| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
1 Central Institute for the Deaf St. Louis, MO
This study was conducted to obtain information on test-retest differences in Articulation Indices (AI) when a clinical threshold-based method is used to calculate AI. The AI was calculated using a nine-frequency-band method on 2 separate days for 209 ears. Standard deviations of test-retest differences were calculated. The standard deviations were greatest when the AI was near 50% and least when the AI was near 0% or 100%. The standard deviations for test-retest AI differences were also dependent on the relationship of the hearing thresholds and speech spectrum. Under the assumption that each dB change in threshold would have an affect on the AI, standard deviations of test-retest differences increased significantly. Test-retest standard deviations for nine-band and five-band methods were approximately equal when only five thresholds were actually measured. A small improvement would be expected if the number of threshold measurements increased from five to nine.
Key Words: Articulation Index hearing thresholds test-retest variability
Submitted on December 27, 1988
Accepted on July 11, 1989
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| All ASHA Journals | AJA | AJSLP | JSLHR | LSHSS |