Slashdot Mirror
Catching Exam Cheats With a Spectrum Analyzer
angry tapir writes "Police in Taiwan have used a set of spectrum analyzers to catch at least three people suspected of cheating on an exam by monitoring them for mobile phone signals. Officers used three FSH4 analyzers specially configured by the German manufacturer Rohde & Schwarz to monitor an exam in south Taiwan for prospective government workers."
Wow, those R&S analyzers are some serious tools! I was just looking at frequency analyzers over at DealExtreme, where they have a dirt-cheap handheld model that sniffs out cellular frequencies for $60. Or they could have hung a cell jammer in the room for about a hundred. Or if they really thought they had to have the fancy gear, they probably could have hired in a contractor who would have sniffed around for maybe $300 per hour, and known what he was doing.
Was it was really worth the $40,000 they probably spent on them?
Oh, that's right. It's a government organization. Spending money is in their job description.
John
I use the FSH4 at work - nice little SA - interesting use for it.
This can't be a bad thing, if it raises the quality and character of prospective government workers...
This issue is a bit more complicated than you think.
How do they manage to use their phones/other wireless devices without being seen. Maybe the classes are huge...
Perhaps they just had Windows 7 phones.
You need a warrant to do things like that (in the U.S.)...right?
vos nescitis quicquam, nec cogitatis quia expedit nobis ut unus moriatur homo pro populo et non tota gens pereat.
I have lived outside our Western culture for a while now, and there is a big difference in the idea of tests and examinations. We have the idea that the test is there to see who is competent to get the job. Simple, right? Nope, it's our own cultural biases that make us think this way. Elsewhere, it's all about getting what comes after the test. Your actual skill is irrelevant, not really a worthy topic of discussion. It's all about the job that you can get, or the university that you can get into, or whatever. The idea that if you don't have the skills then you're not qualified doesn't translate. Eastern cultures have a long history of examinations and take a different view than we do. I know a teacher who, after repeatedly warning against cheating in his class, was fired for daring to catch his students cheating in class. The students lost face, you see, and the teacher (not the students' cheating) was identified as the cause of the problem. True story.
Shutting down free speech with violence isn't fighting fascism. It IS fascism!
Catch more of them!
I'm sick of the widespread mentality that cheating is not only desirable but necessary, and that if done for the purpose of "getting ahead", it's alright. I sure wouldn't want a doctor or a lawyer who cheated their way through. I want one who took every test honestly and demonstrated they actually learned the material.
Maybe if we put back the concept of "Cheat or lie (as an adult) once, career suicide for good", we could eliminate this crap. It's infected everything from police to politicians, and programmers to paramedics. If we can find better ways to ensure people actually know what the hell they're doing, instead of demonstrating they can read letters from a cheat sheet, good.
Though, part of the blame also lies with those who design the tests. Multiple choice and fill in the blank tests are obsolete. The best tests would give the taker a project to do, and should be made difficult enough that collaboration is allowed and encouraged. After all, in real world scenarios, collaboration and the ability to research are important skills at nearly everything. As an alternative, one could at the very least give essay questions that would require careful thought and don't have a single "right answer" that can be copied in.
Of course, that takes more effort to grade than running a bunch of sheets through a reader. Imagine that, giving something actual thought.
To fight the war on terror, stop being afraid.
After all, some other ways work too.
Never trust a spiritual leader who cannot dance -- Mr. Miyagi
Too bad it's not the U.S.
If it were, I'd say: make them take a polygraph, a urine test, and walk through a backscatter machine before entering the test room.
I know those are either nearly useless (backscatter and polygraph) and of questionable value to society (urine test), but government and corporations make us take them... let them do it too.
Administer exams in a low, low basement room, or in an area with no cell service.
Using cell phones to cheat? All we had were a couple cans and a crummy piece of string. But we got by... Next they'll ask for transportation, too, I'll bet.
Disrespectful punks, get off my lawn!
Damn_registrars has no butt-hole. Damn_registrars has no use for a butt-hole.
They didn't find cheaters, they found idiots who didn't turn their phones off!
Using an FS4 to find someone with their phone turned on is nothing other than a waste of money and will most likely be a one of case of killing the chicken to scare the monkey.
They were suspected. That does not mean they actually were. Also how many false negatives?
And why allow phones in the first place? Just do not allow anything that can be used for cheating, including pens. Provide people with what they need. If they need a calculator, then give them a calculator. If people need a pen and paper, then give them pen and paper.
From experience I know that you can open a pen and put paper in it. (Also I know that by re-writing the cheating paper several times, because it was to big, I was actually learning and in the end did not need the cheat sheet, because I knew what was on it.)
Don't fight for your country, if your country does not fight for you.
Test with no collaboration and no open book / Google are not the real world and just lead to people who can pass the test but have no idea on how to do the work.
The tests need be better less about memory and more hands on. Also how many people have jobs when having a book, reference guide, google, a manual, and more is banned and having others working / help with you is a no no?
Wouldn't this falsely convict anyone carrying a Windows Phone 7 phone, which sends tons of data unbidden?
Dog is my co-pilot.
brb, going to text my friends during their finals
I've always wondered what it would be like to take a test ala "Old School" with an earbud in my ear and a van outside sending me the answers.
It could be done!
The actual price isn't that different from retail. The rest are kick backs for the purchasing officer and his superiors.
Our parents failed so miserably that the only way to catch cheaters is with technology. Cause it's too hard to raise them to be upstanding adults.
I bet you they'd blame someone else if we pointed out the kids weren't raised well. (And then fuss about how today's youth won't accept responsibility, while expressing confusion over who they learned that from!)
I'd like to know where the proctor for that test gets their money! Those R&S analyzers are awesome but -pricey- pieces of equipment.
If the world isn't beating a path to your door you're doing something wrong.
talk to it in morse code by pressing buttons in your shoes and get braile like responses tapped into your skin.
no online hookup, no detection without test takers going through a scanner.
From TFA "The devices checked for signals from pagers or mobile phones near the test site." I'm guessing this was in the 90s.
Why exam rooms are simply built as Faraday cages, I'll never understand. I bet they even sell drywall laminated with foil or some such that would make it easy. Put it up, ground it, make sure the doors are steel or aluminum, no cellular or wifi at all. The rooms totally isolated.
My wife writes certification exams for a large IT company. It's not enough that her company has to create 6-7 versions of their exam and pull 100 questions out of a pool of a 1000. In older versions you could go back and check/review/revise an answer before you hit the big "Submit for Grading" button. Now the exam constantly grades you and when it determines that either you cannot pass or you cannot fail, the test is automatically ended and you score is provided. Thereby making it harder for someone to see all 100 questions on their specific test.
Why?
In Asia test takers were paid to go in, sit in the exam and just memorize as many questions as possible. Now if they miss say 31% of the questions, the exam is over (assuming you need a 70% to pass).
Not only is the exam much harder now, but it's more expensive for students to take.
I used to think as you do, and I've come to realize that while it's very true for some things, it's exactly wrong for many.
Test with no collaboration and no open book / Google are not the real world and just lead to people who can pass the test but have no idea on how to do the work.
Well, to start with, if you only know how to solve the problem with Google, do you really know how to solve the problem? Anyone can Google for the answer. What you need to do is solve the problem yourself. Yes, it's boring, it's something everyone else has already solved, and you probably solved at least once on the homework. But if you can solve this problem, you've demonstrated that if someone throws a new problem at you, one no one has solved yet, you'll be able to do it -- and it'll be all the easier when you have Google to help.
The same can be said of the book.
If you test people in groups, you now have all the problems of groupwork -- how do you know who did what? Who actually contributed, who did all the work, and who sat back and collected the credit? Or, if it's a problem that actually needs multiple people to solve in the given time, maybe one of the students you're about to fail was actually competent, just got stuck with a useless group?
Even assuming the students are honest about it, it can be hard to remember how it actually happened.
Add to that the logistical problems. If you allow Google, but not collaboration, how do you prevent students from communicating? What if you want to limit collaboration? How do you know a student hasn't hidden an answer sheet on Google or in their book?
The only way this could possibly work is if you came up with at least significant variations on a project each semester, if not entirely new projects, and you made sure that the key concepts you want students to understand are something Google doesn't know -- which means you're basically going to do original research each semester. Really?
Homework is good on its own, and it has many of the above problems. But a test is one way to make sure a student has actually learned something from your course.
Of course, the better courses include sufficient information on the test to bring it closer to the environment you were supposed to be doing your homework in -- enough to avoid the memorization, but not enough for you to just Google "Hey, everybody, what's the answer?" For instance, for my Data Structures course, the Javadoc for relevant pieces of the Java APIs was included on the test. For my introductory Physics course, relevant equations were included on the test, you just needed to know what they mean and how to apply them.
And guess what?
If you made it through that Physics course without understanding what E=K+U means, or F=ma, or K=(1/2)mv^2, guess what? You don't know what they mean and how to apply them. Sure, on the job, you have access to Google, so maybe you could Google things like the universal gravitational constant or the formula for energy in a sinusoidal oscillator, but if you have to Google for what F=ma means, you didn't learn a single goddamned thing from that course, and you deserve the F you'll be getting. If you don't know that F means Force, you get F for Fail.
You can have the Java Collections APIs -- we don't care if you've memorized something your IDE will tab-complete for you anyway. But if you can't loop through a linked list without Googling "linked lists" or using iterators, you don't really understand linked lists. I don't care that iterators make more sense in the real world -- you fail, see you next semester.
Don't thank God, thank a doctor!
How many times a year are exams held? I count less than twenty, and probably more like ten, for a given course -- and they do try to pack the exams relatively close together.
Should the exam rooms sit idle the rest of the time? I generally see lecture halls being co-opted as exam rooms, but they're useful as lecture halls the rest of the time.
Or are you saying we should cage everything that might ever be used for an exam, so students can't use laptops or cell phones during class? Do not want.
If you can come up with a way to do it easily and cheaply, and just for exam time, I'm all for it. But even then, I'd also rather catch the cheaters and expel the fuckers than just throw up another obstacle for them to work around.
Don't thank God, thank a doctor!
Depends upon the field of study. I had an O Chem prof that took that position. We were allowed to bring in as many books as we liked, as many pages of notes as we liked and any other resources. The only things we couldn't use were the other students and the internet.
The reason being that in the real world, chemists look things up constantly, nobody is expected to know everything, and with a field of any complexity they won't know everything.
Because the internet has all the answers: every right and wrong one. According to google, a "string quartet" is a type of musical form (it's not). Some things require that you look them up, and part of that challenge is being able to find the places that have information that is complete, accurate, and up-to-date. There are also things that you just have to know cold. Unfortunately, most people seem to play at the research thing without actually being able to do legitimate research.
but when they reuse last years test and call it cheating to use the old one as a study guide.
if they are paying people to memorize the test then that shows that the test is to much based on memorizing stuff and will be better off being made so its more hands on.
I presume, since cheating on the exam shows a disregard for rules and a willingness to lie to gain advantage over others, they will now be fast-tracked onto the management course?
They whose government reduces their essential liberties for temporary security, receive neither liberty nor security.
And I can tell you, first hand, that the parent is correct !
In other parts of the world, it is the sheepskin which talks the loudest.
No matter how qualified (or the lack of) the applicant of the job is, the company will, 90% of the time, hire the one with the "best" sheepskin.
That's the fact of life in many many parts in this world we are living.
And that is why many 3rd world countries remain 3rd world, for so long.
Google: define: string quartet. What's wrong here?
Me so solly! Me frappy dickie! Ah, so! Ah, so!
Does this actually happen that much in America that it comes up on every slashdot story about testing and examinations? My university actively encouraged students to look at past exams. They made all of them for all subjects available online.
It is pretty impressive that Taiwan has police officers who even know what a spectrum analyser is!
However, it is surprising that the candidates didn't know that they could disguise their transmissions by matching their carrier modulation to the warp signature - surely everyone knows that.
http://patapata.sourceforge.net/WhyEducationalTechnologyHasFailedSchools.html ... So, there is more to the story of technology than it failing in schools. Modern information and manufacturing technology itself is giving compulsory schools a failing grade. Compulsory schools do not pass in the information age. They are no longer needed. What remains is just to watch this all play out, and hopefully guide the collapse of compulsory schooling so that the fewest people get hurt in the process. ..."
"Ultimately, educational technology's greatest value is in supporting "learning on demand" based on interest or need which is at the opposite end of the spectrum compared to "learning just in case"
based on someone else's demand. Compulsory schools don't usually traffic in "learning on demand", for the most part leaving that kind of activity to libraries or museums or the home or business or the "real world". In order for compulsory schools to make use of the best of educational technology and what is has to offer, schools themselves must change.
A larger elaboration on that theme:
http://www.pdfernhout.net/recognizing-irony-is-a-key-to-transcending-militarism.html
A 21st century issue: the irony of technologies of abundance in the hands of those still thinking in terms of scarcity.
I agree that cheating (as an aside, I actually prefer using lying, as cheating seems to have less weight these days) is a major problem, but I don't think you have it right about testing. Selected response assessments (common name: multiple choice) are still very valuable, and are indeed one of the best ways to assess a wide range of skills or knowledge. The fact that they are easy to score is a bonus, but more because it allows for deeper statistical analysis of items, strands and a means of comparing performance across many classes, schools, years, etc.
It would be more accurate to say that selected response assessments are overused, abused and frequently fail to meet the minimum standards for quality. However, this isn't the fault of the format, but the writers and people who use them. Performance based assessments (common name: project based, problem based assessment) still suffer from many problems: poor problem writing, unclear or incomplete scoring rubrics, bias in scoring, and more. Also, they are poor to determine the extend of student understanding. For instance, if you understand and can apply 80% of the subject material, but the second step of 20 for the solution is in the 20% you don't know, this scores as only completing 10% of the entire problem.
For a complete understanding of what a student knows, understands, has mastery of and can utilize, a variety of assessments are needed in a variety of formats. No single format is sufficient.
FTA:
> Spectrum analyzers cannot conclusively identify specific phone users...
Strictly this isn't quite true (I used to work for HP T&M/Agilent). They actually can.
There are add-on modules to most high-end spectrum analyzers that let you completely decode cellular signals all the way down to burping out the entire raw data exchange of cell tower set-up to full MMS/SMS traffic. It's pretty trivial actually.
How do people imagine cell phones get tested in the factories?
Why do you suspect the #1 buyer of high-end spectrum analyzers has always been government intelligence agencies?