Closing the gap analyser.

## Warning in vtable::sumtable(tt, summ = c("mean(x)", "sd(x)", "pctile(x)[25]", :
## Some labelled variables have unlabeled values. Treating these as numeric
## variables and ignoring labels.

## Warning in vtable::sumtable(tt, summ = c("mean(x)", "sd(x)", "pctile(x)[25]", : Factor variables ignore custom summ options. Cols 1 and 2 are count and percentage.
## Beware combining factors with a custom summ unless factor.numeric = TRUE.

Summary Statistics
Variable	Mean	Sd	Pctile[25]	Pctile[75]
T2_LS_S	18	5.9	14	23
T2_PI_S	8.8	2.3	9	10
T2_WR_S	6.4	3.5	3	10
T2_SP_S	4.8	3.3	1	8
T2_RAN_T	57	16	46	65
T2_DS_S	6.1	1.7	5	7
T3_LW_R	2170
… 0	1941	89%
… 1	229	11%
T3_PI_R	2169
… 0	2035	94%
… 1	134	6%
T3_SP_R	2170
… 0	1988	92%
… 1	182	8%
T3_WR_R	2169
… 0	2015	93%
… 1	154	7%
T3_RC_R	2166
… 0	2025	93%
… 1	141	7%
T1_M_edR	2153
… 1	344	16%
… 2	1809	84%
T2_Risk	2170
… 0	1684	78%
… 1	486	22%

## Warning in checkConv(attr(opt, "derivs"), opt$par, ctrl = control$checkConv, :
## Model failed to converge with max|grad| = 0.00249606 (tol = 0.002, component 1)

## Warning in checkConv(attr(opt, "derivs"), opt$par, ctrl = control$checkConv, :
## Model failed to converge with max|grad| = 0.00216612 (tol = 0.002, component 1)

## Warning in checkConv(attr(opt, "derivs"), opt$par, ctrl = control$checkConv, :
## Model failed to converge with max|grad| = 0.00313584 (tol = 0.002, component 1)

Observations	2106
Dependent variable	T3_LW_R
Type	Mixed effects generalized linear model
Family	binomial
Link	logit

Fixed Effects
	Est.	S.E.	z val.	p
(Intercept)	-2.37	0.28	-8.57	0.00
cond1	-0.63	0.30	-2.08	0.04
T2_LS_S	-0.79	0.14	-5.58	0.00
T2_PI_S	-0.05	0.08	-0.66	0.51
T2_WR_S	-0.34	0.16	-2.07	0.04
T2_SP_S	0.07	0.18	0.37	0.71
T2_RAN_T	0.17	0.07	2.29	0.02
T2_DS_S	-0.18	0.09	-1.95	0.05

Random Effects
Group	Parameter	Std. Dev.
Class:school	(Intercept)	0.30
school	(Intercept)	0.57

Grouping Variables
Group	# groups	ICC
Class:school	101	0.03
school	41	0.09

Observations	2105
Dependent variable	T3_PI_R
Type	Mixed effects generalized linear model
Family	binomial
Link	logit

Fixed Effects
	Est.	S.E.	z val.	p
(Intercept)	-3.54	0.34	-10.37	0.00
cond1	-0.82	0.33	-2.46	0.01
T2_LS_S	-0.75	0.18	-4.15	0.00
T2_PI_S	-0.11	0.10	-1.14	0.26
T2_WR_S	-0.72	0.25	-2.88	0.00
T2_SP_S	-0.04	0.29	-0.14	0.89
T2_RAN_T	0.41	0.09	4.61	0.00
T2_DS_S	-0.13	0.12	-1.02	0.31

Random Effects
Group	Parameter	Std. Dev.
Class:school	(Intercept)	0.58
school	(Intercept)	0.40

Grouping Variables
Group	# groups	ICC
Class:school	101	0.09
school	41	0.04

Observations	2106
Dependent variable	T3_SP_R
Type	Mixed effects generalized linear model
Family	binomial
Link	logit

Fixed Effects
	Est.	S.E.	z val.	p
(Intercept)	-3.38	0.28	-11.99	0.00
cond1	-0.66	0.24	-2.71	0.01
T2_LS_S	-0.42	0.15	-2.83	0.00
T2_PI_S	-0.41	0.08	-4.97	0.00
T2_WR_S	-0.59	0.23	-2.62	0.01
T2_SP_S	-0.52	0.26	-1.96	0.05
T2_RAN_T	0.36	0.08	4.47	0.00
T2_DS_S	-0.20	0.11	-1.73	0.08

Random Effects
Group	Parameter	Std. Dev.
Class:school	(Intercept)	0.00
school	(Intercept)	0.25

Grouping Variables
Group	# groups	ICC
Class:school	101	0.00
school	41	0.02

Observations	2105
Dependent variable	T3_WR_R
Type	Mixed effects generalized linear model
Family	binomial
Link	logit

Fixed Effects
	Est.	S.E.	z val.	p
(Intercept)	-3.01	0.22	-13.84	0.00
cond1	-0.49	0.22	-2.26	0.02
T2_LS_S	-0.55	0.15	-3.60	0.00
T2_PI_S	-0.23	0.08	-2.69	0.01
T2_WR_S	-0.42	0.21	-2.02	0.04
T2_SP_S	0.00	0.23	0.01	0.99
T2_RAN_T	0.37	0.08	4.68	0.00
T2_DS_S	-0.19	0.11	-1.74	0.08

Random Effects
Group	Parameter	Std. Dev.
Class:school	(Intercept)	0.00
school	(Intercept)	0.00

Grouping Variables
Group	# groups	ICC
Class:school	101	0.00
school	41	0.00

Observations	2102
Dependent variable	T3_RC_R
Type	Mixed effects generalized linear model
Family	binomial
Link	logit

Fixed Effects
	Est.	S.E.	z val.	p
(Intercept)	-3.55	0.33	-10.70	0.00
cond1	-0.58	0.32	-1.81	0.07
T2_LS_S	-0.91	0.18	-5.08	0.00
T2_PI_S	-0.19	0.09	-2.04	0.04
T2_WR_S	-0.50	0.25	-2.01	0.04
T2_SP_S	0.08	0.28	0.28	0.78
T2_RAN_T	0.39	0.09	4.49	0.00
T2_DS_S	-0.22	0.13	-1.72	0.09

Random Effects
Group	Parameter	Std. Dev.
Class:school	(Intercept)	0.47
school	(Intercept)	0.40

Grouping Variables
Group	# groups	ICC
Class:school	101	0.06
school	41	0.04

Replikasjonsanalyser

Tables

Dependent variable: T2_R_LW_KP

Observations	680
Dependent variable	T2_R_LW_KP
Type	Mixed effects generalized linear model
Family	binomial
Link	logit

Fixed Effects
	Est.	S.E.	z val.	p
(Intercept)	-1.60	0.19	-8.45	0.00
cond2	0.07	0.30	0.25	0.80
T1_LR_S	-0.35	0.11	-3.24	0.00
T1_PI_S	-0.20	0.13	-1.51	0.13
T1_STM	-0.10	0.11	-0.87	0.39
T1_WR_S	-0.87	0.24	-3.61	0.00
T1_RAN_S	0.32	0.11	3.01	0.00
T1_SP_S	0.48	0.23	2.07	0.04

Random Effects
Group	Parameter	Std. Dev.
class:school	(Intercept)	0.25
school	(Intercept)	0.29

Grouping Variables
Group	# groups	ICC
class:school	41	0.02
school	12	0.03

Dependent variable: T2_R_PI_KP

Observations	680
Dependent variable	T2_R_PI_KP
Type	Mixed effects generalized linear model
Family	binomial
Link	logit

Fixed Effects
	Est.	S.E.	z val.	p
(Intercept)	-2.03	0.21	-9.73	0.00
cond2	-0.16	0.28	-0.58	0.56
T1_LR_S	-0.42	0.12	-3.56	0.00
T1_PI_S	-0.18	0.14	-1.24	0.21
T1_STM	-0.44	0.13	-3.28	0.00
T1_WR_S	-0.39	0.27	-1.44	0.15
T1_RAN_S	0.37	0.11	3.25	0.00
T1_SP_S	-0.38	0.30	-1.28	0.20

Random Effects
Group	Parameter	Std. Dev.
class:school	(Intercept)	0.43
school	(Intercept)	0.00

Grouping Variables
Group	# groups	ICC
class:school	41	0.05
school	12	0.00

Dependent variable: T2_R1_WR_KP

Observations	680
Dependent variable	T2_R1_WR_KP
Type	Mixed effects generalized linear model
Family	binomial
Link	logit

Fixed Effects
	Est.	S.E.	z val.	p
(Intercept)	-2.45	0.31	-7.88	0.00
cond2	-0.04	0.47	-0.09	0.93
T1_LR_S	-0.43	0.13	-3.42	0.00
T1_PI_S	-0.20	0.16	-1.25	0.21
T1_STM	-0.18	0.14	-1.31	0.19
T1_WR_S	-0.71	0.32	-2.19	0.03
T1_RAN_S	0.55	0.12	4.59	0.00
T1_SP_S	-0.11	0.34	-0.32	0.75

Random Effects
Group	Parameter	Std. Dev.
class:school	(Intercept)	0.00
school	(Intercept)	0.61

Grouping Variables
Group	# groups	ICC
class:school	41	0.00
school	12	0.10

Dependent variable: T2_RI_RC_KP

Observations	680
Dependent variable	T2_RI_RC_KP
Type	Mixed effects generalized linear model
Family	binomial
Link	logit

Fixed Effects
	Est.	S.E.	z val.	p
(Intercept)	-2.81	0.32	-8.73	0.00
cond2	0.07	0.46	0.15	0.88
T1_LR_S	-0.42	0.14	-3.05	0.00
T1_PI_S	-0.21	0.18	-1.17	0.24
T1_STM	-0.36	0.16	-2.30	0.02
T1_WR_S	-0.90	0.37	-2.43	0.02
T1_RAN_S	0.46	0.12	3.72	0.00
T1_SP_S	0.18	0.37	0.48	0.63

Random Effects
Group	Parameter	Std. Dev.
class:school	(Intercept)	0.00
school	(Intercept)	0.57

Grouping Variables
Group	# groups	ICC
class:school	41	0.00
school	12	0.09

Dependent variable: T2_RI_SP_KP

Observations	680
Dependent variable	T2_RI_SP_KP
Type	Mixed effects generalized linear model
Family	binomial
Link	logit

Fixed Effects
	Est.	S.E.	z val.	p
(Intercept)	-2.13	0.22	-9.69	0.00
cond2	-0.28	0.29	-0.95	0.34
T1_LR_S	-0.27	0.12	-2.25	0.02
T1_PI_S	-0.46	0.15	-3.04	0.00
T1_STM	-0.48	0.14	-3.46	0.00
T1_WR_S	-0.72	0.31	-2.32	0.02
T1_RAN_S	0.30	0.11	2.61	0.01
T1_SP_S	0.00	0.32	0.00	1.00

Random Effects
Group	Parameter	Std. Dev.
class:school	(Intercept)	0.00
school	(Intercept)	0.24

Grouping Variables
Group	# groups	ICC
class:school	41	0.00
school	12	0.02

A visual aid

I was able to make this graph for the naiv logistic regressions (i.e. without class taken into account). If the parameters for the logistic model with clustering are comparable for the naive, then this graph may be useful:

## Registered S3 methods overwritten by 'broom':
##   method            from  
##   tidy.glht         jtools
##   tidy.summary.glht jtools

## Loading required namespace: broom.mixed

Replicate closing gap with on track

N Foldnes

aug23