St Louis Kenneth O, Williams Mandy J, Ware Mercedes B, Guendouzi Jacqueline, Reichel Isabella K
West Virginia University, Morgantown, WV, USA.
University of South Dakota, Vermillion, SD, USA.
J Commun Disord. 2014 Jul-Aug;50:36-50. doi: 10.1016/j.jcomdis.2014.04.001. Epub 2014 May 5.
In order to estimate instrument validity, attitudes toward stuttering measured by the newly developed Public Opinion Survey of Human Attributes-Stuttering (POSHA-S) and the Woods and Williams (1976) semantic differential scale (referred to herein as the Bipolar Adjective Scale [BAS]) are compared in college students on one occasion as well as before and after coursework on fluency disorders.
Undergraduate and graduate students (n=321) from four universities filled out online versions of the POSHA-S and BAS. Two-thirds were speech-language pathology (SLP) majors; one-third were students in other majors. A subset of the SLP students (n=35) filled out the two instruments again after 8-13 weeks of coursework on fluency disorders.
Correlations between all ratings of the POSHA-S and BAS were run for the 321 students. Only 26% of the correlations were statistically significant (R ≥ ± 0.129), and the large majority of these reflected small relationships. POSHA-S ratings were correlated with up to 77% of the items of the BAS while BAS items were correlated with up to 45% of the POSHA-S ratings. After coursework on stuttering, students' attitudes improved on both instruments, but more on the POSHA-S than the BAS.
Greater evidence of discriminant validity than convergent validity characterized the POSHA-S and BAS. Both measures showed improved attitudes after fluency disorders coursework, but more so for the POSHA-S, confirming previous reports of construct validity. The POSHA-S taps relevant constructs not included in the BAS, which provide advantages for intracultural, international, and other comparisons of public attitudes toward stuttering.
The reader will be able to: (1) describe differentiating characteristics of the POSHA-S and BAS as measures of public attitudes toward stuttering, (2) describe the overlap and lack of overlap in the constructs measured by POSHA-S and BAS, (3) describe discriminant versus convergent validity and (4) describe advantages of the POSHA-S and BAS in various types of comparative studies of stuttering attitudes.
为评估工具的有效性，在大学生群体中一次性比较以及在流畅性障碍课程学习前后，比较用新开发的《人类属性关于口吃的民意调查》（POSHA - S）和伍兹与威廉姆斯（1976）语义差异量表（本文中称为双极形容词量表[BAS]）所测量的对口吃的态度。
来自四所大学的本科和研究生（n = 321）填写了POSHA - S和BAS的在线版本。三分之二是言语语言病理学（SLP）专业的学生；三分之一是其他专业的学生。一部分SLP专业的学生（n = 35）在进行了8 - 13周的流畅性障碍课程学习后再次填写了这两种工具。
对321名学生的POSHA - S和BAS的所有评分进行了相关性分析。只有26%的相关性具有统计学意义（R≥±0.129），并且其中绝大多数反映的是小关系。POSHA - S的评分与BAS高达77%的项目相关，而BAS的项目与POSHA - S高达45%的评分相关。在口吃课程学习后，学生在两种工具上的态度都有所改善，但在POSHA - S上的改善比在BAS上更多。
POSHA - S和BAS的区别效度证据比聚合效度证据更显著。两种测量方法在流畅性障碍课程学习后都显示出态度改善，但POSHA - S更明显，这证实了先前关于结构效度的报告。POSHA - S挖掘了BAS中未包含的相关结构，这为文化内部、国际以及其他关于公众对口吃态度的比较提供了优势。
读者将能够：（1）描述POSHA - S和BAS作为公众对口吃态度测量工具的区别特征，（2）描述POSHA - S和BAS所测量结构中的重叠和非重叠部分，（3）描述区别效度与聚合效度，以及（4）描述POSHA - S和BAS在各种口吃态度比较研究中的优势。
