111 585
## interested_self_eff 898
## intvalpost_courseCOSC111 1245
## intvalpost_genderMale 1621
## intvalpost_self_eff 1654
##
## Family Specific Parameters:
## Estimate Est.Error l-95% CI u-95% CI Rhat Bulk_ESS Tail_ESS
## sigma_frustrated 0.76 0.03 0.71 0.84 1.01 306 350
## sigma_confident 0.66 0.03 0.60 0.73 1.01 324 588
## sigma_interested 0.67 0.05 0.57 0.77 1.01 316 555
## sigma_intvalpost 1.01 0.03 0.96 1.07 1.00 2916 1499
##
## Residual Correlations:
## Estimate Est.Error l-95% CI u-95% CI Rhat
## rescor(frustrated,confident) -0.29 0.06 -0.42 -0.18 1.01
## rescor(frustrated,interested) -0.20 0.10 -0.39 0.01 1.02
## rescor(confident,interested) 0.50 0.07 0.36 0.62 1.01
## rescor(frustrated,intvalpost) -0.19 0.15 -0.46 0.10 1.02
## rescor(confident,intvalpost) 0.36 0.11 0.14 0.56 1.01
## rescor(interested,intvalpost) 0.68 0.06 0.54 0.78 1.01
## Bulk_ESS Tail_ESS
## rescor(frustrated,confident) 240 314
## rescor(frustrated,interested) 210 318
## rescor(confident,interested) 298 658
## rescor(frustrated,intvalpost) 210 304
## rescor(confident,intvalpost) 285 629
## rescor(interested,intvalpost) 302 604
##
## Samples were drawn using sampling(NUTS). For each parameter, Bulk_ESS
## and Tail_ESS are effective sample size measures, and Rhat is the potential
## scale reduction factor on split chains (at convergence, Rhat = 1).