2 ms·
So essentially, the math questions are being revised to involve more reading comprehension and less actual math? I can't help noticing that most of those quest
by logfromblammo 11y ago
So essentially, the math questions are being revised to involve more reading comprehension and less actual math?
I can't help noticing that most of those questions contained at least one bit of completely irrelevant information. Each of those questions could be worded less ambiguously and more concisely--to test math skill, rather than reading skill. You can trick more students into getting a wrong answer by inserting red herrings and ambiguous sentences into the question, but in doing so, you are no longer strictly testing their math skills. There is no particular need to ask these questions as story paragraphs rather than lists of known facts.
The following are the same math problems, with dramatically simplified wording. If you find them easier to understand than the originals, that is because I have removed the part of the problem that involves translating the text into facts relevant to the question.
1. Armand spent 5 hours texting at a rate of m texts/hour.
Tyrone spent 4 hours texting at a rate of n texts/hour.
What expression represents the total number of texts sent?
2. P = 108 - 23d models the weekly workload of an employee.
P is the number of work units remaining for this week.
d is the number of days the employee has spent working this week.
What does the value 108 represent?
3. h = 3a + 28.6 models height (h) in inches at an age (a) in years.
How much taller does the modeled subject grow per year?
4. A vendor sells salads for $6.50 and drinks for $2.00.
The vendor sold 209 items for a total of $836.50.
How many of the items were salads?
5. Cultivar A of a fruit tree produces 20% more fruit than cultivar B.
Cultivar A produces 144 fruits.
How many fruits does cultivar B produce?
- rtkwe 11y agoBy including extra information testers can't just be coached or memorize apply this formula when you see this kind of question and the test shows they're actually understanding more real world problems and how to take what they know and how to apply it to a problem where everything hasn't been stripped away for you. If students are tricked by the simple extraneous information it shows there's a gap in their ability to take information and use math to answer a question about it. That's an important thing to evaluate.
- logfromblammo 11y agoBear in mind that the SAT tests literacy skills and math skills separately. Reading comprehension is unequivocally something that should be included in the reading and writing portion of the test, rather than the math portion. Read some of the comments elsewhere in this discussion. There are some people that noticed how "209 salads and drinks" could be interpreted as "209 salads, and an unspecified quantity of drinks" or as "209 * ( a salad or a drink )". That has nothing to do with one's ability to take a system of two equations and two variables and solve it for either variable, and everything to do with poor question design. Professional pollsters work very hard at tuning the exact wording of poll questions, to eliminate hidden biases in the question itself and to better reveal what the participants themselves think. These samples have non-mathematical bias embedded in the questions.
- rtkwe 11y agoThis is different than the reading comprehension tested in the other sections though. It's straightforward to understand what is being said what the extra information tests is if you can determine what information is important to the question you're being asked to solve. Being able to figure that out is really important to practically applying math and most fields one might get into post HS. Also both of those interpretations are ruled out by a check of the math. 209 salads and 209 salad + drink combos are both more profit than the food van made so if a student answers those they failed to check their work at all. Also the phrasing there is the same as student would have seen tons of times before taking the test. It's not exactly an extremely exotic way to write it out. Really this is one of the more straightforward questions of the batch presented.
- logfromblammo 11y agoSo what percentage of the math score do you think should depend on reading natural language sentences, and what percentage should depend on manipulating equations, numbers, and symbols? I suppose if the test really wanted to be practical, there would be a few questions about student loans, tuition rate increases, and post-graduation employment rates. My core argument is that making the questions wordier do not make them better questions for a math test. Most of the practical math I have ever done (anecdotal) has not started as a paragraph of prose. It most often begins as a chart, table, technical drawing, or list of numbers. If I ever need to translate between English and math, it has almost always been to document the mathematical model, rather than implement a model from a written description. I therefore cannot fathom why mathematical pegagogy has such an infatuation with story problems. Real life just throws a pile of numbers at you, maybe gives you some column headings, and expects you to make sense of it yourself. You want practical? Put up an image of a medical bill next to the insurance statement and ask the student how much they owe. Good luck with parsing that "off vst l3, 2MT38K, diag" and the "negtd max" and the "less copay".