2  IAT: Sexuality

Author

Affiliation

Seungju Kim

University of Illinois at Urbana-Champaign

fs ## Source Description

Full name: Implicit Association Test - Sexuality
Website: Harvard Project Implicit
Data type: Survey
Unit of analysis: State
Geographic Coverage: 50 U.S. states + D.C.
Temporal Coverage: 2010-2025

About the Original Data

Project Implicit (housed within Harvard University) is a web-based platform that allows users to Implicit associations and explicit attitudes towards sexuality.

Data collection for sexuality has been ongoing since 2004 (as of 2026) using non-probability sample of volunteer participants who visit the website through search engines, social media, or as part of school or work.

Warning

Researchers should be aware of one significant limitation surrounding the use of implicit associations for sexuality: it only measures implicit associations between good and bad for gay men.

This means the implicit associations test may not hold meaningful implications for lesbian, bi/pansexual, or otherwise sexually diverse individuals.

Original Data Collection

Feature	Details
Collection method	BLANK (e.g., Policy tracking, Online survey, Phone survey)
Sampling frame	BLANK (e.g., All U.S. states, U.S. adult internet users)
Sample type	BLANK (e.g., Census, Probability, Convenience)
Update frequency	BLANK (e.g., Annual, Continuous)
Geographic coverage	BLANK (e.g., State-level, National)
Temporal coverage	BLANK–present

Note

BLANK: Note any important caveats about how the original investigators collected or coded this data.

Required Citation

If you use this indicator in published research, you must cite the original data source in addition to the stigmaR package:

BLANK: Full citation for original data source here.

---

Access and Download

Feature	Details
Access type	BLANK (e.g., Public download, Registration required, Request)
URL	BLANK
Authentication required	BLANK (Yes/No)
Cost	BLANK (Free/Paid)
Last accessed	BLANK (Month Year)

Download Procedure

BLANK: Step-by-step description of how the data was obtained. Be specific enough that someone could reproduce this in 5 years if the website changes.

# BLANK: Download code here

Warning

BLANK: Note any known instability in access (e.g., website changes, file format shifts across years). Delete this block if not applicable.

---

Cleaning Procedure

Raw Data Structure

• File format: BLANK (e.g., .csv, .xlsx, .sav)
  
• Rows: BLANK (e.g., One row per state-year, One row per respondent)
  
• Key variables used: BLANK

Cleaning Steps

1. BLANK (e.g., Variable Selection)

BLANK: What variables were retained and why. What was excluded and why.

# BLANK

2. BLANK (e.g., Exclusion Criteria)

BLANK: What cases/rows were excluded and why.

# BLANK

3. BLANK (e.g., Recoding / Scoring)

BLANK: Any recoding, reverse scoring, or composite construction.

# BLANK

Important

BLANK: Document any judgment calls here — these are the decisions future you (or a reviewer) will ask about. Delete if not applicable.

---

Aggregation to State Level

Note

This section documents how individual-level or sub-state data was aggregated to the state level. Skip if data is already state-level.

Unit of Analysis in Raw Data

BLANK: What is the original unit of observation? (e.g., Individual respondent, County, Zip code)

Aggregation Method

Feature	Decision	Rationale
Statistic	BLANK (Mean/Median/Sum)	BLANK
Minimum n	BLANK (e.g., n ≥ 50 per state)	BLANK
Weighting	BLANK (Unweighted/Survey-weighted)	BLANK
Missing states	BLANK (Excluded/Imputed)	BLANK

# BLANK: Aggregation code here

Sensitivity Checks

BLANK: Note any alternative aggregation approaches considered and why the chosen approach was preferred. Delete if not applicable.

---

Output Dataset

Variables in Final Output

Variable	Type	Description	Range
state	chr	Two-letter state abbreviation	—
year	int	Calendar year	BLANK–BLANK
BLANK	BLANK	BLANK	BLANK
BLANK	BLANK	BLANK	BLANK

Output File

saveRDS(BLANK, here("data", "clean", "BLANK", "BLANK.rds"))

File location: data/clean/BLANK/BLANK.rds
Rows: BLANK
Last generated: BLANK (Month Year)

---

Known Limitations

• BLANK
• BLANK

---

References

BLANK: Full references for any papers, reports, or technical documentation from the original data collectors.