Reviewing a Practice Test Should Take LONGER Than Taking It
Saturday morning. You sit down at the kitchen table, run a full practice Executive Assessment, and 90 minutes later a number pops up. Say it's a 148. You stare at it and feel something: relief, dread, a little of both. You click through the questions you got wrong, nod at a few explanations, close the laptop and go make lunch. Ten minutes, tops.
That is NOT a review. That's reading the headline and skipping the article.
Here's the rule: a real review of a practice test should take LONGER than the test itself. More than 90 minutes. I'll bet your last review took less than 20. Be honest with yourself. Next time, look at the clock when you start and when you stop.
The score is the headline, not the story
The Executive Assessment is 90 minutes long, split into three 30 minute sections. Integrated Reasoning (charts, tables and multi-part questions) has 12 questions. Verbal Reasoning has 14. Quantitative Reasoning has 14. Each section is scored 0 to 20, and the total runs from 100 to 200, with all three sections counting the same. The test is also adaptive section by section, which means how you perform shapes the difficulty of what you see. You can check all of that on GMAC's page on the test structure.
So that one number, 148, is sitting on top of 40 separate decisions you made. Two people can earn the exact same score for totally different reasons. Say one lost most of her points to arithmetic slips in Quant. Say another ran out of time in Integrated Reasoning and guessed on the last four. Same headline. Different story. Completely different fix.
A practice test is a thermometer, not medicine. It tells you the temperature. It does not tell you why you have a fever, and it sure doesn't cure it. The review is where the diagnosis happens.
What a real review actually looks like
You go through EVERY question. Not just the wrong ones. For each one, you cover the explanation, redo the question on paper from scratch, and then tag it with one of these five labels:
| Tag | What it means | What you do about it |
|---|---|---|
| Right and solid | Correct, confident, reasonable time | Move on |
| Right but lucky | You guessed between two, or can't explain why it's right | Treat it as WRONG |
| Right but slow | Correct, but way over pace | Find the faster route |
| Wrong and silly | You knew how, then slipped: a misread, a sign error, bad arithmetic | Find the exact moment it broke |
| Wrong and didn't know | A rule or method you're missing | Learn it, then redo similar questions |
Quick note on pace. In Quant and Verbal you get 30 minutes for 14 questions, a little over 2 minutes each. In Integrated Reasoning it's 30 minutes for 12, so 2.5 minutes each. Those are AVERAGES, not a stopwatch per question. Some questions should take 40 seconds. Some deserve 3 minutes.
Now stop reading and think about this for a second. Which of those five piles do you think holds most of your missing points?
Most people guess the last one, the hard stuff they didn't know. Then they actually do the review and find the points were bleeding out of the middle three: the lucky guesses, the slow wins that stole time from later questions, and the silly slips. The "right but lucky" pile is the sneakiest. The score gave you credit for those. On Test Day, luck doesn't show up on schedule.
Watch a review happen
Here's a Quant question in the Data Sufficiency format. Data Sufficiency asks whether the facts you're given are ENOUGH to answer the question. You don't need the answer itself, just whether you could get it.
Now, here's the part that matters. Say you picked D on your practice test. The lazy review note is "Wrong. Answer is B." That note will not make you better. The real review note sounds like this: "Assumed statement (1) gave me information. Didn't test values. Next time: on any odd or even statement, test one odd and one even number on paper before deciding."
THAT sentence is the whole point. It names the behavior, not just the result, and it gives you a rule you can practice. Write one of those for every question you tag as anything other than "right and solid." That's why the review takes longer than the test.
Five ways people fake a review
Only looking at the wrong answers. You skip the right ones because they're "done." But your lucky guesses and slow wins are hiding in there.
INSTEAD: Review all 40 questions. Every one gets a tag.
Reading the explanation first. Nodding along to a solution FEELS like learning. It isn't. Of course it makes sense when someone else does it for you.
INSTEAD: Redo the question cold, on paper, before you look at anything. Then compare your path to the explanation.
Calling it a "timing problem." Timing is a symptom, never the disease. Something CAUSED those extra minutes.
INSTEAD: Find your three slowest questions in each section and figure out what you were doing during the extra time. Re-reading the prompt? Working in your head? Taking the long way when testing the answers would have been faster? When 90 minutes is all you get, every wasted minute has a name.
Ignoring your scratch paper. Your notes are the black box recording of the crash.
INSTEAD: Put your scratch work next to each question as you review. If there's almost nothing on the page for a question you missed, you just found the cause.
Taking the next practice test before finishing the last review. Another test is just another headline. It won't fix anything you haven't diagnosed.
INSTEAD: No new practice test until every question from the last one is tagged and you've drilled the fixes. For most people that means a week between tests, not a day.
Do this today
Pull up your most recent practice test. Pick ONE section. Set a timer for 30 minutes, the same time the test gave you for that section. Redo every question on paper before looking at any explanation, tag each one, and write one behavior sentence for every question that isn't "right and solid."
If you finish before the timer goes off, you went too fast. Go back. Then count your piles and tell me where your points REALLY went.
