Overview

This document provides an overview of the final data set used for the meta-analysis. This contains all pairwise parameter pairs as well as our covariates.

The data set used for the meta-analysis is all_pairs. This is available as file all_pairs_core.RData (or all_pairs_core_OLD.RData for using and R version before 3.5). This document provides a general overview of this data set and describes all covariates in this data set.

all_pairs contains all pair-wise group-level parameter estimates, where pairs are defined by the combination of estimation methods.

In total, the data set consists of 121514 observations and 42 columns and considers 9 different methods:

## [1] "Comp MLE"   "Comp Bayes" "No asy"     "No PB"      "No NPB"    
## [6] "No Bayes"   "Beta PP"    "Trait_u PP" "Trait PP"

Estimates and Dependent Variable

The dependent variable for our meta-analysis is the absolute deviation between the two estimates making up a pair, which is contained in variable abs_dev. The data set also contains both individual estimates for each pair as well as the information which method is used in each case. We label the two estimates making up each pair as x and y and the corresponding estimation methods cond_x and cond_y. An overview is provided next:

str(all_pairs[, c("abs_dev", "x", "cond_x", "y", "cond_y") ])

## tibble [121,514 x 5] (S3: tbl_df/tbl/data.frame)
##  $ abs_dev: num [1:121514] 6.31e-05 2.72e-02 1.15e-03 1.07e-02 4.44e-03 ...
##  $ x      : num [1:121514] 0.447 0.695 0.247 0.595 0.529 ...
##  $ cond_x : Factor w/ 9 levels "Comp MLE","Comp Bayes",..: 2 2 2 2 2 2 2 2 2 2 ...
##  $ y      : num [1:121514] 0.447 0.668 0.246 0.585 0.525 ...
##  $ cond_y : Factor w/ 9 levels "Comp MLE","Comp Bayes",..: 7 7 7 7 7 7 7 7 7 7 ...

Note that the data set contains all pair-wise combinations so that each absolute deviation appears twice in the data set. More specifically, for a particular parameter estimate of one data set and two methods A and B, both the combination of A as x and B as y as well as the combination of A as y and B as x appear in the data. An analysis of the absolute deviation (or other DV) therefore usually needs to pick one instance of each such pair.

Metadata

The data set contains several columns with metadata identifying an observations. As discussed above, cond_x and cond_y identify the method used for estimating both estimates of a pair. An overview of the additional metadata is provided next.

str(all_pairs[, c("model", "model2", "dataset", "parameter",
                  "condition", "orig_condition", "parameter_o") ])

## tibble [121,514 x 7] (S3: tbl_df/tbl/data.frame)
##  $ model         : Factor w/ 9 levels "2htsm","c2ht",..: 1 1 1 1 1 1 1 1 1 1 ...
##  $ model2        : Factor w/ 13 levels "2htsm_4","2htsm_5d",..: 1 1 1 1 1 1 1 1 1 1 ...
##  $ dataset       : Factor w/ 164 levels "A2013:2htsm_4",..: 1 1 1 1 1 1 1 1 12 12 ...
##  $ parameter     : Factor w/ 53 levels "2htsm_4:b","2htsm_4:d",..: 1 2 3 4 1 2 3 4 1 2 ...
##  $ condition     : Factor w/ 313 levels "1:2htsm_4","1:pm",..: 75 75 75 75 241 241 241 241 173 173 ...
##  $ orig_condition: chr [1:121514] "encoding" "encoding" "encoding" "encoding" ...
##  $ parameter_o   : chr [1:121514] "2htsm_4:b" "2htsm_4:d" "2htsm_4:D" "2htsm_4:g" ...

model and model2 provide information about the MPT model. An overview of the proportion of observations for each of the different models is shown next:

map(all_pairs[, c("model", "model2")], ~round(prop.table(table(.)), 2))

## $model
## .
## 2htsm  c2ht    pc    pd    pm    hb    rm  real  quad 
##  0.34  0.03  0.15  0.04  0.08  0.04  0.16  0.08  0.08 
## 
## $model2
## .
##  2htsm_4 2htsm_5d 2htsm_6e    c2ht6    c2ht8       pc     pd_s     pd_e 
##     0.23     0.09     0.02     0.02     0.00     0.15     0.02     0.01 
##       pm       hb       rm     real     quad 
##     0.08     0.04     0.16     0.08     0.08

The other variables containing metadata are described next:

dataset: Combination of original dataset name (defined during fitting) and model2. Thus uniquely identifies each data set.
parameter: Name of base parameter nested in model (i.e., model2:parameter). Uniquely identifies each parameter across models. ‘Base parameter’ here means that within-subjects identifiers or conditions are removed from the parameter name.
condition: Uniquely identifies each condition that provides at least one base parameter per model that otherwise would not be unique given the base-model equations. Thus, this includes both within-subjects and between-conditions and also includes the model2 string.
orig_condition: Original condition name from fitting process with 184 unique values.
parameter_o: Original parameter names as used during fitting and with added model2 string to uniquely identify it across models. Has 148 unique values.

Covariates

The data set contains the covariates we had previously agreed on. An overvieew is provided first followed by a description below.

all_pairs %>% 
  select(se_x, se_y, p_hetero:sci_goal) %>% 
  str

## tibble [121,514 x 30] (S3: tbl_df/tbl/data.frame)
##  $ se_x            : num [1:121514] 0.0161 0.1164 0.0245 0.0221 0.0178 ...
##  $ se_y            : num [1:121514] 0.0373 0.1067 0.0284 0.0435 0.0338 ...
##  $ p_hetero        : num [1:121514] 3.87e-17 3.87e-17 3.87e-17 3.87e-17 3.00e-38 ...
##  $ logp_hetero     : num [1:121514] -37.8 -37.8 -37.8 -37.8 -86.4 ...
##  $ log1p_hetero    : num [1:121514] 3.87e-17 3.87e-17 3.87e-17 3.87e-17 3.00e-38 ...
##  $ sd_emp          : num [1:121514] 0.1636 0.0606 0.0206 0.1869 0.1262 ...
##  $ sd_emp_inv      : num [1:121514] 0.4411 0.1702 0.0667 0.5115 0.3267 ...
##  $ rho_max         : num [1:121514] 0.2577 0.0922 0.2577 0.0816 0.1359 ...
##  $ rho_mean        : num [1:121514] 0.1229 0.0483 0.1289 0.0353 0.0602 ...
##  $ rho_fzmean      : num [1:121514] 0.125 0.0484 0.1309 0.0354 0.0605 ...
##  $ rho_med         : num [1:121514] 0.0922 0.0473 0.0816 0.0189 0.0362 ...
##  $ rho_propl5      : num [1:121514] 0 0 0 0 0 ...
##  $ fungi_max       : num [1:121514] 0.1305 0.3106 0.3106 0.0231 0.0734 ...
##  $ fungi_mean      : num [1:121514] 0.0721 0.1275 0.1547 0.017 0.044 ...
##  $ fungi_fzmean    : num [1:121514] 0.0724 0.1311 0.1585 0.017 0.044 ...
##  $ fungi_med       : num [1:121514] 0.0649 0.0649 0.1305 0.0209 0.0453 ...
##  $ fungi_propl5    : num [1:121514] 0 0 0 0 0 0 0 0 0 0 ...
##  $ p_fit_x         : num [1:121514] 0.434 0.434 0.434 0.434 0.492 ...
##  $ p_fit_y         : num [1:121514] 0.428 0.428 0.428 0.428 0.289 ...
##  $ log1p_fit_x     : num [1:121514] 0.361 0.361 0.361 0.361 0.4 ...
##  $ logp_fit_x      : num [1:121514] -0.834 -0.834 -0.834 -0.834 -0.709 ...
##  $ log1p_fit_y     : num [1:121514] 0.357 0.357 0.357 0.357 0.254 ...
##  $ logp_fit_y      : num [1:121514] -0.848 -0.848 -0.848 -0.848 -1.241 ...
##  $ rel_par_weight_x: num [1:121514] 1 0.292 1 0.46 1 ...
##  $ rel_par_weight_y: num [1:121514] 1 0.292 1 0.46 1 ...
##  $ rel_n_x         : num [1:121514] 1536 448 1536 707 1536 ...
##  $ rel_n_y         : num [1:121514] 1536 448 1536 707 1536 ...
##  $ npar            : int [1:121514] 4 4 4 4 4 4 4 4 4 4 ...
##  $ population      : Factor w/ 5 levels "children","college students",..: 2 2 2 2 2 2 2 2 3 3 ...
##  $ sci_goal        : Factor w/ 2 levels "estimation","model_comparison": 1 1 1 1 1 1 1 1 1 1 ...

se_x & se_y: Standard errors (to tap into \(N\)/sample size)
p_hetero, logp_hetero & log1p_hetero: \(p\)-value of non-parametric test of heterogeneity. \(p\)-value is transformed using either log() or log1p() (i.e., log(1+x)).
sd_emp and sd_emp_inv: Hetereogeneity across individual parameter estimates. Calculated as the empirical SD of partial-pooling estimates (as we did not save SD estimate). sd_emp is simply the SD of the estimates, whereas sd_emp_inv is the SD of the estimates after applying the probit link function to the estimates first (i.e., qnorm).
rho_max, rho_mean, rho_fzmean, rho_med, rho_propl5: rho is the correlation estimate from partial pooling model (i.e., correlation of individual-level estimates across parameters). Here, we provide a summary statistic for each parameter based on its absolute correlation with all other model parameters. The idea for rho was to focus on the average absolute correlation with other parameters, rho_mean (idea was to include this only in case this correlation is substantial). rho_max is absolute maximum correlation, rho_fzmean is mean of absolute Fisher-z transormed correlations, rho_med is median of absolute correlation estimates, and rho_propl5 is the proportion of correlation that are larger than absolute .5.
fungi_max, fungi_mean, fungi_fzmean, fungi_med, fungi_propl5: fungi is short for fungibility and describes the across-chain correlation of the group-level model parameters from the partial pooling model (i.e., correlation of group-level parameter estimates across posterior samples). Here, we provide a summary statistic for each parameter based on its absolute correlation with all other model parameters. The idea for fungi was to focus on the maximum absolute correlation with other parameters, fungi_max. fungi_mean is absolute mean correlation, fungi_fzmean is mean of absolute Fisher-z transormed correlations, fungi_med is median of absolute correlation estimates, and fungi_propl5 is the proportion of correlation that are larger than absolute .5
p_fit_x, log1p_fit_x, logp_fit_x, p_fit_y, log1p_fit_y, & logp_fit_y: Model fit based on \(p\)-value of model fit test for methods x and y. \(p\)-value is transformed using either log() or log1p() (i.e., log(1+x)).
rel_par_weight: Proportion of information available for estimating this parameter (or relative parameter weight). Calculated as sum of all branches containing a parameter (where branch is the product of the parameters in the branch).
rel_n: Relative N available for estimating this parameter. rel_par_weight times N.
npar: Number of parameters in a model (not a regsitered or agreed on covariate).
population: Factor with 5 levels describing the population of participants.
sci_goal: Factor with 2 levels describing scientific goal of original paper that used the model. Either estimation (i.e., interest was in parameter values or differences in parameter values across conditions) or model_comparison (i.e., interest was in comparing different models).

Codebook

The following gives a codebook of all_pairs in which some variables with metadata containing many levels are removed

Metadata

Description

Dataset name: codebook_data

The dataset has N=121514 rows and 37 columns. 114180 rows have no missing values on any column.

Metadata for search engines

Date published: 2020-08-27
- keywords: model, model2, abs_dev, x, se_x, cond_x, y, se_y, cond_y, p_hetero, logp_hetero, log1p_hetero, sd_emp, sd_emp_inv, rho_max, rho_mean, rho_fzmean, rho_med, rho_propl5, fungi_max, fungi_mean, fungi_fzmean, fungi_med, fungi_propl5, p_fit_x, p_fit_y, log1p_fit_x, logp_fit_x, log1p_fit_y, logp_fit_y, rel_par_weight_x, rel_par_weight_y, rel_n_x, rel_n_y, npar, population and sci_goal

Variables

model

Distribution

0 missing values.

Summary statistics

name	data_type	ordered	value_labels	n_missing	complete_rate	n_unique	top_counts	label
model	factor	FALSE	1. 2htsm, 2. c2ht, 3. pc, 4. pd, 5. pm, 6. hb, 7. rm, 8. real, 9. quad	0	1	9	2ht: 41592, rm: 19008, pc: 17712, pm: 10304	NA

model2

Distribution

0 missing values.

Summary statistics

name	data_type	ordered	value_labels	n_missing	complete_rate	n_unique	top_counts	label
model2	factor	FALSE	1. 2htsm_4, 2. 2htsm_5d, 3. 2htsm_6e, 4. c2ht6, 5. c2ht8, 6. pc, 7. pd_s, 8. pd_e, 9. pm, 10. hb, 11. rm, 12. real, 13. quad	0	1	13	2ht: 27840, rm: 19008, pc: 17712, 2ht: 10920	NA

abs_dev

Distribution

0 missing values.

Summary statistics

name	data_type	n_missing	complete_rate	min	median	max	mean	sd	hist	label
abs_dev	numeric	0	1	0	0.015	0.97	0.0408976	0.0710231	▇▁▁▁▁	NA

x

Distribution

0 missing values.

Summary statistics

name	data_type	n_missing	complete_rate	min	median	max	mean	sd	hist	label
x	numeric	0	1	0	0.53	1	0.5136987	0.2222714	▂▅▇▆▂	NA

se_x

Distribution

69 missing values.

Summary statistics

name	data_type	n_missing	complete_rate	min	median	max	mean	sd	hist	label
se_x	numeric	69	0.9994322	0	0.033	14	0.0469776	0.1271063	▇▁▁▁▁	NA

cond_x

Distribution

0 missing values.

Summary statistics

name	data_type	ordered	value_labels	n_missing	complete_rate	n_unique	top_counts	label
cond_x	factor	FALSE	1. Comp MLE, 2. Comp Bayes, 3. No asy, 4. No PB, 5. No NPB, 6. No Bayes, 7. Beta PP, 8. Trait_u PP, 9. Trait PP	0	1	9	Com: 13813, Com: 13813, No : 13813, No : 13808	NA

y

Distribution

0 missing values.

Summary statistics

name	data_type	n_missing	complete_rate	min	median	max	mean	sd	hist	label
y	numeric	0	1	0	0.53	1	0.5136987	0.2222714	▂▅▇▆▂	NA

se_y

Distribution

69 missing values.

Summary statistics

name	data_type	n_missing	complete_rate	min	median	max	mean	sd	hist	label
se_y	numeric	69	0.9994322	0	0.033	14	0.0469776	0.1271063	▇▁▁▁▁	NA

cond_y

Distribution

0 missing values.

Summary statistics

name	data_type	ordered	value_labels	n_missing	complete_rate	n_unique	top_counts	label
cond_y	factor	FALSE	1. Comp MLE, 2. Comp Bayes, 3. No asy, 4. No PB, 5. No NPB, 6. No Bayes, 7. Beta PP, 8. Trait_u PP, 9. Trait PP	0	1	9	Com: 13813, Com: 13813, No : 13813, No : 13808	NA

p_hetero

Distribution

0 missing values.

Summary statistics

name	data_type	n_missing	complete_rate	min	median	max	mean	sd	hist	label
p_hetero	numeric	0	1	0	2.6e-23	0.97	0.0313725	0.1367286	▇▁▁▁▁	NA

logp_hetero

Distribution

0 missing values.

Summary statistics

## Warning in inline_hist(., 5): Variable contains Inf or -Inf value(s) that were
## converted to NA.

name	data_type	n_missing	complete_rate	min	median	max	mean	hist	label
logp_hetero	numeric	0	1	-Inf	-52	-0.035	-Inf	▁▁▁▂▇	NA

log1p_hetero

Distribution

0 missing values.

Summary statistics

name	data_type	n_missing	complete_rate	min	median	max	mean	sd	hist	label
log1p_hetero	numeric	0	1	0	2.6e-23	0.68	0.0246592	0.102541	▇▁▁▁▁	NA

sd_emp

Distribution

5152 missing values.

Summary statistics

name	data_type	n_missing	complete_rate	min	median	max	mean	sd	hist	label
sd_emp	numeric	5152	0.9576016	0.00071	0.091	0.39	0.1075554	0.0766734	▇▆▃▁▁	NA

sd_emp_inv

Distribution

5152 missing values.

Summary statistics

name	data_type	n_missing	complete_rate	min	median	max	mean	sd	hist	label
sd_emp_inv	numeric	5152	0.9576016	0.0056	0.3	1.6	0.3485758	0.2578424	▇▅▂▁▁	NA

rho_max

Distribution

5152 missing values.

Summary statistics

name	data_type	n_missing	complete_rate	min	median	max	mean	sd	hist	label
rho_max	numeric	5152	0.9576016	0.016	0.34	0.94	0.3826195	0.216775	▆▇▆▃▂	NA

rho_mean

Distribution

5152 missing values.

Summary statistics

name	data_type	n_missing	complete_rate	min	median	max	mean	sd	hist	label
rho_mean	numeric	5152	0.9576016	0.0087	0.18	0.73	0.2139975	0.1346426	▇▇▃▁▁	NA

rho_fzmean

Distribution

5152 missing values.

Summary statistics

name	data_type	n_missing	complete_rate	min	median	max	mean	sd	hist	label
rho_fzmean	numeric	5152	0.9576016	0.0087	0.19	1	0.2324338	0.1652217	▇▅▂▁▁	NA

rho_med

Distribution

5152 missing values.

Summary statistics

name	data_type	n_missing	complete_rate	min	median	max	mean	sd	hist	label
rho_med	numeric	5152	0.9576016	0.0068	0.17	0.79	0.2062475	0.1492717	▇▅▂▁▁	NA

rho_propl5

Distribution

5152 missing values.

Summary statistics

name	data_type	n_missing	complete_rate	min	median	max	mean	sd	hist	label
rho_propl5	numeric	5152	0.9576016	0	0	1	0.0927087	0.1870121	▇▁▁▁▁	NA

fungi_max

Distribution

5152 missing values.

Summary statistics

name	data_type	n_missing	complete_rate	min	median	max	mean	sd	hist	label
fungi_max	numeric	5152	0.9576016	0.0061	0.25	0.93	0.3061428	0.2097439	▇▇▃▂▁	NA

fungi_mean

Distribution

5152 missing values.

Summary statistics

name	data_type	n_missing	complete_rate	min	median	max	mean	sd	hist	label
fungi_mean	numeric	5152	0.9576016	0.0047	0.11	0.75	0.1177429	0.0743623	▇▂▁▁▁	NA

fungi_fzmean

Distribution

5152 missing values.

Summary statistics

name	data_type	n_missing	complete_rate	min	median	max	mean	sd	hist	label
fungi_fzmean	numeric	5152	0.9576016	0.0047	0.11	1.1	0.1240729	0.0885148	▇▁▁▁▁	NA

fungi_med

Distribution

5152 missing values.

Summary statistics

name	data_type	n_missing	complete_rate	min	median	max	mean	sd	hist	label
fungi_med	numeric	5152	0.9576016	0.0014	0.074	0.79	0.0936153	0.0826948	▇▂▁▁▁	NA

fungi_propl5

Distribution

5152 missing values.

Summary statistics

name	data_type	n_missing	complete_rate	min	median	max	mean	sd	hist	label
fungi_propl5	numeric	5152	0.9576016	0	0	1	0.0280125	0.0833873	▇▁▁▁▁	NA

p_fit_x

Distribution

0 missing values.

Summary statistics

name	data_type	n_missing	complete_rate	min	median	max	mean	sd	hist	label
p_fit_x	numeric	0	1	0	0.38	1	0.3852175	0.3389751	▇▂▅▂▃	NA

p_fit_y

Distribution

0 missing values.

Summary statistics

name	data_type	n_missing	complete_rate	min	median	max	mean	sd	hist	label
p_fit_y	numeric	0	1	0	0.38	1	0.3852175	0.3389751	▇▂▅▂▃	NA

log1p_fit_x

Distribution

0 missing values.

Summary statistics

name	data_type	n_missing	complete_rate	min	median	max	mean	sd	hist	label
log1p_fit_x	numeric	0	1	0	0.32	0.69	0.29651	0.241329	▇▂▅▅▃	NA

logp_fit_x

Distribution

0 missing values.

Summary statistics

## Warning in inline_hist(., 5): Variable contains Inf or -Inf value(s) that were
## converted to NA.

name	data_type	n_missing	complete_rate	min	median	max	mean	hist	label
logp_fit_x	numeric	0	1	-Inf	-0.97	0	-Inf	▁▁▁▁▇	NA

log1p_fit_y

Distribution

0 missing values.

Summary statistics

name	data_type	n_missing	complete_rate	min	median	max	mean	sd	hist	label
log1p_fit_y	numeric	0	1	0	0.32	0.69	0.29651	0.241329	▇▂▅▅▃	NA

logp_fit_y

Distribution

0 missing values.

Summary statistics

## Warning in inline_hist(., 5): Variable contains Inf or -Inf value(s) that were
## converted to NA.

name	data_type	n_missing	complete_rate	min	median	max	mean	hist	label
logp_fit_y	numeric	0	1	-Inf	-0.97	0	-Inf	▁▁▁▁▇	NA

rel_par_weight_x

Distribution

319 missing values.

Summary statistics

name	data_type	n_missing	complete_rate	min	median	max	mean	sd	hist	label
rel_par_weight_x	numeric	319	0.9973748	0.0021	0.33	1	0.4322523	0.2899259	▇▇▇▁▅	NA

rel_par_weight_y

Distribution

319 missing values.

Summary statistics

name	data_type	n_missing	complete_rate	min	median	max	mean	sd	hist	label
rel_par_weight_y	numeric	319	0.9973748	0.0021	0.33	1	0.4322523	0.2899259	▇▇▇▁▅	NA

rel_n_x

Distribution

319 missing values.

Summary statistics

name	data_type	n_missing	complete_rate	min	median	max	mean	sd	hist	label
rel_n_x	numeric	319	0.9973748	5.1	1260	1e+05	3167.097	7886.651	▇▁▁▁▁	NA

rel_n_y

Distribution

319 missing values.

Summary statistics

name	data_type	n_missing	complete_rate	min	median	max	mean	sd	hist	label
rel_n_y	numeric	319	0.9973748	5.1	1260	1e+05	3167.097	7886.651	▇▁▁▁▁	NA

npar

Distribution

0 missing values.

Summary statistics

name	data_type	n_missing	complete_rate	min	median	max	mean	sd	hist	label
npar	numeric	0	1	2	4	13	5.059977	2.633611	▇▂▁▁▁	NA

population

Distribution

1440 missing values.

Summary statistics

name	data_type	ordered	value_labels	n_missing	complete_rate	n_unique	top_counts	label
population	factor	FALSE	1. children, 2. college students, 3. older adults, 4. other, 5. patients	1440	0.9881495	5	col: 88270, pat: 12816, old: 10032, oth: 5320	NA

sci_goal

Distribution

1440 missing values.

Summary statistics

name	data_type	ordered	value_labels	n_missing	complete_rate	n_unique	top_counts	label
sci_goal	factor	FALSE	1. estimation, 2. model_comparison	1440	0.9881495	2	est: 116156, mod: 3918	NA

Missingness report

Codebook table

JSON-LD metadata

The following JSON-LD can be found by search engines, if you share this codebook publicly on the web.

{
  "name": "codebook_data",
  "datePublished": "2020-08-27",
  "description": "The dataset has N=121514 rows and 37 columns.\n114180 rows have no missing values on any column.\n\n\n## Table of variables\nThis table contains variable names, labels, and number of missing values.\nSee the complete codebook for more.\n\n|name             |label | n_missing|\n|:----------------|:-----|---------:|\n|model            |NA    |         0|\n|model2           |NA    |         0|\n|abs_dev          |NA    |         0|\n|x                |NA    |         0|\n|se_x             |NA    |        69|\n|cond_x           |NA    |         0|\n|y                |NA    |         0|\n|se_y             |NA    |        69|\n|cond_y           |NA    |         0|\n|p_hetero         |NA    |         0|\n|logp_hetero      |NA    |         0|\n|log1p_hetero     |NA    |         0|\n|sd_emp           |NA    |      5152|\n|sd_emp_inv       |NA    |      5152|\n|rho_max          |NA    |      5152|\n|rho_mean         |NA    |      5152|\n|rho_fzmean       |NA    |      5152|\n|rho_med          |NA    |      5152|\n|rho_propl5       |NA    |      5152|\n|fungi_max        |NA    |      5152|\n|fungi_mean       |NA    |      5152|\n|fungi_fzmean     |NA    |      5152|\n|fungi_med        |NA    |      5152|\n|fungi_propl5     |NA    |      5152|\n|p_fit_x          |NA    |         0|\n|p_fit_y          |NA    |         0|\n|log1p_fit_x      |NA    |         0|\n|logp_fit_x       |NA    |         0|\n|log1p_fit_y      |NA    |         0|\n|logp_fit_y       |NA    |         0|\n|rel_par_weight_x |NA    |       319|\n|rel_par_weight_y |NA    |       319|\n|rel_n_x          |NA    |       319|\n|rel_n_y          |NA    |       319|\n|npar             |NA    |         0|\n|population       |NA    |      1440|\n|sci_goal         |NA    |      1440|\n\n### Note\nThis dataset was automatically described using the [codebook R package](https://rubenarslan.github.io/codebook/) (version 0.8.2).",
  "keywords": ["model", "model2", "abs_dev", "x", "se_x", "cond_x", "y", "se_y", "cond_y", "p_hetero", "logp_hetero", "log1p_hetero", "sd_emp", "sd_emp_inv", "rho_max", "rho_mean", "rho_fzmean", "rho_med", "rho_propl5", "fungi_max", "fungi_mean", "fungi_fzmean", "fungi_med", "fungi_propl5", "p_fit_x", "p_fit_y", "log1p_fit_x", "logp_fit_x", "log1p_fit_y", "logp_fit_y", "rel_par_weight_x", "rel_par_weight_y", "rel_n_x", "rel_n_y", "npar", "population", "sci_goal"],
  "@context": "http://schema.org/",
  "@type": "Dataset",
  "variableMeasured": [
    {
      "name": "model",
      "value": "1. 2htsm,\n2. c2ht,\n3. pc,\n4. pd,\n5. pm,\n6. hb,\n7. rm,\n8. real,\n9. quad",
      "@type": "propertyValue"
    },
    {
      "name": "model2",
      "value": "1. 2htsm_4,\n2. 2htsm_5d,\n3. 2htsm_6e,\n4. c2ht6,\n5. c2ht8,\n6. pc,\n7. pd_s,\n8. pd_e,\n9. pm,\n10. hb,\n11. rm,\n12. real,\n13. quad",
      "@type": "propertyValue"
    },
    {
      "name": "abs_dev",
      "@type": "propertyValue"
    },
    {
      "name": "x",
      "@type": "propertyValue"
    },
    {
      "name": "se_x",
      "@type": "propertyValue"
    },
    {
      "name": "cond_x",
      "value": "1. Comp MLE,\n2. Comp Bayes,\n3. No asy,\n4. No PB,\n5. No NPB,\n6. No Bayes,\n7. Beta PP,\n8. Trait_u PP,\n9. Trait PP",
      "@type": "propertyValue"
    },
    {
      "name": "y",
      "@type": "propertyValue"
    },
    {
      "name": "se_y",
      "@type": "propertyValue"
    },
    {
      "name": "cond_y",
      "value": "1. Comp MLE,\n2. Comp Bayes,\n3. No asy,\n4. No PB,\n5. No NPB,\n6. No Bayes,\n7. Beta PP,\n8. Trait_u PP,\n9. Trait PP",
      "@type": "propertyValue"
    },
    {
      "name": "p_hetero",
      "@type": "propertyValue"
    },
    {
      "name": "logp_hetero",
      "@type": "propertyValue"
    },
    {
      "name": "log1p_hetero",
      "@type": "propertyValue"
    },
    {
      "name": "sd_emp",
      "@type": "propertyValue"
    },
    {
      "name": "sd_emp_inv",
      "@type": "propertyValue"
    },
    {
      "name": "rho_max",
      "@type": "propertyValue"
    },
    {
      "name": "rho_mean",
      "@type": "propertyValue"
    },
    {
      "name": "rho_fzmean",
      "@type": "propertyValue"
    },
    {
      "name": "rho_med",
      "@type": "propertyValue"
    },
    {
      "name": "rho_propl5",
      "@type": "propertyValue"
    },
    {
      "name": "fungi_max",
      "@type": "propertyValue"
    },
    {
      "name": "fungi_mean",
      "@type": "propertyValue"
    },
    {
      "name": "fungi_fzmean",
      "@type": "propertyValue"
    },
    {
      "name": "fungi_med",
      "@type": "propertyValue"
    },
    {
      "name": "fungi_propl5",
      "@type": "propertyValue"
    },
    {
      "name": "p_fit_x",
      "@type": "propertyValue"
    },
    {
      "name": "p_fit_y",
      "@type": "propertyValue"
    },
    {
      "name": "log1p_fit_x",
      "@type": "propertyValue"
    },
    {
      "name": "logp_fit_x",
      "@type": "propertyValue"
    },
    {
      "name": "log1p_fit_y",
      "@type": "propertyValue"
    },
    {
      "name": "logp_fit_y",
      "@type": "propertyValue"
    },
    {
      "name": "rel_par_weight_x",
      "@type": "propertyValue"
    },
    {
      "name": "rel_par_weight_y",
      "@type": "propertyValue"
    },
    {
      "name": "rel_n_x",
      "@type": "propertyValue"
    },
    {
      "name": "rel_n_y",
      "@type": "propertyValue"
    },
    {
      "name": "npar",
      "@type": "propertyValue"
    },
    {
      "name": "population",
      "value": "1. children,\n2. college students,\n3. older adults,\n4. other,\n5. patients",
      "@type": "propertyValue"
    },
    {
      "name": "sci_goal",
      "value": "1. estimation,\n2. model_comparison",
      "@type": "propertyValue"
    }
  ]
}`

Data Set and Covariates for Meta-Analysis (i.e., Parameter Pairs across Methods)

MPT Reanalysis Project (code and text current document: Henrik Singmann)

Results from: 2020-08-27

Overview

Estimates and Dependent Variable

Metadata

Covariates

Codebook

Metadata

Description

Variables

model

Distribution

Summary statistics

model2

Distribution

Summary statistics

abs_dev

Distribution

Summary statistics

x

Distribution

Summary statistics

se_x

Distribution

Summary statistics

cond_x

Distribution

Summary statistics

y

Distribution

Summary statistics

se_y

Distribution

Summary statistics

cond_y

Distribution

Summary statistics

p_hetero

Distribution

Summary statistics

logp_hetero

Distribution

Summary statistics

log1p_hetero

Distribution

Summary statistics

sd_emp

Distribution

Summary statistics

sd_emp_inv

Distribution

Summary statistics

rho_max

Distribution

Summary statistics

rho_mean

Distribution

Summary statistics

rho_fzmean

Distribution

Summary statistics

rho_med

Distribution

Summary statistics

rho_propl5

Distribution

Summary statistics

fungi_max

Distribution

Summary statistics

fungi_mean

Distribution

Summary statistics

fungi_fzmean

Distribution

Summary statistics

fungi_med

Distribution

Summary statistics