The correlation matrix of 17 anthropometric measures from the United States Air Force survey of 2420 airmen. The data are taken from the Anthropometry package and included here as a demonstration of a hierarchical factor structure suitable for analysis by the omega
or omegaSem
.
data("USAF")
The format is: num [1:17, 1:17] 1 0.1148 -0.0309 -0.028 -0.0908 ... - attr(*, "dimnames")=List of 2 ..$ : chr [1:17] "age" "weight" "grip" "height" ... ..$ : chr [1:17] "age" "weight" "grip" "height" ...
The original data were collected by the USAF and reported in Churchill et al, 1977. They are included as a data file of 2420 participants and 202 variables (the first being an id) in the Anthropometry package. The list of variable names may be found in Churchill et al, on pages 99-103.
The three (correlated) factor structure shows a clear height, bulk, and head size structure with an overall general factor (g) which may be interpreted as body size.
The variables included (and their variable numbers in Anthropometry) are:
age | V1 |
weight | V2 |
grip strength | V12 |
height (stature) | V13 |
leg length | V26 |
knee height | V37 |
upper arm | V42 |
thumb tip reach | V47 |
in sleeve | V49 |
chest breadth | V52 |
hip breadth | V55 |
waist circumference | V71 |
thigh circumference | V97 |
scye circumference | V103 |
head circumference | V141 |
bitragion coronal | V145 |
head length | V150 |
glabella to wall | V181 |
external canthus to wall | V183 |
Note that these numbers are equivalant to the numbers in Churchill et al. The numbers in Anthropometry are these + 1.
Edmund Churchill, Thomas Churchill, Paul Kikta (1977) The AMRL anthropmetric data bank library, volumes I-V. (Technical report AMRL-TR-77-1) ) https://apps.dtic.mil/dtic/tr/fulltext/u2/a047314.pdf
Guillermo Vinue, Anthropometry: An R Package for Analysis of Anthropometric Data, Journal of Statistical Software, (2017), 77, 6.
data(USAF)
psych::corPlot(USAF,xlas=3)
psych::omega(USAF[c(4:8,10:19),c(4:8,10:19)]) #just the size variables
Run the code above in your browser using DataLab