专注分享 http://blog.sciencenet.cn/u/laifly 真正的快乐来自心灵的富足,来自于一种教养,来自于对理想的憧憬,也来自于良朋益友的切磋与交流。

博文

思考:4次因子分析

已有 4140 次阅读 2012-6-30 17:37 |个人分类:学习心得|系统分类:科研笔记| 因子分析

     最近比较烦,学习spss非常耗费时间,一个晚上可能折腾过去,就忙了一个分析,还没弄明白,觉得非常累。因子分析本身并不是很难,但是我搞不清楚:何时需要进行2次因子分析,如果2次因子分析还是出现需要删除的变量,那然后继续删除,做第三次因子分析吗??书上很多就是一次因子分析。这个我也拿捏不准。先记录下来,期待以后解决。
 
 
举个例子,这是我做了4次因子分析的结果。
 
分析过程 因子数量 解释总变异量 KMO值 Bartlett's Test球型检验显著性 删除的变量
第一次因子分析 8 63.906% 0.807 0 VC3、VF2、VF1、VF4、VA4、VG5
第二次因子分析 6 66.787% 0.782 0 VB2、VG1、VG6
第三次因子分析 6 69.404% 0.796 0 VD2
第四次因子分析 6 71.300% 0.805 0  
 
 

第一次旋转成份矩阵a

 

成份

1

2

3

4

5

6

7

8

VB6

.740

.147

-.085

.081

.051

.064

.174

.122

VB1

.728

.304

.217

.186

.034

.257

-.174

-.084

VB3

.719

.179

.319

.095

.013

.178

.108

.129

VB2

.698

.321

.216

.020

.116

.308

-.046

-.051

VB5

.697

-.024

.097

.154

.013

.014

.436

.056

VB4

.671

.461

-.100

.154

.035

.209

.010

-.056

VD1

.214

.768

.112

.167

-.072

.149

.020

.109

VC2

.153

.766

.056

.122

.221

.000

.090

.093

VC1

.338

.731

.044

.082

.146

.087

-.048

.092

VC3

.110

.473

.038

-.059

.419

.135

.446

.071

VF2

.235

.442

.337

.361

.210

.176

-.284

.152

VE4

-.014

.003

.766

-.115

.229

.105

.227

-.032

VE3

.244

.055

.760

.138

.009

.072

.159

.110

VF1

.151

.259

.493

.413

.161

-.099

-.056

.320

VF4

.110

.368

.438

.407

.108

.195

.127

.092

VG4

.339

.189

.137

.695

.264

.140

.067

.047

VG2

.031

.360

-.040

.634

.109

.072

.396

.087

VG6

.166

-.030

.019

.596

.224

.034

.152

.468

VG3

.134

.047

.005

.446

.727

.034

.094

-.175

VF3

.136

.153

.350

.028

.684

.049

.041

.113

VG5

-.019

.050

-.112

.202

.608

.089

-.030

.581

VG1

-.105

.212

.209

.155

.534

.032

.279

.198

VA2

.180

.017

-.019

.082

.186

.822

.161

.125

VA1

.360

.232

.066

-.121

-.018

.667

-.125

.120

VA3

.114

.314

.229

.321

-.139

.663

.056

-.145

VA4

.410

-.174

.178

.113

.205

.483

.318

.139

VE2

.072

-.057

.192

.098

.063

.055

.807

-.014

VE1

.269

.149

.198

.220

.167

.076

.632

.228

VD2

.075

.274

.228

.091

.019

.129

.142

.785

 
第2次因子分析,

 

2次因子分析旋转成份矩阵a

 

成份

1

2

3

4

5

6

VB1

.773

.242

-.103

.152

.291

.136

VB2

.735

.261

-.062

.098

.274

.223

VB4

.725

.337

.050

.129

.236

-.141

VB6

.715

.107

.248

.082

.070

-.071

VB3

.710

.212

.204

.028

.153

.321

VB5

.674

-.099

.515

.080

.025

.093

VD1

.294

.771

.047

.013

.179

.033

VC2

.211

.760

.042

.244

-.001

.079

VC1

.394

.662

-.038

.181

.165

.045

VD2

-.035

.503

.465

-.029

.200

.245

VE2

.084

-.149

.718

.063

-.006

.218

VE1

.229

.159

.683

.208

.115

.231

VG2

.067

.404

.510

.421

.115

-.114

VG6

.077

.189

.484

.462

.060

-.060

VG3

.135

-.044

.076

.876

.075

.054

VG4

.344

.266

.265

.602

.181

.031

VF3

.097

.187

.005

.563

.000

.468

VG1

-.126

.271

.282

.480

-.015

.359

VA2

.146

-.035

.204

.193

.828

.055

VA1

.373

.148

-.079

-.083

.697

.102

VA3

.193

.266

.074

.075

.695

.093

VE4

.004

-.021

.117

.100

.111

.840

VE3

.261

.109

.232

.037

.101

.697

 

 
又做了第三次因子分析。

 

第三次旋转成份矩阵a

 

 

旋转成份矩阵a

 

 

成份

1

2

3

4

5

6

VB6

.777

.142

.091

.147

-.015

.080

VB5

.720

-.049

.050

.466

.121

.073

VB1

.704

.300

.310

-.149

.182

.206

VB3

.692

.246

.173

.139

.349

.018

VB4

.689

.383

.277

.025

-.143

.118

VD1

.210

.800

.186

.051

.034

.004

VC2

.147

.782

.009

.055

.059

.213

VC1

.355

.689

.175

-.073

.064

.187

VD2

.006

.497

.161

.341

.353

-.099

VA2

.141

-.014

.837

.212

.050

.161

VA1

.356

.154

.707

-.133

.133

-.077

VA3

.107

.310

.699

.089

.115

.096

VE2

.090

-.104

.009

.775

.197

.060

VE1

.190

.226

.115

.731

.237

.210

VG2

.095

.436

.113

.537

-.144

.335

VE4

-.023

-.031

.119

.147

.790

.109

VE3

.253

.122

.088

.158

.756

.055

VG3

.092

.019

.082

.162

.019

.898

VF3

.026

.225

-.005

.025

.477

.594

VG4

.324

.329

.173

.229

.095

.585

 

4次因子分析

旋转成份矩阵a

 

成份

1

2

3

4

5

6

VB6

.787

.140

.089

.133

.098

-.031

VB5

.723

-.042

.051

.461

.071

.121

VB3

.692

.249

.170

.133

.031

.349

VB1

.686

.336

.307

-.133

.199

.181

VB4

.668

.422

.274

.046

.087

-.132

VD1

.188

.814

.177

.073

-.022

.063

VC2

.119

.804

.002

.087

.183

.086

VC1

.344

.696

.166

-.063

.189

.068

VA2

.162

-.033

.836

.191

.190

.027

VA1

.373

.140

.703

-.160

-.035

.120

VA3

.059

.377

.699

.142

.022

.160

VE2

.090

-.109

.012

.778

.041

.218

VE1

.200

.202

.113

.724

.214

.240

VG2

.067

.465

.112

.581

.270

-.123

VG3

.085

.039

.084

.190

.889

-.023

VF3

.047

.189

-.009

.007

.658

.434

VG4

.318

.337

.171

.247

.576

.067

VE4

-.044

-.010

.122

.166

.099

.815

VE3

.247

.121

.087

.161

.066

.764

 

终于没有可以删除的变量了。

 

看了别人的一些文章,发现还有可以删除的变量。

比如因子6,因为只有2个变量,我们可以说实际可代表性,比较差。

所以又可以删除因子6了。



https://blog.sciencenet.cn/blog-635998-587560.html

上一篇:2010年10月《结构方程模型—amos的操作与应用》(吴明隆)数据
下一篇:学习amos所犯的一个低级错误
收藏 IP: 180.125.97.*| 热度|

1 彭真明

该博文允许注册用户评论 请点击登录 评论 (0 个评论)

数据加载中...

Archiver|手机版|科学网 ( 京ICP备07017567号-12 )

GMT+8, 2024-3-28 20:30

Powered by ScienceNet.cn

Copyright © 2007- 中国科学报社

返回顶部