期刊名称：IMPACT : International Journal of Research in Humanities, Arts and Literature
印刷版ISSN：2347-4564
电子版ISSN：2321-8878
出版年度：2017
卷号：5
期号：9
页码：171-184
语种：English
出版社：IMPACT Journals
摘要：The practice of combining both positively and negatively worded statements on a scale is one of the suggested means of reducing acquiescence bias. The literature, however showed mixed results with regard to the outcomes of this practice and called for further validation studies. Based on these premises, this study examined the item wording effect on the factor structure of the Arabic version of the UCLA Loneliness Scale. The participants included samples from six populations in the United Arab Emirates who were differentiated into groups according to sex, age, and marital status. Exploratory Factor Analysis (EFA) was used to determine the underlying structure of the scale and Confirmatory Factor Analysis (CFA) was used to ascertain the validity of the outcomes. The internal reliability of the scale was confirmed in all the three groups of participants and the results showed that the scale reflected two main factors clearly divided by positively and negatively worded items. To avoid the effect of distorting the structure on account of item wording, it is incumbent on the designer of a scale to prudently determine the necessity of using negatively worded items in consideration of the context of the research and the evaluation setting.
