Skip to main content

Table 3 P-values of the pooled variables after log-transformation and calculation of the median

From: A simple pooling method for variable selection in multiply imputed datasets outperformed complex methods

Dataset

Variables

Pooling Method D1

Pooling Method D2

Pooling Method D3

Pooling Method MPR

Complete Dataset*

N = 200 Corr 0.2 P-out < 0.1

Noise

-1.309554

-1.3096763#

-1.229329

-1.9106885

-1.32294506

Cont4

-2.012191#

-1.8803549

-1.643659

-2.8330669

-2.27920822

Cat1

-1.930734

-1.8166761

-1.840571

-2.6829073#

-2.36633468

Cat2

-2.150953

-1.9567753

-2.043423

-3.0906292#

-2.95750666

Dich1

-1.755433#

-1.746318

-1.696342

-1.8969296

-1.75720332

N = 200 Corr 0.6 P-out < 0.1

Noise

-1.439446

-1.3603044#

-1.285094

-1.7425542

-1.338895

Cont4

-1.83904#

-1.7223828

-1.630539

-2.5040535

-1.95430019

Cat1

-1.773127#

-1.6700756

-1.705565

-2.1858457

-1.92590193

Cat2

-2.182335

-2.0396509

-2.282796

-2.6709277#

-2.54638103

Dich1

-1.610444

-1.5757474

-1.597061

-1.8110715#

-1.72596463

N = 200 Corr 0.2 P-out < 0.05

Noise

-1.873396#

-1.7555967

-1.59476

-2.5543024

-1.90936438

Cont4

-2.262126#

-2.0474303

-1.870487

-3.1615994

-2.53263423

Cat1

-2.143072

-1.9870785

-2.058752

-2.88706#

-2.54509389

Cat2

-2.434034

-2.1368796

-2.237235

-3.1390634#

-3.0642774

Dich1

-1.984843#

-1.9703728

-1.923421

-2.166719

-2.002823

N = 200 Corr 0.6 P-out < 0.05

Noise

-1.683931#

-1.6379195

-1.607315

-2.3293738

-1.86372761

Cont4

-2.161964#

-2.0390992

-1.900202

-2.8912924

-2.374376

Cat1

-2.078314

-1.920269

-2.029932

-2.4134016#

-2.29490206

Cat2

-2.465616

-2.2333715

-2.489852

-2.8188999#

-2.71046546

Dich1

-1.95608

-1.8382816

-1.854454

-2.0588553#

-2.09515079

N = 500 Corr 0.2 P-out < 0.1

Noise

-1.260527

-1.3075#

-1.240598

-1.639111

-1.38193069

Cont4

-2.936592

-2.7997339

-2.435983

-4.1197571#

-4.13946275

Cat1

-3.194703

-3.6122543

-3.598083

-4.8961963#

-5.85087529

Cat2

-3.713544

-4.3224707

-4.399027

-5.60206#

-7.19565311

Dich1

-1.951632#

-1.9122966

-1.833822

-2.0774095

-1.93383386

N = 500 Corr 0.6 P-out < 0.1

Noise

-1.348488#

-1.4031749

-1.401532

-1.9028052

-1.2418635

Cont4

-2.418733

-2.3622861

-2.145894

-3.3509581#

-2.95032364

Cat1

-2.729985

-2.8352247

-2.758217

-4.0065819#

-4.15967193

Cat2

-4.064997

-4.3178549

-4.49222

-5.1426675#

-5.75557227

Dich1

-1.816346

-1.7937226#

-1.764417

-1.9024322

-1.7891513

N = 500 Corr 0.2 P-out < 0.05

Noise

-1.443512

-1.4313625

-1.388748

-1.8756295#

-1.66712132

Cont4

-3.027566

-2.859925

-2.478627

-4.2321024#

-4.1460952

Cat1

-3.320076

-3.6443612

-3.662341

-4.9232755#

-5.84393201

Cat2

-3.729321

-4.3001623

-4.416825

-5.60206#

-7.11509176

Dich1

-2.16806

-2.1131203#

-2.052601

-2.3234185

-2.13648909

N = 500 Corr 0.6 P-out < 0.05

Noise

-1.683931#

-1.6379195

-1.607315

-2.3293738

-1.86372761

Cont4

-2.161964#

-2.0390992

-1.900202

-2.8912924

-2.374376

Cat1

-2.078314

-1.920269

-2.029932

-2.4134016#

-2.29490206

Cat2

-2.465616

-2.2333715

-2.489852

-2.8188999#

-2.71046546

Dich1

-1.95608

-1.8382816

-1.854454

-2.0588553#

-2.09515079

  1. N Number of observations, Corr Correlation, P-out P-value for excluding a variable from the prognostic model, Noise Noise variable, Cont4 Continuous variable 4, Cat1 Categorical variable, Cat2 Categorical variable 2, Dich1 Dichotomous variable 1, D1 D1 method, D2 D2 method, D3 D3 method, MPR Median-P-rule pooling method, complete dataset analyses in complete dataset (reference values for the pooling methods); * = reference values for comparison the pooling methods with the complete data; # = value that is closest to the reference value