Jain etal. 1 examined knowledgeand beliefsaboutautismspectrumdisorders (ASD) in healthcare professionals in India. Theseauthors used an instrument that wasadapted froman instrument developed by an Israeliteam. 2 Theadapted version excluded theself-efficacy component ofthe originalinstrument because participants were notcomfortable with thesection, particularlywith thesection on self-competence. However, selfefficacy is important because knowledgeand beliefsarelikely to influenceself-efficacy,and discomfort in answering questionsaboutself-efficacy suggests lack ofself-competence. It would have beenmoreimportant to revisethesection to match thecomfortand understanding level of participants than to removethesection fromtheadapted version. Further, theauthors replaced theself-efficacy sectionwith a 26-itemsection testing knowledgeabout DSM-5 criteriafor ASD. DSM-5 separates social deficitcriteriafromrepetitive behaviorcriteria;so, the 26-itemsection ofthestudy instrument was not unidimensional. Sinceinternal consistency asa measure ofreliability ofan instrumentassumes unidimensionality, when theassumption is violated, reliabilitywill below. This is the likely reasonwhy theauthors obtained alowCronbach’salphafor thesection. It would have beenmoreappropriatefor theauthors to have examined Cronbach’salphaseparately for thesocial deficit itemsand for therepetitive behavior items. 3.
展开▼
