LibGuides: Scholarly Publishing: Data Management

Best Practices

When conducting research, it is important to not only keep certain information confidential but to also retain data securely and make it available when needed. In order to effectively do this, there are a number of best practices you can follow:

Keep a detailed record of what data you have
Keep relevant metadata related to that data
Have proper, secure storage
Keep a backup of the data
Observe privacy and ethics principles
Know the college's policies and legal obligations
Plan for access and sharing of data

How do I store my data?

You should store and backup the raw data that you collect. You should also store documents and analyses that you've produced or conducted. The following are high value data to store:

Raw data
Metadata
Documentation
Final results and analysis
References cited
Submitted manuscripts
Final paper/publication

It's advisable to create a file naming system when creating and storing data. Consider creating a naming convention including:

Project name or acronym
Researcher name or initials
Date
Type of data
Version number

Example: learningassessment_wilson_111623_testscores_2.csv

File Formats

You may ask yourself what format you should store your data in. This can vary depending on the type of data you are storing. This chart from Ohio State University, provides a good breakdown of recommended formats.

Type of Data	Recommended Formats
Text	Plain text (.txt) Portable Document Format (.pdf) LaTeX documents (.tex) Hypertext Markup Language (.html) Open Document Format (.odt) Extensible Markup Language (.xml)
Tables, spreadsheets, and databases	Tab-separated tables (.txt — sometimes .tsv or .tab) Comma-separated tables (.csv or .txt) Other standard delimiter (e.g. colon, pipe) Fixed-width OpenDocument Spreadsheet (.ods) OpenDocument Database (.odb)
Image Files	TIFF (.tiff or .tif) JPEG (.jpg or .jp2) Portable Network Graphics (.png) Scalable Vector Graphics (.svg) Portable Document Format (.pdf) Graphics Interchange Format (.gif) Microsoft Windows Bitmap Format (.bmp)
Sound Files	WAVE (.wav) FLAC (.flac) MPEG-3 (.mp3 — usually suitable for human voice and moderate-quality audio, but may not be suitable for high-fidelity audio) Audio Interchange File Format (.aiff)
Video Files	MPEG-4 (.mp4) Material Exchange Format (.mxf)
Databases	Extensible Markup Language (.xml) Comma-separated tables (.csv)
Geospatial Data	Geo-Referenced TIFF (.tiff) ESRI Shapefile (.shp, .shx, .dbf) Keyhole Markup Language (.kml) Network Common Data Format (.nc)
Web Data	Javascript Object Notation (.json) Extensible Markup Language (.xml) Hypertext Markup Language (.html)
Web Archive	WebARChive (.warc)
Multidimensional Arrays	Common Data Format (.cdf) Network Common Data Format (.nc) Hierarchical Data Format (usually .hdf or .h5)
E-books	Electronic Publication (.epub)

(File Format Basics, Ohio State University)

In general, you want to opt to using more open, preservable formats that aren't tied to single programs. For example, you could use:

CSV instead of Excel
XML instead of SPV
ASCII instead of Excel
TIFF instead of JPEG (TIFF is a lossless compression file)

Metadata

Metadata provides documentation and context that allows your data to be re-usable. This includes file-level information and project information. It's important to collect metadata throughout the research process. Metadata often includes:

research rationale
experiment descriptions
research methods and protocols
units of measurement, dates, and times
analyzes
locations where data is stored
definitions for acronyms or codes

You can store metadata in a file naming schema, readme file, data dictionary, research notebooks, Standard Operation Procedures (SOPs), or project reports.