Datasets

These services can help you find datasets held in different repositories:

• Google Dataset Search: search for datasets using keywords, then follow the results to the organisations or repositories providing them.
• DataCite Commons: search descriptions of research outputs with persistent identifiers. Use the work-type filter to focus on datasets.
• Data Citation Index: search for research datasets and their connections to published research. Open Web of Science through Library Resources A–Z, then select Data Citation Index.
• OpenAIRE Explore: find research data and related publications from different sources.

Finding a dataset record does not necessarily mean that its files are immediately available. Follow the record to the provider and check the access information.

If you know your subject area, a specialist source may offer more relevant data and more detailed search options:

• UK Data Service: social, economic and population data, including surveys and census data.
• CESSDA Data Catalogue: descriptions of social science datasets held by European data services.
• NERC Data Catalogue Service: environmental research data.
• EMBL-EBI: molecular biology data resources.

Use re3data to identify repositories by subject. It is a directory of repositories, so follow its links to search within the relevant repository.

You can also search repositories that hold data from different subjects, including Zenodo, Figshare, Mendeley Data, Dryad and public projects on the Open Science Framework.

Check each dataset’s description, documentation and reuse conditions before choosing it.

For data about populations, economies, societies and other subjects, try:

• UK public data: data published by government departments and other public bodies.
• Scottish Government statistics: data about Scotland and its local areas.
• Office for National Statistics: data on the economy, population and society. Check the geographical coverage of each dataset.
• National Records of Scotland: population, migration, household and other demographic data for Scotland.
• Eurostat: European statistical data.
• World Bank Open Data: data and indicators for comparing countries.
• WTO Data: international trade data.
• UN Data: data from the United Nations system.
• European Data Portal: public datasets from European institutions and countries.

For help finding published figures and summary tables, see Statistics.

Abertay’s Research Portal includes records of datasets from University research. Records may provide files directly or link to data held in another repository.

If you are interested in a particular researcher’s work, check their institutional research profile and the data links in their publications.

Read the dataset’s documentation before committing to using it. Check:

• Purpose and coverage: does it contain the information you need, for the relevant population, place and time period?
• Collection methods: who produced the data, how were they collected, and what limitations could affect your analysis?
• Documentation: are the variables, units, codes and file contents explained? Look for a README file, codebook or other supporting information.
• Completeness and quality: are missing values, errors, exclusions or other limitations identified?
• Format and version: can you work with the files using the software and skills available to you? Record which version you use.

A dataset can be useful for one question and unsuitable for another. Its availability or presence in a repository does not establish that it is appropriate for your project.

For fuller guidance, see Evaluate information.

Some datasets can be downloaded directly. Others require registration, an application or approval before access is granted. Read the access conditions before planning your work around a dataset.

Check the dataset’s licence and any access conditions before using, adapting or sharing it. Being able to download data does not necessarily mean that every use is permitted.

If you combine datasets, check the terms for each one. Do not assume that choosing the most restrictive licence makes the combination permissible.

For datasets about people, read any privacy, consent or confidentiality information supplied with the data and follow your project’s requirements.

For help understanding dataset licences and reuse conditions, email repository@abertay.ac.uk.

Check your module or assessment guidance and use the referencing style it requires.

Cite any dataset you use, whether you analyse it, extract particular values or use it to create another dataset.

Keep a record of:

• the creator or responsible organisation

• the dataset title and publication year

• the repository or publisher

• the version or edition used

• its DOI or another permanent link

Use Cite Them Right Online for dataset citation and reference examples. Search for datasets and select your required referencing style.

If the repository supplies a suggested citation, use it as a starting point and check it against your required referencing style.

Describe any selection, cleaning or other changes you make to the data so that readers can understand your analysis.

For wider guidance on referencing principles, when to cite and where to get help, see Referencing.