“…Previous research, using a test-level approach, has shown that particularly two basic cognitive abilities are related to students' word problem solving performance: (1) spatial ability, covering the visual-spatial processing domain (Boonen et al, 2013;Hegarty & Kozhevnikov, 1999;Van Garderen, 2006); and (2) reading comprehension, covering the semanticlinguistic processing domain (Bernardo, 1999;Van der Schoot, Bakker Arkema, Horsley, & Van Lieshout, 2009;VileniusTuohimaa, Aunola, & Nurmi, 2008). Based on the increasingly accepted idea that word problem solving taps into both domains (Boonen et al, 2013;Krawec, 2010Krawec, , 2012, we aimed to replicate the findings of prior studies regarding the importance of spatial ability and reading comprehension by using analyses at the item-level rather than at the test-level. In doing so, we were able to overcome the aforementioned limitations regarding test-level analyses and prevent ourselves from falling for the ecological fallacy.…”