Perl gets an explicit citation in my (Computational Psychology) MSc dissertation because it was so heavily used in data preparation and translation. Some of the tasks perl was used for:

• Translating between alpha letters and binary representations for a PDP model
• An interactive script to assign 'features' to patterns
• CGI scripts to collect & collate input from participants in a side study
• Converting data from format used by SNNS to format used by Xerion when I switched modelling software
• Converting output data from Xerion's output format to something the stats software could handle
• interpolating human readable 'tags' for the output, in place of the index numbers of the data from the Xerion input file (hurray for the Hash!)

If I could have done the PDP modelling (Neural networks, for those not familiar with the term) in perl I would have done. I tried, the results of which are on CPAN. Likewise, if I could have done the 6 way ANOVA in perl that would have been easier, since I could have changed the code to recognise a within subjects study when it saw one! (Maybe a project for the future).

Perl is the most powerful and flexible tool in my toolbox, both professionally as a code monkey and academically as a student of psychology & language.

One of the ways Perl scores, and this might be worth bearing in mind for teaching/promoting it in your field, is the ability to both write simple linear scripts to perform a given task, and to develop complex fully functional applications, in the same language, and with full interoperability.

The other thing I would stress is CPAN - if something can be done, someones probably written a module to do it