MyMathLab Grading Error: "When You See It...

[ComplexVar89]

Okay, so, this is over two years old, and I don't need help with anything, so I'm going to post this here (instead of one of the other forums) for laughs.

The lovely computer software my former college used for all math classes back in 2012, MyMathLab, made an error in grading a question. I, of course, immediately took a screenshot and posted it to Facebook for my more mathematically-inclined brethren to scoff at.

When you see it...

https://fbcdn-sphotos-e-a.akamaihd.net/hphotos-ak-xaf1/v/t1.0-9/487508_4725981957330_747481942_n.jpg?oh=462e66c7d4fe208f76e59ce6bdfdffe7&oe=551F6D8B&__gda__=1425932204_838ba47d02047e4771bcd15ea257b35b

 

[Vanadium 50]

I don't understand. Plug in 0, and you get -24, not +24.

 

[ComplexVar89]

I multiplied my result by -1 to rid it of negative leading coefficient (thus changing all of the signs, which I did correctly,) which is allowable by the rules of algebra, but the software wasn't able to tell that. Normally, when it detects an equivalent answer that is not of the form they want, it will say so, but this time it didn't. It was just marked wrong.

 

[ComplexVar89]

I didn't explain my reasoning in the beginning (which has been independently verified both back then, and as recently as today, by a number of people who could hold their own here easily, including my professor at the time,) because I thought people might find it insulting if I did explain it. I guess I really misjudged the situation.

This is kind of why I'm a 20-year supporter, but I don't post much--I feel stupid around so many bright people. Perhaps I should be encouraged that I saw something on first inspection a staff member here didn't catch, though. Perhaps I've been looking at things the wrong way.

As Silvanus Thompson wrote in Calculus Made Easy (which I love for his wit more than anything):

What one fool can do, another can.

Maybe, just maybe, I'm one of the fools who can, where I've long thought I was one of the fools who couldn't.

Should I have multiplied that expression through by -1 and risked ridicule from a bit of software that doesn't know any better, just because I didn't like the way the original answer looked? I guess not. But because I changed all of the signs accordingly, I was within the rules, and the computer should have been able to verify that and at least complain about the answer being equivalent but not in the expected form, but it didn't do that. Did the people who programmed the software think that the possibility someone might dislike the form of the answer they preferred to be unsavory too remote to take into consideration? I wonder...

Anyway, that's a curiosity I unearthed on Facebook today from a couple of years back, and I thought I'd share it on here.

 

[Bandersnatch]

I multiplied my result by -1 to rid it of negative leading coefficient (thus changing all of the signs, which I did correctly,) which is allowable by the rules of algebra,

If you do that, you get -fg(x), rather than fg(x) that is asked for.
Is there something I'm missing?

 

[ComplexVar89]

Perhaps that's what it was, @Bandersnatch. Although, how would anyone know that, since the program didn't say anything, period? Like, could they seriously not expect someone might look at that (ugly, in my opinion) negative leading exponent and decide to get rid of it? That, in and of itself wasn't wrong, but in when you explain it that way, in the entire context of the problem, I can see why it would kick the answer I gave back at me. Legal algebraic operation, but rendering a function that wasn't asked for.

What bothers me about this whole thing, is that I didn't take this out of context (I showed the exact same screenshot) then, or today, and no one elsewhere said anything about it, including my professor. And almost all of the people I consulted, I would consider to be of the same abilities as most of the people here. Do you think they just looked at the expressions and didn't read the whole problem? You have to remember that my mathematical sophistication and "maturity" two years ago was very much lacking. It's still lacking comparatively, but not nearly as much as it was then.

Also, I must thank you for not ripping me apart. I've kinda been afraid of that happening, too, to be honest. I can sure as heck admit when I'm wrong, and it appears I didn't have all the pieces to the puzzle, and so I was wrong in this case. Silver lining: I now know why I was wrong, which is even more important than just knowing that I was wrong. Thanks!

 

[Bandersnatch]

No worries. If I had a penny for each time I missed something, I'd never have to work anymore.

I have no insight into other people's minds, so I can't tell you why they didn't say anything.

Anyway, just remember that the operation of multiplying by -1(or whatever, for that matter) is legal only if you perform it on both sides of the equation.

 

[Vanadium 50]

since the program didn't say anything, period?

It is easy to write a program to compare two expressions and see if they are equal. It is very very difficult to write a program that can tell you what error you made.

A good thing to do with algebraic manipulation is to test by plugging in easy numbers like 0 or 1. If they do not match, you know you made a mistake.

 

[mfb]

I don't see the point. The answer was wrong, the program told you the answer was wrong.

If the problem is "What is 4-7?" and you multiply the correct answer of -3 with -1 to get a nicer 3, it is wrong in the same way.

It is wrong in a special way, and apparently the program is not designed to recognize this special way of being wrong. So what?