Shows how to encrypt data held in public space
Switch branches/tags
Nothing to show
Clone or download
Permalink
Failed to load latest commit information.
data linked to OSF version Aug 11, 2017
.gitignore
LICENSE
README.md

README.md

Encrypt data example

Shows how to encrypt data held in public space, for analysis in R

The example data is from my Master's thesis. You can obtain all the details at the GitHub repository:

https://github.com/richarddmorey/absoluteJudgmentCalibration

Here, I've only used the minimum content needed for a working example.

What am I assuming?

  • You would like to save data in a online repository for convenience, safety, and potential future sharing
  • You would like to keep these data private for a time
  • You might like to share your analysis scripts with others, so they can be checked, but maybe not the data itself
  • You'd like to easily share the data with colleagues/editors/reviewers, even though they are private
  • You don't need heavy-duty security/encryption
  • You analyze data in R

If these are true, this example is for you.

The GitHub version is here: https://github.com/richarddmorey/encrypt_data_example

The OSF version is here: https://osf.io/vqfw8/

What does this example show?

  • How to encrypt a file for sharing: See data/encrypt_it.R, which shows how to create an encrypted file with a passphrase, which you can then share on GitHub or OSF. The file that is encrypted, exp1.txt, is linked in the script.
  • How to decrypt a file for analysis: See data/load_data.R for an example data loading script that decrypts the file directly from GitHub. Note that in this file, passphrase is defined so that you can load the data. You would not share this until you were ready. This repository is in a "public-sharing-ready" state.
  • You can share analyses and meta-data without sharing data: See data/analysis.R and data/exp1_columns.md. These files contain a (basic) analysis and meta-data for the encrypted data set. These could be shared with someone --- for instance, someone who might want the details of an analysis --- without revealing the data.

Once the data is loaded using the load_data.R script, one can analyse it as normal. For someone simply running the script with the correct passphrase, the encryption is completely transparent.