vik09
Hello,
Hope everyone is doing great. I have a question regarding the accuracy of GMAT club quant tests, I recently gave two tests in one I scored 45 with 23 correct and 14 wrong and in the other one I scored 50 with 25 correct and 12 wrong. I am literally surprised by the results. So my question is regarding the accuracy and reliability of these tests. Can someone please shed light on this.
Many thanks in advance
Regards,
Vikas
Dear Vikas,
I'm happy to respond.
First, I'll say that I think that one consequence of the Industrial Revolution and the modern computer age is that real human being confuse themselves for precisely designed machines. Human beings are messy --- we have all kinds of rhythms and cycles, depending what you eat, how you sleep, what you dream about, any minor colds or infections your body is fighting, your love life, your emotional ups & downs, etc. etc. If you took the GMAT every day for a month (not that I recommend doing so) you would see wild fluctuations in your score day by day. We are far more variable creatures than most people appreciate.
So, does the difference of 50 vs. 45 say more about the GC Quant tests, or more about inherent human variability? That's hard to say. I can say that the GMAT Quant tests are a collection of excellent questions. I don't know, but I suspect that Bunuel had a strong hand in designing the math, and he's a very intelligent individual. I have been impressed with the individual questions I have seen. Now, having said that, how are these tests overall
as tests? It's not clear to me that any scientific design went into the structure of the tests overall --- for example, what is the distribution of easy vs. hard questions, are two different tests rigorously set at the same difficulty, etc. etc. You see, the GMAT itself is a rigorously designed test at every level. The GC Quant Tests are certainly a collection of excellent problems, but a well-designed tests is much more than a collection of excellent problems. I just don't know how much planning and research went into the design of these tests, but I suspect that there's not a major amount of psychometric data underlying their design.
I would not be inclined to attach tremendous significance to this score difference. I would say --- learn deeply from your mistakes and focus on the future.
Does all this make sense?
Mike