This README describes a dataset for state-level implicit and explicit bias index against Asian Americans across the United States that are in the collection of New York University Libraries and were examined by the NYU Silver School of Social Work. For more information about this object, please contact the Nari Yoo (nari.yoo@nyu.edu).
The dataset includes a dataset with state FIPS codes, code for analysis. The individual-level raw data were downloaded from Project Implicit demo website datasets. For the sample, we dropped non-Hispanic East Asian (N=57,914) and South Asian (N=29,490) cases based on their self-reported race/ethnicity in accordance with our purpose of visualization of out-group biases. The original project implicit dataset does not include the Southeast Asian category for race/ethnicity questions. As the population of non-Whites is overrepresented in Asian IAT datasets, we further calculated a weighted mean based on the size of the race and ethnicity populations at the state-level in the 2021 American Community Survey.
The IAT score was computed by comparing the time it took participants to categorize words or images when they were paired with stimuli that were congruent with their implicit biases (e.g., European Americans + American-Foreign) versus when they were paired with stimuli that were incongruent with their implicit biases (e.g., Asian Americans + American-Foreign).
The Stata dta file can be opened in the Stata software.
| Variable | Label | | --- | --- | | statefips | State FIPS code | | state_abbrev | State Abbreviation | | state_name | State Name | | state_exp_asian | Explicit Bias | | state_imp_asian | Implicit Bias | | state_exp_asian_sd | (sd) Explicit Bias | | state_imp_asian_sd | (sd) Implicit Bias | | state_asian_exp_count | (count) Explicit Bias | | state_asian_imp_count | (count) Implicit Bias |
The Comma-separated values (CSV) file is available to be imported into Microsoft Excel, SPSS, or other software.
Lastly, this Stata do file includes the code used in analysis from the raw data downloaded from Project Implicit demo website datasets to data visualization.
This code uses several external data sources, primarily the Project Implicit data available through OSF at the URL https://osf.io/cpmfk/, and data from the US Census' ACS data.
Instructions for accessing the ACS data for use with Stata can be found here: https://nariyoo.com/stata-getcensus-package-for-american-community-survey-datasets/
For step 5 in the code, the file "macro_state_merged.dta" draws on external data sources, including ACS and MIT Election Lab data (https://electionlab.mit.edu), which are not available in this repository.