In which scenario below would a standardized effect size (e.g. Cohen's d, Pearson's correlation coefficient) be most preferable to a unit-specific effect size (i.e. the effect expressed in original units of measure)? Question 25 options: A researcher reports the effects of an exercise intervention on depression using a highly used and well-known depression scale. A researcher reports predicted weight loss (in kilograms) by number of hours exercised per week. A researcher compares effects of exercise on depression from several different studies, many of which use different measurement scales. A researcher compares children's growth (in centimeters) by treatment to which they were randomized in a clinical trial (growth delay being a possible side effect of some treatments)