Adaptation of the knowledge about childhood autism among health workers questionnaire aimed for usage in Turkey

To the Editor,

We would like to read and contribute to the research article titled “Adaptation of the Knowledge about Childhood Autism among Health Workers (KCAHW) Questionnaire aimed for usage in Turkey” written by Özdemir et al. [1] published in North Clinics of Istanbul. The KCAHW is a scale developed by Bakare et al. [2] consisting of four areas and 19 items which measure knowledge level of health workers on autism.

During the scale adaptation process, it was seen that the writers excluded the 1^st item since all participants answered it accurately and the 4^th area as a whole (6 items as a whole) for a reason that is not easily understandable such as to avoid misunderstandings of the physicians about autism spectrum disorder and the scale was reduced to three areas. As a result, the scale consisting of four areas and 19 items were reduced to three areas and 12 items without the presentation of sufficient psychometric reasons.

In the process of developing a new scale, adding or excluding an item in line with analysis results might be in question [3]. However, while doing scale validity and reliability studies, excluding items can cause the loss of certain concepts and thus “the loss of content validity.” A scale which does not have content validity cannot claim to measure anything. Excluding items from a scale can only be possible in psychometric studies only with an exploratory approach and require objective hypothesis and accurate and sufficient psychometric methods (IRT and Differential Item Functioning, etc.). It is apparent that the objective in Özdemir et al.’s study is not to test the content validity of the original scale. In fact, there is a need to develop a new scale rather than test the content validity of a scale which has previously been developed, because items are prepared within the framework of theoretical structure (conceptual content) while developing a scale. In addition, the exclusion of any item requires certain statistical approaches. The approaches used here to exclude items from a scale are insufficient.

In the other cultural adaptation and validity study of the same scale published by Gürbüz Özgür et al. [4], sticking to the original version of the scale, the validity and reliability of the KCAHW, which includes all four areas and all 19 items, without needing any items to be removed, has been demonstrated for Turkish culture and Turkish.

It will be a more scientifically correct approach to use the Turkish version of this scale adapted by sticking to the original, in the field of health services.