Over the course of a few weeks in December and January I received a number of practice test results from students preparing to take the GMAT. As the fourth or fifth result came my way I noticed that all of these students happened to be struggling to break the 35-point barrier on the verbal section. The realization gave me pause (I mean, I’m a test prep geek, why wouldn’t it?). Over the years I’d come to realize that many students at one point or another struggle to maximize their verbal score, and more importantly, these struggles were usually tied to reading comprehension.
I wanted to see if I could quantify this a bit for folks, so I started by going back through those student’s results to see if there were any trends. Sure enough, an initial investigation revealed what looked to be a pretty strong correlation between RC and Verbal subscore.
My interest piqued, I started to dig a little deeper. Bear with me, there’s gonna be a few numbers with this. Numbers? you ask. In a blogpost about verbal? Yup. Think of it like an integrated reasoning blog post.
So, with the help of one of our interns (big shout out to Cannie L!) we tabulated the results of about 220 GMATPrep, and broke out their respective accuracy %s on CR, RC, and SC. After grappling MMA-style with those percentages (grappling I’m not going to bore you with – let’s just say it was ugly…almost as ugly as this knockout), were able to see some pretty clear – and surprising – trendlines. Let’s take a look at our sample:
Total GMATPrep Results: 216
# with Verbal 35 or greater: 92
# with Verbal 34 or less: 124
Essentially, I aggregated and compared these results to see if, in fact, there was some strong correlation between RC accuracy and Verbal score. Turns out, there kind of is. Of the 92 results that scored 35 or above on the verbal, only 3 of those had managed that score with an RC-accuracy below 71%. It’s important to note that we just studied results where test-takers answered all of the Verbal questions. Reading comprehension totals 14 questions of the verbal, and 71% is 10 out of 14 questions correct. So basically, only three of the 92 results had fewer than 10/14 of the RC questions correct. That’s approximately 3.2% of the sample. 96.8% of the results that were 35 or higher had 10 or more RC correct! The breakdown is illustrated here:
