Replication of Stellar, Gordon, Anderson, Piff, McNeil, & Keltner (2018; Study 3)

We are completing a direct replication of Awe and Humility (Stellar, et. al) as part of a group project for our Research Methods in Social Psychology course at McGill University. In the initial study they hypothesized, “We predicted that compared with a neutral control condition, momentary experiences of awe will lead to a more balanced disclosure of one’s personal strengths and weakness” which we also will be aiming to falsify with our additional question predicting that there will be a negative relationship between high levels of trait narcissism and behaviourally measured state humility levels after exposure to awe-inducing stimuli as compared to non-narcissistic trait possessing individuals. Replication Overview: The experimental manipulation of awe will be induced by using a standardized video induction. Participants’ state humility is assessed by a behavioral measure: the number of strengths and weaknesses listed, as well as the ratio between the two. Replication Methods The replication methods used in this study are identical to the methods used in the original Stellar et al. study. The participants are randomly assigned by the Qualtrics computer program to one of two conditions where they will watch a video that is two minutes in length. Control: a nonemotional video intended to elicit only feelings of relaxation and calmness in participants. Experimental: a perspective from Earth zooming out into the cosmos, should elicit awe in participants. After the video, participants will write for two minutes about their strengths and weaknesses, starting with their strengths. They are instructed to imagine that they would discuss these strengths and weaknesses with someone they just met. There will be a visible timer counting down from two minutes to zero while they write. Participants are then asked to rank the extent to which they feel happiness, fear, awe, wonder, and amazement from 1 (not at all) to 7 (very much). After the data is collected, two coders who are blind to participant conditions will read through the responses and count the number of strengths and weaknesses as well as the balance. Additional Hypothesis Methods: We are conducting a direct replication, so there will be no deviations from the original study. The only addition to the original study will be the NPI and follow-up analysis. The materials relevant to our additional question will only be presented to participants after they have completed the measures from the direct replication. Participants will complete steps one to three from the original research methods. Afterwards, they will complete a narcissistic personality inventory (NPI) After the data is collected, two coders who are blind to participant conditions will read through the responses and count the number of strengths and weaknesses as well as the balance. The data will be sorted into narcissistic and non-narcissistic categories based on their scores on the NPI for both treatment and control groups. Analysis: For the direct replication portion of the study, we will directly follow the analysis outlined in the Stellar paper where they first removed participants that met the criteria for exclusion. They log transformed each of their dependent variables (number of strengths, number of weaknesses) which were both positively skewed, they calculated the balance between the two and then log transformed that data as well to meet the assumptions of normality. After, they conducted independent samples t-tests to compare the balance of disclosing strengths vs weaknesses between awe and neutral conditions. They then conducted a multiple regression analysis with awe and happiness as independent variables that predict humility, controlling for condition (awe and control). Analysis of our Additional Research Question: Data Transformation: We will follow the same procedure as performed in initial study by Stellar, first seeing if any participants meet criteria for exclusion and then we will perform a log transformation on each of our DVs, if our data is positively skewed to ensure the assumption of normality is met. For our balance we will be calculating the difference between number of strengths and weaknesses and then likely log transforming those values. This ensures our data is normally distributed in order to conduct our multivariant analysis of variance (MANOVA). Key Effects: We will be conducting a two-way MANOVA to examine 2 factors: Awe Treatment Control Narcissism Narcissistic Non-Narcissistic We will be studying the effects these factors have on our three dependent variables quantifying our state humility: Number of Strengths Number of Weaknesses The Difference between the number of Strengths and Weaknesses We chose to do a two-way MANOVA rather than multiple two-way ANOVAs to control for familywise error rate that we would encounter in conducting multiple two-way ANOVAs. We will be testing for a main effect of inducing awe, main effect of narcissism, and an interaction effect between inducing awe and narcissism across all dependent variables simultaneously. If our hypothesis is correct, the interaction effect will be significant, this would suggest the impact of awe on humility is impacted by levels of trait narcissism. If the two-way MANOVA indicates significant main effects or interactions we will follow up with univariate ANOVAs for each DV (number of strengths, number of weaknesses, and balance). This will help clarify which specific variables contributed to the multivariate effect. If the interaction is significant in our MANOVA or univariate ANOVAs we will conduct simple main effects analyses for both factors to explore how one factor affects the DVs at each level of the other factor. In the cases where the univariate ANOVAs show significant main effects or interactions, we will perform post-hoc tests to make pairwise comparisons between levels and determine which specific groups differ from each other. We will use Tukey’s Honest Significant Difference (HSD) test to control for multiple comparisons, minimizing the risk of Type I error. Since we’re analyzing both the simple main effects and conducting post-hoc tests, we will report effect sizes using partial eta squared to provide insight into the magnitude of the effects. Target Sample: N = 100 Recruiting participants from SONA - McGill Psychology Human Participant Pool Our study will be completed on a computer within a lab setting.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.