[ Contents | Search |Previous | Up ]
From: Ken Logan
Date: 22 Apr 2010
Time: 14:03:30 -0500
Remote Name: 128.227.5.94
Hi Ashley, this is a very good question. The answer I think, first, depends upon the type of rate measurement you are using. Articulation rate (the number of syllables per second in stretches of perceptibly fluent speech) is one way to go. I tend to think of this as a measure of communicative competence, as it tells you something about how the speaker functions when speech production processes seem to be going well. Alternately, speech rate(the number of syllables per second in all types of utterances, including those with disfluent speech)is a measure of performance (i.e., how long it takes to communicate particular intentions). Speech rate tends to be slower than articulation rate, because disfluencies take up time that would otherwise be spent in productive communication. As far as a cut-off for "normal limits" -- I'm not sure there is one right answer to that question. +/- 1 SD would be a relatively strict criterion, +/- 2 SDs would be a relatively lax criterion (i.e., you'd really have to be functioning much differently from normal to be flagged as "atypical"). For clinical reporting, you could just simply report how many standard deviations from the mean the particular score is, and then discuss what percentage of the population would be likely to score below that point. (For research purposes, of course, you'd want something more specific.) One caution about rate date, ...the numbers you get can be influenced significantly by the task. So, be sure that your task closely matches the task upon which your norms were based.