CAUSEweb.org webinar, facilitated by http://www.ReadyTalk.com.
“Helping Students Understand the Meaning of Random: Addressing Lexical Ambiguity”
with Diane Fisher, University of Louisiana at Lafayette; Jennifer Kaplan, Michigan State University; Neal Rogness, Grand Valley State University
2:00 to 2:30 pm Eastern time, August 10th, 2010
Note: I did not get audio from the webcast (I had to call in to get sound).
This was a presentation of a project on Language and Lexical Ambiguity. The purpose was to identify problem words, design instructional interventions, and assess effectiveness.
The hope is to reduce student confusion about words, particularly in statistics, that have colloquial and multiple definitions.
Student surveys were used to identify problem words with questions like:
Write a sentence with the word random. Define random, etc.
Some problem words that occur in statistics are: Association, Average, confidence, random, spread; normal significant, bias, independent, error.
Students have to get used to multiple definitions of random. There may be one or more statistical definitions, and a colloquial definition as well.
Student poll at the beginning of the semester about common use of random showed that 49% defined it as an occurrence that is unplanned unexpected or haphazard.
Post instruction before the intervention showed random defined as:
by chance 4%
without reason 39%
without bias 23%
equally likely 8%
One way of getting the statistical meaning of random across was juxtaposing a picture with people dressed as zebras on a street in Shanghai (haphazard, colloquial) vs. a picture of picking names out of a hat (blind selection).
Structural interventions were developed (without adding material to course) to increase student understanding.
One of these was having students guess the average word size in the Gettysburg Address, and comparing this to two other methods: Choosing 10 representative words and taking the average, (doesn’t work as well), and using a computer to generate 10 words and taking the average (this worked the best). This did manage to get the idea across.
After the intervention, the percentage of students who said random meant equally likely rose from 8% to 40%.
This group is in the process of developing teaching modules for future use. This talk was really aimed specifically at teachers of statistics. There are several different ways random is used in statistics: random variable, random assignment, random sampling, which slightly different, but specific meanings. There are other problem words as well (as noted above). There was a lot of interest in this project from other stat teachers in the audience.