Our aim is to construct a general measurement framework for analyzing the effects of measurement errors in multivariate measurement scales. We define a measurement model, which forms the core of the framework. The measurement scales in turn are often produced by methods of multivariate statistical analysis. As a central element of the framework, we introduce a new, general method of estimating the reliability of measurement scales. It is more appropriate than the classical procedures, especially in the context of multivariate analyses. The framework provides methods for various topics related to the quality of measurement, such as assessing the structural validity of the measurement model, estimating the standard errors of measurement, and correcting the predictive validity of a measurement scale for attenuation. A proper estimate of reliability is a requisite in each task. We illustrate the idea of the measurement framework with an example based on real data.

With data on children’s dental state from 1976 to 1993, we evaluated how the size of the high-caries group has changed concurrently with simultaneously decreasing mean numbers of teeth with past or current caries. Information related to all dental check-ups done for 5- and 15-year-olds in Helsinki, some 4,000 subjects of each age by year, consisted of numbers of teeth with caries experience (dmft or DMFT) and of all decayed teeth (dt+DT). Polarization of dental caries was described as the proportion of high-caries groups in each year, both in terms of caries experience and current untreated caries, diagnosed at subjects’ annual clinical dental check-ups. For 5-year-olds, the high-caries group by caries experience included patients with their dmft ≥ 3. For 15-year-olds the limits were set at DMFT ≥ 6 and DMFT ≥ 15. The high-caries group in terms of untreated caries was similar for both age groups: dt+DT ≥ 3. Furthermore, polarization of caries was calculated as the share of numbers of both dmf or DMF teeth and dt+DT in each high-caries group of the total number of such teeth in the entire age cohort. During the 17 years, mean dmft for 5-year-olds decreased from 4.6 to 0.8 and mean dt+DT from 0.9 to 0.6. In 1993, 78% had their dmft = 0, whereas only 8% of the patients accounted for 76% of all decayed teeth. For 15-year-olds the decrease was even greater: their DMFT fell from 12.1 to 3.0 and their dt+DT from 3.1 to 0.8. However, only 26% had their DMFT = 0 in 1993, with 55% of all dt+DT occurring in 10% of the patients. The present results confirmed a strong polarization in caries for both age cohorts, showing the need for renewed strategies in preventive caries treatment.

A vocabulary for the description of the sensory properties of Finnish rye breads was developed as a basis for the design of a practical quality control/product development system to be used in a later study.Six samples, selected from thirty-two commercial rye breads, were evaluated for the intensity of twenty-nine attributes and for similarity between sample pairs. Significant differences were found between samples in the intensity of twenty-eight of these descriptors. The data were combined by a multidimensional scaling procedure (KYST 2A and PROFIT), which revealed three discriminative dimensions among the samples:(1) from pleasant rye-like to unpleasant flour-like; (2) from sweet to bitter; and (3) the presence/absence of a characteristic flavour caused by specific baking conditions (low temperature, long baking time).The PROFIT procedure identified 20 attributes as important for quality of rye bread samples using multiple correlation coefficients. There were five appearance attributes, two odour attributes, six taste attributes, five texture attributes and overall impression of pleasantness and freshness.

No abstract