FloSchuberth
commented
3 years ago That VIFs are not reported is strange. @Manuel: can you please check this.? For that reason, could you please send me your R code with the data attached.
Considering the FL, it is correctly calculated. You calculated the squared correlation between the composite scores, i.e., the composites. However, since you specified your constructs as common factors, we report the squared correlation between the latent variables.
Best regards and many thanks again, Florian
Von: srk7774 @.***> Gesendet: Mittwoch, 25. August 2021 20:35:19 An: M-E-Rademaker/cSEM Cc: Subscribed Betreff: [M-E-Rademaker/cSEM] VIF's not not reported (#448)
VIF's are not reported. Also, cross check the values in fornell-lacker in smartpls.
library(readr)
> Warning: package 'readr' was built under R version 4.0.5
demo_1 <- read_csv("https://raw.githubusercontent.com/srk7774/data/master/demo_1_nk.csv")
> Rows: 243 Columns: 27
> -- Column specification --------------------------------------------------------
> Delimiter: ","
> dbl (27): age, prof, income, gender, resi, quali, quality1, quality2, qualit...
>
> i Use spec() to retrieve the full column specification for this data.
> i Specify the column types or set show_col_types = FALSE to quiet this message.
model <- "
Structural model
sat ~ qual
(Reflective) measurement model
qual =~ quality1 + quality2 + quality3 + quality4 + quality5 + quality6 sat =~ sat1 + sat2 + sat3 + sat4 "
Estimate
library(cSEM)
> Warning: package 'cSEM' was built under R version 4.0.5
>
> Attaching package: 'cSEM'
> The following object is masked from 'package:stats':
>
> predict
out <- csem(demo_1, model, .approach_weights = "PLS-PM", .disattenuate = TRUE, .resample_method = 'bootstrap') assess(out)
> ____
>
> Construct AVE R2 R2_adj
> qual 0.5802 NA NA
> sat 0.5991 0.4046 0.4021
>
> -------------- Common (internal consistency) reliability estimates -------------
>
> Construct Cronbachs_alpha Joereskogs_rho Dijkstra-Henselers_rho_A
> qual 0.8946 0.8913 0.8981
> sat 0.8520 0.8558 0.8620
>
> ----------- Alternative (internal consistency) reliability estimates -----------
>
> Construct RhoC RhoC_mm RhoC_weighted
> qual 0.8913 0.8761 0.8981
> sat 0.8558 0.8587 0.8620
>
> Construct RhoC_weighted_mm RhoT RhoT_weighted
> qual 0.8981 0.8946 0.8968
> sat 0.8620 0.8520 0.8427
>
> --------------------------- Distance and fit measures --------------------------
>
> Geodesic distance = 0.1448631
> Squared Euclidian distance = 0.2236872
> ML distance = 0.7148407
>
> Chi_square = 172.9915
> Chi_square_df = 5.087984
> CFI = 0.8948454
> CN = 68.99049
> GFI = 0.843506
> IFI = 0.8957133
> NFI = 0.8734316
> NNFI = 0.8608248
> RMSEA = 0.1299711
> RMS_theta = 0.0730421
> SRMR = 0.06377335
>
> Degrees of freedom = 34
>
> --------------------------- Model selection criteria ---------------------------
>
> Construct AIC AICc AICu
> sat -122.9953 122.1051 -120.9871
>
> Construct BIC FPE GM
> sat -116.0092 0.6028 252.9861
>
> Construct HQ HQc Mallows_Cp
> sat -120.1814 -120.0674 3.0000
>
> ----------------------- Variance inflation factors (VIFs) ----------------------
>
> -------------------------- Effect sizes (Cohen's f^2) --------------------------
>
> Dependent construct: 'sat'
>
> Independent construct f^2
> qual 0.6795
>
> ------------------------------ Validity assessment -----------------------------
>
> Heterotrait-monotrait ratio of correlations matrix (HTMT matrix)
>
> qual sat
> qual 1.0000000 0
> sat 0.6318983 1
>
>
> Fornell-Larcker matrix
>
> qual sat
> qual 0.5802354 0.4045816
> sat 0.4045816 0.5990676
>
>
> ------------------------------------ Effects -----------------------------------
>
> Estimated total effects:
> ========================
> Total effect Estimate Std. error t-stat. p-value
> sat ~ qual 0.6361 0.0456 13.9444 0.0000
> ____
latent variable correlation
cor(out$Estimates$Construct_scores[,1], out$Estimates$Construct_scores[,2])
> [1] 0.5596549
Created on 2021-08-26 by the reprex packagehttps://reprex.tidyverse.org (v2.0.1)
— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHubhttps://github.com/M-E-Rademaker/cSEM/issues/448, or unsubscribehttps://github.com/notifications/unsubscribe-auth/AD5KRRALYQJV22C4S4QWG5TT6UZWPANCNFSM5CZW7JNQ. Triage notifications on the go with GitHub Mobile for iOShttps://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675 or Androidhttps://play.google.com/store/apps/details?id=com.github.android&utm_campaign=notification-email.
srk7774
VIF's are not reported. Also, please cross check the values in fornell-lacker in smartpls.
Created on 2021-08-26 by the reprex package (v2.0.1)