Calculating average citation scores or average proportion cited for sets of articles
If you just want to know the average citation score (or altmetric/webometric score) or the average proportion cited for a set of articles then you only need to collect these articles and do not need to collect any extra. You should only take this approach if your document set is from a single field and year. This is because articles have a different average number of citations in different fields and years and so it does not make sense to calculate the average citation impact of a set of documents from multiple fields and years.
Calculating world normalised indicators
If you would like to calculate world normalised indicators - in other words indicators that are normalised so that they can be compared between fields and/or years, then you will need to collect extra sets of documents to form the reference set. The following steps are recommended, assuming that you are starting with an initial set of documents to analyse.
- Split your document set by field and year. For the fields, use a field categorisation scheme from a database that you have access to, such as Scopus or the Web of Science.
- As a practical step to reduce the amount of data needed, and also to help the NPC indicator, discard any small field/year combinations so that the remaining field/year combinations are all approximately of the same size, if possible.
- For each field and year in your collection, in the next step you will need to download the complete set of documents as the world reference set. In some cases, you may need a random sample or balanced sample (e.g., of 500 articles) instead of the complete set. Details of how to do this are in the next step of the instructions.
If you adopt this strategy, then Webometric Analyst will be able to work out whether the articles in your collection have higher values than comparable articles from the same fields and years, giving world (field and year) normalised indicators.
Back to the overview page.