Charles Darwin University

CDU eSpace
Institutional Repository

 
CDU Staff and Student only
 

Structural Equation Modelling and Amos: Factor Analysis of Torque, Work and Power

Heazlewood, Ian and Boutagy, Tony (2010). Structural Equation Modelling and Amos: Factor Analysis of Torque, Work and Power<br />. In: Lyons, Keith, Baca, Arnold and Lebedew, Alexis 7th International IACSS Symposium on Computer Science in Sport, Canberra, 22-25 September 2009.

Document type: Conference Paper
Citation counts: Google Scholar Search Google Scholar

Author Heazlewood, Ian
Boutagy, Tony
Title Structural Equation Modelling and Amos: Factor Analysis of Torque, Work and Power
Conference Name 7th International IACSS Symposium on Computer Science in Sport
Conference Location Canberra
Conference Dates 22-25 September 2009
Conference Publication Title Proceedings of the seventh International IACSS Symposium
Editor Lyons, Keith
Baca, Arnold
Lebedew, Alexis
Place of Publication Canberra, ACT
Publisher National Institute of Sport Studies
Publication Year 2010
ISBN 9780980716009   (check CDU catalogue open catalogue search in new window)
Start Page 67
End Page 74
Total Pages 8
Abstract The software AMOS, a second generation statistical method based on structural equation modelling (SEM), can address complex problems and interactions in sport and exercise science, and especially factor analytic research relevant to measurement theory in
biomechanics, exercise physiology and motor skill learning. Sixty healthy and active participants (32 males; 28 females) were involved in the study located in an exercise laboratory. AMOS was applied as a method of confirmatory factor analysis (CFA) confirm a measurement model for torque-work and power based on isokinetic leg extension at isokinetic speeds of 60o s-1 (5 reps), 180os-1 (15 reps) and 300os-1 (15 reps) using leg extension/flexion with CYBEX technology and measures of power using vertical jump concepts of height, flight time, contact time and 10m sprint acceleration. The hypothesis was that many dependent va-
riables are only measuring a small set of underlying factors or latent constructs. The results indicated that the two factor model was partially supported using SEM goodness of fit indices, however more substantively by arguments based on construct, convergent and discriminant validity. The discussion promotes CFA-SEM as methods to understand more completely sports specific fitness tests based on multiple measurements and measurement parsimony.
Keyword Factor analysis
Amos Software
Statstical models
Goodness of fit
Description for Link Link to published version
URL http://api.ning.com/files/f3hD5ZeJUnXIJnxCApyD3h73jPBeMRwRoLC2pnCL4pIB*qMLDVJpDZMPeyQ7qcvOojfjfL9z2LInlJE-6T-xzzQrKoXMtHrh/090922IACSS09Proceedingsrfs.pdf
 
Versions
Version Filter Type
Access Statistics: 21 Abstract Views  -  Detailed Statistics
Created: Wed, 17 Feb 2016, 11:37:39 CST by Marion Farram