Author(s) This pre-registration is currently anonymous to enable blind peer-review. It has 3 authors.
Pre-registered on 2019/04/16 01:06 (PT)
1) Have any data been collected for this study already? No, no data have been collected for this study yet.
2) What's the main question being asked or hypothesis being tested in this study? This project focuses on interindividual differences in Machiavellianism, which is defined as a tendency toward cynicism and manipulativeness, and the belief that ends justify means. We hypothesized that some situations are more conducive than others for Machiavellianism to translate into behavior. In particular, Construal Level Theory holds that individuals construe social situations on a concrete level, or an abstract level, and that an abstract construal level triggers values and value-related traits to be more influential on behavior. Against this background, we hypothesized that differences in Machiavellianism produce differences in dishonest monetary behavior when situations are construed abstractly.
3) Describe the key dependent variable(s) specifying how they will be measured. Cheating paradigm: The likelihood of participants winning a bonus for themselves serves as dependent variable on a group level. On individual level this variable can have two outcomes: 0 for no win and 1 for a win.
4) How many and which conditions will participants be assigned to? Independent Variable 1: Construal level (manipulated)
Participants will be randomly assigned to a concrete or abstract construal level condition and asked to either think about why (abstract level) or how (concrete level) they would try to reach an objective of improving and maintaining one’s physical health (Freitas, Gollwitzer, & Trope, 2004).
Independent Variable 2: Machiavellianism (measured, not manipulated)
Participants’ Machiavellianism score will be assessed with the MACH-IV scale (Christie & Geis, 1970).
5) Specify exactly which analyses you will conduct to examine the main question/hypothesis. We will conduct the following analyses:
• A binomial test to investigate whether the observed probability of winning a bonus was higher than expected by chance (.50).
• Furthermore, we will use the RRreg package (Heck & Moshagen, 2017) to estimate the cheating prevalence in our sample while taking into account the baseline probability of winning and the proportion of participants claiming to win a bonus.
• Next, we will use the RRreg package to calculate a logistic regression on probability of winning the bonus (DV) with the following predictors: construal level (IV1, dummy-coded as 0 = abstract construal level and 1 = concrete construal level), Machiavellianism score (IV2, mean-centered), and the interaction between the two predictors.
• If a two-way interaction is found, we will apply a recentering strategy and will recalculate the logistic regression model with the construal level factor coded as 0 = concrete and 1 = abstract, to test whether Machiavellianism predicts the DV under a concrete construal level as well.
6) Describe exactly how outliers will be defined and handled, and your precise rule(s) for excluding observations. Previous research revealed that participants’ age and Machiavellianism are negatively correlated (the older, the less Machiavellian), and variance on Machiavellianism decreases with age. As age is not our variable of interest in this line of research, we will only recruit participants aged < 40, to increase the likelihood for high and well distributed Machiavellianism scores in our sample.
As in our previous studies we will only recruit US-American, male participants, who have not participated in our previous study.
From the sample obtained, we will exclude all participants who did not fully complete the study.
7) How many observations will be collected or what will determine sample size? No need to justify decision, but be precise about exactly how the number will be determined. Based on the effects found in Study 2, a desired power of .80, and an alpha-error of .05, we collected data from 484 participants in Study 3 and plan to collect data from 484 participants in this study, too. Furthermore, to ensure that we will reach this target even when applying our predefined exclusion criteria, we will add 10% to this number and therefore collect data from 532 participants.
Note: Because we are not aware of a power estimation function for the RRreg package, this total N is an approximation. This approximation is derived by calculating power for two alternative ways of data analysis (using correlations), both of which are suggested as biased in the literature, but allow the use of standard power calculations procedures (G*Power).
8) Anything else you would like to pre-register? (e.g., secondary analyses, variables collected for exploratory purposes, unusual analyses planned?) We might exploratorily investigate the correlations between age and Machiavellianism scores.
