Final exam
Winter 2007Seven or eight of the following
fifteen questions will
appear on the final exam, and you will answer each of these questions (probably
about two paragraphs each)
- In one
or two sentences each, discuss which corpus (or other electronic source) of
English you would use for each of the following tasks, and why:
- historical
changes from the 1300s-1600s
- changes from the 1960s-1990s
- collocations
for a given word
- differences between 5-10 different types of written
English
- data from learners of English
- the evolution of the meaning of a
given word over the past 400 years
-
Provide a brief overview of the history of corpus linguistics from the
1950s-1900s
-
Imagine that someone has asked you to create a 10 million word corpus of a
language of your choice, or a particular aspect of English. What are 5-6 design criteria that might be
important, and how would you address each of these?
- What
are 5-6 of the most important principles governing the use of concordances?
- When
would each of the following be most helpful:
- WordCruncher (the version
we’ve been using)
- WordSmith
- relational databases
- What
are the uses of each of the following:
- proportions
- chi-square
- z-scores
- factor analysis
-
Briefly discuss Biber’s “multi-dimensional” approach to register variation
- What
are 5-6 of the major challenges and issues facing the designers or historical
corpora and/or challenges in using these corpora to investigate language
change
- What
are the major uses of parallel corpora?
- How
can corpora most effectively be used in the classroom?
- What
are some general principles governing the use of data from the Web?
-
Comment on the idea that “corpora have made language analysis more simple, as
well as more complex”
- Give
some examples where corpora provide (even native speakers) with insights that
otherwise might not be available
- How
does the approach to data taken in most corpus-based studies compare to that
of many more formalist-based studies?
-
Briefly discuss the idea that language is best analyzed (and learned) at the
lexical / word level, rather than as abstract grammatical rules
|