GMAT Focus Algorithm - importance of the first 6 Questions is GONE

­
Sure. Here is a whole long discussion about it in Dec 2023 with links to YouTube videos where contradictory statements were said. This in turn triggered me checking with the GMAC team about what I thought would be a simple question, which lead to several not very clear answers and then finally they said that that they do not disucss specifics of their algorithm.  Let me know if you have further thoughts/ideas about it:
https://gmatclub.com/forum/gmat-focus-e ... l#p3312759­

­Hi wolfof6thstreet,

The GMAT is question-adaptive. This is a fact, and the carry-over that you mentioned between sections isn't actually related to the issue at all (you're looking at the term section-adaptive in a new, non-standard way).

I've tried to address this issue here: Myth Buster: Is the GMAT Section-Adaptive?. The short version is that a question-adaptive test chooses each individual question, whereas a section-adaptive test chooses entire sections. That is, the terms question-adaptive and section-adaptive do not refer to the basis on which the algorithm adapts.

Hey bud, not sure why you keep debating semantics on the term section-adaptive, but it really is a mute point and makes it seem like you’re just trying to feed traffic to your website. If anything, you’re just confusing people that come to this thread and confuse your opposition as validation for the contrary.

The fact of the matter is that the only section that isn’t influenced by other sections is the 1st section. I believe GMAC has flat out said this at one point. Whether this aligns with your definition of section-adaptive relative to how it is used in the GRE is besides the point. I think I speak for most when people are more focused on whether or not there is spillover between sections.

If you are so focused on creating a perfect term to describe how the GMAT uses performance in prior sections in the next section, be my guest, but it really is a pointless hill to die on.

Posted from my mobile device

­Hi bb,

The post that you linked to is among the ones that (grossly) misinterpret the statements made by GMAC personnel. It has already been edited 63 times, but the latest version of that post is still a sensationalist mischaracterisation of the information GMAC has shared with us.

To be clear, I don't think we should punish GMAC for actually making an effort to communicate with the test prep industry. They've tried to be as clear and consistent as possible: The GMAT is a question-adaptive test, and performance on a section influences the difficulty level of the first question on the next section. These are not contradictory statements.

I can't comment on your correspondence with GMAC, or on their decision not to discuss certain aspects of the algorithm with external parties. However, it's worth noting that I have been in touch with them as well, and they have provided clear answers to my questions.­

­Hi wolfof6thstreet,

This may be my hill, but you're welcome to be here too.

1. This is not a moot point, at least in my opinion. Some forum members either don't know what section-adaptive means, or they are creating a new, non-standard definition of the term that they then use to claim that GMAC is being contradictory. This is unfair. GMAC has been far more willing to engage with us than some other testing organisations. We can choose to use that against them, or we can appreciate it. Having been in the industry for some time, I appreciate their willingness to engage with us.

2. I would not mind more visitors at all! However, I did give you the short version if you don't want to read my entire post.

3. Again, it is a fact that the GMAT is question-adaptive. Any contrary opinions are in fact the ones that are confusing the issue.

4. You're welcome to think that most people are more interested in whether the first question of subsequent sections is no longer of "mid-level" difficulty. I'd be just as happy reaching out to the minority of test takers who actually want the right information.

5. Again, I am not the one creating the term here. If anything, it is a few (maybe just 1-2?) other forum members who may not be very clear on what a section-adaptive test does.

Thanks for taking the time to put your thoughts down. I'd be happy to address any follow-up questions you may have.

Not to be rude, but for someone trying to sell people that you can help them on logical reasoning preparation, you sure don’t represent yourself well.

I do find it funny that you seem to be stuck on this premise that GMAT is question-adaptive. No one is debating that and for you to use it in your argument is not helping your case and makes it seem like you can’t structure a proper argument. Kind of ironic since reasoning is supposed to be what you’re selling to people.

If you are dead set on debating the correct term to describe how the GMAT considers performance on prior sections in choosing your level on the future section… by all means continue what you’re doing. I just don’t understand how you think people care more about that more than just knowing that the test is spilling over between sections.

Again, you do you, but it really looks silly and completely against the purpose of this thread to debate irrelevant semantics.

Posted from my mobile device

Now that we've reached ad hominem, we should perhaps move further attacks on my logical ability to PM. Thank you for contributing to this discussion.

­
Great  - did your source confirm anything about the section/question adaptive aspect or algorithm?
Can you share what they said and who confirmed it? If you can't disclose the name, can you share a screenshot of their answer with the name blurred out perhaps so we have some kind of official statement on this.

In my humble opinion I think you took a position which I don't think you can defend. Not because of any of your personal qualities, but simply because facts about the algorithm have not been provided (and will not be) and you were forced to make assumptions which are only assumptions. And as you know from CR, when there is even a bit of unknown, it is easy to unravel an argument because we just don't know what small impact of previous section on the next section's starting question means. I can similarly argue that if the impact is really really minor or negligible then why the heck did they bother building this in esp. if I take quant, verbal, DI sequence but again we don't know (and don't yet have enough data in GMAT prep simulations) to answer it. I have to admit I am yet to read your article (just no time right not unfortunately). Maybe it will change my mind but until then these are my thoughts. 

PS. Let's all cool down. There is no reason to be aggressive with a person just because you think they don't agree with you. ­

­Hi bb,

I prefer not to share screenshots of private correspondence publicly without a compelling reason. I did, however, make the relevant information public here.

Again, I respect your right to hold whatever opinion you wish, including about the content or quality of my reasoning. However, I see no reason why GMAC personnel should be punished for sharing (correct!) information with us.

To anyone looking for this correct information:

1. The GMAT is question-adaptive. Your test will be adaptive even within a section.

2. Performance on a previous section is used to "slightly tweak" the starting point of the next section.

These statements are not contradictory. That is, (2) does not mean that the GMAT is somehow section-adaptive. There is no controversy here, and that's why posts that say that GMAC is flip-flopping or confusing us are incorrect. People like Manish and Seema are doing their best, and I appreciate their efforts.

3. To anyone looking for a clarification on any of my points but not looking to post publicly, please feel free to reach out to me (politely or aggressively ) on PM or at ajitesh@delv.ac.

AjiteshArun is right. I attended an online session with GMAC representative, who was explaining the GMAT FE. Someone asked the GMAC representative this question (whether a candidate's performance on one section impacts the difficulty level of the initial questions that the candidate receives on the next section).

Frankly, from what I could assess, the GMAC representative was caught slightly off-guard by this question; she definitely did not deny this, but tried to downplay it by stating that this issue is quite trivial and should not bother the candidates.

­Hi Experts -  MartyMurray @mcelroytutorin GMATCoachBen AjiteshArun vv65 Maheshkv carcass gmatophobia @educationaisle


I need help regarding time management strategy.

-My target scoe is 605-625 with Quant being my strength.

Issue - I cannot get my time per question down to 2 min mark. 

If I rush through the process - my accuracy is all over the place. 

If I go at my pace my accuracy is decent but I end up running out of time for 3-4 questions in each of the 3 section ( Q, V , DI) and end up marking them randomly. 

How will these strategies work - 

1) I take my time on each question and try to get maximum accuracy - and end up random marking last 3-4 questions. 
Pro - Don't have to worry about thinking what to skip 
Cons - Mistakes are clubbed together

2) I time box each section and then solve however many question I can solve within the timebox and then randomly mark and move on to catchup on time. Ex. First 4 questions I take X time. I take my time to solve the first 3 question and marks the 4th randomly <2 secs. Do this for all quarters.  

Pros - Mistakes will be spread out 
Cons - I will end up marking easy question as they are blind guess and loose lot of points. 

Issue is I cannot identify difficulty of a question especially on verbal and IR part of DI to skip difficult questions smartly.

Someone who have done a lot of if-then analysis or seen a lot of score reports can help.

Also where can i see others score reports on the forum @bb?­