This component has the materials, data, and analysis script for the ML:5 measures pretest described in the ML:5 Revised Protocol ---------- **Short Version:** We wanted to make sure that the effects of our new manipulation could be picked up by our new Free-Will scale, even among a non-college-student sample. So, we put everything on mTurk. Looks like it worked well enough, though not with the effect sizes picked up in either the original study or in the paper from which we took the revised manipulation. **Long Results (from the Revised Protocol):** In our 200-person mTurk sample, the FAD-Plus Free Will subscale showed good reliability: alpha = 0.84 (95% CI: 0.80-0.87). We did find a main effect of Condition, such that those given the Anti-Free-Will passages scored lower (M = 25.99, SD = 4.53) on the Free Will belief scale than those given the Control passages (M = 27.6, SD = 5.39); F(1, 184) = 6.30, p = 0.013, d = 0.36 (95% CI: 0.07 - 0.64). However, the main effect was qualified by a Condition:# of Studies interaction, such that those who participated in more studies showed the effect more strongly: F(1,184) = 5.66, p = 0.018. For those who had completed more than 300 psychology studies on mTurk (Anti n = 25, Control n = 30), the manipulation was effective: Anti M = 25.52, SD = 5.53; Control M = 29.03, SD = 5.59; F(1, 44) = 4.90, p = 0.032, d = 0.6 [0.04 - 1.15]. Participants with fewer than 300 studies (Anti n = 62, Control n = 82) did not show the effect of the manipulation: Anti M = 26.18, SD = 4.10, Control M = 26.99, SD = 5.21; F(1, 131) = 1.36, p = 0.25. No Condition or Condition-interacted effects were seen in either Positive or Negative Affect, as measured by the PANAS (all F’s < 2, all p’s > 0.15)