jump to navigation

Final grades posted December 18, 2006

Posted by Ron in 3790 Administrivia.
add a comment

I’ve just posted the final grades for the course. For the two midterms and quiz totals I used a straight scale (90+=A, 80+=B, 60+=C, 50+=D). Because the homework total and the final both had lower than expected median values, I lowered the scale for both by 5 points (85+=A, 75+=B, 55+=C, 45+=D). This is a slightly higher scale than that posted for the homework earlier, but produced a better balance of grades than my earlier estimate. This produces a scale for the final grades of (87.5+=A, 77.5+=B, 62.5+=C, 47.5+=D). There were 20 A’s, 15 B’s, 14 C’s, 5 D’s, and 1 F. This gives a median grade of B, which is right around the GT average.

The scores on the final were interesting.  Almost everyone had trouble on the short answer section and did well on the long answer section, which is the reverse of what usually happens. In general, the questions about the last few lectures (Brooks and Wilson) were most frequently answered incorrectly.  We’ve broken out the scores for the two parts on the test on WebCT, if you’d like to see for yourself.

Final lectures have been posted December 9, 2006

Posted by Ron in 3790 Administrivia.
add a comment

All the lectures for the term should be up.

Robot ethics December 4, 2006

Posted by Ron in 3790 Administrivia, Cognitive Science News.
add a comment

Those of you who were intrigued by the behavior-based robotics models might want to take Ron Arkin’s course on Robot Ethics (CS 4002). This is a class that can be substituted for CS 4001, the standard CS ethics course. It should be quite interesting.

Homework scores December 4, 2006

Posted by Ron in 3790 Administrivia.

Some of you may be concerned that your homework scores are low. In general, the homework have been graded to maximize the spread of scores (e.g., if there are 15 points in the assignment and everyone gets 13-15, there isn’t must space to discriminate between thefair,  good and very good). As a result, however, the median value for the homework are pretty low (Median = 73, Mean= 66).

Here’s a WebCT graph that makes this clearer:

Homework Histogram

Rest assured, we’ll be curving the homework grade to make up for the spread. Right now, I estimate that we would probably use the scale of 85+=A, 70+=B, 55+=C, and 40+=D. This may vary a little once we go over all the grades at the end of the term.

You can check your current homework score by looking on WebCT, where I’ve justed added “Total Homework” and “Total Quiz” columns. All the columns should have the statistics enabled, which will let you check the spread.

Readings for this week are up December 3, 2006

Posted by Ron in Uncategorized.
add a comment

The readinngs on situated cognition and affordances are now up in the reading list:

Sorry for the delay.