“…However, there is no common, widely accepted research taxonomy or common methods to assess learning outcomes (Feinstein & Cannon, 2002; Gosenpud, 1990). Measuring the positive effects of using such games can be done in the following way: - post-experience surveys of the students (Farrell, 2005; Lainema & Hilmola, 2005; Chang et al, 2009; Tao et al, 2009; Hernández-Lara et al, 2019; Buil et al, 2019),
- post-experience reports or essays written by the students (Siewiorek et al, 2012; Sai, 2017),
- post-experience interviews of the students (Lin and Tu, 2012),
- pre- and post-experience surveys of the students (Grimes & Willey, 1990; Tompson & Dass, 2000; Romme, 2003; Olhager & Persson, 2006; Cousens et al, 2009; Guillén-Nieto & Aleson-Carbonell, 2012; Hwang & Cruthirds, 2017),
- comparing the post-experience of the students with other teaching method experiences (Betts & Knaus, 2006; Cook & Swift, 2006; Kenworthy & Wong, 2005; Li et al, 2007; Arias-Aranda & Bustinza-Sánchez, 2009; Mitchell, 2004),
- measuring student satisfaction with the simulation course (Léger, 2006),
- mail surveys sent to Professors (Pongpanich et al, 2009; Liu et al, 2009; Chang, 2003),
- comparing students’ results, such as final scores or marks to those from traditional teaching method courses (Wellington & Faria, 1991; Tao et al, 2012),
- studying mistakes students made during the simulation game (Pasin & Giroux, 2011).
…”