RISING STAR ! The ultimate source to ace your NYPD Sergeant, Lieutenant, and Captain Exam Visit www.RisingStarPromotion.com to subscribe to our mailing list and get info on the next Sgt, Lt. or Captain Exam!
So everyone is bitching that they did bad on this test. So if 23 got A. And 19 got B. And 6 got C. Whose to say that the smart people weren't the ones to figure it out and pick C and that's the true answer
I understand the mechanics of the answer key, but the test wasn't straight forward. The answer key may have worked in the past with an "acronym test", but this test was different.... I can see that key being off by 10 points +.
I wouldn't call the 2013 CAPT test an acronym test and the Gamblor was once again right on point, as he was for the 2011 Lt test (I personally was off by about 3 with his key). It's the best option available until the protest key so I'd take it for now.
I wouldn't call the 2013 CAPT test an acronym test and the Gamblor was once again right on point, as he was for the 2011 Lt test (I personally was off by about 3 with his key). It's the best option available until the protest key so I'd take it for now.
So everyone is bitching that they did bad on this test. So if 23 got A. And 19 got B. And 6 got C. Whose to say that the smart people weren't the ones to figure it out and pick C and that's the true answer
You are a pot stirrer. No one is forcing you to accept Gamblor's key as gospel. The man has been consistently right with several versions of promotional exams to within a few negligible percentage points. If you don't like it, don't use or take it into account. As above poster stated; it's what we have until the protest session. I wanna see what you have contributed so far other than trolling for questions on this exam that you have yet to take.
It is obvious Beast Lesnar failed and is not too bright.
I don't like my results either, but you can't ignore facts of mathematics. I bet you never took calculus, statistics and/complex algebra at a college level. It is nonsense calling something nonesense when you know nothing of the subject!
This answer key is a great gauge to see how well you did compared to others. I just don't understand how the "Proposed Score" can be so accurate +- 3/4 points when we have no clue the research questions, throw outs and double answers. I know I'm not the only one who is thinking this and vented In a post. I think it's a great job and gauge tool. For me it's a proposed score of 65 but a 75 for the other score which is driving me nuts. I studied my butt off
This answer key is a great gauge to see how well you did compared to others. I just don't understand how the "Proposed Score" can be so accurate +- 3/4 points when we have no clue the research questions, throw outs and double answers. I know I'm not the only one who is thinking this and vented In a post. I think it's a great job and gauge tool. For me it's a proposed score of 65 but a 75 for the other score which is driving me nuts. I studied my butt off
I totally agree with you, there is no way to predict a "Proposed Score"... if the test was a 100 question test, then maybe I could see a "Proposed Score". However, I can't see how they can include the 20 research questions into the "Proposed Score"...
I wouldn't go crazy with the "potential score"! That's just the percentile out of 120. For example, you got 60 correct/matched answers to the popular answer key, meaning your "potential score" is 50, as in 50%. More emphasis should be put on the total correct/matched choices. Anyway, unless you got 30 or less wrongs, you should be ****tin brick... with these research questions in play... never know...never know...
If people with over a 70% are only ones safe the list will be like 150-200 people. There has to be a decent amount of double answers or throwouts. Not to mention a few hard test questions suddenly becoming research ones
All Gamblor's key does is give you an approximate range of how you did compared to the rest of your fellow test takers. It was fairly accurate for me and many of the others that took the 2013 Sgt's test and my "mostly likely" score ended up being only off by 1.
If you're looking for your absolute answer, just wait until the protest. If you want an approximation on how well you did compared to others who've studied, then use gamblor's key.
The key is off here and there....for some reason people here refuse to accept letter b is wrong for question 120....it's been two days look up the damn procedure you will see the second part if clearly wrong.....
-- Edited by AhouseRocker on Monday 20th of April 2015 10:27:24 PM
I find myself explaining how this key works over and over. I'll try it again, at the very least, to dispel the misinformation that is being thrown out here.
The key does two things in principle:
1) It attempts to determine what the correct answers are based on essentially "crowd sourcing" the answers.
I breakdown the answers into three categories of confidence (high, moderate, and low). High is obvious: most people chose this answer so it is likely the correct one. Historically, 2 or 3 of these will still be wrong, but a vast majority of them are correct. Moderate is a middle ground. It means that, while there is still one answer that is clearly ahead of the others, there is another one that is reasonably close, at least enough that it should be considered. Mathematically, though, I do not treat these any different than the ones labeled as high. The low confidence ones are where it gets interesting. These are the situations where two or more answers are very close and it could swing either way. In some of these instances, particularly when the two answers are incredibly close, there is some logic that adjusts the potential score to accommodate it.
2) It estimates your score by looking at how many questions it thinks you have wrong, and making a statistical estimation on how many of those should be research questions.
In substance, with a 120 question test, 1/6th of the exam is research and does not count. So, in an example, if you have 30 questions wrong, 1/6th of them will likely be research (here meaning 5 questions). That would put you at having only 25 wrong and leave you with a score of 75. I also have to factor in the potential doubles (these are the low confidence answers that are very, very close, not necessarily all of them). It will give you 50% of the points the total number of doubles where you have at least one of the possible answers.
Now, in the above example, could you have 15 of your 30 wrong answers as research? Of course, but that is highly improbable. It works like a bell curve. The potential score it displays is the center of that curve. +/- 3 is about 2 standard deviations from the center, which is about the most you would see absent being one of the very rare few that are on the extreme ends of the curve.
All of this is dependent on the accuracy of the key, which historically is between 90 and 95%. Yes, this test is controversial, but not more so than the 2011 Lieutenants exam and the 2013 Captains exam, both of which I took and ran this spreadsheet for, and the accuracy for both fall within this range.
At the end of the day, the score that is estimated here is something that tends to be fairly accurate to what you will have in about 5 weeks when the official key is given at the protest session. Doubles and throwouts, if any, will improve your score beyond what is estimated here. However, if the sheet says you have a 75 or higher, then I would venture to guess that the odds of you passing are well over 90%. Conversely, if the sheet puts you lower than a 65, you will likely need some good protests and commensurate throwouts to push you over the 70 mark.
2300 hundred people signed up for the test, lets say 1000-1300 studied hardcore and half of them even harder. as of now this list consists of 90 answer keys. thats not enough, if hundreds of us participated then we could get better understanding.
2300 hundred people signed up for the test, lets say 1000-1300 studied hardcore and half of them even harder. as of now this list consists of 90 answer keys. thats not enough, if hundreds of us participated then we could get better understanding.
That's the most critical aspects so far. Getting people to submit keys. Some people refuse to hand em in to Gamblor.
-- Edited by Su Madre 24 on Monday 20th of April 2015 10:47:47 PM
I find myself explaining how this key works over and over. I'll try it again, at the very least, to dispel the misinformation that is being thrown out here.
The key does two things in principle: 1) It attempts to determine what the correct answers are based on essentially "crowd sourcing" the answers.
I breakdown the answers into three categories of confidence (high, moderate, and low). High is obvious: most people chose this answer so it is likely the correct one. Historically, 2 or 3 of these will still be wrong, but a vast majority of them are correct. Moderate is a middle ground. It means that, while there is still one answer that is clearly ahead of the others, there is another one that is reasonably close, at least enough that it should be considered. Mathematically, though, I do not treat these any different than the ones labeled as high. The low confidence ones are where it gets interesting. These are the situations where two or more answers are very close and it could swing either way. In some of these instances, particularly when the two answers are incredibly close, there is some logic that adjusts the potential score to accommodate it.
2) It estimates your score by looking at how many questions it thinks you have wrong, and making a statistical estimation on how many of those should be research questions.
In substance, with a 120 question test, 1/6th of the exam is research and does not count. So, in an example, if you have 30 questions wrong, 1/6th of them will likely be research (here meaning 5 questions). That would put you at having only 25 wrong and leave you with a score of 75. I also have to factor in the potential doubles (these are the low confidence answers that are very, very close, not necessarily all of them). It will give you 50% of the points the total number of doubles where you have at least one of the possible answers.
Now, in the above example, could you have 15 of your 30 wrong answers as research? Of course, but that is highly improbable. It works like a bell curve. The potential score it displays is the center of that curve. +/- 3 is about 2 standard deviations from the center, which is about the most you would see absent being one of the very rare few that are on the extreme ends of the curve.
All of this is dependent on the accuracy of the key, which historically is between 90 and 95%. Yes, this test is controversial, but not more so than the 2011 Lieutenants exam and the 2013 Captains exam, both of which I took and ran this spreadsheet for, and the accuracy for both fall within this range.
At the end of the day, the score that is estimated here is something that tends to be fairly accurate to what you will have in about 5 weeks when the official key is given at the protest session. Doubles and throwouts, if any, will improve your score beyond what is estimated here. However, if the sheet says you have a 75 or higher, then I would venture to guess that the odds of you passing are well over 90%. Conversely, if the sheet puts you lower than a 65, you will likely need some good protests and commensurate throwouts to push you over the 70 mark.
Thanks for explaining it in full details,...hopefully it will bring some closure to some on here,..question, where would high 60s (e.g, 67%,68%,69%) put you?
some people just don't want to bother and rather wait till the end of may. with 20 throw outs, though factored in, the numbers could sway either way. i know couple of guys who were 58 and 59 on 2011 test and ended up with 70/72. and there is another one who needed 3 points to make the list and only got 2.
as i look at my answer sheet i noticed an interesting pattern. forget about in-basket, but starting from 24 till 120. i have columns like from 60-66 all wrong. then another column lets say 80-85 all wrong.
Same here.. After both versions I am still at a 64. The only thing is that I don't know how to read the rest of the boxes of the spreadsheet. Don't know what everything else means. I attached a picture of it
Same here.. After both versions I am still at a 64. The only thing is that I don't know how to read the rest of the boxes of the spreadsheet. Don't know what everything else means. I attached a picture of it
So youd most likey need 10-12 research questions in your favor
Same here.. After both versions I am still at a 64. The only thing is that I don't know how to read the rest of the boxes of the spreadsheet. Don't know what everything else means. I attached a picture of it
So youd most likey need 10-12 research questions in your favor
What about my answer key (got a 69%),...not sure how to read this,...thanks for your help
Same here.. After both versions I am still at a 64. The only thing is that I don't know how to read the rest of the boxes of the spreadsheet. Don't know what everything else means. I attached a picture of it
So youd most likey need 10-12 research questions in your favor
What about my answer key (got a 69%),...not sure how to read this,...thanks for your help
Well i have a 70%. We need 6-7 research to go in our favor
Same here.. After both versions I am still at a 64. The only thing is that I don't know how to read the rest of the boxes of the spreadsheet. Don't know what everything else means. I attached a picture of it
So youd most likey need 10-12 research questions in your favor
What about my answer key (got a 69%),...not sure how to read this,...thanks for your help
Well i have a 70%. We need 6-7 research to go in our favor
I have 74 right out of 120, but I have a 60%. The formula doesn't make sense. I guess is for comfort or an idea where ppl stand, but it doesn't make much sense.
I have 74 right out of 120, but I have a 60%. The formula doesn't make sense. I guess is for comfort or an idea where ppl stand, but it doesn't make much sense.
It's weighted for a 120 question test. Each question is worth .83 % give or take. 74 x .83 is around a 60-61%. so when people see 65 per say they don't need 5 questions to go there way it's more like 6-7.
