lexicoder

  Home
  Development
  Research
  Download

lexicoder

 

Download

Sections below describe how to get Lexicoder, the Lexicoder Sentiment Dictionary and some related topic dictionaries, Supporting Files useful for those learning how to use Lexicoder with R, and additional files and links useful for anyone downloading and reformatting content for use with Lexicoder, or other automated content-analytic software.

Lexicoder Software

Lexicoder 2.0 was freely available upon request to academic researchers from summer 2009 to summer 2015. As of August 2015, we have replaced the old software with Lexicoder 3.0. The new software is more reliable, it includes a greater number of functions, and it can deal with larger bodies of data. The interaface is very different, however: 2.0 was accessible only through a basic Java-based interface; 3.0 is designed to be accessible through R (and eventually STATA), and from the Command Line.

For the freely-available version of Lexicoder, please fill out our request form by clicking here:

Lexicoder Software Request

Please note that requests should include an email address at an academic institution. Users who have already been using Lexicoder 2.0 should fill out the form again to recieve the newest version. Shortly after recieving the form, we will send you Lexicoder and associated files by email. The Lexicoder manual and practise datasets are available separately, below.

Citation: Daku, M., Soroka, S., and Young, L. 2015. Lexicoder, version 3.0. Available at: www.lexicoder.com.

The Lexicoder Sentiment Dictionary

Like Lexicoder software, the LSD has been freely available upon request to academic researchers from fall 2011. As of August 2015, we have updated the format of the LSD so that it works with Lexicoder 3.0. Users who have already been using the Lexicoder 2.0-formatting version of the LSD should fill out this form again to recieve the newest version.

For this most current version of the LSD, please fill out our request form by clicking here. Users will be provided with the dictionary in Lexicoder format, easily adapted to other software as well.

Lexicoder Sentiment Dictionary Request

Please note that requests should include an email address at an academic institution. Shortly after recieving the form, we will send you the LSD and associated files by email. The LSD manual will be included in the distributed zip file.

Citation: Young, L. and Soroka, S. 2012. Lexicoder Sentiment Dictionary. Available at lexicoder.com.

A French-language version of the LSD is also available, produced by Duval and Petry, here.

Lexicoder Topic Dictionaries

We are currently working on a series of topic dictionaries, aimed at capturing topics in news content, legislative debates, and policy documents. Our goal to is to capture Major Topic codes from the comparative Policy Agendas project, in multiple languages. Currently, we are working on dictionaries in English, Dutch, and Hebrew, in conjunction with the INFOPOL project.

Development of these dictionaries is ongoing, but we welcome any and all testers and contributors. (Indeed, our hope is that we can crowd-source the development of topic dictionaries.) The current versions of English and Dutch dictionaries can be downloaded below; the most recent tests of these dictionaries are described in this paper. If you are interested in adding to our work on topic coding, please do let us know at lexicoder@me.com.

Lexicoder Topic Dictionaries (English and Dutch, June 2013, Lexicoder 2.0 versions)

Citation: Albugh, Quinn, Julie Sevenans and Stuart Soroka. 2013. Lexicoder Topic Dictionaries, June 2013 versions, McGill University, Montreal, Canada. (available at lexicoder.com)

Supporting Files

Lexicoder 3.0 Users Manual

And example workflow in R. This zip file includes an R script file, sample dataset and dictionary files. It is intended to provide some simple examples of how to integrate Lexicoder into a R-based content analysis.

Additional Supporting Scripts, Software, Links

Simon Kiss has developed scripts that can be integrated in R (but needn't be), designed to reformat media data downloaded from ProQuest archives. You can get the script at GitHub here.

Also see Will Lowe's website, with information on a range of tools for automated content analysis.