Resumo

COSTA, Ariela Raissa Lima  e  HAUCK FILHO, Nelson. Methods for the control of extreme response styles in self-report instruments: a review. Temas psicol. [online]. 2019, vol.27, n.2, pp. 309-323. ISSN 1413-389X.  http://dx.doi.org/10.9788/TP2019.2-02.

Response styles are systematic ways of responding to self-report items that may impact the validity and the precision of scores from instruments. One of these biases is extreme responding (ER), which occurs when a person tends to use only extreme rating categories from a response scale (e.g., totally disagree or totally agree), irrespective of item content. Many different methods were developed that aim to identify and control extreme responses to provide a more accurate assessment of an individual's trait. The aim of this study is to perform a systematic review of these main techniques for statistical control of extreme responses in psychometric instruments of self-report. We identified several analytical approaches, which we organized into seven clusters, from simple count of the numbers of extreme response to the use of modern statistics methods, as Item Response Theory uni and multidimensional. Advantages and limitations of each method are discussed. We also present a general diagram that summarizes the distinct available methods we found.

Palavras-chave : Response style; Likert; response bias; measurement.

        · resumo em Português | Espanhol     · texto em Português     · Português ( pdf )