the numbers are in...

Just for funzies...for all of my MAT267 classes for the last couple of years, over 200 students,  I plotted everybody's total homework score vs their final grade.  The data look like this:


Clearly there is a strong trend that relates performance on the homework to overall performance in the course.  The red line is the linear least squares fit; hopefully you've heard of it in one of your lab classes, it's an important tool for data analysis--Engineers use it a lot, so you should feel comfortable with its use.   The precise linear least squares description is this:

%Final Score=8.90 + 0.80 %HW + "Noise"

In this model "Noise" indicates a random effect with a Standard Deviation of 10 points in the final score and an average of zero--this is to say that homework scores don't absolutely determine the final score in this model, but they nevertheless have a strong effect on the outcome.  Of course by "random" we mean really mean all the effects that I don't have data on, for instance how much of the homework was actually done by each students in question.

Note however, that the homework is only 15% of the final, while the exams are 75% of the final score. If the only effect of the homework on the final score was from the homework itself, we would expect a slope of the least squares line to be at most 0.15.  This is to say that the homework has an effect on your grade much larger than just the numerical contribution from the homework scores.

This course has an average homework score of 73% with a standard deviation of 25%.  Clearly there is a lot of room for a lot of people to do a lot better.

Moral of the story: work hard on your homework, it will help you do better in this course.