Paketlerin Yüklenmesi
library(dplyr)
library(lavaan)
library(semPlot)
library(psych)cast_cov_mat <- lav_matrix_lower2full(
scan("castejon.txt")
)colnames(cast_cov_mat) <- paste0("item", 1:22)
rownames(cast_cov_mat) <- paste0("item", 1:22)model_1 <- '
dogal =~ item1 + item2 + item3 + item4 + item5 + item6
bedensel =~ item7 + item8 + item9 + item10
uzamsal =~ item11 + item12 + item13
muziksel =~ item14 + item15 + item16
mantiksal =~ item17 + item18 + item19
dilsel =~ item20 + item21 + item22
'
fit_1 <- cfa(model_1, sample.cov = cast_cov_mat, sample.nobs = 393)
summary(fit_1, fit.measures=TRUE, standardized=TRUE)## lavaan 0.6-21 ended normally after 51 iterations
##
## Estimator ML
## Optimization method NLMINB
## Number of model parameters 59
##
## Number of observations 393
##
## Model Test User Model:
##
## Test statistic 680.542
## Degrees of freedom 194
## P-value (Chi-square) 0.000
##
## Model Test Baseline Model:
##
## Test statistic 3962.137
## Degrees of freedom 231
## P-value 0.000
##
## User Model versus Baseline Model:
##
## Comparative Fit Index (CFI) 0.870
## Tucker-Lewis Index (TLI) 0.845
##
## Loglikelihood and Information Criteria:
##
## Loglikelihood user model (H0) -10616.331
## Loglikelihood unrestricted model (H1) -10276.060
##
## Akaike (AIC) 21350.663
## Bayesian (BIC) 21585.117
## Sample-size adjusted Bayesian (SABIC) 21397.912
##
## Root Mean Square Error of Approximation:
##
## RMSEA 0.080
## 90 Percent confidence interval - lower 0.073
## 90 Percent confidence interval - upper 0.086
## P-value H_0: RMSEA <= 0.050 0.000
## P-value H_0: RMSEA >= 0.080 0.496
##
## Standardized Root Mean Square Residual:
##
## SRMR 0.104
##
## Parameter Estimates:
##
## Standard errors Standard
## Information Expected
## Information saturated (h1) model Structured
##
## Latent Variables:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## dogal =~
## item1 1.000 0.641 0.642
## item2 0.909 0.088 10.333 0.000 0.583 0.583
## item3 1.095 0.091 12.089 0.000 0.702 0.703
## item4 1.225 0.093 13.209 0.000 0.785 0.786
## item5 1.367 0.095 14.319 0.000 0.877 0.878
## item6 1.406 0.096 14.576 0.000 0.902 0.903
## bedensel =~
## item7 1.000 0.682 0.683
## item8 1.048 0.110 9.512 0.000 0.715 0.716
## item9 0.600 0.091 6.577 0.000 0.409 0.410
## item10 0.693 0.093 7.432 0.000 0.472 0.473
## uzamsal =~
## item11 1.000 0.873 0.874
## item12 0.978 0.047 20.591 0.000 0.854 0.855
## item13 0.939 0.048 19.570 0.000 0.820 0.821
## muziksel =~
## item14 1.000 0.605 0.606
## item15 1.523 0.212 7.192 0.000 0.921 0.922
## item16 0.556 0.094 5.941 0.000 0.336 0.337
## mantiksal =~
## item17 1.000 0.530 0.530
## item18 1.587 0.159 10.012 0.000 0.841 0.842
## item19 1.526 0.153 9.964 0.000 0.808 0.809
## dilsel =~
## item20 1.000 0.817 0.818
## item21 1.004 0.115 8.714 0.000 0.820 0.821
## item22 0.361 0.070 5.175 0.000 0.295 0.295
##
## Covariances:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## dogal ~~
## bedensel 0.132 0.030 4.364 0.000 0.303 0.303
## uzamsal 0.201 0.035 5.674 0.000 0.359 0.359
## muziksel 0.078 0.025 3.146 0.002 0.201 0.201
## mantiksal 0.109 0.023 4.648 0.000 0.321 0.321
## dilsel 0.142 0.034 4.211 0.000 0.271 0.271
## bedensel ~~
## uzamsal 0.268 0.044 6.110 0.000 0.450 0.450
## muziksel 0.159 0.035 4.478 0.000 0.384 0.384
## mantiksal 0.148 0.029 5.031 0.000 0.410 0.410
## dilsel 0.165 0.041 4.069 0.000 0.296 0.296
## uzamsal ~~
## muziksel 0.105 0.034 3.115 0.002 0.198 0.198
## mantiksal 0.283 0.040 7.091 0.000 0.612 0.612
## dilsel 0.168 0.045 3.776 0.000 0.236 0.236
## muziksel ~~
## mantiksal 0.096 0.025 3.927 0.000 0.301 0.301
## dilsel 0.100 0.033 3.004 0.003 0.202 0.202
## mantiksal ~~
## dilsel 0.080 0.028 2.832 0.005 0.185 0.185
##
## Variances:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## .item1 0.586 0.045 13.172 0.000 0.586 0.588
## .item2 0.658 0.049 13.398 0.000 0.658 0.660
## .item3 0.504 0.039 12.832 0.000 0.504 0.505
## .item4 0.381 0.032 12.027 0.000 0.381 0.382
## .item5 0.229 0.024 9.731 0.000 0.229 0.230
## .item6 0.184 0.022 8.479 0.000 0.184 0.185
## .item7 0.533 0.057 9.386 0.000 0.533 0.534
## .item8 0.487 0.057 8.514 0.000 0.487 0.488
## .item9 0.830 0.064 13.010 0.000 0.830 0.832
## .item10 0.774 0.062 12.568 0.000 0.774 0.776
## .item11 0.235 0.028 8.325 0.000 0.235 0.236
## .item12 0.268 0.029 9.203 0.000 0.268 0.269
## .item13 0.325 0.031 10.428 0.000 0.325 0.326
## .item14 0.632 0.064 9.870 0.000 0.632 0.633
## .item15 0.149 0.105 1.422 0.155 0.149 0.149
## .item16 0.884 0.065 13.557 0.000 0.884 0.887
## .item17 0.717 0.055 12.963 0.000 0.717 0.719
## .item18 0.291 0.043 6.760 0.000 0.291 0.291
## .item19 0.344 0.043 8.040 0.000 0.344 0.345
## .item20 0.331 0.075 4.430 0.000 0.331 0.332
## .item21 0.325 0.075 4.328 0.000 0.325 0.326
## .item22 0.910 0.066 13.704 0.000 0.910 0.913
## dogal 0.411 0.060 6.906 0.000 1.000 1.000
## bedensel 0.465 0.073 6.328 0.000 1.000 1.000
## uzamsal 0.762 0.073 10.472 0.000 1.000 1.000
## muziksel 0.366 0.071 5.125 0.000 1.000 1.000
## mantiksal 0.280 0.054 5.217 0.000 1.000 1.000
## dilsel 0.667 0.098 6.832 0.000 1.000 1.000YORUM:
Ki-kare = 680.542, df = 194, p < .001
Model ile veri arasında fark var, yani mükemmel uyum yok
diyebiliriz.
CFI = 0.870, TLI = 0.845
Kabul edilebilir sınır ≥ 0.90’dır. Burada değerler sınırın altında
olduğundan iyi uyum olmadığını söyleyebiliriz. B
RMSEA = 0.080 (CI: 0.073–0.086)
Tam sınırda olduğundan ≤ 0.08 kabul edilebilir uyum olarak
yorumlanabilir.
SRMR = 0.104
→ İyi uyum için ≤ 0.08 beklenir. 0.104 yüksek olduğundan kötü uyum
yorumu yapılabilir.
İkinci Dereceli Altı Faktörlü Modelin Uyumu
model_2 <- '
dogal =~ item1 + item2 + item3 + item4 + item5 + item6
bedensel =~ item7 + item8 + item9 + item10
uzamsal =~ item11 + item12 + item13
muziksel =~ item14 + item15 + item16
mantiksal =~ item17 + item18 + item19
dilsel =~ item20 + item21 + item22
ikinci_derece =~ dogal + bedensel + uzamsal + muziksel + mantiksal + dilsel
'
fit_2 <- cfa(model_2, sample.cov = cast_cov_mat, sample.nobs = 393)
summary(fit_2, fit.measures=TRUE, standardized=TRUE)## lavaan 0.6-21 ended normally after 57 iterations
##
## Estimator ML
## Optimization method NLMINB
## Number of model parameters 50
##
## Number of observations 393
##
## Model Test User Model:
##
## Test statistic 708.967
## Degrees of freedom 203
## P-value (Chi-square) 0.000
##
## Model Test Baseline Model:
##
## Test statistic 3962.137
## Degrees of freedom 231
## P-value 0.000
##
## User Model versus Baseline Model:
##
## Comparative Fit Index (CFI) 0.864
## Tucker-Lewis Index (TLI) 0.846
##
## Loglikelihood and Information Criteria:
##
## Loglikelihood user model (H0) -10630.544
## Loglikelihood unrestricted model (H1) -10276.060
##
## Akaike (AIC) 21361.087
## Bayesian (BIC) 21559.778
## Sample-size adjusted Bayesian (SABIC) 21401.129
##
## Root Mean Square Error of Approximation:
##
## RMSEA 0.080
## 90 Percent confidence interval - lower 0.073
## 90 Percent confidence interval - upper 0.086
## P-value H_0: RMSEA <= 0.050 0.000
## P-value H_0: RMSEA >= 0.080 0.470
##
## Standardized Root Mean Square Residual:
##
## SRMR 0.105
##
## Parameter Estimates:
##
## Standard errors Standard
## Information Expected
## Information saturated (h1) model Structured
##
## Latent Variables:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## dogal =~
## item1 1.000 0.641 0.642
## item2 0.910 0.088 10.328 0.000 0.583 0.584
## item3 1.095 0.091 12.068 0.000 0.702 0.702
## item4 1.225 0.093 13.189 0.000 0.785 0.786
## item5 1.369 0.096 14.310 0.000 0.877 0.878
## item6 1.407 0.097 14.557 0.000 0.902 0.903
## bedensel =~
## item7 1.000 0.674 0.675
## item8 1.066 0.115 9.290 0.000 0.718 0.719
## item9 0.605 0.093 6.492 0.000 0.407 0.408
## item10 0.711 0.096 7.443 0.000 0.479 0.480
## uzamsal =~
## item11 1.000 0.871 0.873
## item12 0.980 0.048 20.488 0.000 0.854 0.855
## item13 0.943 0.048 19.557 0.000 0.822 0.823
## muziksel =~
## item14 1.000 0.628 0.629
## item15 1.376 0.189 7.263 0.000 0.864 0.865
## item16 0.593 0.095 6.241 0.000 0.372 0.373
## mantiksal =~
## item17 1.000 0.534 0.535
## item18 1.573 0.156 10.061 0.000 0.840 0.841
## item19 1.511 0.151 10.018 0.000 0.807 0.808
## dilsel =~
## item20 1.000 0.849 0.850
## item21 0.930 0.115 8.090 0.000 0.789 0.790
## item22 0.349 0.069 5.085 0.000 0.296 0.296
## ikinci_derece =~
## dogal 1.000 0.476 0.476
## bedensel 1.373 0.244 5.618 0.000 0.622 0.622
## uzamsal 2.154 0.331 6.518 0.000 0.755 0.755
## muziksel 0.887 0.200 4.445 0.000 0.431 0.431
## mantiksal 1.303 0.228 5.723 0.000 0.745 0.745
## dilsel 0.982 0.223 4.413 0.000 0.353 0.353
##
## Variances:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## .item1 0.587 0.045 13.175 0.000 0.587 0.588
## .item2 0.658 0.049 13.398 0.000 0.658 0.659
## .item3 0.505 0.039 12.837 0.000 0.505 0.507
## .item4 0.382 0.032 12.033 0.000 0.382 0.383
## .item5 0.228 0.023 9.692 0.000 0.228 0.228
## .item6 0.184 0.022 8.472 0.000 0.184 0.185
## .item7 0.543 0.058 9.431 0.000 0.543 0.545
## .item8 0.482 0.058 8.259 0.000 0.482 0.483
## .item9 0.831 0.064 12.993 0.000 0.831 0.834
## .item10 0.768 0.062 12.466 0.000 0.768 0.770
## .item11 0.238 0.028 8.358 0.000 0.238 0.239
## .item12 0.268 0.029 9.160 0.000 0.268 0.269
## .item13 0.322 0.031 10.331 0.000 0.322 0.322
## .item14 0.603 0.066 9.205 0.000 0.603 0.605
## .item15 0.251 0.094 2.673 0.008 0.251 0.252
## .item16 0.859 0.065 13.270 0.000 0.859 0.861
## .item17 0.712 0.055 12.922 0.000 0.712 0.714
## .item18 0.292 0.044 6.686 0.000 0.292 0.293
## .item19 0.346 0.043 7.971 0.000 0.346 0.347
## .item20 0.277 0.085 3.245 0.001 0.277 0.278
## .item21 0.375 0.077 4.881 0.000 0.375 0.376
## .item22 0.910 0.066 13.704 0.000 0.910 0.912
## .dogal 0.317 0.048 6.636 0.000 0.773 0.773
## .bedensel 0.278 0.053 5.283 0.000 0.613 0.613
## .uzamsal 0.327 0.053 6.227 0.000 0.431 0.431
## .muziksel 0.321 0.060 5.324 0.000 0.814 0.814
## .mantiksal 0.127 0.029 4.341 0.000 0.445 0.445
## .dilsel 0.630 0.098 6.410 0.000 0.875 0.875
## ikinci_derece 0.093 0.026 3.648 0.000 1.000 1.000YORUM:
Ki-kare = 708.967, df = 203, p < .001
Model ile veri arasında fark var, mükemmel uyum yok diyebiliriz. Ancak
altı faktörlü modelden biraz daha yüksektir.
CFI = 0.864, TLI = 0.846
Altı faktörlü modelde CFI = 0.870, TLI = 0.845 idi. Burada CFI daha
düşük, TLI aynı, yine kötü uyum var yorumu yapılabilir.
RMSEA = 0.080 (CI: 0.073–0.086)
RMSEA aynı değeri almış. Bu sınırda kabul edilebilir uyum yorumu
yapılabilir.
SRMR = 0.105
Altı faktörlü modele göre küçük bir artış var ama yine dekabul
edilebilir sınırın üzerinde diyebiliriz.
Model-3
model_3 <- '
dogal =~ item1 + item2 + item3 + item4 + item5 + item6
bedensel =~ item7 + item8 + item9 + item10
uzamsal =~ item11 + item12 + item13
muziksel =~ item14 + item15 + item16
mantiksal =~ item17 + item18 + item19
dilsel =~ item20 + item21 + item22
bilissel =~ dogal + mantiksal + dilsel + uzamsal
bilissel_olmayan =~ muziksel + bedensel
'
fit_3 <- cfa(model_3, sample.cov = cast_cov_mat, sample.nobs = 393)
summary(fit_3, fit.measures=TRUE, standardized=TRUE)## lavaan 0.6-21 ended normally after 64 iterations
##
## Estimator ML
## Optimization method NLMINB
## Number of model parameters 51
##
## Number of observations 393
##
## Model Test User Model:
##
## Test statistic 700.020
## Degrees of freedom 202
## P-value (Chi-square) 0.000
##
## Model Test Baseline Model:
##
## Test statistic 3962.137
## Degrees of freedom 231
## P-value 0.000
##
## User Model versus Baseline Model:
##
## Comparative Fit Index (CFI) 0.867
## Tucker-Lewis Index (TLI) 0.847
##
## Loglikelihood and Information Criteria:
##
## Loglikelihood user model (H0) -10626.070
## Loglikelihood unrestricted model (H1) -10276.060
##
## Akaike (AIC) 21354.140
## Bayesian (BIC) 21556.805
## Sample-size adjusted Bayesian (SABIC) 21394.983
##
## Root Mean Square Error of Approximation:
##
## RMSEA 0.079
## 90 Percent confidence interval - lower 0.073
## 90 Percent confidence interval - upper 0.086
## P-value H_0: RMSEA <= 0.050 0.000
## P-value H_0: RMSEA >= 0.080 0.427
##
## Standardized Root Mean Square Residual:
##
## SRMR 0.107
##
## Parameter Estimates:
##
## Standard errors Standard
## Information Expected
## Information saturated (h1) model Structured
##
## Latent Variables:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## dogal =~
## item1 1.000 0.641 0.642
## item2 0.909 0.088 10.324 0.000 0.582 0.583
## item3 1.094 0.091 12.064 0.000 0.701 0.702
## item4 1.225 0.093 13.188 0.000 0.785 0.786
## item5 1.370 0.096 14.313 0.000 0.878 0.879
## item6 1.407 0.097 14.557 0.000 0.902 0.903
## bedensel =~
## item7 1.000 0.678 0.678
## item8 1.058 0.112 9.441 0.000 0.717 0.718
## item9 0.614 0.092 6.646 0.000 0.416 0.416
## item10 0.695 0.094 7.384 0.000 0.471 0.471
## uzamsal =~
## item11 1.000 0.872 0.873
## item12 0.980 0.048 20.530 0.000 0.854 0.855
## item13 0.942 0.048 19.562 0.000 0.821 0.822
## muziksel =~
## item14 1.000 0.608 0.609
## item15 1.497 0.212 7.053 0.000 0.911 0.912
## item16 0.565 0.094 6.017 0.000 0.344 0.344
## mantiksal =~
## item17 1.000 0.532 0.533
## item18 1.576 0.157 10.028 0.000 0.839 0.840
## item19 1.521 0.152 9.990 0.000 0.809 0.810
## dilsel =~
## item20 1.000 0.852 0.853
## item21 0.922 0.116 7.959 0.000 0.786 0.787
## item22 0.346 0.069 5.050 0.000 0.295 0.295
## bilissel =~
## dogal 1.000 0.473 0.473
## mantiksal 1.316 0.231 5.689 0.000 0.750 0.750
## dilsel 0.963 0.223 4.316 0.000 0.343 0.343
## uzamsal 2.240 0.347 6.464 0.000 0.779 0.779
## bilissel_olmayan =~
## muziksel 1.000 0.480 0.480
## bedensel 1.873 0.441 4.243 0.000 0.808 0.808
##
## Covariances:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## bilissel ~~
## bilissel_olmyn 0.064 0.017 3.777 0.000 0.724 0.724
##
## Variances:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## .item1 0.587 0.045 13.175 0.000 0.587 0.588
## .item2 0.658 0.049 13.399 0.000 0.658 0.660
## .item3 0.506 0.039 12.839 0.000 0.506 0.507
## .item4 0.382 0.032 12.033 0.000 0.382 0.383
## .item5 0.227 0.023 9.679 0.000 0.227 0.228
## .item6 0.184 0.022 8.473 0.000 0.184 0.185
## .item7 0.538 0.057 9.446 0.000 0.538 0.540
## .item8 0.483 0.058 8.401 0.000 0.483 0.484
## .item9 0.825 0.064 12.961 0.000 0.825 0.827
## .item10 0.776 0.062 12.568 0.000 0.776 0.778
## .item11 0.238 0.028 8.367 0.000 0.238 0.238
## .item12 0.268 0.029 9.167 0.000 0.268 0.269
## .item13 0.323 0.031 10.365 0.000 0.323 0.324
## .item14 0.627 0.065 9.637 0.000 0.627 0.629
## .item15 0.168 0.106 1.588 0.112 0.168 0.168
## .item16 0.879 0.065 13.490 0.000 0.879 0.881
## .item17 0.714 0.055 12.935 0.000 0.714 0.716
## .item18 0.294 0.044 6.752 0.000 0.294 0.295
## .item19 0.343 0.043 7.903 0.000 0.343 0.344
## .item20 0.271 0.088 3.096 0.002 0.271 0.272
## .item21 0.380 0.078 4.892 0.000 0.380 0.381
## .item22 0.910 0.066 13.708 0.000 0.910 0.913
## .dogal 0.319 0.048 6.638 0.000 0.776 0.776
## .bedensel 0.160 0.065 2.452 0.014 0.347 0.347
## .uzamsal 0.299 0.054 5.499 0.000 0.393 0.393
## .muziksel 0.285 0.055 5.214 0.000 0.769 0.769
## .mantiksal 0.124 0.029 4.247 0.000 0.438 0.438
## .dilsel 0.641 0.101 6.364 0.000 0.883 0.883
## bilissel 0.092 0.025 3.619 0.000 1.000 1.000
## bilissel_olmyn 0.085 0.032 2.634 0.008 1.000 1.000YORUM:
Ki-kare = 700.020, df = 202, p < .001
Model anlamlı, mükemmel uyum yok ama tek genel faktörlü modele göre
biraz daha iyi diyebiliriz.
CFI = 0.867, TLI = 0.847
Tek genel faktörlü modelde CFI = 0.864, TLI = 0.846 idi. Burada CFI
biraz yükselmiş, TLI aynı kalmıştır. Yani iki üst faktörlü yapının
veriyi daha iyi açıkladığı yorumunu yapabiliriz.
RMSEA = 0.079 (CI: 0.073–0.086)
0.079 sınırda kabul edilebilir uyum, ikinci modele göre daha iyi
diyebiliiriz.
SRMR = 0.107: Hala yüksek, modelin bazı hataları açıklamada sorun var yorumu yapılabilir.
İki Faktörlü model
model_4 <- '
genel =~ item1 + item2 + item3 + item4 + item5 + item6 +
item7 + item8 + item9 + item10 +
item11 + item12 + item13 +
item14 + item15 + item16 +
item17 + item18 + item19 +
item20 + item21 + item22
dogal =~ item1 + item2 + item3 + item4 + item5 + item6
bedensel =~ item7 + item8 + item9 + item10
uzamsal =~ item11 + item12 + item13
muziksel =~ item14 + item15 + item16
mantiksal =~ item17 + item18 + item19
dilsel =~ item20 + item21 + item22
'
fit_4 <- cfa(model_4, sample.cov = cast_cov_mat, sample.nobs = 393)
summary(fit_4, fit.measures=TRUE, standardized=TRUE)## lavaan 0.6-21 ended normally after 75 iterations
##
## Estimator ML
## Optimization method NLMINB
## Number of model parameters 87
##
## Number of observations 393
##
## Model Test User Model:
##
## Test statistic 299.543
## Degrees of freedom 166
## P-value (Chi-square) 0.000
##
## Model Test Baseline Model:
##
## Test statistic 3962.137
## Degrees of freedom 231
## P-value 0.000
##
## User Model versus Baseline Model:
##
## Comparative Fit Index (CFI) 0.964
## Tucker-Lewis Index (TLI) 0.950
##
## Loglikelihood and Information Criteria:
##
## Loglikelihood user model (H0) -10425.832
## Loglikelihood unrestricted model (H1) -10276.060
##
## Akaike (AIC) 21025.663
## Bayesian (BIC) 21371.385
## Sample-size adjusted Bayesian (SABIC) 21095.335
##
## Root Mean Square Error of Approximation:
##
## RMSEA 0.045
## 90 Percent confidence interval - lower 0.037
## 90 Percent confidence interval - upper 0.053
## P-value H_0: RMSEA <= 0.050 0.826
## P-value H_0: RMSEA >= 0.080 0.000
##
## Standardized Root Mean Square Residual:
##
## SRMR 0.050
##
## Parameter Estimates:
##
## Standard errors Standard
## Information Expected
## Information saturated (h1) model Structured
##
## Latent Variables:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## genel =~
## item1 1.000 0.757 0.758
## item2 0.668 NA 0.505 0.506
## item3 0.839 NA 0.635 0.636
## item4 1.114 NA 0.843 0.844
## item5 1.395 NA 1.056 1.057
## item6 1.280 NA 0.969 0.970
## item7 0.380 NA 0.287 0.288
## item8 0.303 NA 0.229 0.229
## item9 -0.062 NA -0.047 -0.047
## item10 0.473 NA 0.358 0.358
## item11 0.566 NA 0.428 0.429
## item12 0.528 NA 0.399 0.400
## item13 0.495 NA 0.375 0.375
## item14 0.015 NA 0.011 0.011
## item15 0.144 NA 0.109 0.109
## item16 1.115 NA 0.844 0.845
## item17 -0.079 NA -0.060 -0.060
## item18 0.488 NA 0.369 0.369
## item19 0.533 NA 0.403 0.404
## item20 0.406 NA 0.307 0.308
## item21 0.331 NA 0.250 0.250
## item22 0.448 NA 0.339 0.340
## dogal =~
## item1 1.000 0.697 0.698
## item2 1.149 NA 0.801 0.802
## item3 1.361 NA 0.949 0.950
## item4 1.334 NA 0.930 0.932
## item5 1.368 NA 0.954 0.955
## item6 1.510 NA 1.053 1.054
## bedensel =~
## item7 1.000 0.614 0.615
## item8 1.140 NA 0.700 0.701
## item9 0.831 NA 0.510 0.511
## item10 0.554 NA 0.340 0.341
## uzamsal =~
## item11 1.000 0.637 0.637
## item12 1.013 NA 0.645 0.646
## item13 0.985 NA 0.627 0.628
## muziksel =~
## item14 1.000 0.560 0.561
## item15 1.814 NA 1.015 1.017
## item16 0.402 NA 0.225 0.226
## mantiksal =~
## item17 1.000 0.664 0.665
## item18 1.020 NA 0.678 0.679
## item19 0.933 NA 0.620 0.620
## dilsel =~
## item20 1.000 0.745 0.746
## item21 1.210 NA 0.901 0.902
## item22 0.298 NA 0.222 0.222
##
## Covariances:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## genel ~~
## dogal -0.322 NA -0.610 -0.610
## bedensel -0.008 NA -0.017 -0.017
## uzamsal 0.147 NA 0.306 0.306
## muziksel -0.011 NA -0.027 -0.027
## mantiksal 0.100 NA 0.200 0.200
## dilsel -0.091 NA -0.162 -0.162
## dogal ~~
## bedensel 0.061 NA 0.143 0.143
## uzamsal -0.035 NA -0.079 -0.079
## muziksel 0.055 NA 0.142 0.142
## mantiksal -0.004 NA -0.009 -0.009
## dilsel 0.145 NA 0.280 0.280
## bedensel ~~
## uzamsal 0.105 NA 0.268 0.268
## muziksel 0.117 NA 0.340 0.340
## mantiksal 0.111 NA 0.273 0.273
## dilsel 0.103 NA 0.225 0.225
## uzamsal ~~
## muziksel 0.042 NA 0.117 0.117
## mantiksal 0.165 NA 0.390 0.390
## dilsel 0.036 NA 0.076 0.076
## muziksel ~~
## mantiksal 0.087 NA 0.235 0.235
## dilsel 0.069 NA 0.166 0.166
## mantiksal ~~
## dilsel 0.027 NA 0.054 0.054
##
## Variances:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## .item1 0.582 NA 0.582 0.583
## .item2 0.594 NA 0.594 0.595
## .item3 0.428 NA 0.428 0.430
## .item4 0.377 NA 0.377 0.378
## .item5 0.201 NA 0.201 0.202
## .item6 0.194 NA 0.194 0.195
## .item7 0.544 NA 0.544 0.545
## .item8 0.460 NA 0.460 0.461
## .item9 0.734 NA 0.734 0.736
## .item10 0.758 NA 0.758 0.760
## .item11 0.242 NA 0.242 0.242
## .item12 0.264 NA 0.264 0.265
## .item13 0.320 NA 0.320 0.321
## .item14 0.684 NA 0.684 0.686
## .item15 -0.040 NA -0.040 -0.040
## .item16 0.245 NA 0.245 0.245
## .item17 0.569 NA 0.569 0.570
## .item18 0.302 NA 0.302 0.303
## .item19 0.351 NA 0.351 0.352
## .item20 0.423 NA 0.423 0.424
## .item21 0.196 NA 0.196 0.196
## .item22 0.857 NA 0.857 0.860
## genel 0.573 NA 1.000 1.000
## dogal 0.486 NA 1.000 1.000
## bedensel 0.377 NA 1.000 1.000
## uzamsal 0.405 NA 1.000 1.000
## muziksel 0.313 NA 1.000 1.000
## mantiksal 0.441 NA 1.000 1.000
## dilsel 0.555 NA 1.000 1.000YORUM:
Ki‑kare= 299.543, df = 166, p < .001
Önceki modellerde ki-kare çok daha yüksekti (700 civarı). Burada ciddi
bir düşüş var, yani model veriye daha iyi uyum sağlıyor yorumu
yapılabilir.
CFI = 0.964, TLI = 0.950
Bu değerler artık 0.95’in üzerinde olduğundan mükemmele yakın uyum var
diyebiliriz.
RMSEA = 0.045 (CI: 0.037–0.053)
0.05’in altında, iyi uyum sağlıyor diyebiliriz.
SRMR = 0.050
0.08 sınırının altında, iyi uyum sağlıyor yorumu yapılabilir.
Veri Setinin YÜklenmesi
mot <- readRDS("mot.Rds")Model Kurulumu
tetra <- tetrachoric(mot)$rhomodel <- '
Disssal =~ ext1 + ext2 + ext3 + ext4 + ext5 + ext6 + ext7 + ext8 + ext9 + ext10 + ext11 + ext12
Icsel =~ int1 + int2 + int3 + int4 + int5
'
fit <- cfa(model, data=mot, estimator="WLSMV", ordered=colnames(mot))
summary(fit, fit.measures=TRUE, standardized=TRUE)## lavaan 0.6-21 ended normally after 42 iterations
##
## Estimator DWLS
## Optimization method NLMINB
## Number of model parameters 35
##
## Used Total
## Number of observations 794 852
##
## Model Test User Model:
## Standard Scaled
## Test Statistic 492.422 464.935
## Degrees of freedom 118 118
## P-value (Unknown) NA 0.000
## Scaling correction factor 1.136
## Shift parameter 31.654
## simple second-order correction
##
## Model Test Baseline Model:
##
## Test statistic 4421.778 3011.080
## Degrees of freedom 136 136
## P-value NA 0.000
## Scaling correction factor 1.491
##
## User Model versus Baseline Model:
##
## Comparative Fit Index (CFI) 0.913 0.879
## Tucker-Lewis Index (TLI) 0.899 0.861
##
## Robust Comparative Fit Index (CFI) NA
## Robust Tucker-Lewis Index (TLI) NA
##
## Root Mean Square Error of Approximation:
##
## RMSEA 0.063 0.061
## 90 Percent confidence interval - lower 0.058 0.055
## 90 Percent confidence interval - upper 0.069 0.067
## P-value H_0: RMSEA <= 0.050 0.000 0.001
## P-value H_0: RMSEA >= 0.080 0.000 0.000
##
## Robust RMSEA NA
## 90 Percent confidence interval - lower NA
## 90 Percent confidence interval - upper NA
## P-value H_0: Robust RMSEA <= 0.050 NA
## P-value H_0: Robust RMSEA >= 0.080 NA
##
## Standardized Root Mean Square Residual:
##
## SRMR 0.119 0.119
##
## Parameter Estimates:
##
## Parameterization Delta
## Standard errors Robust.sem
## Information Expected
## Information saturated (h1) model Unstructured
##
## Latent Variables:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## Disssal =~
## ext1 1.000 0.391 0.391
## ext2 1.018 0.173 5.881 0.000 0.398 0.398
## ext3 0.529 0.148 3.562 0.000 0.207 0.207
## ext4 1.494 0.263 5.684 0.000 0.585 0.585
## ext5 0.181 0.175 1.034 0.301 0.071 0.071
## ext6 0.419 0.173 2.425 0.015 0.164 0.164
## ext7 1.422 0.208 6.846 0.000 0.557 0.557
## ext8 2.045 0.274 7.470 0.000 0.801 0.801
## ext9 1.893 0.273 6.930 0.000 0.741 0.741
## ext10 1.577 0.229 6.899 0.000 0.617 0.617
## ext11 2.030 0.276 7.363 0.000 0.794 0.794
## ext12 1.835 0.250 7.330 0.000 0.718 0.718
## Icsel =~
## int1 1.000 0.821 0.821
## int2 0.924 0.066 14.092 0.000 0.758 0.758
## int3 0.997 0.050 19.793 0.000 0.818 0.818
## int4 0.906 0.057 15.980 0.000 0.744 0.744
## int5 1.100 0.050 21.863 0.000 0.903 0.903
##
## Covariances:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## Disssal ~~
## Icsel 0.047 0.017 2.744 0.006 0.148 0.148
##
## Thresholds:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## ext1|t1 0.041 0.045 0.922 0.356 0.041 0.041
## ext2|t1 0.584 0.047 12.327 0.000 0.584 0.584
## ext3|t1 0.493 0.047 10.591 0.000 0.493 0.493
## ext4|t1 1.541 0.070 21.954 0.000 1.541 1.541
## ext5|t1 -1.128 0.056 -19.964 0.000 -1.128 -1.128
## ext6|t1 -1.059 0.055 -19.296 0.000 -1.059 -1.059
## ext7|t1 0.242 0.045 5.386 0.000 0.242 0.242
## ext8|t1 1.042 0.055 19.122 0.000 1.042 1.042
## ext9|t1 1.481 0.068 21.880 0.000 1.481 1.481
## ext10|t1 0.622 0.048 13.016 0.000 0.622 0.622
## ext11|t1 1.048 0.055 19.180 0.000 1.048 1.048
## ext12|t1 0.013 0.045 0.284 0.777 0.013 0.013
## int1|t1 -0.657 0.048 -13.632 0.000 -0.657 -0.657
## int2|t1 -1.242 0.060 -20.865 0.000 -1.242 -1.242
## int3|t1 0.194 0.045 4.324 0.000 0.194 0.194
## int4|t1 -0.873 0.051 -17.028 0.000 -0.873 -0.873
## int5|t1 -0.272 0.045 -6.022 0.000 -0.272 -0.272
##
## Variances:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## .ext1 0.847 0.847 0.847
## .ext2 0.841 0.841 0.841
## .ext3 0.957 0.957 0.957
## .ext4 0.658 0.658 0.658
## .ext5 0.995 0.995 0.995
## .ext6 0.973 0.973 0.973
## .ext7 0.690 0.690 0.690
## .ext8 0.359 0.359 0.359
## .ext9 0.451 0.451 0.451
## .ext10 0.619 0.619 0.619
## .ext11 0.369 0.369 0.369
## .ext12 0.484 0.484 0.484
## .int1 0.327 0.327 0.327
## .int2 0.425 0.425 0.425
## .int3 0.331 0.331 0.331
## .int4 0.447 0.447 0.447
## .int5 0.185 0.185 0.185
## Disssal 0.153 0.038 3.984 0.000 1.000 1.000
## Icsel 0.673 0.051 13.305 0.000 1.000 1.000YORUM:
X2/df = 464.935 / 118 = 3.94 Kabul edilebilir sınırda diyebiliriz(≤5). CFI = 0.913 0.90 üzeri kabul edilebilir aralıktadır. TLI = 0.899 sınırda bir değerdir. RMSEA = 0.061 (CI: 0.055–0.067) Kabul edilebilir aralıkta olduğu yorumunu yapaniliriz. SRMR = 0.119 bu değer sınır değerinden yüksek olduğundan kötü uyum yorumu yapılabilir.
Modifikasyon Önerileri Düşük yükleri olan maddeler (ext3, ext5, ext6) modelden çıkarılabilir. Modification Indices (MI) çıktısına bakarak aynı faktördeki maddeler arasında hata kovaryansı eklenebilir. SRMR değerini düşürmek için özellikle dışsal faktörde madde ilişkilerini gözden geçirmek iyi olabilir.
Yeni model
model_yeni <- '
DisMot =~ ext1 + ext2 + ext4 + ext7 + ext8 + ext9 + ext10 + ext11 + ext12
IcMot =~ int1 + int2 + int3 + int4 + int5
'
fit_yeni <- cfa(model_yeni, data=mot, estimator="WLSMV", ordered=colnames(mot))
summary(fit_yeni, fit.measures=TRUE, standardized=TRUE)## lavaan 0.6-21 ended normally after 39 iterations
##
## Estimator DWLS
## Optimization method NLMINB
## Number of model parameters 29
##
## Used Total
## Number of observations 796 852
##
## Model Test User Model:
## Standard Scaled
## Test Statistic 246.192 251.341
## Degrees of freedom 76 76
## P-value (Unknown) NA 0.000
## Scaling correction factor 1.055
## Shift parameter 17.925
## simple second-order correction
##
## Model Test Baseline Model:
##
## Test statistic 4117.723 2850.807
## Degrees of freedom 91 91
## P-value NA 0.000
## Scaling correction factor 1.459
##
## User Model versus Baseline Model:
##
## Comparative Fit Index (CFI) 0.958 0.936
## Tucker-Lewis Index (TLI) 0.949 0.924
##
## Robust Comparative Fit Index (CFI) 0.726
## Robust Tucker-Lewis Index (TLI) 0.672
##
## Root Mean Square Error of Approximation:
##
## RMSEA 0.053 0.054
## 90 Percent confidence interval - lower 0.046 0.047
## 90 Percent confidence interval - upper 0.061 0.061
## P-value H_0: RMSEA <= 0.050 0.238 0.186
## P-value H_0: RMSEA >= 0.080 0.000 0.000
##
## Robust RMSEA 0.172
## 90 Percent confidence interval - lower 0.149
## 90 Percent confidence interval - upper 0.196
## P-value H_0: Robust RMSEA <= 0.050 0.000
## P-value H_0: Robust RMSEA >= 0.080 1.000
##
## Standardized Root Mean Square Residual:
##
## SRMR 0.105 0.105
##
## Parameter Estimates:
##
## Parameterization Delta
## Standard errors Robust.sem
## Information Expected
## Information saturated (h1) model Unstructured
##
## Latent Variables:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## DisMot =~
## ext1 1.000 0.377 0.377
## ext2 1.052 0.182 5.772 0.000 0.397 0.397
## ext4 1.591 0.282 5.644 0.000 0.600 0.600
## ext7 1.474 0.225 6.563 0.000 0.556 0.556
## ext8 2.161 0.302 7.149 0.000 0.814 0.814
## ext9 1.946 0.291 6.684 0.000 0.734 0.734
## ext10 1.669 0.251 6.651 0.000 0.629 0.629
## ext11 2.101 0.300 7.011 0.000 0.792 0.792
## ext12 1.898 0.270 7.029 0.000 0.715 0.715
## IcMot =~
## int1 1.000 0.820 0.820
## int2 0.926 0.066 14.099 0.000 0.758 0.758
## int3 1.000 0.051 19.772 0.000 0.820 0.820
## int4 0.906 0.057 15.938 0.000 0.743 0.743
## int5 1.100 0.050 21.880 0.000 0.901 0.901
##
## Covariances:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## DisMot ~~
## IcMot 0.040 0.017 2.410 0.016 0.130 0.130
##
## Thresholds:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## ext1|t1 0.041 0.044 0.921 0.357 0.041 0.041
## ext2|t1 0.582 0.047 12.312 0.000 0.582 0.582
## ext4|t1 1.542 0.070 21.982 0.000 1.542 1.542
## ext7|t1 0.242 0.045 5.379 0.000 0.242 0.242
## ext8|t1 1.044 0.054 19.163 0.000 1.044 1.044
## ext9|t1 1.483 0.068 21.910 0.000 1.483 1.483
## ext10|t1 0.624 0.048 13.069 0.000 0.624 0.624
## ext11|t1 1.049 0.055 19.221 0.000 1.049 1.049
## ext12|t1 0.009 0.044 0.213 0.832 0.009 0.009
## int1|t1 -0.655 0.048 -13.616 0.000 -0.655 -0.655
## int2|t1 -1.244 0.060 -20.901 0.000 -1.244 -1.244
## int3|t1 0.193 0.045 4.319 0.000 0.193 0.193
## int4|t1 -0.874 0.051 -17.074 0.000 -0.874 -0.874
## int5|t1 -0.274 0.045 -6.085 0.000 -0.274 -0.274
##
## Variances:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## .ext1 0.858 0.858 0.858
## .ext2 0.843 0.843 0.843
## .ext4 0.640 0.640 0.640
## .ext7 0.691 0.691 0.691
## .ext8 0.337 0.337 0.337
## .ext9 0.462 0.462 0.462
## .ext10 0.604 0.604 0.604
## .ext11 0.373 0.373 0.373
## .ext12 0.488 0.488 0.488
## .int1 0.328 0.328 0.328
## .int2 0.425 0.425 0.425
## .int3 0.328 0.328 0.328
## .int4 0.448 0.448 0.448
## .int5 0.188 0.188 0.188
## DisMot 0.142 0.038 3.785 0.000 1.000 1.000
## IcMot 0.672 0.051 13.286 0.000 1.000 1.000YORUM:
x2/df = 246.192 / 76 ≈ 3.31 Kabul edilebilir uyum (≤5). İlk modelde 3.94. yani iyileşme var diyebiliriz.
CFI = 0.958, TLI = 0.949 iyi uyum (≥0.95 mükemmel, ≥0.90 kabul edilebilir). İlk modelde CFI 0.913, TLI 0.899. iyileşme olduğu görülmektedir.
RMSEA = 0.053 (CI: 0.047–0.061) İyi uyum (≤0.08 kabul edilebilir). İlk modelde 0.061. Daha iyi uyum olduğunu söyleyebiliriz.
SRMR = 0.105 ilk modelde 0.119. Düşmüş ama 0.08’in üzerinde, hah kötü uyum gösteriyor diyebiliriz.
Veri Setinin Yüklenmesi
aidiyet <- readRDS("aidiyet.Rds")
str(aidiyet)## 'data.frame': 794 obs. of 12 variables:
## $ kurumsal1 : num 1 0 0 1 1 1 1 1 0 0 ...
## $ kurumsal2 : num 0 0 1 0 0 1 0 1 0 1 ...
## $ kurumsal3 : num 0 0 1 1 1 0 0 0 0 0 ...
## $ kurumsal4 : num 0 0 0 0 0 0 0 1 0 0 ...
## $ bireysel1 : num 1 0 1 1 1 1 1 1 1 1 ...
## $ bireysel2 : num 0 0 1 0 1 0 0 1 0 0 ...
## $ bireysel3 : num 1 1 1 0 1 0 0 1 0 1 ...
## $ bireysel4 : num 0 0 1 1 1 1 0 1 1 1 ...
## $ katilimsal1: num 1 0 1 1 1 1 1 1 0 1 ...
## $ katilimsal2: num 1 0 1 1 1 1 1 1 1 1 ...
## $ katilimsal3: num 0 0 1 0 0 1 0 0 1 1 ...
## $ katilimsal4: num 0 1 1 1 1 1 1 1 0 1 ...
## - attr(*, "na.action")= 'omit' Named int [1:58] 5 14 28 54 59 77 81 95 99 107 ...
## ..- attr(*, "names")= chr [1:58] "6" "22" "44" "101" ...colnames(aidiyet)## [1] "kurumsal1" "kurumsal2" "kurumsal3" "kurumsal4" "bireysel1"
## [6] "bireysel2" "bireysel3" "bireysel4" "katilimsal1" "katilimsal2"
## [11] "katilimsal3" "katilimsal4"Tetrakorik Korelasyon Matrisi
tetra <- tetrachoric(aidiyet)$rhomodel_3_f <- '
Kurumsal =~ kurumsal1 + kurumsal2 + kurumsal3 + kurumsal4
Katilimsal =~ katilimsal1 + katilimsal2 + katilimsal3 + katilimsal4
Bireysel =~ bireysel1 + bireysel2 + bireysel3 + bireysel4
'
fit_3_f <- cfa(model_3_f, data=aidiyet, estimator="WLSMV", ordered=colnames(aidiyet))
summary(fit_3_f, fit.measures=TRUE, standardized=TRUE)## lavaan 0.6-21 ended normally after 43 iterations
##
## Estimator DWLS
## Optimization method NLMINB
## Number of model parameters 27
##
## Number of observations 794
##
## Model Test User Model:
## Standard Scaled
## Test Statistic 161.148 189.809
## Degrees of freedom 51 51
## P-value (Unknown) NA 0.000
## Scaling correction factor 0.882
## Shift parameter 7.113
## simple second-order correction
##
## Model Test Baseline Model:
##
## Test statistic 2506.101 1939.031
## Degrees of freedom 66 66
## P-value NA 0.000
## Scaling correction factor 1.303
##
## User Model versus Baseline Model:
##
## Comparative Fit Index (CFI) 0.955 0.926
## Tucker-Lewis Index (TLI) 0.942 0.904
##
## Robust Comparative Fit Index (CFI) 0.816
## Robust Tucker-Lewis Index (TLI) 0.762
##
## Root Mean Square Error of Approximation:
##
## RMSEA 0.052 0.059
## 90 Percent confidence interval - lower 0.043 0.050
## 90 Percent confidence interval - upper 0.061 0.068
## P-value H_0: RMSEA <= 0.050 0.331 0.053
## P-value H_0: RMSEA >= 0.080 0.000 0.000
##
## Robust RMSEA 0.126
## 90 Percent confidence interval - lower 0.101
## 90 Percent confidence interval - upper 0.151
## P-value H_0: Robust RMSEA <= 0.050 0.000
## P-value H_0: Robust RMSEA >= 0.080 0.999
##
## Standardized Root Mean Square Residual:
##
## SRMR 0.088 0.088
##
## Parameter Estimates:
##
## Parameterization Delta
## Standard errors Robust.sem
## Information Expected
## Information saturated (h1) model Unstructured
##
## Latent Variables:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## Kurumsal =~
## kurumsal1 1.000 0.361 0.361
## kurumsal2 0.655 0.248 2.636 0.008 0.237 0.237
## kurumsal3 1.424 0.374 3.807 0.000 0.515 0.515
## kurumsal4 0.627 0.381 1.647 0.100 0.227 0.227
## Katilimsal =~
## katilimsal1 1.000 0.833 0.833
## katilimsal2 0.988 0.067 14.710 0.000 0.823 0.823
## katilimsal3 0.870 0.068 12.758 0.000 0.725 0.725
## katilimsal4 0.947 0.061 15.551 0.000 0.789 0.789
## Bireysel =~
## bireysel1 1.000 0.602 0.602
## bireysel2 0.834 0.100 8.311 0.000 0.502 0.502
## bireysel3 1.412 0.132 10.717 0.000 0.850 0.850
## bireysel4 1.421 0.132 10.730 0.000 0.855 0.855
##
## Covariances:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## Kurumsal ~~
## Katilimsal 0.145 0.038 3.780 0.000 0.480 0.480
## Bireysel 0.121 0.030 4.011 0.000 0.557 0.557
## Katilimsal ~~
## Bireysel 0.232 0.033 6.932 0.000 0.462 0.462
##
## Thresholds:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## kurumsal1|t1 0.041 0.045 0.922 0.356 0.041 0.041
## kurumsal2|t1 0.592 0.047 12.465 0.000 0.592 0.592
## kurumsal3|t1 0.489 0.046 10.522 0.000 0.489 0.489
## kurumsal4|t1 1.541 0.070 21.954 0.000 1.541 1.541
## katilimsal1|t1 -0.665 0.048 -13.769 0.000 -0.665 -0.665
## katilimsal2|t1 -1.256 0.060 -20.956 0.000 -1.256 -1.256
## katilimsal3|t1 0.184 0.045 4.112 0.000 0.184 0.184
## katilimsal4|t1 -0.868 0.051 -16.963 0.000 -0.868 -0.868
## bireysel1|t1 -0.725 0.049 -14.785 0.000 -0.725 -0.725
## bireysel2|t1 0.025 0.045 0.567 0.570 0.025 0.025
## bireysel3|t1 -0.385 0.046 -8.419 0.000 -0.385 -0.385
## bireysel4|t1 -0.493 0.047 -10.591 0.000 -0.493 -0.493
##
## Variances:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## .kurumsal1 0.869 0.869 0.869
## .kurumsal2 0.944 0.944 0.944
## .kurumsal3 0.735 0.735 0.735
## .kurumsal4 0.949 0.949 0.949
## .katilimsal1 0.306 0.306 0.306
## .katilimsal2 0.322 0.322 0.322
## .katilimsal3 0.475 0.475 0.475
## .katilimsal4 0.378 0.378 0.378
## .bireysel1 0.638 0.638 0.638
## .bireysel2 0.748 0.748 0.748
## .bireysel3 0.278 0.278 0.278
## .bireysel4 0.268 0.268 0.268
## Kurumsal 0.131 0.052 2.519 0.012 1.000 1.000
## Katilimsal 0.694 0.058 11.960 0.000 1.000 1.000
## Bireysel 0.362 0.059 6.123 0.000 1.000 1.000YORUM: x2 = 189.809, df = 51 x2/df ≈ 3.72 kabul edilebilir uyum göstermektedir (≤5). CFI = 0.955 iyi uyum gösterdiği görülmektedir (≥0.95 çok iyi, ≥0.90 kabul edilebilir). TLI = 0.942 kabul edilebilir uyum aralığında değer almıştır. RMSEA = 0.052 (CI: 0.043–0.061) ≤0.08 olduğu için iyi uyum gösterdiği söylenebilir. SRMR = 0.088 Kabul edilebilir aralıkta olduğu yorumu yapılabilir ( ≤0.10 kabul edilebilir).
İkinci Dereceli Model
model_4_f <- '
Kurumsal =~ kurumsal1 + kurumsal2 + kurumsal3 + kurumsal4
Katilimsal =~ katilimsal1 + katilimsal2 + katilimsal3 + katilimsal4
Bireysel =~ bireysel1 + bireysel2 + bireysel3 + bireysel4
Aidiyet =~ Kurumsal + Katilimsal + Bireysel
'
fit_4_f <- cfa(model_4_f, data=aidiyet, estimator="WLSMV", ordered=colnames(aidiyet))
summary(fit_4_f, fit.measures=TRUE, standardized=TRUE)## lavaan 0.6-21 ended normally after 59 iterations
##
## Estimator DWLS
## Optimization method NLMINB
## Number of model parameters 27
##
## Number of observations 794
##
## Model Test User Model:
## Standard Scaled
## Test Statistic 161.148 189.809
## Degrees of freedom 51 51
## P-value (Unknown) NA 0.000
## Scaling correction factor 0.882
## Shift parameter 7.113
## simple second-order correction
##
## Model Test Baseline Model:
##
## Test statistic 2506.101 1939.031
## Degrees of freedom 66 66
## P-value NA 0.000
## Scaling correction factor 1.303
##
## User Model versus Baseline Model:
##
## Comparative Fit Index (CFI) 0.955 0.926
## Tucker-Lewis Index (TLI) 0.942 0.904
##
## Robust Comparative Fit Index (CFI) 0.816
## Robust Tucker-Lewis Index (TLI) 0.762
##
## Root Mean Square Error of Approximation:
##
## RMSEA 0.052 0.059
## 90 Percent confidence interval - lower 0.043 0.050
## 90 Percent confidence interval - upper 0.061 0.068
## P-value H_0: RMSEA <= 0.050 0.331 0.053
## P-value H_0: RMSEA >= 0.080 0.000 0.000
##
## Robust RMSEA 0.126
## 90 Percent confidence interval - lower 0.101
## 90 Percent confidence interval - upper 0.151
## P-value H_0: Robust RMSEA <= 0.050 0.000
## P-value H_0: Robust RMSEA >= 0.080 0.999
##
## Standardized Root Mean Square Residual:
##
## SRMR 0.088 0.088
##
## Parameter Estimates:
##
## Parameterization Delta
## Standard errors Robust.sem
## Information Expected
## Information saturated (h1) model Unstructured
##
## Latent Variables:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## Kurumsal =~
## kurumsal1 1.000 0.361 0.361
## kurumsal2 0.655 0.248 2.636 0.008 0.237 0.237
## kurumsal3 1.424 0.374 3.807 0.000 0.515 0.515
## kurumsal4 0.627 0.381 1.647 0.100 0.227 0.227
## Katilimsal =~
## katilimsal1 1.000 0.833 0.833
## katilimsal2 0.988 0.067 14.710 0.000 0.823 0.823
## katilimsal3 0.870 0.068 12.758 0.000 0.725 0.725
## katilimsal4 0.947 0.061 15.551 0.000 0.789 0.789
## Bireysel =~
## bireysel1 1.000 0.602 0.602
## bireysel2 0.834 0.100 8.311 0.000 0.502 0.502
## bireysel3 1.412 0.132 10.717 0.000 0.850 0.850
## bireysel4 1.421 0.132 10.730 0.000 0.855 0.855
## Aidiyet =~
## Kurumsal 1.000 0.761 0.761
## Katilimsal 1.910 0.493 3.872 0.000 0.631 0.631
## Bireysel 1.602 0.450 3.561 0.000 0.732 0.732
##
## Thresholds:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## kurumsal1|t1 0.041 0.045 0.922 0.356 0.041 0.041
## kurumsal2|t1 0.592 0.047 12.465 0.000 0.592 0.592
## kurumsal3|t1 0.489 0.046 10.522 0.000 0.489 0.489
## kurumsal4|t1 1.541 0.070 21.954 0.000 1.541 1.541
## katilimsal1|t1 -0.665 0.048 -13.769 0.000 -0.665 -0.665
## katilimsal2|t1 -1.256 0.060 -20.956 0.000 -1.256 -1.256
## katilimsal3|t1 0.184 0.045 4.112 0.000 0.184 0.184
## katilimsal4|t1 -0.868 0.051 -16.963 0.000 -0.868 -0.868
## bireysel1|t1 -0.725 0.049 -14.785 0.000 -0.725 -0.725
## bireysel2|t1 0.025 0.045 0.567 0.570 0.025 0.025
## bireysel3|t1 -0.385 0.046 -8.419 0.000 -0.385 -0.385
## bireysel4|t1 -0.493 0.047 -10.591 0.000 -0.493 -0.493
##
## Variances:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## .kurumsal1 0.869 0.869 0.869
## .kurumsal2 0.944 0.944 0.944
## .kurumsal3 0.735 0.735 0.735
## .kurumsal4 0.949 0.949 0.949
## .katilimsal1 0.306 0.306 0.306
## .katilimsal2 0.322 0.322 0.322
## .katilimsal3 0.475 0.475 0.475
## .katilimsal4 0.378 0.378 0.378
## .bireysel1 0.638 0.638 0.638
## .bireysel2 0.748 0.748 0.748
## .bireysel3 0.278 0.278 0.278
## .bireysel4 0.268 0.268 0.268
## .Kurumsal 0.055 0.036 1.535 0.125 0.420 0.420
## .Katilimsal 0.418 0.072 5.797 0.000 0.602 0.602
## .Bireysel 0.168 0.051 3.312 0.001 0.464 0.464
## Aidiyet 0.076 0.035 2.180 0.029 1.000 1.000YORUM:
x2= 251.341, df = 76 → x2/df= 3.31
İlk modelde bu oran ≈ 3.72 idi. iyileşme olduğu görülmektedir. CFI =
0.955 TLI = 0.942 RMSEA = 0.052 (CI: 0.047–0.061) İlk model ile yaklaşık
sonuçlar aldığını görürüz. SRMR = 0.105 İlk modelde 0.088. burada biraz
kötüleştiğini görüyoruz. SRMR, gözlenen korelasyonlarla modelin tahmin
ettiği korelasyonlar arasındaki farkı ölçer. Karmaşık modellerde bu fark
büyüyebilir. Üst faktör eklenince parametre sayısı arttığından bu değer
düşmüş olabilir.
GENEL YORUM: İkinci düzey DFA sonuçları, Aidiyet yapısının Kurumsal, Katılımsal ve Bireysel boyutlardan oluşan hiyerarşik bir yapı olarak doğrulandığını göstermektedir. Uyum indeksleri modelin kabul edilebilir ve iyi düzeyde uyuma sahip olduğunu göstermektedir (CFI=.955, TLI=.942, RMSEA=.052). Bununla birlikte robust uyum indekslerinin düşük olması modelde bazı uyumsuzlukların bulunduğuna işaret etmektedir yorumu yapılabilir.