The current study evaluated the reliability and validity of the Arabic version of the revised general nutrition knowledge questionnaire (GNKQ-R) for adults. The eighty-eight-item English GNKQ-R was adapted into an eighty-six-item Arabic version. Four validation studies were conducted for internal (n 805) and external (n 106) reliability, construct validity between participants with (n 84) and without (n 88) nutrition background, convergent validity for associations between nutrition knowledge and demographic characteristics (n 750) and responsiveness to online nutrition information (n 55). United Arab Emirates University in United Arab Emirates and Hashemite University in Jordan. Undergraduate students aged 18 years and above, enrolled in any programme at the two universities, were recruited. Overall, internal reliability (Cronbach's α = 0·91) and external reliability (P = 0·350; intra-class correlation coefficient = 0·84) were high. Significantly higher GNKQ-R scores of students with (66·0 (10·6)) v. without (38·0 (10·7), P < 0·001; d = 2·6) nutrition background indicated high construct validity. Significantly higher GNKQ-R scores among females v. males, older and senior students v. younger and junior students and students in health discipline v. non-health discipline reflected good convergent validity. Significant differences in GNKQ-R scores with nutrition information (time 1 = 37·8 (10·5) and time 2 = 47·7 (9·1), P < 0·001; d = 1·0) indicated high responsiveness to nutrition intervention. The Arabic GNKQ-R showed high reliability and validity in the young adult Arab population. Besides the reliability of the overall questionnaire, each section demonstrated adequate reliability. Further studies are warranted to establish the generalisability and applicability of the Arabic GNKQ-R in older adults and in different middle-eastern Arab countries.