Choosing between micro and macro nonverbal measurement: Application to selected vocalic and kinesic indices

Selecting the appropriate size of measurement unit in nonverbal research requires empirical as well as conceptual analysis. A definition is offered to distinguish macroscopic from microscopic measurement. Merits of each measurement approach are reviewed briefly in terms of cost, efficiency, precision, and analysis flexibility, and four additional criteria are advocated as central considerations in choosing measurement units: (1) isomorphism between the form of measurement and the phenomenological experience of interactants, (2) reliability, (3) concurrent validity, and (4) predictive validity. These criteria are applied to the analysis of 20 vocalic and kinesic nonverbal behaviors measured microscopically and macroscopically. Reliability and validity results demonstrate that either approach may be acceptable for many behaviors, but that relatively molecular coding may be preferable for highly dynamic and "objective" behaviors, while relatively molar coding may be preferable for static behaviors and perceptual judgments.