CFA
# Modelo
senna1o<-"
C =~ C_0_c + C_1_c
A =~ A_0_c + A_1_c
O =~ O_0_c + O_1_c
N =~ N_0_c + N_1_c
E =~ E_0_c + E_1_c
"
senna1o.fit <- cfa(senna1o, data = scores, std.lv = TRUE)
summary(senna1o.fit , fit.measure=TRUE, standardized=TRUE, rsquare=TRUE)
## lavaan (0.5-23.1097) converged normally after 40 iterations
##
## Number of observations 168
##
## Estimator ML
## Minimum Function Test Statistic 48.790
## Degrees of freedom 25
## P-value (Chi-square) 0.003
##
## Model test baseline model:
##
## Minimum Function Test Statistic 618.233
## Degrees of freedom 45
## P-value 0.000
##
## User model versus baseline model:
##
## Comparative Fit Index (CFI) 0.958
## Tucker-Lewis Index (TLI) 0.925
##
## Loglikelihood and Information Criteria:
##
## Loglikelihood user model (H0) -1540.325
## Loglikelihood unrestricted model (H1) -1515.930
##
## Number of free parameters 30
## Akaike (AIC) 3140.651
## Bayesian (BIC) 3234.370
## Sample-size adjusted Bayesian (BIC) 3139.383
##
## Root Mean Square Error of Approximation:
##
## RMSEA 0.075
## 90 Percent Confidence Interval 0.043 0.106
## P-value RMSEA <= 0.05 0.092
##
## Standardized Root Mean Square Residual:
##
## SRMR 0.044
##
## Parameter Estimates:
##
## Information Expected
## Standard Errors Standard
##
## Latent Variables:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## C =~
## C_0_c 0.549 0.059 9.298 0.000 0.549 0.685
## C_1_c 0.734 0.054 13.668 0.000 0.734 0.959
## A =~
## A_0_c 0.438 0.045 9.634 0.000 0.438 0.739
## A_1_c 0.478 0.049 9.773 0.000 0.478 0.749
## O =~
## O_0_c 0.452 0.052 8.721 0.000 0.452 0.675
## O_1_c 0.586 0.061 9.608 0.000 0.586 0.744
## N =~
## N_0_c 0.630 0.064 9.848 0.000 0.630 0.770
## N_1_c 0.688 0.065 10.577 0.000 0.688 0.828
## E =~
## E_0_c 0.339 0.064 5.267 0.000 0.339 0.538
## E_1_c 0.286 0.064 4.488 0.000 0.286 0.417
##
## Covariances:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## C ~~
## A 0.714 0.065 10.942 0.000 0.714 0.714
## O 0.704 0.070 10.088 0.000 0.704 0.704
## N 0.505 0.074 6.793 0.000 0.505 0.505
## E 0.844 0.133 6.354 0.000 0.844 0.844
## A ~~
## O 0.646 0.085 7.572 0.000 0.646 0.646
## N 0.551 0.083 6.660 0.000 0.551 0.551
## E 0.884 0.146 6.068 0.000 0.884 0.884
## O ~~
## N 0.677 0.077 8.799 0.000 0.677 0.677
## E 0.980 0.153 6.412 0.000 0.980 0.980
## N ~~
## E 0.600 0.137 4.378 0.000 0.600 0.600
##
## Variances:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## .C_0_c 0.341 0.045 7.516 0.000 0.341 0.531
## .C_1_c 0.047 0.047 1.011 0.312 0.047 0.081
## .A_0_c 0.160 0.027 5.944 0.000 0.160 0.454
## .A_1_c 0.179 0.031 5.725 0.000 0.179 0.439
## .O_0_c 0.244 0.035 6.990 0.000 0.244 0.544
## .O_1_c 0.276 0.048 5.704 0.000 0.276 0.446
## .N_0_c 0.273 0.054 5.051 0.000 0.273 0.408
## .N_1_c 0.217 0.059 3.687 0.000 0.217 0.314
## .E_0_c 0.282 0.044 6.442 0.000 0.282 0.711
## .E_1_c 0.388 0.048 8.117 0.000 0.388 0.826
## C 1.000 1.000 1.000
## A 1.000 1.000 1.000
## O 1.000 1.000 1.000
## N 1.000 1.000 1.000
## E 1.000 1.000 1.000
##
## R-Square:
## Estimate
## C_0_c 0.469
## C_1_c 0.919
## A_0_c 0.546
## A_1_c 0.561
## O_0_c 0.456
## O_1_c 0.554
## N_0_c 0.592
## N_1_c 0.686
## E_0_c 0.289
## E_1_c 0.174
fitMeasures( senna1o.fit,
fit.measures = c("chisq", "df", "cfi", "tli", "rmsea","srmr")
)
## chisq df cfi tli rmsea srmr
## 48.790 25.000 0.958 0.925 0.075 0.044
CFA
names(scores)
## [1] "banco" "cod_suj" "Data de nasc."
## [4] "data.aplic" "Idade0" "idade1"
## [7] "Termo" "sujeito" "serie"
## [10] "escola" "turma" "Sexo"
## [13] "Esc. Mãe" "port" "mat"
## [16] "cloze" "A_o" "C_o"
## [19] "E_o" "N_o" "O_o"
## [22] "A_c" "C_c" "E_c"
## [25] "N_c" "O_c" "A_z"
## [28] "C_z" "E_z" "N_z"
## [31] "O_z" "antonym.rc" "antonym.cntrst_A"
## [34] "antonym.cntrst_C" "antonym.cntrst_E" "antonym.cntrst_N"
## [37] "antonym.cntrst_O" "mean_A" "mean_C"
## [40] "mean_E" "mean_N" "mean_O"
## [43] "sd_A" "sd_C" "sd_E"
## [46] "sd_N" "sd_O" "antonym.rc_A"
## [49] "antonym.rc_C" "antonym.rc_E" "antonym.rc_N"
## [52] "antonym.rc_O" "nse" "A_1_o"
## [55] "C_1_o" "E_1_o" "N_1_o"
## [58] "O_1_o" "A_1_c" "C_1_c"
## [61] "E_1_c" "N_1_c" "O_1_c"
## [64] "A_1_z" "C_1_z" "E_1_z"
## [67] "N_1_z" "O_1_z" "A_0_o"
## [70] "C_0_o" "E_0_o" "N_0_o"
## [73] "O_0_o" "A_0_c" "C_0_c"
## [76] "E_0_c" "N_0_c" "O_0_c"
## [79] "A_0_z" "C_0_z" "E_0_z"
## [82] "N_0_z" "O_0_z" "cexp_o"
## [85] "comlrn_o" "coplrn_o" "cstrat_o"
## [88] "effper_o" "elab_o" "insmot_o"
## [91] "intmat_o" "intrea_o" "memor_o"
## [94] "scacad_o" "scmath_o" "scverb_o"
## [97] "selfef_o" "cexp_c" "comlrn_c"
## [100] "coplrn_c" "cstrat_c" "effper_c"
## [103] "elab_c" "insmot_c" "intmat_c"
## [106] "intrea_c" "memor_c" "scacad_c"
## [109] "scmath_c" "scverb_c" "selfef_c"
## [112] "cexp_z" "comlrn_z" "coplrn_z"
## [115] "cstrat_z" "effper_z" "elab_z"
## [118] "insmot_z" "intmat_z" "intrea_z"
## [121] "memor_z" "scacad_z" "scmath_z"
## [124] "scverb_z" "selfef_z" "means"
## [127] "sd"
v <- names(scores)[c(14:15)]
corr.test(scores[ , v])
## Call:corr.test(x = scores[, v])
## Correlation matrix
## port mat
## port 1.00 0.51
## mat 0.51 1.00
## Sample Size
## [1] 76
## Probability values (Entries above the diagonal are adjusted for multiple tests.)
## port mat
## port 0 0
## mat 0 0
##
## To see confidence intervals of the correlations, print with the short=FALSE option
describe(scores[ , v])
## vars n mean sd median trimmed mad min max range skew kurtosis
## port 1 76 69.74 18.55 75 70.81 22.24 25 100 75 -0.46 -0.64
## mat 2 76 61.05 23.31 65 62.02 22.24 0 100 100 -0.48 -0.10
## se
## port 2.13
## mat 2.67
source("http://www.labape.com.br/rprimi/R/cria_quartis.R")
ggplot(data = scores, aes(y = port, x = mat)) + geom_point(alpha=.5) +
geom_smooth(method = "lm")
senna1o<-"
C =~ C_0_c + C_1_c
A =~ A_0_c + A_1_c
O =~ O_0_c + O_1_c
N =~ N_0_c + N_1_c
E =~ E_0_c + E_1_c
gc =~ port + mat
gc ~ C + A + O + N + O
"
senna1o.fit <- sem(senna1o, data = scores, std.lv = TRUE)
summary(senna1o.fit , fit.measure=TRUE, standardized=TRUE, rsquare=TRUE)
## lavaan (0.5-23.1097) converged normally after 137 iterations
##
## Used Total
## Number of observations 76 168
##
## Estimator ML
## Minimum Function Test Statistic 54.741
## Degrees of freedom 40
## P-value (Chi-square) 0.060
##
## Model test baseline model:
##
## Minimum Function Test Statistic 384.450
## Degrees of freedom 66
## P-value 0.000
##
## User model versus baseline model:
##
## Comparative Fit Index (CFI) 0.954
## Tucker-Lewis Index (TLI) 0.924
##
## Loglikelihood and Information Criteria:
##
## Loglikelihood user model (H0) -1300.792
## Loglikelihood unrestricted model (H1) -1273.421
##
## Number of free parameters 38
## Akaike (AIC) 2677.584
## Bayesian (BIC) 2766.152
## Sample-size adjusted Bayesian (BIC) 2646.373
##
## Root Mean Square Error of Approximation:
##
## RMSEA 0.070
## 90 Percent Confidence Interval 0.000 0.112
## P-value RMSEA <= 0.05 0.236
##
## Standardized Root Mean Square Residual:
##
## SRMR 0.062
##
## Parameter Estimates:
##
## Information Expected
## Standard Errors Standard
##
## Latent Variables:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## C =~
## C_0_c 0.582 0.091 6.431 0.000 0.582 0.700
## C_1_c 0.762 0.071 10.789 0.000 0.762 1.081
## A =~
## A_0_c 0.391 0.059 6.625 0.000 0.391 0.738
## A_1_c 0.458 0.068 6.734 0.000 0.458 0.749
## O =~
## O_0_c 0.395 0.073 5.381 0.000 0.395 0.605
## O_1_c 0.477 0.078 6.125 0.000 0.477 0.674
## N =~
## N_0_c 0.700 0.106 6.602 0.000 0.700 0.833
## N_1_c 0.675 0.099 6.804 0.000 0.675 0.864
## E =~
## E_0_c 0.427 0.079 5.411 0.000 0.427 0.727
## E_1_c 0.359 0.083 4.310 0.000 0.359 0.543
## gc =~
## port 8.352 2.506 3.333 0.001 12.120 0.658
## mat 12.491 3.916 3.189 0.001 18.126 0.783
##
## Regressions:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## gc ~
## C 0.307 0.324 0.947 0.344 0.212 0.212
## A -1.109 1.006 -1.102 0.271 -0.764 -0.764
## O 1.653 1.220 1.355 0.176 1.139 1.139
## N -0.139 0.364 -0.382 0.702 -0.096 -0.096
##
## Covariances:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## C ~~
## A 0.632 0.097 6.498 0.000 0.632 0.632
## O 0.666 0.105 6.331 0.000 0.666 0.666
## N 0.318 0.106 2.995 0.003 0.318 0.318
## E 0.426 0.124 3.428 0.001 0.426 0.426
## A ~~
## O 0.834 0.105 7.924 0.000 0.834 0.834
## N 0.386 0.133 2.906 0.004 0.386 0.386
## E 0.677 0.141 4.809 0.000 0.677 0.677
## O ~~
## N 0.541 0.133 4.080 0.000 0.541 0.541
## E 0.861 0.129 6.684 0.000 0.861 0.861
## N ~~
## E 0.261 0.155 1.686 0.092 0.261 0.261
##
## Variances:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## .C_0_c 0.352 0.071 4.950 0.000 0.352 0.510
## .C_1_c -0.084 0.074 -1.137 0.255 -0.084 -0.169
## .A_0_c 0.128 0.030 4.232 0.000 0.128 0.456
## .A_1_c 0.164 0.040 4.077 0.000 0.164 0.439
## .O_0_c 0.271 0.049 5.489 0.000 0.271 0.634
## .O_1_c 0.272 0.053 5.108 0.000 0.272 0.545
## .N_0_c 0.217 0.107 2.036 0.042 0.217 0.307
## .N_1_c 0.155 0.097 1.604 0.109 0.155 0.254
## .E_0_c 0.163 0.053 3.071 0.002 0.163 0.471
## .E_1_c 0.308 0.060 5.169 0.000 0.308 0.705
## .port 192.513 46.619 4.130 0.000 192.513 0.567
## .mat 207.839 84.441 2.461 0.014 207.839 0.387
## C 1.000 1.000 1.000
## A 1.000 1.000 1.000
## O 1.000 1.000 1.000
## N 1.000 1.000 1.000
## E 1.000 1.000 1.000
## .gc 1.000 0.475 0.475
##
## R-Square:
## Estimate
## C_0_c 0.490
## C_1_c NA
## A_0_c 0.544
## A_1_c 0.561
## O_0_c 0.366
## O_1_c 0.455
## N_0_c 0.693
## N_1_c 0.746
## E_0_c 0.529
## E_1_c 0.295
## port 0.433
## mat 0.613
## gc 0.525
fitMeasures( senna1o.fit,
fit.measures = c("chisq", "df", "cfi", "tli", "rmsea","srmr")
)
## chisq df cfi tli rmsea srmr
## 54.741 40.000 0.954 0.924 0.070 0.062
semPaths(senna1o.fit,
whatLabels = "std"
)
semPaths(senna1o.fit,
whatLabels = "std",
edge.label.cex=.6,
node.width = 1.1
)
ggplot(data = scores,
aes(y = port+mat, x = C_c, color = cria_quartis(A_c))) +
geom_point(alpha=.5) + geom_smooth(method = "lm")
