Executable Environment for OSF Project mc26t

This repository was automatically generated as part of a project to test the reproducibility of open science projects hosted on the Open Science Framework (OSF).

Project Title: Count regression models for keyness analysis

Project Description:

This OSF project accompanies a research paper that explores an approach to keyword analysis based on regression modeling. Specifically, we use a form of negative binomial regression, which offers a number of advantages compared to existing methods for identifying typical items in a target corpus. Thus, it is responsive to the multidimensional nature of keyness and can address multiple aspects of typicalness simultaneously, using a single statistical model. Further, metrics of interest can be enriched with confidence intervals, which allows us to isolate descriptive and inferential indicators of keyness. Finally, all quantities are based on a text-level analysis, which accounts for the fact that corpora consist of text files and adjusts statistical estimates accordingly. As an illustrative case study, we use data from COCA to identify key verbs in academic writing. To assess the performance of our method, we monitor the coverage rate of the 95% confidence intervals and observe that, for our analysis task, this model seems to be adequate for purposes of statistical inference. Due consideration is also given to the limitations of this procedure, and we conclude by outlining the kinds of keyness analyses for which count regression models may be a worthwhile approach.

Original OSF Page: https://osf.io/mc26t/

---

Important Note: The contents of the mc26t_src folder were cloned from the OSF project on 12-03-2025. Any changes made to the original OSF project after this date will not be reflected in this repository.

The DESCRIPTION file was automatically added to make this project Binder-ready. For more information on how R-based OSF projects are containerized, please refer to the osf-to-binder GitHub repository: https://github.com/Code-Inspect/osf-to-binder

flowR Integration

This version of the repository has the flowR Addin preinstalled. flowR allows visual design and execution of data analysis workflows within RStudio, supporting better reproducibility and modular analysis pipelines.

To use flowR, open the project in RStudio and go to Addins > flowR.

How to Launch:

Launch in your Browser:

🚀 MyBinder:

• This will launch the project in an interactive RStudio environment in your web browser.
• Please note that Binder may take a few minutes to build the environment.

🚀 NFDI JupyterHub:

• This will launch the project in an interactive RStudio environment on the NFDI JupyterHub platform.

Access Downloaded Data: The downloaded data from the OSF project is located in the mc26t_src folder.

Run via Docker for Long-Term Reproducibility

In addition to launching this project using Binder or NFDI JupyterHub, you can reproduce the environment locally using Docker. This is especially useful for long-term access, offline use, or high-performance computing environments.

Pull the Docker Image

docker pull meet261/repo2docker-mc26t-f:latest

Launch RStudio Server

Run the container (with a name, e.g. rstudio-dev):

docker run -it --name rstudio-dev --platform linux/amd64 -p 8888:8787 --user root meet261/repo2docker-mc26t-f bash

Inside the container, start RStudio Server with no authentication:

/usr/lib/rstudio-server/bin/rserver --www-port 8787 --auth-none=1

Then, open your browser and go to: http://localhost:8888

Note: If you're running the container on a remote server (e.g., via SSH), replace localhost with your server's IP address. For example: http://<your-server-ip>:8888

Looking for the Base Version?

For the original Binder-ready repository without flowR, visit: osf_mc26t