Randall R. Robey

Decisions in Single Group Repeated Measures Analysis: Statistical Tests and Three Computer Packages

Abstract Repeated measures analysis is a widely used tool of applied statisticians. However, its users are confronted with what seems to be a myriad of decisions, even in its simple application. Should a univariate or a multivariate mode be used? Should an adjusted univariate test be used, and if so, which one? Should a pooled or partitioned error term be used? How important is the sphericity test? Given answers to these questions, which of three major statistical packages, and which programs within a package, could be used? These questions are discussed and resolved in this presentation.

Changes in Lingual Sensitivity as a Function of Age and Stimulus Exposure Time

Effects of duration of stimulus exposure on lingual vibrotactile thresholds were examined across three groups of 10 subjects each (n = 30). Subjects were grouped according to age (child group, mean age = 10.1 yr.; young adult group, mean age = 21.9 yr.; elderly group, mean age = 76.0 yr.). Lingual vibrotactile threshold measurements were obtained for all subjects under 5 conditions of exposure (1, 2, 3, 4, and 5 sec.). Results showed statistically significant differences in threshold among all three age groups. As age increased, thresholds of lingual sensitivity increased (became poorer). Stimulus duration also created significant differences in threshold for all age groups. As stimulus duration increased, thresholds of lingual sensitivity decreased (became better). The children appeared to be the most stable across conditions whereas the elderly group appeared to be the most affected by stimulus duration.

AUDITORY MASKING EFFECTS ON LINGUAL VIBROTACTILE THRESHOLDS AS A FUNCTION OF AGE

Auditory masking effects on lingual vibrotactile thresholds were examined across three groups of 10 subjects each. The first group included children 13 yr. of age or younger, the second group young college age adults, and the third group elderly individuals 66 yr. of age or older. Lingual vibrotactile thresholds were obtained from each group at a frequency of 250 Hz, under experimental conditions of no-masking, narrow-band masking, and wide-band masking. Results showed statistically significant threshold differences between the elderly group and the other two groups for all three experimental conditions. A difference was also detected between experimental conditions when an over-all statistical analysis was performed, but this difference disappeared under more stringent post hoc examination. Results are discussed with respect to current literature on aging and appropriate simplified procedures are recommended for future lingual vibrotactile threshold testing.

The Phonetic Structure of Errors in the Perception of Fluent Speech

Publisher Summary This chapter describes the focal errors and presents the general observations concerning the implications of the data. The limitations of uncontrolled sampling from naturalistic observation are well known. Yet it is believed that these data are valuable because they provide information about language comprehension in the most natural of situations—ordinary conversation. The classification scheme, primarily based on traditional phonetic categories, is well suited for characterizing the data. The phonetic categories might be interpreted in terms of either acoustic parameters or a system of distinctive features. However, it is interpreted that the phonetic categories provide an appropriate system for dealing with speech errors. Furthermore, although the belief is not tested, the impression is that the errors tend to be made along dimensions of phonetic similarity. The perceptual errors of children and of adults are very similar. For the most part, therefore, perceptual strategies of children should be relatively similar to those of adults. There are two possible exceptions to this generalization. Children make few errors that involve compensating for phonological reductions—errors that lead to perceived, though not produced consonants predictable from English phonology.

Calculating the Statistical Power of the Univariate and the Multivariate Repeated Measures Analyses of Variance for the Single Group Case Under Various Conditions

Repeated measures designs find frequent application in behavioral research. Researchers using such designs must choose between the univariate approach and the multivariate approach for a test statistic. The statistical power of these two approaches has been shown to be varyingly different depending upon the structure of the variance-covariance matrix relative to the pattern of mean differences. However, the guidelines for choosing the most powerful test for a given situation have not been well defined. A FORTRAN program is described which calculates the statistical power of the omnibus tests for the univariate and the multivariate approaches to repeated measures in the single group case. Program input, output, applications, and availability are discussed.

Comparison of midline and off-midline lingual vibrotactile threshold responses in men and women

Lingual vibrotactile thresholds were obtained from 10 young adult men and 10 young adult women on three different lingual test sites at a frequency of 250 Hz. The lingual test sites chosen for study were: (1) the dorsal midline of the anterior portion of the tongue, (2) a point halfway between the dorsal midline and the right upper lateral border of the tongue, and (3) a point halfway between the dorsal midline and the left upper lateral border of the tongue. Results showed that thresholds obtained on the dorsal midline of the tongue were not statistically significantly different for the men and women, but thresholds obtained off midline in either direction were statistically significantly different for the two groups of subjects. Compared with the midline thresholds, the off-midline thresholds were seen to become higher for the male subjects, but they were seen to become lower for the female subjects. Implications of these results for future testing are discussed, and further research dealing with off-midline lingual vibrotactile measurement is recommended.

Auditory masking effects on lingual vibrotactile thresholds as a function of age

Auditory masking effects on lingual vibrotactile thresholds were examined across three groups of 10 subjects each. The first group included children 13 yr. of age or younger, the second group young college age adults, and the third group elderly individuals 66 yr. of age or older. Lingual vibrotactile thresholds were obtained from each group at a frequency of 250 Hz, under experimental conditions of no-masking, narrow-band masking, and wide-band masking. Results showed statistically significant threshold differences between the elderly group and the other two groups for all three experimental conditions. A difference was also detected between experimental conditions when an over-all statistical analysis was performed, but this difference disappeared under more stringent post hoc examination. Results are discussed with respect to current literature on aging and appropriate simplified procedures are recommended for future lingual vibrotactile threshold testing.