In conventional canonical analysis, canonical parameters such as weights and loadings, which typically are not tested individually for statistical significance, are commonly used for interpreting the contributions of particular predictor and criterion variables to interset relationships. As an alternative to the traditional use of arbitrary cutoff values for such interpretations, the generalized jackknife statistic may be used to give estimated standard errors for canonical parameters, thereby enabling these parameters to be tested for statistical significance. The authors (a) address major scaling considerations encountered in jackknifing canonical parameters, (b) report experience in using the jackknife statistic to test the significance of canonical weights, loadings, and cross-loadings, (c) explore effects of different omitted subgroup sizes and subsamples, and (d) review advantages and limitations of jackknifing canonical parameters.
Get full access to this article
View all access options for this article.
References
1.
AlpertMark I., and PetersonRobert A. (1972), “On the Interpretation of Canonical Analysis,”Journal of Marketing Research, 9(May), 187–92.
2.
BarcikowskiRobert S., and StevensJames P. (1975), “A Monte Carlo Study of the Stability of Canonical Correlations, Canonical Weights and Canonical Variate-Variable Correlations,”Multivariate Behavorial Research, 10(July), 353–64.
3.
CattellRaymond B. (1978), The Scientific Use of Factor Analysis in Behavioral and Life Sciences.New York: Plenum Press.
4.
ClarksonDouglas B. (1979), “Estimating the Standard Errors of Rotated Factor Loadings by Jackknifing,”Psychometrika, 44(September), 297–314.
5.
CliffNorman, and KrusDavid J. (1976), “Interpretation of Canonical Analysis: Rotated Vs. Unrotated Solutions,”Psychometrika, 41(March), 35–42.
6.
CooleyWilliam W., and LohnesPaul R. (1971), Multivariate Data Analysis.New York: John Wiley & Sons, Inc.
7.
CraskMelvin R., and PerreaultWilliam D.Jr. (1977), “Validation of Discriminant Analysis in Marketing Research,”Journal of Marketing Research, 14(February), 60–8.
8.
FenwickIan (1979), “Techniques in Market Measurement: The Jackknife,”Journal of Marketing Research, 16(August), 410–4.
9.
GrayH. L., and SchucanyW. R. (1972), The Generalized Jackknife Statistic.New York: Marcel Dekker, Inc.
10.
IrelandM. Edwin, and UseltonGene G. (1976), “A Distribution-Free Statistic for Management Scientists,” in Eleventh Annual Meeting, Southeastern Chapter, The Institute of Management Sciences, MenkeW. W., and WhitehurstC. H.Jr., eds. (October), 25–6.
11.
JohanssonJohny K., and ShethJagdish N. (1977), “Canonical Correlation and Marketing Research,” in Multivariate Methods for Market and Survey Research, ShethJagdish N., ed. Chicago: American Marketing Association, 111–31.
12.
JöreskogK. G. (1973), “Analysis of Covariance Structures,” in Multivariate Analysis-III, KrishnaiahParuchuri R., ed. New York: Academic Press 1973, 263–85.
13.
LambertZarrel V., and DurandRichard M. (1975), “Some Precautions in Using Canonical Analysis,”Journal of Marketing Research, 12(November), 468–75.
14.
LeeS. Y., and JennrichR. I. (1979), “A Study of Algorithms for Covariance Structure Analysis with Specific Comparisons Using Factor Analysis,”Psychometrika, 44(March), 99–113.
15.
MarriottF. H. C. (1974), The Interpretation of Multiple Observations.New York: Academic Press.
16.
MorrisonDonald F. (1976), Multivariate Statistical Methods, 2nd ed. New York: McGraw-Hill Book Company.
17.
MostellerFrederick (1971), “The Jackknife,”Review of the International Statistical Institute, 39, 363–8.
18.
MostellerFrederick, and TukeyJohn W. (1968), “Data Analysis, Including Statistics,” in The Handbook of Social Psychology, Vol. 2, LindzeyG., and AronsonE., eds. Reading, MA: Addison-Wesley, 80–203.
19.
QuenouilleM. H. (1956), “Notes on Bias in Estimation,”Biometrika, 43(December), 353–60.
20.
SchaningerCharles M., and Parker LessigV., and PantonDon B. (1980), “The Complementary Use of Multivariate Procedures to Investigate Nonlinear and Interactive Relationships Between Personality and Product Usage,”Journal of Marketing Research, 17(February), 119–24.
21.
SparksDavid L., and TuckerW. T. (1971), “A Multivariate Analysis of Personality and Product Use,”Journal of Marketing Research, 8(February), 67–70.
22.
TatsuokaMaurice M. (1971), Multivariate Research: Techniques for Educational and Psychological Research.New York: John Wiley & Sons, Inc.
23.
ThorndikeRobert M. (1976), “Studying Canonical Analysis: Comments on Barcikowski and Stevens,”Multivariate Behavorial Research, 11(April), 249–53.
24.
ToussaintGodfried T. (1974), “Bibliography on Estimation of Misclassification,”IEEE Transactions on Information Theory, IT-20 (July), 472–9.
25.
TukeyJohn W. (1958), “Bias and Confidence in Not-Quite Large Samples,”Annals of Mathematical Statistics, 29(June), 614.
26.
van den WollenbergArnold L. (1977), “Redundancy Analysis: An Alternative for Canonical Correlation Analysis,”Psychometrika, 42(June), 207–19.
