Premium
The Analysis of an Unbalanced Paired Comparison Experiment by Multiple Regression
Author(s) -
Johnston B.,
Schwartz C.
Publication year - 1977
Publication title -
journal of the royal statistical society: series c (applied statistics)
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.205
H-Index - 72
eISSN - 1467-9876
pISSN - 0035-9254
DOI - 10.2307/2347020
Subject(s) - statistics , regression analysis , mathematics , regression , econometrics
Summary A method of analysing paired comparisons is presented for an experiment in which panellists test two samples and numerically evaluate the difference. The model is first developed for a balanced experiment and it is extended to include other multi‐classification variables. The model is converted into a multiple regression equation which can be analysed with unequal numbers of observations per treatment pair and with missing pairs. The coding procedure for the regression variables allows one to estimate important treatment differences and their variances, to test various hypotheses and to observe any confounding of effects arising from the unbalanced experiment.