Last month, the CSIC Cybermetrics Lab (Spain) published its Top Institutional Repositories Ranking. In total, 1440 institutional repositories were ranked. It is worth mentioning that there are very few requirements in order to get your repository added to the ranking. For starters, inclusion is free of charge. The most important requirement is having a unique (sub-level) URL attributed to your repository. Even for small or new repositories, it is useful to be included in the ranking because the ranking basically offers a free, twice-yearly web-exposure audit. There are 42 DSpace installations in the top 100 compared to 43 from January of 2012.

A short introduction to the Webometrics Repository Ranking initiative

The aim of the ranking is to support Open Access initiatives and therefore the free and open access to online scientific publications and other scholarly materials. The web indicators are used here to measure the global visibility and impact of scientific repositories. Read a full description of the methodology

Evolutions in the ranking reflect following changes

• Changes to the ranking algorithm
• Changes in the internal search engine algorithms (necessary since the ranking is based on visibility in search engines)
• Improvements or defects on the level of the repository itself: more content, search engine optimization, link building, downtime …

Because of the many variables, it is nearly impossible to put your finger on a strong causal relation between fluctuations of your repository's ranking and one of these possible causes. However, witnessing a sizeable drop of your repository in the ranking can motivate a more in-depth investigation and search engine optimization.

Changes to the ranking algorithm in July 2012

In January, Majestic SEO was introduced as a new database used in the ranking to keep track of incoming links to repositories. The number of incoming links is very important for the ranking as they determine the visibility indicator that counts for 50% of the total weight. In order to make sure that the ranking is not relying on a single source of data, the ranking now introduces AHREFS as another source of incoming link data in order to increase the reliability of the visbility indicator.

The inclusion of AHREFS data is not the only change to the visibility indicator. Repositories are now rewarded to get those incoming links from different domains. The effect is substantial, because the number of incoming links gets multiplied by the number of domains where the backlinks originate.

Changes in the ranking algorithm always make it very tedious to compare this ranking to previous editions, but as the author, Dr. Isidro F. Aguillo, states:

"[The] Web is still evolving, I think in the future, the web will reflect even better the reality, in the meantime I am fueling the virtuous circle. If the ranking push people to web publish then web indicators become more trusted."

DSpace Repositories in the top 200

The following list represents a subset of the official repository ranking in which we only include DSpace instances. @mire's clients are highlighted in blue. The evolution presented here is the rise or drop in positions compared to the January 2012 ranking.

Trending Climbers

Instead of restricting our report to the improvements between January and July 2012, we aimed to evaluate positive fluctuations over a longer period, taking into account evolutions over the last 4 rankings. Therefore the following list contains those repositories that improved their positions between January 2011 and July 2012. @mire's clients are highlighted in blue.

There are important caveats to consider before you decide to attach any value to these evolutions or to the position of your repository in the ranking in general. First of all, looking at trends over a larger period still does not address the weakness that it is entirely relative to the position of the repository in one particular ranking, in this case, the January 2011 ranking. If your repository was experiencing technical difficulties in January 2011, causing a big drop in that particular ranking, you might look very good in this table of trends if you were able to rectify the technical difficulties in future rankings. On the other hand, if you already had a high position prior to January 2011 that you were able to maintain or slightly improve, your efforts are not reflected in this list.

The detailed list of repositories suffering from large setbacks in the ranking since January 2011 is not included here, but we did evaluate it. Fifteen of the top 20 repositories that lost the most ground were repositories that are based out of Asia. Whether they were incorrectly attributed higher positions in January 2011 or whether they are now suffering from region or language biased adjustments remains a topic of debate.

Contact us if you are interested in more extensive data or recommendations on how to improve search engine exposure for your repository.