Parallel analysis
Parallel analysis, also known as Horn's parallel analysis, is a statistical method used to determine the number of components to keep in a
Monte-Carlo simulated matrix created from random data of the same size.[2]
Evaluation and comparison with alternatives
Parallel analysis is regarded as one of the more accurate methods for determining the number of factors or components to retain.[3] Since its original publication, multiple variations of parallel analysis have been proposed.[4][5] Other methods of determining the number of factors or components to retain in an analysis include the scree plot, Kaiser rule, or Velicer's MAP test.[6]
Implementation
Parallel analysis has been implemented in
STATA, and MATLAB[8][9][10] and in multiple packages for the R programming language, including the psych[11][12] multicon,[13] hornpa,[14] and paran packages.[15][16]
See also
- Scree plot
- Exploratory factor analysis § Selecting the appropriate number of factors
- Marchenko-Pastur distribution
References
- S2CID 19663974.
- ISBN 978-1-4833-8142-8.
- .
- S2CID 123508406.
- S2CID 63269411.
- S2CID 122907389.
- S2CID 143051337.
- S2CID 61286653.
- ^ O'Connor, Brian. "Programs for Number of Components and Factors". people.ok.ubc.ca.
- PMID 11029811.
- ^ Revelle, William (2007). "Determining the number of factors: the example of the NEO-PI-R" (PDF).
{{cite journal}}
: Cite journal requires|journal=
(help) - ^ Revelle, William (8 January 2020). "psych: Procedures for Psychological, Psychometric, and PersonalityResearch".
- ^ Sherman, Ryne A. (2 February 2015). "multicon: Multivariate Constructs".
- ^ Huang, Francis (3 March 2015). "hornpa: Horn's (1965) Test to Determine the Number of Components/Factors".
- ^ Dinno, Alexis. "Gently Clarifying the Application of Horn's Parallel Analysis to Principal Component Analysis Versus Factor Analysis" (PDF).
{{cite journal}}
: Cite journal requires|journal=
(help) - ^ Dinno, Alexis (14 October 2018). "paran: Horn's Test of Principal Components/Factors".
{{cite journal}}
: Cite journal requires|journal=
(help)