The SAT is back. But I understand that since it is electronic, and can be scored in real time the test will change depending on how well or poorly a student is doing. In other words, easier questions will pop up in the test if the student is doing poorly and vice-versa.
In essence, each student can end up taking a different test. It is also being "dumbed-down" with certain sections left out for everybody.
"easier questions will pop up in the test if the student is doing poorly and vice-versa"
This isn't necessarily indicative of dumbing down. What may be happening here is a "staircase" algorithm that allows an online test to zero in on the score of a test taker by giving a range of questions of varying difficulties, taking the answers into account, and giving more questions around that level. It allows the test to zero in on the test taker's true level faster than if every test taker has to answer every question. If this is what the SAT does, then not every test taker will get the same number of questions, or the same set of questions, and two people who both got, say, 80% correct would end up with different scores based on the level of difficulty of the questions they were given. It sounds unfair but it is a valid scientific approoach used on some types of tests.
Thank you Michael. So, the level of difficulty would be factored into the final score? So, 2 people can get 80% correct on SAT and one would end up with 600, let's say, and the other with 700?
The dumbing down wasn't referring to the question but either the leaving out of a section, or shortening the reading.....but I can't remember.
Think of it like an eye exam. An eye doctor doesn't make everyone start at the hugest letters and work their way all the way down the chart until the letters are too small for them to read. Instead, he starts on a row somewhere in the middle of the chart, and if the person can read those letters easily, he moves to a lower row, while if the person cannot read the letters at all, he moves to a higher row. Everyone he tests in a day may attempt to read a different number of letters, and different sizes of letters, but at the end of the day the doctor will have fitted everyone with the right prescription.
Something like that. And it is the test taker's own right or wrong answers that lead to the level of difficultly of their exam as a whole. No need to give all of the easy questions to a person who is acing the hard ones, or all of the hard questions to a person who is struggling with the easier ones. But the person struggling with the eaiser ones is getting a lower score.
I haven't heard of removing sections (/topics?) from the SAT. I really hope that if that is true, it is due to validity-data, not capitalistic desires to improve its "public image" at the expense of testing integrity. Is it along the lines of reacting to the worsening of children's reading and writing abilities?
To the electronic-enabled questions. That was already standard in for example the GRE - though when I took it, you still had to come into a testing location. The psychometric idea tends to be, among other reasons, that rather than rewarding random guesses on all the visible items, if there is evidence of struggling with a certain difficulty of items, the test taker is taken down a lower difficulty avenue (which should have mechanisms for returning to higher difficulties based on performance) which also carries less score-value and potentially less items. I'm not certain that this is how the SAT is now working, but that is what the testing-precedent would be.
The SAT is back. But I understand that since it is electronic, and can be scored in real time the test will change depending on how well or poorly a student is doing. In other words, easier questions will pop up in the test if the student is doing poorly and vice-versa.
In essence, each student can end up taking a different test. It is also being "dumbed-down" with certain sections left out for everybody.
What have you all heard about these things?
"easier questions will pop up in the test if the student is doing poorly and vice-versa"
This isn't necessarily indicative of dumbing down. What may be happening here is a "staircase" algorithm that allows an online test to zero in on the score of a test taker by giving a range of questions of varying difficulties, taking the answers into account, and giving more questions around that level. It allows the test to zero in on the test taker's true level faster than if every test taker has to answer every question. If this is what the SAT does, then not every test taker will get the same number of questions, or the same set of questions, and two people who both got, say, 80% correct would end up with different scores based on the level of difficulty of the questions they were given. It sounds unfair but it is a valid scientific approoach used on some types of tests.
Thank you Michael. So, the level of difficulty would be factored into the final score? So, 2 people can get 80% correct on SAT and one would end up with 600, let's say, and the other with 700?
The dumbing down wasn't referring to the question but either the leaving out of a section, or shortening the reading.....but I can't remember.
Think of it like an eye exam. An eye doctor doesn't make everyone start at the hugest letters and work their way all the way down the chart until the letters are too small for them to read. Instead, he starts on a row somewhere in the middle of the chart, and if the person can read those letters easily, he moves to a lower row, while if the person cannot read the letters at all, he moves to a higher row. Everyone he tests in a day may attempt to read a different number of letters, and different sizes of letters, but at the end of the day the doctor will have fitted everyone with the right prescription.
Something like that. And it is the test taker's own right or wrong answers that lead to the level of difficultly of their exam as a whole. No need to give all of the easy questions to a person who is acing the hard ones, or all of the hard questions to a person who is struggling with the easier ones. But the person struggling with the eaiser ones is getting a lower score.
I haven't heard of removing sections (/topics?) from the SAT. I really hope that if that is true, it is due to validity-data, not capitalistic desires to improve its "public image" at the expense of testing integrity. Is it along the lines of reacting to the worsening of children's reading and writing abilities?
To the electronic-enabled questions. That was already standard in for example the GRE - though when I took it, you still had to come into a testing location. The psychometric idea tends to be, among other reasons, that rather than rewarding random guesses on all the visible items, if there is evidence of struggling with a certain difficulty of items, the test taker is taken down a lower difficulty avenue (which should have mechanisms for returning to higher difficulties based on performance) which also carries less score-value and potentially less items. I'm not certain that this is how the SAT is now working, but that is what the testing-precedent would be.
Hi BennyBobDixson, look at the thread with my original comment because others have responded with good helpful responses.