Biomedical subjects

M I Messinger

Publications and source records attributed to M I Messinger.

4 recordsLinked to original sources

PubMed1996

Factoring handedness data: II. Geschwind's multidimensional hypothesis.

The challenge in this journal by Peters and Murphy to the validity of two published factor analyses of handedness data because of bimodality was dealt with in Part I by identifying measures to normalize the handedness item distributions. A new survey using Oldfield's questionnaire format had 38 bell-shaped (unimodal) handedness-item distributions and 11 that were only marginally bimodal out of the 55 items used in Geschwind's 1986 study. Yet they were still non-normal and the factor analysis was unsatisfactory; bimodality is not the only problem. By choosing a transformation for each item that was optimal as assessed by D'Agostino's K2 statistic, all but two items could be normalized. Seven factors were derived that showed high congruence between maximum likelihood and principal components extractions before and after varimax rotation. Geschwind's assertion that handedness is not unidimensional is therefore supported.

Analysis of Variance↗

PubMed1995

Factoring handedness data: I. Item analysis.

Recently in this journal Peters and Murphy challenged the validity of factor analyses done on bimodal handedness data, suggesting instead that right- and left-handers be studied separately. But bimodality may be avoidable if attention is paid to Oldfield's questionnaire format and instructions for the subjects. Two characteristics appear crucial: a two-column LEFT-RIGHT format for the body of the instrument and what we call Oldfield's Admonition: not to indicate strong preference for handedness item, such as write, unless "... the preference is so strong that you would never try to use the other hand unless absolutely forced to...". Attaining unimodality of an item distribution would seem to overcome the objections of Peters and Murphy. In a 1984 survey in Boston we used Oldfield's ten-item questionnaire exactly as published. This produced unimodal item distributions. With reflection of the five-point item scale and a logarithmic transformation, we achieved a degree of normalization for the items. Two surveys elsewhere based on Oldfield's 20-item list but with changes in the questionnaire format and the instructions, yielded markedly different item distributions with peaks at each extreme and sometimes in the middle as well.

Australia↗

PubMed1994

Handedness and headache.

A California handedness study involved 199 male and 74 female cluster headache sufferers as well as 477 migraineurs. Earlier reported data indicated that cluster headache patients had a left-handedness prevalence of over 15%. With the current data the prevalence in males was 11.0% and in females, 8.2%. The corresponding migraine figures were 11.8% and 8.1%. The cluster and migraine headache groups did not differ significantly from each other or from the expected 10% frequency of left-handedness in either sex.

Adult↗

PubMed1988

Migraine and left-handedness: is there a connection?

In 1982, Geschwind and Behan reported an association between migraine headache and left-handedness. The present study was an attempt to test this hypothesis by comparing the frequency of left-handedness in migraine and tension headache patients at a headache center. Cluster headache cases were also included because Geschwind and Behan suspected that the association might be even stronger in this disorder. A special scoring method for handedness was devised by Geschwind and Behan to help identify a possible higher risk in mixed-handedness subjects. No significant associations emerged in any of these tests. A modest association of cluster headache and left-handedness disappeared when adjustment was made for the strong predilection of cluster headache for the male sex.

Adult↗