Estimates the number of communities (dimensions) of a dataset or correlation matrix using a network estimation method (Golino & Epskamp, 2017; Golino et al., 2020). After, a community detection algorithm is applied (Christensen et al., 2023) for multidimensional data. A unidimensional check is also applied based on findings from Golino et al. (2020) and Christensen (2023)
Arguments
- data
Matrix or data frame. Should consist only of variables to be used in the analysis. Can be raw data or a correlation matrix
- n
Numeric (length = 1). Sample size if
data
provided is a correlation matrix- corr
Character (length = 1). Method to compute correlations. Defaults to
"auto"
. Available options:"auto"
— Automatically computes appropriate correlations for the data using Pearson's for continuous, polychoric for ordinal, tetrachoric for binary, and polyserial/biserial for ordinal/binary with continuous. To change the number of categories that are considered ordinal, useordinal.categories
(seepolychoric.matrix
for more details)"cor_auto"
— Usescor_auto
to compute correlations. Arguments can be passed along to the function"pearson"
— Pearson's correlation is computed for all variables regardless of categories"spearman"
— Spearman's rank-order correlation is computed for all variables regardless of categories
For other similarity measures, compute them first and input them into
data
with the sample size (n
)- na.data
Character (length = 1). How should missing data be handled? Defaults to
"pairwise"
. Available options:"pairwise"
— Computes correlation for all available cases between two variables"listwise"
— Computes correlation for all complete cases in the dataset
- model
Character (length = 1). Defaults to
"glasso"
. Available options:"BGGM"
— Computes the Bayesian Gaussian Graphical Model. Set argumentordinal.categories
to determine levels allowed for a variable to be considered ordinal. See?BGGM::estimate
for more details"glasso"
— Computes the GLASSO with EBIC model selection. SeeEBICglasso.qgraph
for more details"TMFG"
— Computes the TMFG method. SeeTMFG
for more details
- algorithm
Character or
igraph
cluster_*
function (length = 1). Defaults to"walktrap"
. Three options are listed below but all are available (seecommunity.detection
for other options):"leiden"
— Seecluster_leiden
for more details"louvain"
— By default,"louvain"
will implement the Louvain algorithm using the consensus clustering method (seecommunity.consensus
for more information). This function will implementconsensus.method = "most_common"
andconsensus.iter = 1000
unless specified otherwise"walktrap"
— Seecluster_walktrap
for more details
- uni.method
Character (length = 1). What unidimensionality method should be used? Defaults to
"louvain"
. Available options:"expand"
— Expands the correlation matrix with four variables correlated 0.50. If number of dimension returns 2 or less in check, then the data are unidimensional; otherwise, regular EGA with no matrix expansion is used. This method was used in the Golino et al.'s (2020) Psychological Methods simulation"LE"
— Applies the Leading Eigenvector algorithm (cluster_leading_eigen
) on the empirical correlation matrix. If the number of dimensions is 1, then the Leading Eigenvector solution is used; otherwise, regular EGA is used. This method was used in the Christensen et al.'s (2023) Behavior Research Methods simulation"louvain"
— Applies the Louvain algorithm (cluster_louvain
) on the empirical correlation matrix. If the number of dimensions is 1, then the Louvain solution is used; otherwise, regular EGA is used. This method was validated Christensen's (2022) PsyArXiv simulation. Consensus clustering can be used by specifying either"consensus.method"
or"consensus.iter"
- plot.EGA
Boolean (length = 1). Defaults to
TRUE
. Whether the plot should be returned with the results. Set toFALSE
for no plot- verbose
Boolean (length = 1). Whether messages and (insignificant) warnings should be output. Defaults to
FALSE
(silent calls). Set toTRUE
to see all messages and warnings for every function call- ...
Additional arguments to be passed on to
auto.correlate
,network.estimation
,community.detection
,community.consensus
, andcommunity.unidimensional
Value
Returns a list containing:
- network
A matrix containing a network estimated using
link[EGAnet]{network.estimation}
- wc
A vector representing the community (dimension) membership of each node in the network.
NA
values mean that the node was disconnected from the network- n.dim
A scalar of how many total dimensions were identified in the network
- correlation
The zero-order correlation matrix
- n
Number of cases in
data
- dim.variables
An ordered matrix of item allocation
- TEFI
link[EGAnet]{tefi}
for the estimated structure- plot.EGA
Plot output if
plot.EGA = TRUE
References
Original simulation and implementation of EGA
Golino, H. F., & Epskamp, S. (2017).
Exploratory graph analysis: A new approach for estimating the number of dimensions in psychological research.
PLoS ONE, 12, e0174035.
Current implementation of EGA, introduced unidimensional checks, continuous and dichotomous data
Golino, H., Shi, D., Christensen, A. P., Garrido, L. E., Nieto, M. D., Sadana, R., & Thiyagarajan, J. A. (2020).
Investigating the performance of Exploratory Graph Analysis and traditional techniques to identify the number of latent factors: A simulation and tutorial.
Psychological Methods, 25, 292-320.
Compared all igraph community detection algorithms, introduced Louvain algorithm, simulation with continuous and polytomous data
Also implements the Leading Eigenvalue unidimensional method
Christensen, A. P., Garrido, L. E., Pena, K. G., & Golino, H. (2023).
Comparing community detection algorithms in psychological data: A Monte Carlo simulation.
Behavior Research Methods.
Comprehensive unidimensionality simulation
Christensen, A. P. (2023).
Unidimensional community detection: A Monte Carlo simulation, grid search, and comparison.
PsyArXiv.
Compared all igraph
community detection algorithms, simulation with continuous and polytomous data
Christensen, A. P., Garrido, L. E., Guerra-Pena, K., & Golino, H. (2023).
Comparing community detection algorithms in psychometric networks: A Monte Carlo simulation.
Behavior Research Methods.
See also
plot.EGAnet
for plot usage in EGAnet
Author
Hudson Golino <hfg9s at virginia.edu>, Alexander P. Christensen <alexpaulchristensen at gmail.com>, Maria Dolores Nieto <acinodam at gmail.com> and Luis E. Garrido <garrido.luiseduardo at gmail.com>
Examples
# Obtain data
wmt <- wmt2[,7:24]
# Estimate EGA
ega.wmt <- EGA(
data = wmt,
plot.EGA = FALSE # No plot for CRAN checks
)
# Print results
print(ega.wmt)
#> Model: GLASSO (EBIC with gamma = 0.5)
#> Correlations: auto
#> Lambda: 0.0648639582532287 (n = 100, ratio = 0.1)
#>
#> Number of nodes: 18
#> Number of edges: 96
#> Edge density: 0.627
#>
#> Non-zero edge weights:
#> M SD Min Max
#> 0.082 0.060 -0.013 0.363
#>
#> ----
#>
#> Algorithm: Walktrap
#>
#> Number of communities: 2
#>
#> wmt1 wmt2 wmt3 wmt4 wmt5 wmt6 wmt7 wmt8 wmt9 wmt10 wmt11 wmt12 wmt13
#> 1 1 1 1 1 2 2 2 2 2 2 2 2
#> wmt14 wmt15 wmt16 wmt17 wmt18
#> 2 2 2 2 2
#>
#> ----
#>
#> Unidimensional Method: Louvain
#> Unidimensional: No
#>
#> ----
#>
#> TEFI: -11.171
# Estimate EGAtmfg
ega.wmt.tmfg <- EGA(
data = wmt, model = "TMFG",
plot.EGA = FALSE # No plot for CRAN checks
)
# Estimate EGA with Louvain algorithm
ega.wmt.louvain <- EGA(
data = wmt, algorithm = "louvain",
plot.EGA = FALSE # No plot for CRAN checks
)
# Estimate EGA with an {igraph} function (Fast-greedy)
ega.wmt.greedy <- EGA(
data = wmt,
algorithm = igraph::cluster_fast_greedy,
plot.EGA = FALSE # No plot for CRAN checks
)