Work Description

Separate and Unequal: Moral Domains Differ in Corresponding Social Judgments of Others

======================================================================================

This set of studies were run between February of 2022 and July 2023 at the University of Michigan using participants from Prolific Academic.

PI: Savannah Adams (contact: snladams@umich.edu)
Research Supervisor: Dr. Oscar Ybarra

This research was conducted to investigate potential differences in the weight given to behaviors within various moral domains when forming social impressions of others. More specifically, we aimed to see if impressions of a social target's moral character differed based on whether participants were given information about a target's actions in line with or in violation of various moral domains.

Key Findings:
- In Study 1, judgments of a target's moral character differed as a function of moral domain, such that behaviors related to fairness or respect for property were judged in more extremes (e.g., higher judgments of a target's moral character following positive behaviors in these domains, and lower judgments of a target's moral character following negative behaviors in these domains). Attributions made about behaviors also differed such that behaviors in the fairness and property domains were more likely to be attributed to the target rather than the situation.

- In Study 2, the results from Study 1 were replicated and we also found that willingness to cooperate with a target differed as a function of moral domain. Behaviors in the fairness and property domains led to stronger judgments about whether someone would be willing to cooperate with a social target.

- In Study 3, we replicated the results of Study 1 and introduced a digit span task. Judgments about a social target's moral character were robust to cognitive load.

The following data files were analyzed in SPSS primarily, but there are some R files provided for obtaining specific descriptives (i.e., repeated measures correlations). All information about the variables can be found in the variables tab within SPSS for each dataset, with any additional created variables explained here or within the syntax.

NOTE: All files provided in this package are either primary data files, meaning that they appear as they would when first downloaded from Qualtrics, or processed files that have been altered to make analyses easier. However, all identifiable information has been removed in both the primary and processed data files. Long format processed files have also been included for ease of import into R for any rmcorr analyses.

In some cases, there is also an additional column included in the data labeled "COGCHECK_PASS". This is not included in the original Qualtrics but allows for quicker filtering of participants who did or did not pass the cognition check.

For Studies 1 and 2, the researchers did not include demographic questions in the surveys themselves. Demographic data was instead obtained via Prolific and was later incorporated into the datasets using Prolific ID information. This created some limits around the breadth of our demographic categories. Due to anonymity reasons, the full extent of the information collected and reported by Prolific could not be included in these datasets, but the specific age, ethnicity, and gender information as reported by Prolific are included in the processed data files.

List of Files
-------

*All survey files for the three studies in our manuscript (Moral_Domains_Social_Judgments_Study1_Survey.docx, Moral_Domains_Social_Judgments_Study2_Survey.docx, Moral_Domains_Social_Judgments_Study3_Survey.docx). A codebook is included with this package under the file name (Moral_Domains_Social_Judgments_Codebook.csv). In this file, all variables are listed for each of the three processed datasets.

-------

* Study 1 primary data (Moral_Domains_Social_Judgments_Study_1_primary.sav). This dataset corresponds to the first study in our manuscript. This file is the primary data, meaning it is as it appears when first downloaded from Qualtrics (with minor alterations as detailed above). All identifiable information has been stripped, and the steps taken to transform it into the processed data file are detailed in the provided Study 1 syntax file as comments above the first breakpoint.

*Study 1 processed data (Moral_Domains_Social_Judgments_Study_1_processed.sav). This dataset is a ready-to-analyze version of the primary dataset provided for convenience. The syntax file includes all changes made to the primary dataset to create the processed dataset, and includes code for all analyses included in our manuscript.

*Study 1 syntax file (Moral_Domains_Social_Judgments_Syntax_Study_1.sps). The SPSS syntax file for dataset cleaning and the analyses included in our manuscript.

*Study 1 long format processed data (Moral_Domains_Social_Judgments_Study_1_long.sav). This dataset includes the same information as the normal processed dataset but has already been converted to long format for use in R or in any additional analyses.

-------

* Study 2 primary data (Moral_Domains_Social_Judgments_Study_2_primary.sav). This dataset corresponds to the first study in our manuscript. This file is the primary data, meaning it is as it appears when first downloaded from Qualtrics (with minor alterations as detailed above). All identifiable information has been stripped, and the steps taken to transform it into the processed data file are detailed in the provided Study 1 syntax file as comments above the first breakpoint.

*Study 2 processed data (Moral_Domains_Social_Judgments_Study_2_processed.sav). This dataset is a ready-to-analyze version of the primary dataset provided for convenience. The syntax file includes all changes made to the primary dataset to create the processed dataset, and includes code for all analyses included in our manuscript.

*Study 2 syntax file (Moral_Domains_Social_Judgments_Syntax_Study_2.sps). The SPSS syntax file for dataset cleaning and the analyses included in our manuscript.

*Study 2 long format processed data (Moral_Domains_Social_Judgments_Study_2_long.sav). This dataset includes the same information as the normal processed dataset but has already been converted to long format for use in R or in any additional analyses.
-------

Preregistration for Study 3 can be found as a pdf document in the files.

* Study 3 primary data (Moral_Domains_Social_Judgments_Study_3_primary.sav). This dataset corresponds to the first study in our manuscript. This file is the primary data, meaning it is as it appears when first downloaded from Qualtrics (with minor alterations as detailed above). All identifiable information has been stripped, and the steps taken to transform it into the processed data file are detailed in the provided Study 1 syntax file as comments above the first breakpoint.

*Study 3 processed data (Moral_Domains_Social_Judgments_Study_3_processed.sav). This dataset is a ready-to-analyze version of the primary dataset provided for convenience. The syntax file includes all changes made to the primary dataset to create the processed dataset, and includes code for all analyses included in our manuscript.

*Study 3 syntax file (Moral_Domains_Social_Judgments_Syntax_Study_3.sps). The SPSS syntax file for dataset cleaning and the analyses included in our manuscript.

*Study 3 long format processed data (Moral_Domains_Social_Judgments_Study_3_long.sav). This dataset includes the same information as the normal processed dataset but has already been converted to long format for use in R or in any additional analyses.

Use and Access:
This data set is made available under a Creative Commons Attribution-NonCommercial 4.0 International license (CC BY-NC 4.0).