Posted by knockels on 22 June, 2010
Possibly two posts in one day breaks a blogging convention, but here is the second thing from Innovations in Reference Management 2, while I am thinking about it!
Plenty to think about in Kevin Ashley’s talk about citing data. In addition to talking about how to cite it, Kevin spent some time persuading us (if we needed to be persuaded) that we needed to care about data and its use. I thought of (my favourite!) critical appraisal.
If you have the underlying data, it makes it easier to see where authors have been selective, or used a particular technique to make the data “prove” something or other. It adds new dimensions to the evaluation of a paper, or a figure or table in a paper. Kevin’s examples included a survey that said that “nine out of ten cats preferred the Open University”, but where an examination of the underlying data showed that you could only say this if you sorted the data in a particular way. There are all the ruses outlined in books like “How to lie with statistics”, and being aware of those ruses is part of critical appraisal, but having the data adds a new dimension to this.
Kevin also gave some instances of where data might be valuable on its own, without any accompanying papers. A researcher might be interested in the data, and not the story being told about it by one particular person or group:
- Researchers in one area might be able to use data gathered by others for other purposes;
- Having access to all the separate data sets relating to, say, the distribution of individual species, might lead to work on biodiversity on a larger scale;
- Having access to all the separate data sets on a subject could lead to being able to identify where the gaps are.