**We will be conducting the study on MTurk with American participants and in our lab in the Netherlands with Dutch participants.** Notes for MTurk Replication =========================== Changes to methods from original study -------------------------------------- - The original study was conducted on computers in a lab. This version will be conducted on computers on the internet. - The original wording was `"Now, please answer the following questions about the room's environment."` - The new wording is `"Now, please answer the following questions about the environment of the room you are currently in."` - The authors were not clear about the exact contents of the filler measures. They provided two examples and so these two examples are included in the study. - Compared to the demographics included in the original study, I added a few choices to the ethnicity measure and measures of income, education, currently religiosity, religiosity growing up, and liberal/conservative identification. - Also included a measure of the importance of being moral to test as a possible moderator. This measure (and the demographics described above) will come after the IV and primary DVs. - Because much of the two studies are identical (IVs, filler tasks, mood measures, demographics) we will run the study as if it were one study. Participants will be randomly assigned to the ethical vs. unethical recall manipulation, complete the filler tasks, be randomly assigned to Study 1 vs. Study 2 DVs, complete a mood measure, and finally complete the demographic information (including a question about the importance of being a moral person). Power Analyses (conducted with G*Power) --------------------------------------- - Because the environments people experience on MTurk are far from standardized (and thus have greater error) we will assume that the effect sizes are .75 times their original size. We should note that this is a hunch. - Study 1, d = .65, .65 * .75 = .4875, For 95% power n = 184. - Study 2, (multiple DVs, power based on smallest d), d = .64, .64 * .75 = .48, For 95% power n = 190. - Data will be collected until we have enough participants in the study. The aim will be 220 in each study to allow for the possibility of missing data and other forms of incomplete responses. Notes for Dutch Lab Replication =========================== Changes to methods from original study -------------------------------------- - The original study was conducted in English. This version will be conducted in Dutch. The materials were back translated by two research assistants at Tilburg University who are fluent in both Dutch and English. - The study will be conducted on computers in the lab. Each participant in an individual cubicle. The study will be presented via Qualtrics. - Compared to the demographics included in the original study and the MTurk replication, the demographic measures were altered to reflect that this study is for Dutch participants. For example, the ethnicity measures and the Liberal-Conservative measure were changed for this sample. - We will run each study as a separate study for the lab (instead of randomly assigning people to Study 1 DVs or Study 2 DVs). This is because it will be difficult to have enough statistical power for the two combined studies with our lab setup, whereas it will be relatively straightforward to have enough statistical power for each individual study. - The other changes to the original study described above, were also instituted in this version. - UPDATE Feb 13, 2013: We slightly changed the experimental manipulation of the ethical/unethical deed so that participants are asked to really think about the deed. This makes the instructions more explicit and will make the prime more salient to participants. The final materials for the Dutch lab studies can are the files Is_it_light_or_dark_-_Dutch_Lab_-_Study_1.docx and Is_it_light_or_dark_-_Dutch_Lab_-_Study_2.docx Power Analyses (conducted with G*Power) --------------------------------------- - We based the power analyses on the effect sizes from the original study. - Study 1, d = .65, For 95% power n = 126. - Study 2, (multiple DVs, power based on smallest d), d = .64, For 95% power n = 130. - Per lab norms, data will be collected for one week. The typical N from one week of data collection is between 120 and 150, so this should be adequate for 95% Power. If we do not collect data from enough participants we will run the study a second time in a different week.