2013-01-26: The paper is now in press! The final draft can be downloaded at EsPal Processing in press.

2012-11-13: Phonology issues have been corrected. The most common lemma information has been added to all three databases so you can now search for lemma information in "Words to Properties" and "Constraints to Words".

2012-10-23: We finally have a FAQ page!

2012-10-16: Working on adding the most frequent part of speech category to the word tables, so you can search for "NOUN" or "VERB" at the same time as frequency data.

2012-09-04: Added the subtitle contextual diversity database.

2012-08-09: Submitted journal paper about the EsPal website. Started paper on validation studies.

EsPal is designed to be a source of information containing all the possible properties of Spanish words.

EsPal is primarily designed for psycholinguists looking for words with certain properties for their experiments.

There are currently two corpora: 1) A large collection of written data from the Web, government sources, newspapers, and literature. 2) Subtitles kindly provided by opensubtitles.org.

The first version of EsPal available to the public was made in August, 2012. But we are always working to improve it, so please note the date of the database you are using in case you report results or find issues with the data.

You can email us at espal@bcbl.eu if you have any suggestions, questions, or find issues with the data. We are always trying to make it better and more accurate!

There are two basic ways to search.

1) If you have a list of words that you would like to get properties for, you can upload it to get word properties or to get part of speech and lemma properties.

2) If you have a set of constraints on the words you are looking for, you can set those constraints either on the word properties or on the part of speech and lemma properties.

If you click the "Collapse/Expand All" button on any of the search pages, it will open up all of the sections and you'll see all the variables. Look under the variable's "Help" button to read a description of that variable.

Three separate databases are available: the written database, the subtitle database using token frequency, and the subtitle database using contextual diversity (number of different movies or shows). You must choose one of them. All results will be based on the frequencies in the chosen source. Note for the contextual diversity that the frequencies are "percent of movies" not "per million tokens".

Spanish has fairly regular rules for phonetic transciption of orthography. We apply the rules from the SAGA project for Castillian pronunciation. For Latin American pronunciation, the only difference at this time is that 'z' and 'c' (followed by 'e' or 'i') are transcribed as /T/ in Castilian and /s/ in Latin American Spanish. You must choose one of them. All phonology results will be based on the chosen phonology (orthographic and other data will not be different).

Contextual diversity is the number of different documents a word appears in. In EsPal we have the contextual diversity of the subtitle data, which basically means: How many different movies did this word appear in? We shorten this to CDM: contextual diversity:movie-based. Note for the CDM data that the counts are the number of different movies, and that the frequencies are "percent of movies" not "per million tokens".

When doing a constraint search, you should not only enter the values of the constraint, but also check the box next to it.