D0 <- Dincl <- D$ankeedikeel ==1& D$LV_ankeedikeel ==1D <-subset(D, incl)# Di <- Dj <- Dij <- D
Keeping the respondents if both the parent and the child responded in Estonian. A few 18- and 19-year-olds are included (even though they are not in the target age group).
Descriptive analyses
Background
Chidren’s gender and age (including only those who responded in Estonian)
# Correlations in 2nd vs 3rd age groupc1 <-cor(subset(D, D$vanus_grupp==2, select=psc_l), use="pai")c2 <-cor(subset(D, D$vanus_grupp==3, select=psc_l), use="pai")plot(lt(c1), lt(c2), xlab="Correlations in the 2nd age group", ylab="Correlations in the 3rd age group")
PSC means by gender and age
Table 4 (first part). Descriptive analysis for gender and age groups and the Cohen d’s of PSC-17-Y and PSC-17-P.
Column labels: lowercase part: a = alpha, o = omega, avg = average interitem correlation, min = minimum interitem correlation, max = maximum interitem correlation UPPERCASE part_: T = total, B=boys, G=girls
We used the WLSMV estimator due to its suitability for ordered categorical indicators and its less stringent sample size requirements compared to alternatives (Brown, 2015). Following the recommendations of Brosseau-Liard and Savalei (2014), we applied robust corrections to the fit indices (CFI, TLI, RMSEA, and SRMR). To interpret the invariance tests, we used the following cutoff values for noninvariance based on Chen (2007): (1) for metric invariance (testing for equality of loadings), a CFI decrease of .010 or greater along with an RMSEA increase of .015 or greater or an SRMR increase of .030 or greater; (2) for scalar invariance (testing for equality of intercepts), a CFI decrease of .010 or greater combined with an RMSEA increase of .015 or greater or an SRMR increase of .010 or greater. Scaled chi-square and df are reported.
Brosseau-Liard, P. E., & Savalei, V. (2014). Adjusting Incremental Fit Indices for Nonnormality. Multivariate Behavioral Research, 49(5), 460–470. https://doi.org/10.1080/00273171.2014.933697
Brown, T. A. (2015). Confirmatory factor analysis for applied research (2nd ed.). Guilford Press. [ ISBN 978-1-4625-1536-3]
Chen, F. F. (2007). Sensitivity of goodness of fit indexes to lack of measurement invariance. Structural Equation Modeling, 14(3), 464–504. https://doi.org/10.1080/10705510701301834
Scaled Chi-Squared Difference Test (method = "satorra.2000")
lavaan->lavTestLRT():
lavaan NOTE: The "Chisq" column contains standard test statistics, not the
robust test that should be reported per model. A robust difference test is
a function of two standard (not robust) statistics.
Df AIC BIC Chisq Chisq diff Df diff Pr(>Chisq)
fit2mg 348 278.63
fit2mg.el 376 345.00 28.486 28 0.4389
Scaled Chi-Squared Difference Test (method = "satorra.2000")
lavaan->lavTestLRT():
lavaan NOTE: The "Chisq" column contains standard test statistics, not the
robust test that should be reported per model. A robust difference test is
a function of two standard (not robust) statistics.
Df AIC BIC Chisq Chisq diff Df diff Pr(>Chisq)
fit2mg.el 376 345.00
fit2mg.ei 404 365.43 32.543 28 0.253
Scaled Chi-Squared Difference Test (method = "satorra.2000")
lavaan->lavTestLRT():
lavaan NOTE: The "Chisq" column contains standard test statistics, not the
robust test that should be reported per model. A robust difference test is
a function of two standard (not robust) statistics.
Df AIC BIC Chisq Chisq diff Df diff Pr(>Chisq)
fit2mgs 232 195.56
fit2mgs.el 246 226.77 13.365 14 0.498
Scaled Chi-Squared Difference Test (method = "satorra.2000")
lavaan->lavTestLRT():
lavaan NOTE: The "Chisq" column contains standard test statistics, not the
robust test that should be reported per model. A robust difference test is
a function of two standard (not robust) statistics.
Df AIC BIC Chisq Chisq diff Df diff Pr(>Chisq)
fit2mgs.el 246 226.77
fit2mgs.ei 260 259.96 50.723 14 4.618e-06 ***
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Included: only respondents in 2 older age groups if both child and parent responded in Estonian
Code
#dimensions of data frame: rows x columnsdim(subset(D, vg2%in%2:3))
[1] 384 480
Code
ct <-corr.test(D[, c("att", "int", "ext")], D[, c("attlv", "intlv", "extlv")])res <-data.frame(var =rownames(ct$r), r =round(diag(ct$r),3), p =round(diag(ct$p), 5), N =if(length(ct$n)==1) ct$n elsediag(ct$n))qflextable(res)
var
r
p
N
att
0.272
0.00001
254
int
0.344
0.00000
254
ext
0.271
0.00001
254
Correlations between single items (the order of items was different in child’s and parental questionnaires: in the table, the order used in child’s version is used).
Code
ct <-corr.test(D[, pk$laps], D[, pk$lv])res <-data.frame(var =rownames(ct$r), item = pk$laps_item, r =round(diag(ct$r),3), p =round(diag(ct$p), 5), N =if(length(ct$n)==1) ct$n elsediag(ct$n))qflextable(res)
var
item
r
p
N
PSC1
1. Olen rahutu, ei suuda paigal püsida
0.324
0.00000
255
PSC2
2. Olen kurb, õnnetu
0.168
0.00717
255
PSC3
3. Unistan liiga palju
0.073
0.24758
254
PSC4
4. Keeldun oma asju jagamast
0.170
0.00670
254
PSC5
5. Ma ei mõista teiste inimeste tundeid
0.141
0.02442
255
PSC6
6. Tunnen lootusetust
0.327
0.00000
250
PSC7
7. Mul on raske keskenduda
0.324
0.00000
253
PSC8
8. Kaklen teiste lastega
0.117
0.06290
254
PSC9
9. Olen enda suhtes kriitiline, karm
0.252
0.00005
254
PSC10
10. Süüdistan oma muredes teisi
0.233
0.00018
253
PSC11
11. Mul ei ole nii lõbus kui varem
0.156
0.01290
254
PSC12
12. Ma ei järgi reegleid
0.218
0.00049
252
PSC13
13. Tunnen vajadust pidevalt tegutseda, nagu oleksin üles keeratud
0.105
0.09414
255
PSC14
14. Narrin teisi
0.186
0.00306
253
PSC15
15. Muretsen palju
0.234
0.00017
254
PSC16
16. Võtan asju, mis ei kuulu mulle
0.223
0.00034
254
PSC17
17. Minu tähelepanu kaldub kergesti kõrvale
0.277
0.00001
255
Prediction of health problems
Parental questionnaire included a question about different diseases / disorders diagnosed in the child; among these: anxiety disorders, mood disorders, eating disorders, learning difficulties, ADHD, autism spectrum disorders.
Using the following variables from the data set (see end of each row for English summary:)
Descriptive stats of parent-reported diagnoses
Table 7. Descriptive statistics of parent-reported health problems
Call:
glm(formula = LV_terv_prob_9 ~ factor(sugu2) + vanus, family = "binomial",
data = subset(D, vg2 %in% 2:3))
Coefficients:
Estimate Std. Error z value Pr(>|z|)
(Intercept) -1.0086 1.7940 -0.562 0.574
factor(sugu2)2 -0.3457 0.5590 -0.618 0.536
vanus -0.1188 0.1301 -0.913 0.361
(Dispersion parameter for binomial family taken to be 1)
Null deviance: 108.02 on 250 degrees of freedom
Residual deviance: 106.65 on 248 degrees of freedom
(133 observations deleted due to missingness)
AIC: 112.65
Number of Fisher Scoring iterations: 6
