This repository contains the data and code to accompany "Perception of PIN-PEN merger by Southern and Non-Southern Speakers of American English," which has been accepted to JASA Express Letters. The `plots` folder contains an image of Figure 1 from the paper (`JASA_EL_Figure1.png`), the stimuli formant tracks (`formants.png`), and a map showing what was counted as a "Southern state" and what was counted as a "non-Southern state" in the experiment. The `scripts` folder contains all of the analysis code (`data_analysis.Rmd`, `data_analysis.html`), which includes many additional plots. It also contains the script used for data cleaning (`preprocess_data.R`), and code to fit various models (`mod_bayes.R`, which is the model reported in the manuscript, as well as `mod_bayes_mean.R` and `mod_bayes_max.R`, which are fit with mean and max F1, respectively, instead of step as a predictor). The `models` folder contains the brms models fit from each of the three model scripts. It also contains the model table for the model reported in the paper. The `data` folder includes both the (anonymized) raw data files (the `.csv` files) as well as the cleaned data files (the `.rds` files) output by `preprocess_data.R`, which are read into `data_analysis.Rmd`). It also contains the text for the reading task (`reading_task_text.txt`) and a folder with all of the participant recordings (`recordings`). The `stimuli` folder contains both the ElevenLabs-generated stimuli used in the experiment (`ElevenLabs.zip`) and the Amazon Polly tokens that we initially tried (`Amazon_Polly.zip`). Extract both zip files into this folder in order to replicate the original file structure. For the experiment, we only used the odd-numbered stimuli.