Data Standards

Your dataset can be standardised in various aspects. Standardisation, in general, makes data comparable and interpretable. In other words, your data becomes interoperable by applying standards. Datasets can be combined, compared or are simply easier to reuse. You have to plan standardisation, as it is for many aspects hard or impossible to apply afterwards.

Standardise as much as possible between you and your collaborators or research group. If there are standards established and used in your field of research you are advised to use these.

Here is a list of things you can standardise in your research.

• Standardise how, what and when you measure things by standardising your protocol, or methods and materials. For instance, is there a standard set of questions for ‘quality of life’? Is there a standard procedure to house mice for your purpose? What aspects do you measure? At what parameter values (age, concentration, etc.)? When do you measure (every two hours, every gram of weight gain, etc.)?
• Standardise your file formats so you can easily exchange results without technical difficulties. Check for standard taxonomies or coding systems within your research discipline.
• Standardise the units in which you note down your results. For instance, do you use mm, cm, m? It is extra work to transform units between experiments.
• Standardise the metadata you use to describe your records or study. What fields will fill in by default, and according to what standard do you define the fields’ names? Will you design a metadata spreadsheet where you specify all things that you will note down?
• Standardise the vocabulary you use. If everyone has the same terminology, it can avoid confusion or misinterpretation. Check for standard taxonomies or coding systems within your research discipline.