For example, if a program to reduce stress and improve health after a long career is instituted shortly after academy graduation, a randomized sample may not yield results sufficient to measure program outcomes. What if several officers leave the force before retirement or a significant number of years? As noted earlier in research paper by Dumville et al. (2006), attrition is a major concern of randomized trials, with bias being a likely outcome. A before-and-after design may be more appropriate for this type of program. More specifically, a one-group time-series design may reduce the possibility for selection bias and gives evaluators the ability to review data from both before and after the program was implemented (Vito & Higgins, 2015). What are your thoughts here on this topic and the best methodology that may be implemented for measurement and why