Locker David, Berka Eva, Jokovic Aleksandra, Tompson Bryan
Faculty of Dentistry, University of Toronto, Toronto, Ontario, Canada.
Community Dent Oral Epidemiol. 2007 Feb;35(1):35-43. doi: 10.1111/j.1600-0528.2007.00317.x.
To determine if self-weighting of the items in an oral health-related quality of life questionnaire improves its psychometric properties.
The Surgical Orthodontic Outcome Questionnaire (SOOQ) was designed to assess the oral health-related quality of life of individuals before and after corrective surgery. Each of its 33 'items' consists of two questions: a question concerning the frequency with which a given functional or psychosocial problem had been experienced and a 'weighting' question which asked about how much the individual was bothered by that problem. The questionnaire was completed by three groups of individuals: (i) pretreatment; (ii) immediate (i.e. 2-6 months) postsurgery and (iii) postsurgery (i.e. more than 2 years after surgery). Unweighted scale scores were obtained by summing the response codes to the frequency question and weighted scores by summing the products of the frequency and bother questions. These scores were calculated for the full questionnaire and a short form consisting of 15 items. The discriminative and correlational construct validity of these scores was compared along with internal consistency reliability. The sensitivity to change and longitudinal construct validity of unweighted and weighted scores was assessed in a simulated evaluative study in which pretreatment and postsurgery subjects were paired.
For both the long and short forms of the questionnaire, unweighted and weighted scores discriminated between the groups enrolled in the study. Correlations with a general health rating were similar, as were Cronbach's alpha values and test-retest reliabilities. The simulated evaluative study suggested no differences in sensitivity to change or longitudinal construct validity. When subscale scores were examined, there was a suggestion that weighting improved their reliability.
Self-weighting of items did not substantially improve the performance of the SOOQ. Domain weights should be developed and tested to determine if they have an effect on its properties.
确定口腔健康相关生活质量问卷中项目的自我加权是否能改善其心理测量特性。
外科正畸结果问卷（SOOQ）旨在评估个体在矫正手术前后与口腔健康相关的生活质量。其33个“项目”中的每一个都由两个问题组成：一个关于特定功能或心理社会问题出现频率的问题，以及一个询问个体受该问题困扰程度的“加权”问题。三组个体完成了该问卷：（i）术前；（ii）术后即刻（即2 - 6个月）；（iii）术后（即手术超过2年后）。未加权量表得分通过对频率问题的回答代码求和获得，加权得分通过对频率问题和困扰问题的乘积求和获得。对完整问卷和由15个项目组成的简表计算这些得分。比较了这些得分的区分性和相关性结构效度以及内部一致性信度。在一项模拟评估研究中，将术前和术后受试者配对，评估未加权和加权得分对变化的敏感性和纵向结构效度。
对于问卷的长表和简表，未加权和加权得分均能区分参与研究的各组。与总体健康评级的相关性相似，克朗巴哈α值和重测信度也相似。模拟评估研究表明，在对变化的敏感性或纵向结构效度方面没有差异。检查分量表得分时，有迹象表明加权提高了其信度。
项目的自我加权并未显著改善SOOQ的性能。应开发并测试领域权重，以确定它们是否对其特性有影响。
