Hierarchical Models of Self-Concept across Genders and Sciences/Humanities for College Students in Taiwan
Table 4
Path coefficients and goodness-of-Fit indexes obtained by SEM based on the model of GSC as the predictor of correlated FSCs (Model B in Figure 1).
Sample
Path coefficients of GSC on
Goodness of fit
RSC
ISC
ASC
SSC
ESC
CSC
df
CFI
NNFI
RMSEA
All
.124
.173
.281
.391
.378
.192
1044.232
69
.993
.989
.022
All_mgg
.124
.173
.281
.391
.378
.192
17928.499
204
.842
.838
.078
All_F
.084
.161
.286
.409
.384
.208
457.890
69
.994
.991
.019
All_M
.160
.190
.293
.379
.372
.190
565.475
69
.992
.988
.023
Engineering
.229
.217
.265
.380
.351
.212
344.444
69
.991
.986
.026
Engineering_mgg
.229
.217
.265
.380
.351
.212
1227.970
204
.964
.963
.041
Engineering_F
.162
.260
.224
.425
.396
.239
85.757
69
.995
.992
.017
Engineering_M
.242
.211
.275
.375
.345
.211
320.597
69
.990
.986
.027
Natural sci.
.117
.304
.250
.320
.349
.108
102.413
69
.991
.986
.023
Natural sci._mgg
.117
.304
.250
.320
.349
.108
392.127
204
.939
.937
.045
Natural sci._F
.170
.295
.144
.361
.394
.199
73.458
69
.994
.991
.014
Natural sci._M
.063
.310
.307
.283
.321
.083
92.617
69
.991
.986
.024
Education
.172
.240
.378
.467
.405
.159
119.943
69
.990
.985
.027
Education_mgg
.172
.240
.378
.467
.405
.159
1011.146
204
.845
.841
.088
Education_F
.142
.243
.381
.514
.400
.145
109.807
69
.987
.980
.029
Education_M
.275
.328
.337
.388
.417
.210
92.040
69
.986
.978
.033
Business
.122
.177
.287
.398
.411
.207
260.989
69
.993
.990
.020
Business_mgg
.122
.177
.287
.398
.411
.207
4877.513
204
.853
.849
.083
Business_F
.086
.169
.294
.396
.395
.244
198.855
69
.993
.989
.020
Business_M
.114
.152
.299
.407
.429
.179
133.818
69
.993
.989
.022
SEM: structural equation modeling; GSC: general self-concept; FSCs: field-specific self-concepts, including RSC, ISC, ASC, SSC, ESC, and SSC here; RSC: realistic self-concept; ISC: investigative self-concept; ASC: artistic self-concept; SSC: social self-concept; ESC: enterprising self-concept; CSC: conventional self-concept; F: female; M: male; mgg: multigroup SEM with all parameter estimates being set invariant across genders; mgf: multigroup SEM with all parameter estimates being set invariant across four fields of study (engineering, natural sciences, education, and business). Natural sci.: natural sciences. The bold path coefficients and are not significant at the .05 level.