Introduction and Background
Recently, I’ve been interested in evaluating my firstyear student’s grasp of cranial nerves in my Neuroanatomy (OPT 113) class. I’ve been teaching this class since 2008, when I designed the class curriculum out of whole cloth as an inexperienced instructor straight out of Masterdegree work. Since then, I’ve made minor and major updates to the course, by rearranging content, and adding selfpaced modules. Most extensively, in 2011 I massively reorganized and rewrote much of the course material in response to problems that had become obvious to me over my first several years of teaching it.
Of most importance to our current discussion on cranial nerves was my decision to begin teaching cranial nerve material after the first midterm examination, so that the second midterm examination was predominantly composed of cranial nerve items. This was intentional; my previous habit of leaving the cranial nerves till last meant they often were not covered in more than a cursory way.
With the development of Understanding the Cranial Nerves (UCN) this Summer for my Ed. D. work, I addressed a felt need I had observed: even with the increased emphasis I had placed upon them in lecture and on examinations, students were still not grasping the clinical importance and implications of the cranial nerves from OPT 113, compared to the rest of the course material. I hoped the selfpaced, onlinehosted, clinicallybased discussion of the subject in UCN, and initial results from formative evaluative groups reinforced my hopes.
However, my initial move to create this module was founded on nothing more than a personal whim, supported by some post hoc qualitative surveys on voluntary participants. Not exactly the stuff experimental research is made of.
Research Question
If for no other reason than my own curiosity, I reviewed examination grades from OPT 113 courses past with the following research question:
Is student performance on OPT 113 examinations correlated to the percentage of cranial nerve questions on examinations?
An inverse correlation would support my felt need that students were not understanding cranial nerves well.
Methods
To study this question, I had a look at my data and make some decisions what I would omit from my analysis. As stated above, in 2011 the course was massively redesigned so that (among other things) cranial nerves would be emphasized more thoroughly. This shows in the percentage of exam questions about cranial nerves: prior to 2011 only one examination had a percentage in double digits (midterm 1, 2008; my first examination I’d ever administered. It had 3 questions about cranial nerves from the basic introduction lecture I gave, and only 30 total questions on the examination). From 2011 on, one semester midterm examination per semester had at least 41.94% of items about cranial nerves (reaching a high of 75% in the second midterm examination of 2013).
Realizing that the variations in my teaching ability and organization were simply too varied to account for in the first two years, I removed them from my analysis, looking only at results from 2011 to 2014.
I also decided not to include final examination grades in my analysis. These were too different than midterm examination, with the confounding factors of comprehensiveness of material, the glut of studying for six examinations in one week, and the very real phenomenon of student’s not studying as hard as they might due to their need to only make a certain mark to receive their desired grade. Final examination data was simply too different, and so I removed it from my analysis.
Finally, I removed outlying results, based upon course grade. The examination grades of anyone making a D, F, or withdrawing from OPT 113 were removed from the analysis. In general, if you passed the course, your midterm examination scores made the cut.
I plugged the numbers into SPSS and ran Pearson product moment correlation.
Results
Check out table 1 for the SPSS output.
Correlations^{a} 

% of questions on Cranial Nerves 
Mean % Score 

% of questions on Cranial Nerves 
Pearson Correlation 
1 
.877^{**} 
Sig. (1tailed) 
.002 

Sum of Squares and Crossproducts 
6353.893 
685.428 

Covariance 
907.699 
97.918 

N 
8 
8 

Mean % Score 
Pearson Correlation 
.877^{**} 
1 
Sig. (1tailed) 
.002 

Sum of Squares and Crossproducts 
685.428 
96.157 

Covariance 
97.918 
13.737 

N 
8 
8 
Table 1. SPSS output for correlation between percentage of cranial nerve examination items and mean percentage examination score.
Well, well, well!
Let’s do it formally: There was a negative correlation between the mean percentage examination score and the percentage of cranial nerve examination items, r = .877, n = 8, p = .05
Also, see those two asterisks beside the correlation? It doesn’t show up here, but in SPSS that leads to a footnote that reads “Correlation was significant at the 0.01 level (1tailed).”
That’s about as good as it gets. Check out figure 1 to see the scatterplot.
Once again, pretty obvious (I drew the bestfit line myself, but I think I did pretty good).
Conclusions
There’s a strong negative correlation in my OPT 113 classes since 2011 between the percentage an examination covers cranial nerve materials and the mean examination percentage score.
This means my felt need wasn’t just whistling Dixie.
—Daniel