Abstract
During the early steps of the construction of composite health measures, principal component analysis (PCA) is commonly used to identify ‘latent’ factors that underlie observed variables and to determine the dimensionality of the instruments. The determination of the number of components to retain is critical to PCA: it markedly influences the factorial model identified and further conditions the validity of the constructed instrument. However, many researchers developing composite health measures seem to be unaware of the importance of this determination. The purposes of the paper are to illustrate (1) the variability of the factorial models obtained by using different published rules (n=10) for determining the number of components to retain in PCA applied to two quality-of-life datasets, and (2) the value of a careful and diversified approach to the problem of the number of components to retain in PCA that we suggest, instead of the unsatisfactory ‘rule-of-thumb’ that many researchers use. This involves: (1) using robust rules (including parallel analysis and minimum average partial procedure) to generate a set of possible values for the number of components to retain, (2) repeating the analysis across samples, (3) comprehensively assessing the models obtained, and (4) considering complementary methods to PCA and especially confirmatory factor analysis.
Similar content being viewed by others
References
AR Feinstein (1987) Clinimetrics. Yale University Press New Haven
DL Streiner GR. Norman (1989) Health Measurement Scales: A Practical Guide to Their Development and Use. Oxford University Press Oxford
B Kirshner GH Guyatt (1985) ArticleTitleA methodological framework for assessing health indices. J Clin Epidemiol 38 27–36
GH Guyatt B Kirshner R Jaeschke (1992) ArticleTitleMeasuring health status: What are the necessary measurement properties? J Clin Epidemiol 45 1341–1345
DR Cox R Fitzpatrick AE Fletcher et al. (1992) ArticleTitleQuality of life assessment: Can we keep it simple? J Royal Stat Soc 155 353–393
J Coste J Fermanian A Venot (1995) ArticleTitleMethodological and statistical problems in the contruction of composite measurement scales. A survey of six medical and epidemiological journals Stat Med 14 331–345
JC Nunnally (1978) Psychometric Theory. Mac Graw-Hill New York
JE Jackson (1991) A User’s Guide to Principal Components. Wiley New York
RF Vellis ParticleDe (1991) Scale Development: Theory and Applications. Sage Newbury Park, Calif.
IT Jolliffe (2002) Principal Components Analysis EditionNumber2 Springer Verlag New York
JL Fava WF Velicer (1992) ArticleTitleThe effect of overextraction on factor and component analysis. Multivar Behav Res 27 387–415
JM Wood DJ Tataryn RL Gorsuch (1996) ArticleTitleEffects of under- and overextraction on principal factor analysis with varimax rotation. Psychol Meth 1 354–365
FR Lawrence GR Hancock (1999) ArticleTitleConditions affecting integrity of a factor solution under varying degrees of overextraction. Educ Psychol Meas 59 549–579
JL Fava WF Velicer (1996) ArticleTitleThe effect of underextraction in factor and component analysis. Educ Psychol Meas 56 907–929
R Hubbard SJ Allen (1987) ArticleTitleAn empirical comparison of alternative methods for principal component extraction. J Bus Res 15 173–190
DA Jackson (1993) ArticleTitleStopping rules in principal components analysis: A comparison of heuristical and statistical approaches. Ecology 74 2204–2214
L Guttman (1954) ArticleTitleSome necessary conditions for common factor analysis. Psychometrika 19 149–161
HF Kaiser (1960) ArticleTitleThe application of electronic computers to factor analysis. Educ Psychol Meas 20 141–151
L Legendre P Legendre (1983) Numerical Ecology Elsevier Amsterdam 406
RB Cattell (1966) ArticleTitleThe scree test for the number of factors. Multivar Behav Res 1 245–276
WF Velicer (1976) ArticleTitleDetermining the number of components from the matrix of partial correlations. Psychometrika 41 321–327
BP O’Connor (2000) ArticleTitleSPSS and SAS programs for determining the number of components using parallel analysis and Velicer’s MAP test Behav Res Meth Instrum Comput 32 393–402
MS Bartlett (1950) ArticleTitleTests of significance in factor analysis. Br J Psychol (Statistical Section) 3 77–85
JL Horn (1965) ArticleTitleA rationale test for the number of factors in factor analysis. Psychometrika 30 179–185
RS Longman AA Cota RR Holden et al. (1989) ArticleTitleA regression equation for the parallel analysis criterion in principal component analysis: Mean and 95th percentile eigenvalues. Multivar Behav Res 24 59–69
M Kendall A Stuart (1976) The Advanced Theory of Statistics EditionNumber3 Griffin London
B Efron R Tibshirani (1986) ArticleTitleBootstrap methods for standard errors, confidence intervals and other methods of statistical accuracy. Stat Sci 1 54–75
JJ Daudin C Duby P Trecourt (1988) ArticleTitleStability of principal component analysis studied by the bootstrap method. Statistics 19 241–258
ZV Lambert AR Widlt RM Durand (1990) ArticleTitleAssessing sampling variation relative to number-of-factors criteria. Educ Psychol Meas 50 33–48
WJ Krzanowski (1987) ArticleTitleCross-validation in principal component analysis. Biometrics 43 575–584
S Wold (1978) ArticleTitleCross-validatory estimation of the number of components in factor and principal component models. Technometrics 20 397–405
WF Velicer CF Eaton JL Fava (2000) Construct explication through factor or component analysis DN Jackson RG Goffin E Helmes (Eds) Problems and Solutions in Human Assessment–Honoring Douglas N. Jackson at Seventy Kluwer Academic Publishers Boston 41–71
LG Humphreys RG Montanelli (1975) ArticleTitleAn investigation of the parallel analysis criterion for determining the number of common factors. Multivar Behav Res 10 193–206
JR King DA Jackson (1999) ArticleTitleVariable selection in large environmental data sets using principal components analysis. Environmetrics 10 67–77
RJ Linn (1968) ArticleTitleA Monte Carlo approach to the number of factors problem. Psychometrika 83 37–71
RB Cattell S Vogelmann (1977) ArticleTitleA comprehensive trial of the scree and KG criteria for determining the number of factors. Multivar Behav Res 12 289–325
WR Zwick WF Velicer (1982) ArticleTitleFactors influencing four rules for determining the number of components to retain. Multivar Behav Res 17 253–269
R Zwick F Velicer (1986) ArticleTitleComparison of five rules for determining the number of components to retain. Psychol Bull 99 432–442
L Ferre (1995) ArticleTitleSelection of components in principal component analysis. A comparison of methods Comput Stat Data Anal 19 669–682
JE Ware ParticleJr CD Sherbourne (1992) ArticleTitleThe MOS 36-item short-form health survey (SF-36). I. Conceptual framework and item selection Med Care 30 473–483 Occurrence Handle1593914
SD Keller JE Ware ParticleJr PM Bentler et al. (1998) ArticleTitleUse of structural equation modeling to test the construct validity of the SF-36 Health Survey in ten countries: Results from the IQOLA Project. J Clin Epidemiol 51 1179–1188
JE Ware ParticleJr M Kosinski B Gandek et al. (1998) ArticleTitleThe factor structure of the SF-36 Health Survey in 10 countries: Results from the IQOLA Project. J Clin Epidemiol 51 1159–1165 Occurrence Handle10.1016/S0895-4356(98)00107-3 Occurrence Handle9817133
A Leplege E Ecosse A Verdier et al. (1998) ArticleTitleThe French SF-36 Health Survey: Translation, cultural adaptation and preliminary psychometric evaluation. J Clin Epidemiol 51 1013–1023
JR Reddon (1985) ArticleTitleMAPF and MAPS: Subroutines for the number of principal components Appl Psychol Meas 9 97
RL Gorsuch (1974) Factor Analysis. Saunders Philadelphia
InstitutionalAuthorNameSAS Institute Inc (1990) SAS User’s Guide Statistics Version 6 ed. SAS Institute Inc. Cary, North Carolina
PM Bentler P Dudgeon (1996) ArticleTitleCovariance structure analysis: Statistical practice, theory and direction. Ann Rev Psychol 47 541–570
PM Bentler (1990) ArticleTitleComparative fit indexes in structural models. Psychol Bull 107 238–246 Occurrence Handle10.1037//0033-2909.107.2.238 Occurrence Handle1:STN:280:By%2BB3cjls1Q%3D Occurrence Handle2320703
InstitutionalAuthorNameSAS Institute Inc (1992) SAS Technical Report P-229 (Release 6.07). SAS Institute Inc. Cary, North Carolina
WF Velicer DN Jackson (1990) ArticleTitleComponent analysis versus common factor analysis: Some issues in selecting an appropriate procedure. Multivar Behav Res 25 1–28
RL Gorsuch (1990) ArticleTitleCommon factor analysis versus component analysis: Some well and little known facts. Multivar Behav Res 25 33–39
PM Bentler JA Stein (1992) ArticleTitleStructural equation models in medical research. Stat Meth Med Res 1 27–55
KA Bollen (1989) Structural Equations with Latent Variables. Wiley New York
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Coste, J., Bouée, S., Ecosse, E. et al. Methodological issues in determining the dimensionality of composite health measures using principal component analysis: Case illustration and suggestions for practice. Qual Life Res 14, 641–654 (2005). https://doi.org/10.1007/s11136-004-1260-6
Accepted:
Issue Date:
DOI: https://doi.org/10.1007/s11136-004-1260-6