WebExp Demo Paradigm:   Magnitude Estimation
Go to the text experiment >>
Go to the audio experiment >>

Description

Magnitude estimation a technique standardly applied in psychophysics to measure judgments of sensory stimuli (Stevens 1975). The magnitude estimation procedure requires subjects to estimate the magnitude of physical stimuli by assigning numerical values proportional to the stimulus magnitude they perceive. Highly reliable judgments can be achieved for a whole range of sensory modalities, such as brightness, loudness, or tactile stimulation.

Bard et al. (1996) and Cowart (1997) showed that linguistic judgments can be elicited in the same way as judgments of sensory stimuli. In contrast to the 5- or 7-point scale conventionally used to measure human intuitions, magnitude estimation employs a continuous numerical scale. It provides fine-grained measurements of linguistic acceptability, which are robust enough to yield statistically significant results, while being highly replicable both within and across speakers. Since magnitude estimation provides data on an interval scale, parametric statistics can be used for evaluation.

Magnitude estimation requires subjects to assign numbers to a series of linguistic stimuli proportional to the acceptability they perceive. First, subjects are exposed to a modulus item, to which they assign an arbitrary number. Then, all other stimuli are rated proportional to the modulus, i.e., if a sentence is three times as acceptable as the modulus, it gets three times the modulus number, etc.

References

  • Stevens, S. S. 1975. Psychophysics: Introduction to its Perceptual, Neural, and Social Prospects. New York: John Wiley.
  • Bard, Ellen Gurman, Robertson, Dan, and Sorace, Antonella. 1996. Magnitude Estimation of Linguistic Acceptability. Language 72: 32-68.
  • Cowart, Wayne. 1997. Experimental Syntax: Applying Objective Methods to Sentence Judgments. Thousand Oaks, CA: Sage.

The Experiment

This applet shows a sample Magnitude Estimation experiment, where a reference rating is taken and is reused on subsequent slides. There are two versions, making use of the versions feature of WebExp. There is a single experiment description, but different values and stimuli are imported according to the version specified in the applet tag.

The experiment consists of a practice stage where line lengths must be estimated with respect to a reference rating, followed by an experiment stage where sentence acceptability must be judged. Depending on the version, the sentences are presented as either text or sound.
WebExp Demonstrations 2008 The University of Edinburgh