Edukacja 2015, 3(134), 48-62 ISSN 0239-6858 R isk taking can be perceived as environ- mental adaptation. People successfully adapt by systematically taking certain risks whilst avoiding others (Byrnes, 1998). From such a perspective, test-taking strategy can be conceptualised as an adaptation to a test- ing situation. In the test situation, student knowledge and problem solving skills are pitted against scoring rules, which may either penalise them or not for guessing and item omission. For example, if the scoring rules penalise an incorrect answer, a stu- dent might be less likely to take the risk of providing an uncertain answer. However, if the student can make an educated guess, the risk of providing an incorrect answer is reduced and might be considered as worth taking. Omitting test items may be consid- ered as an indicator of the tendency to avoid this risk. When a student has insufficient knowledge and skills, and nevertheless answers an item, while this answer turns © Educational Research Institute * Address: ul. Górczewska 8, 01-180 Warszawa, Poland. E-mail: k.swist @ibe.edu.pl Sex differences in guessing and item omission Karolina Świst, Paulina Skórska, Maciej Koniewski, Aleksandra Jasińska-Maciążek Educational Research Institute* Guessing and item omission may be regarded as risk-taking or risk-avoidance strategies – sex speciic adap- tations to testing situations. In this article, these phenomena were analysed by (a) percentage of omissions by sex, (b) negative binomial regression to asses sex diferences in the number of omissions, (c) c-DIF analysis using IRT-LR test and (d) linear regression using item attributes, to assess whether the c-parameter is sex diferentiated by the percentage of omits (controlling item diiculty). he data set analysed were from the 2012–2014 Polish lower-secondary schools inal exams, comprising tests in maths, language, science and humanities. Contrary to the vast body of literature, boys omitted items slightly more frequently than girls. Possible explanations of this inding – speciic to the Polish examination system – were provided. he hypothesis of a higher c-parameter for boys did not ind strong support from this study. It was shown that the c-parameter should not only be interpreted as resulting from item non-omission. his supports the modern concept of the c-parameter as a consequence not only of random guessing, but also problem solv- ing, creative guessing or cheating. Keywords: sex diferences, guessing, item omission, c-DIF. The article is an extended version of the presentation “Gender differences in pseudo(guessing) and item omission”, presented by the authors at the EduMetric 2014 International Seminar on Educational Research and Measurement, Cracow, December 6–8. The seminar was organised by the Jagiellonian University, as part of the system level project “Quality and effectiveness of educa- tion – strengthening institutional research capabilities” carried out by the Educational Research Institute and co- -financed by the European Social Fund (Human Capital Operational Programme 2007–2013, Priority III High quality of the education system).