TY - JOUR
T1 - Stratification of American hearing aid users by age and audiometric characteristics
T2 - A method for representative sampling
AU - Aronoff, Justin M.
AU - Yoon, Yang Soo
AU - Soli, Sigfrid D.
PY - 2010/6
Y1 - 2010/6
N2 - OBJECTIVES: Stratified sampling plans can increase the accuracy and facilitate the interpretation of a dataset characterizing a large population. However, such sampling plans have found minimal use in hearing aid (HA) research, in part because of a paucity of quantitative data on the characteristics of HA users. The goal of this study was to devise a quantitatively derived stratified sampling plan for HA research, so that such studies will be more representative and generalizable, and the results obtained using this method are more easily reinterpreted as the population changes. DESIGN: Pure-tone average (PTA) and age information were collected for 84,200 HAs acquired in 2006 and 2007. The distribution of PTA and age was quantified for each HA type and for a composite of all HA users. RESULTS: Based on their respective distributions, PTA and age were each divided into three groups, the combination of which defined the stratification plan. The most populous PTA and age group was also subdivided, allowing greater homogeneity within strata. Finally, the percentage of users in each stratum was calculated. CONCLUSIONS: This article provides a stratified sampling plan for HA research, based on a quantitative analysis of the distribution of PTA and age for HA users. Adopting such a sampling plan will make HA research results more representative and generalizable. In addition, data acquired using such plans can be reinterpreted as the HA population changes.
AB - OBJECTIVES: Stratified sampling plans can increase the accuracy and facilitate the interpretation of a dataset characterizing a large population. However, such sampling plans have found minimal use in hearing aid (HA) research, in part because of a paucity of quantitative data on the characteristics of HA users. The goal of this study was to devise a quantitatively derived stratified sampling plan for HA research, so that such studies will be more representative and generalizable, and the results obtained using this method are more easily reinterpreted as the population changes. DESIGN: Pure-tone average (PTA) and age information were collected for 84,200 HAs acquired in 2006 and 2007. The distribution of PTA and age was quantified for each HA type and for a composite of all HA users. RESULTS: Based on their respective distributions, PTA and age were each divided into three groups, the combination of which defined the stratification plan. The most populous PTA and age group was also subdivided, allowing greater homogeneity within strata. Finally, the percentage of users in each stratum was calculated. CONCLUSIONS: This article provides a stratified sampling plan for HA research, based on a quantitative analysis of the distribution of PTA and age for HA users. Adopting such a sampling plan will make HA research results more representative and generalizable. In addition, data acquired using such plans can be reinterpreted as the HA population changes.
UR - http://www.scopus.com/inward/record.url?scp=77951939764&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=77951939764&partnerID=8YFLogxK
U2 - 10.1097/AUD.0b013e3181cda9ee
DO - 10.1097/AUD.0b013e3181cda9ee
M3 - Article
C2 - 20090531
AN - SCOPUS:77951939764
SN - 0196-0202
VL - 31
SP - 401
EP - 406
JO - Ear and hearing
JF - Ear and hearing
IS - 3
ER -