“…As in earlier studies which selected suspect firms, i.e., ROA from zero to 0.01 (Ugrin, Mason, & Emley, 2017) and from zero to 0.005 (Roychowdhury, 2006;Yuliana, Anshori, & Alim, 2015), firms with one or more years of negative ROA were omitted. Then, the average ROA for each firm was calculated (ROA for 2013, 2014, and 2015, divided by 3) and placed in ascending order to identify the 300 firms with lower than average ROA (Al-Absy, Ismail, & Chandren, 2021;Al-Absy et al, 2020;Al-Absy, Ku Ismail, & Chandren, 2018, 2019a, 2019b, 2019c, 2019d. 18 firms were eliminated from the sample because of incomplete data, leading to a final sampling size of 282 firms, or 846 firm-observations over the three years.…”