SEMinR (Ray & Danks) is a domaign specific language for modeling and estimating structural equation models. This package serves to provide additional extra methods and functions that can be used to analyze PLS-SEM models.

New features implemented with this package CVPAT - assess_cvpat() - assess_cvpat_compare()

The Example model: Corporate Reputation

We are applying our CV-PAT to the corporate reputation example bundled with SEMinR.

plot(established_model)

plot(competing_model)

Example


# Create measurement model ----
corp_rep_mm_ext <- constructs(
  composite("QUAL", multi_items("qual_", 1:8), weights = mode_B),
  composite("PERF", multi_items("perf_", 1:5), weights = mode_B),
  composite("CSOR", multi_items("csor_", 1:5), weights = mode_B),
  composite("ATTR", multi_items("attr_", 1:3), weights = mode_B),
  composite("COMP", multi_items("comp_", 1:3)),
  composite("LIKE", multi_items("like_", 1:3)),
  composite("CUSA", single_item("cusa")),
  composite("CUSL", multi_items("cusl_", 1:3))
)

alt_mm <- constructs(
  composite("QUAL", multi_items("qual_", 1:8), weights = mode_B),
  composite("PERF", multi_items("perf_", 1:5), weights = mode_B),
  composite("CSOR", multi_items("csor_", 1:5), weights = mode_B),
  composite("ATTR", multi_items("attr_", 1:3), weights = mode_B),
  composite("COMP", multi_items("comp_", 1:3)),
  composite("LIKE", multi_items("like_", 1:3)),
  composite("CUSA", single_item("cusa")),
  composite("CUSL", multi_items("cusl_", 1:3))
)

# Create structural model ----
corp_rep_sm_ext <- relationships(
  paths(from = c("QUAL", "PERF", "CSOR", "ATTR"), to = c("COMP", "LIKE")),
  paths(from = c("COMP", "LIKE"), to = c("CUSA", "CUSL")),
  paths(from = c("CUSA"),         to = c("CUSL"))
)
alt_sm <- relationships(
  paths(from = c("QUAL", "PERF", "CSOR", "ATTR"), to = c("COMP", "LIKE")),
  paths(from = c("COMP", "LIKE"), to = c("CUSA")),
  paths(from = c("CUSA"),         to = c("CUSL"))
)


# Estimate the models ----
established_model <- estimate_pls(
  data = corp_rep_data,
  measurement_model = corp_rep_mm_ext,
  structural_model  = corp_rep_sm_ext,
  missing = mean_replacement,
  missing_value = "-99")

competing_model <- estimate_pls(
  data = corp_rep_data,
  measurement_model = alt_mm,
  structural_model  = alt_sm,
  missing = mean_replacement,
  missing_value = "-99")

# Function to compare the Loss of two models
compare_results <- assess_cvpat_compare(established_model = established_model,
                                        alternative_model = competing_model,
                                        testtype = "two.sided",
                                        nboot = 2000,
                                        technique = predict_DA,
                                        seed = 123,
                                        noFolds = 10,
                                        reps = 10,
                                        cores = 1)


print(compare_results,
      digits = 3)

# Assess the base model ----
assess_results <- assess_cvpat(established_model,
                               seed = 123, 
                               cores = 1)
print(assess_results$CVPAT_compare_LM,
      digits = 3)
print(assess_results$CVPAT_compare_IA,
      digits = 3)

Now we compare the results:

# Function to compare the Loss of two models
compare_results <- assess_cvpat_compare(established_model = established_model,
                                        alternative_model = competing_model,
                                        testtype = "two.sided",
                                        nboot = 2000,
                                        technique = predict_DA,
                                        seed = 123,
                                        noFolds = 10,
                                        reps = 10,
                                        cores = 1)

print(compare_results, 
      cores = 1,
      digits = 3)
#>         Base Model Loss Alt Model Loss   Diff Boot T value Boot P Value
#> COMP              1.198          1.195  0.003       -0.460        0.645
#> LIKE              1.923          1.933 -0.010        0.883        0.378
#> CUSA              0.988          0.992 -0.004        0.809        0.419
#> CUSL              1.562          1.715 -0.152        3.286        0.001
#> Overall           1.418          1.459 -0.041        3.293        0.001
#> 
#> CV-PAT as per Sharma et al. (2023).
#>   Both models under comparison have identical endoogenous constructs with identical measurement models.
#>   Purely exogenous constructs can be differ in regards to their relationships with both nomological
#>   partners and measurement indicators.

# Assess the base model ----
assess_results <- assess_cvpat(established_model,
                               seed = 123, 
                               cores = 1)
print(assess_results$CVPAT_compare_LM,
      digits = 3)
#>         PLS Loss LM Loss   Diff Boot T value Boot P Value
#> COMP       1.196   1.238 -0.042        1.308        0.192
#> LIKE       1.915   1.929 -0.014        0.247        0.805
#> CUSA       0.994   0.774  0.220       -2.982        0.003
#> CUSL       1.560   1.713 -0.153        3.838        0.000
#> Overall    1.416   1.414  0.003       -0.104        0.918
#> 
#> CV-PAT as per Sharma et al. (2023).
print(assess_results$CVPAT_compare_IA,
      digits = 3)
#>         PLS Loss IA Loss   Diff Boot T value Boot P Value
#> COMP       1.196   2.023 -0.827        8.580        0.000
#> LIKE       1.915   3.103 -1.187        8.293        0.000
#> CUSA       0.994   1.374 -0.379        5.004        0.000
#> CUSL       1.560   2.663 -1.102        7.572        0.000
#> Overall    1.416   2.290 -0.874       10.301        0.000
#> 
#> CV-PAT as per Sharma et al. (2023).

References

Sharma, P. N., Liengaard, B. D., Hair, J. F., Sarstedt, M., & Ringle, C. M. (2022). Predictive model assessment and selection in composite-based modeling using PLS-SEM: extensions and guidelines for using CVPAT. European journal of marketing, 57(6), 1662-1677.
Liengaard, B. D., Sharma, P. N., Hult, G. T. M., Jensen, M. B., Sarstedt, M., Hair, J. F., & Ringle, C. M. (2021). Prediction: coveted,yet forsaken? Introducing a cross‐validated predictive ability test in partial least squares path modeling. Decision Sciences, 52(2), 362-392.
Ray, S., Danks, N.P., Calero Valdez, A. (2024) SEMinR: Domain-specific language for building, estimating, and visualizing structural equation models in R.

Also a quote using >:

“He who gives up [code] safety for [code] speed deserves neither.” (via)

seminrExtras

S. Ray and N.P. Danks

2025-07-21

The Example model: Corporate Reputation

Example

References