#Pearson mastering physics plus#
#Pearson mastering physics download#

The total course grade conflates a lot of stuff, however, so perhaps it would be better to look at the final, the final problems that are from the homework or the final problems that are not from the homework:įinal (200 points): A B delta Mean 153.1 145.0 8.1 Median 161.0 151.5 9.5įinal from the homework (70 points or 35%): A B delta Mean 52.4 48.1 4.3 Median 56.0 52.0 4.0įinal not from the homework (130 points or 65%): A B delta Mean 97.1 89.4 7.7 Median 103.0 98.5 4.5 I'll leave it to those who grade on a curve to calculate what the grade effect would be. To look at it another way, these correspond to the student who used the electronic homework being about 40% of a standard deviation ahead of those who don't. These differences are significant at 2.7 and 2.2 sigma, respectively. To give just the highlights of the numbers: In particular, the unit grading scheme can cause the median to 'stick' at a particular value even when the distribution changes a lot. I also wanted to look at medians and distributions, but as their significance is harder to quantify, I'm just looking for trends in those.

#Pearson mastering physics plus#

Because students using MP were given three tries to do each part, plus some opportunities for extra credit by not using hints, both the students in groups A and B essentially maxed out their homework scores if they tried.īefore the semester started, I decided that the primary measures of the success of the homework would be the means of scores on the course as a whole (because students care), scores on the three exams (as they aren't conflated with homework as the total score is), and partial exam scores for the parts of the exam that do/don't come from the homework (to check that any effect is just due to familiarity with specific problems). The lowest HW after the first week was dropped. The homework was assigned 10% of the credit in the course. Exams were graded using the 'holistic grading' 5 points scale separately for each part of each problem. The students were told that approximately 1/3 of problems on the exams (two midterms and the final) would come straight from the homework, that approximately 1/3 would be very similar to problems they would see in D/L and lecture, and approximately 1/3 would be quite new to them, but using techniques they'd been practicing. I comment on this second alternative near the end, because it complicated the analysis a little.

#Pearson mastering physics download#

Students submitting their homework on paper could download a copy for printing straight from the web, or could do it on MP and then submit it on paper. A post-experiment factor analysis determined that there were no significant section-to-section performance differences within the groups that would effect the results by more than 0.12 sigma, so this method of assigning students is probably OK. All odd numbered sections (N=87 students) would do the homework on paper. All even numbered sections (N=85 students at the end of the course) would do the homework online. Grading consisted of only checking that something was handed in there was no check that it was correct, original, or anything else like that.īecause I want to encourage students to work together with their colleagues in their D/L sections, the students were assigned to two experimental groups by section. I did an experiment in my section of 8B to test the hypothesis that there was no difference in effectiveness between two forms of homework:Ī) Online homework, specifically Mastering Physics, assigned weekly and graded by the machine interactively.ī) The same problems, to be handed in weekly on paper. I've appended below a report from when I was teaching 8B a few years back. There's no question that Mastering Physics is imperfect, even terribly frustrating, but there is some evidence that it _works_. I'm not sure that's a great way to learn the material, which is the expressed goal of the homework in the first place.