Two independent studies compare a new tutoring method to traditional instruction. Both report statistically significant improvements ($p < 0.05$). However, Study A (with $n = 1200$) reports Cohen's $d = 0.12$ (small effect), while Study B (with $n = 80$) reports Cohen's $d = 0.95$ (very large effect). Which conclusion is most appropriate?