assumptionCheck: Assumption Checking Function

Description

Checks some basic assumptions about the suitability of network analysis on your data

Usage

assumptionCheck(
  data,
  type = c("network", "impact"),
  percent = 20,
  split = c("median", "mean", "forceEqual", "cutEqual", "quartiles"),
  plot = FALSE,
  binary.data = FALSE,
  na.rm = TRUE
)

Arguments

data: dataframe or matrix of observational data (rows: observations, columns: nodes)
type: which assumptions to check? "network" tests the suitability for network analysis in general. "impact" tests the suitability for analyzing impact
percent: percent difference from grand mean that is acceptable when comparing variances.
split: if type="impact", specifies the type of split to utilize
plot: logical. Should histograms each variable be plotted?
binary.data: logical. Defaults to FALSE
na.rm: logical. Should missing values be removed?

Details

Network analysis rests on several assumptions. Among these: - Variance of each node is (roughly) equal - Distributions are (roughly) normal

Comparing networks in impact rests on additional assumptions including: - Overall variances are (roughly) equal in each half

This function checks these assumptions and notifies any violations. This function is not intended as a substitute for careful data visualization and independent assumption checks.

See citations in the references section for further details.

References

Terluin, B., de Boer, M. R., & de Vet, H. C. W. (2016). Differences in Connection Strength between Mental Symptoms Might Be Explained by Differences in Variance: Reanalysis of Network Data Did Not Confirm Staging. PLOS ONE, 11(11), e0155205. Retrieved from https://doi.org/10.1371/journal.pone.0155205