Cut Score...

Professional Engineer & PE Exam Forum

Help Support Professional Engineer & PE Exam Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

JoeysVee

MASTER SPAMMER!
Joined
Apr 25, 2006
Messages
662
Reaction score
0
Location
South Carolina
Warning...if you don't like cut score threads you may want to skip this one... :beerchug:

Do you think it's ever gone from a 56 to a 52? In other words if my score is indeed a 52, do I have any shot that the cut score is also a 52 or is that too low? I know nobody knows but do you thionk it has ever been down around a 52?

Thanks!

 
I think it is unlikely. but not impossible. 52 would be 65%. You know, I passed PE on 4th attempt. The first 3 tries, I wanted to convince myself I could have passed. The 4th time, I knew I could have passed.

 
Warning...if you don't like cut score threads you may want to skip this one... :beerchug:
Do you think it's ever gone from a 56 to a 52? In other words if my score is indeed a 52, do I have any shot that the cut score is also a 52 or is that too low? I know nobody knows but do you thionk it has ever been down around a 52?

Thanks!
I think it might have gotten that low - for a hard test. Also, you may have done better than you think.

 
Last edited by a moderator:
I took SE1 in April and didn't pass. (pass rate: 48%)

Based on the report, I got around 45~46 and my score was 65.

My guess for the cut score at that time would be 50.

 
Just guess and wishes the cut score may around 44-48 (55% to 60%)

 
I heard that it is 50% correct to pass FE, if so it may be close 50% correct to pass for PE

 
I think the cut score is a 56 but is often lowered due to several things like...

- If large groups of people miss a problem they will throw it out and lower the cut score by 1

- If they have an error they will lower the cut score by 1 per flawed question. Every sample exam I have that the board has published has at least 2 or 3 errors so this happens quite often...at least this is my guess.

- If the committees determine a problem was too difficult they will throw it out. My guess is this is typically only on the new exam questions they throw in just to see how people do on them.

So they start at 56 and come down from there….my guess is the chances of the cut score being a 56 is higher than a 55 and the chance the cut score is a 55 is higher than a 54 and so on.

These are just my :2cents:

 
Last edited by a moderator:
It's a breadth and depth exam. This means it is probably an xy graph. Breadth section is horizontal and depth section is vertical.

If you get 100% correct on both you fill the xy universe. If you get 70% correct on morning and 70 correct on afternoon, you are at 0.7 x 0.7 = 0.49 = 49 percent of the full area. Which is about 50% of the full area and passing.

Now, they may allow you to be a bit below on the morning if you are bit higher on the afternoon test, but generally all passers must fit the general mold of 0.7X0.7 to 1x1 area.

 
I know everybody has theories, but NCEES tells you how they score the exam. I assume nobody has any inside information.

http://www.ncees.org/exams/scoring/scoring_method.php#curve

You can read up on equating on google but from what I understand it means using a group of questions that they have a fairly good idea are representative. Over repeat exam administrations they get a baseline on how well a group of examinees answer this subset of questions. From this they have a method to tell whether any particular exam is harder or easier than any other particular exam. I don't know all the details, and don't know exactly how any of these other theories fit into this.

They may very well start with the 56 correct questions, but nobody really knows.

By the way - not every exam is breadth and depth. The electrical exam isn't.

 
Last edited by a moderator:
Just to add, from all the people who have posted diagnostics on here over the years my GUESS is that the cutscore is generally somewhere between 54 and 56 points. I assume in rare instances it might go down to 52 or 53, and in even rarer instances (I think I saw one such posting) I saw a person fail with an esitmated 56 correct.

Once again this is guesswork based on anecdotal evidence. I have no voodoo or ouija board to figure this out.

 
Last edited by a moderator:
It's a breadth and depth exam. This means it is probably an xy graph. Breadth section is horizontal and depth section is vertical.
If you get 100% correct on both you fill the xy universe. If you get 70% correct on morning and 70 correct on afternoon, you are at 0.7 x 0.7 = 0.49 = 49 percent of the full area. Which is about 50% of the full area and passing.

Now, they may allow you to be a bit below on the morning if you are bit higher on the afternoon test, but generally all passers must fit the general mold of 0.7X0.7 to 1x1 area.
Not sure what you are talking about... there has never been any indication the passing process cares how you do in the depth vs. the breadth. If 56 is the magic number (and who the hell knows if it is?), you can get 40/40 in the morning and 16/40 in the afternoon, 16/40 in the morning and 40/40 in the afternoon, or 28/40 in the morning and 28/40 in the afternoon... it doesn't matter.

 
What I don't understand about the NCEES process is that they claim to have a huge bank of exam questions that the guard like the gold vault at Ft. Knox. They guard these questions so that they can still be used on future exams without everybody having seen them in a practice exam. Yet they still have to evaluate every question for fairness every year? It seems like they have been doing this long enough to be able to select from their bank of questions in such a way that the exam has a consistent level of difficulty every time, and no "curving" is necessary. If they did that, then they could give the exam on computers and give you the results instantly...or at least just run the scantron sheets through a machine and mail out the results in a week.

 
What I don't understand about the NCEES process is that they claim to have a huge bank of exam questions that the guard like the gold vault at Ft. Knox. They guard these questions so that they can still be used on future exams without everybody having seen them in a practice exam. Yet they still have to evaluate every question for fairness every year? It seems like they have been doing this long enough to be able to select from their bank of questions in such a way that the exam has a consistent level of difficulty every time, and no "curving" is necessary. If they did that, then they could give the exam on computers and give you the results instantly...or at least just run the scantron sheets through a machine and mail out the results in a week.
I think the theory (or at least the general theory of equating) is that every exam has a combination of old and new questions. For the old questions they have statistics on how well people in the past have answered those questions. That way if they get a particularly low score on the exam's new questions they can look at how the group did on the known question pool and tell if the problem is the exam or a particularly poor pool of examinees. Plus I'm sure they look at people's complaints about questions.

Also, they want to release all the exams at the same time, so if they finish the electricals first I don't think they want to release those until the civils are done. So they give themselves a long time.

But I agree all that is probably just an excuse and they could get the answers somewhat more quickly.

 
This is a good description of Equating! It's worth the read.

http://en.wikipedia.org/wiki/Equating

:eek:ld-025:

In classical test theory, mean equating simply adjusts the distribution of scores so that the mean of one form is comparable to the mean of the other form. While mean equating is attractive because of its simplicity, it lacks flexibility, namely accounting for the possibility that the standard deviations of the forms differ. [1]

Linear equating adjusts so that the two forms have a comparable mean and standard deviation. There are several types of linear equating that differ in the assumptions and mathematics used to estimate parameters. The Tucker and Levine Observed Score methods estimate the relationship between observed scores on the two forms, while the Levine True Score method estimates the relationship between true scores on the two forms.[1]

Equipercentile equating determines the equating relationship as one where a score could have an equivalent percentile on either form. This relationship can be nonlinear.

Unlike with item response theory, equating based on classical test theory is somewhat distinct from scaling. Equating is a raw-to-raw transformation in that it estimates a raw score on Form B that is equivalent to each raw score on the base Form A. Any scaling transformation used is then applied on top of, or with, the equating.

 
Last edited by a moderator:
What about the if the test is to easy ---is the score higher than 56? I would seem logical that if they can lower it by a point or two could they not also raise it by a point or two??

 
Just to add, from all the people who have posted diagnostics on here over the years my GUESS is that the cutscore is generally somewhere between 54 and 56 points. I assume in rare instances it might go down to 52 or 53, and in even rarer instances (I think I saw one such posting) I saw a person fail with an esitmated 56 correct.
Once again this is guesswork based on anecdotal evidence. I have no voodoo or ouija board to figure this out.

54-56 cut score is for Civil. I don't think that Electical or Mechanical share the same cut score.

 

Latest posts

Back
Top