Exam Time Limits and Anxiety

[Hadoblado]

I'm doing an assignment on anxiety at the moment that is specific to university students. While justifying the requirement for a scale of anxiety to screen for those in need, I wrote a paragraph on people high in anxiety requiring more time on exams.

This seems unfair on other students, but it's also unfair to expect someone with anxiety issues to perform under such massive pressure.

Solution: Give everyone more time.

Why impose such an arbitrary and artificial restraint? We acknowledge how it affects people, and come up with ways to stop it affecting people in too severe a manner, but this still leaves non-disorder level anxious individuals at a massive disadvantage. Pretty much everyone gets anxious at exams.

Am I missing something?

 

[Ex-User (5841)]

I agree. The goal of exams should be to test for knowledge, not to test speed of completion. Those who finish early are free to leave or do other things.

However this raises an additional problem, imho: How do we prevent or reduce the distraction of those who finish early? This is what has always driven me nuts.

 

[Hadoblado]

Hmmm... I've never had that problem, and I'm extremely distractable.

I don't know how you do things outside of Tasmania, but we are free to leave any time except in the last 15 minutes. Most people don't, since ordinarily leaving that early implies you had no clue what you were doing and didn't know enough to answer the questions satisfactorily.

Assuming this procedure is standard, offering more time than absolutely necessary is still a strict improvement. Yes people will be leaving at random intervals which could cause distractions, but those distractions will occur during time you ordinarily wouldn't have.

Time with potential distractions > No time

Though yes, it's a limitation, though I wonder if that distractability would be less of an issue if you weren't under such strict time pressure?

 

[The Gopher]

The reason there is a time limit is more so that the examiner can get home in time for dinner. How annoying of them.

 

[DelusiveNinja]

How will they know how much time to give? The way I thought about this was:
What if the time they gave you was different? Wouldn't the anxiety still be there for some group of people who can't choose the answers with the amount of time? And what if the number of people in that group ended up being equal to the number of people who couldn't choose the answers with the amount of time given before the time changed? Time gives a drive to answer the question (I think). Imagine what the exam would be like without pressure or with the only pressure being that you need to finish. Ever heard of the infinite monkey theorem?

Then again, all this depends on what the exam is testing, right? If the exam was testing how much information someone knew about a subject infinite time would be an option (I would assume). If the exam was testing your ability to apply knowledge to a problem under pressure then time (or some other type of arbitrary artificial pressure) might be necessary to gauge it and compare it with others to determine competence(?).

Why gauge or compare the score? I think they do this because they know that every human is capable of solving the problem given a certain amount of time and most careers don't give you that much time to solve problems.

Exams that gauge a person's capabilities under pressure should be used as information to determine what the person needs to improve. Then with that information that person's skills should be immediately tackled by some group of people be it a school, college, or personal trainer. What I want to know is why people readily accept the scores on the exams as the 'best they can do' or 'a perfect IQ measurement'. What people need to realize is that everyone doesn't have access to the information necessary to improve that score (let me stop myself before I rant and go off topic).

 

[Ex-User (5841)]

It's the same here except anyone can leave at any time. Sometimes more than enough time is provided, sometimes the opposite. It depends on the individual professor.

Using myself as a case study, anything I pick up in my visual or hearing range while I'm trying to focus pulls me in that direction. People leaving, foot/pencil tapping, proctors moving around, the font of the exam text on occasion... The irony of this is that I can make all the noise and movement I want, and in fact am most focused while listening to music and/or driving.

The reason I'm edging toward this derail is that these (anxiety & distractability) should theoretically be related, in my case specifically through PTSD, but in the general population through the HPA axis. http://www.ncbi.nlm.nih.gov/pubmed/8463202

I think they're separate in the sense that a time limit shouldn't increase distractability. Note that individuals with anxiety have increased cortisol levels, while the opposite is true in individuals with ADHD. So ideally there's an optimum amount of cortisol involved in testing well.

Is the key to exam performance therefore something that takes place long before the exam; long enough prior to alter cortisol levels? Studying comes to mind, but, cutting to the end of that logical rope, what if it's in the teaching...?

*EDIT: Another.... interesting study: http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2685476/?too

 

[Hadoblado]

in fact am most focused while listening to music and/or driving.

This. I'll likely respond later, but am about to head home.

I can't do anything without music, and all the music I listen to has to be thick enough to block out the surrounding noise. I can't concentrate elsewise.

 

[Montresor]

I can't do anything without music, and all the music I listen to has to be thick enough to block out the surrounding noise. I can't concentrate elsewise.

I was like this (to a lesser extent) in Uni but not any more.

It's just scheduling and red tape (re: time limits)

 

[NullPointer]

A timed exam allows immediate recall of skills and information to be examined. If you allow too much time, it's often possible to use deduction to patch over gaps in knowledge. If you take the purpose of an exam to be the assessment of "how well you know the subject", then immediacy of recall becomes an important aspect. As an example, suppose person A wants to assess whether person B can make a paper crane, is it good enough if person B requires 50 sheets of paper and trial and error to produce the crane? With unlimited time, it's a feasible approach, but if a time limit of 5 minutes was placed on it, person A could correctly determine that person B actually has only a vague understanding of how to do it.

There's also the element of cost, which has been brought up before. The cost is both monetary and temporal; a lot of people must be employed to oversee the exams, and if every subject needs an exam, removing time limits (or extending them drastically) would extend the exam season over many more weeks, and possibly months, unless a lot more space is allocated. I assume there would be a very strong correlation between scores in timed exams, and un-timed exams. If that is true, then this extreme cost is hard to justify.

If, instead of treating students uniformly and allowing everybody unlimited time, you allowed students to apply for a finite amount of additional time, perhaps a good balance could be achieved. From a timetabling perspective, students with the additional time allocation could sit their exams in a separate room, which would allow the rest of the exams to be scheduled as tightly as they are. There would need to be a well-defined set of criteria to prevent potential abuse of this system, but it does at least allow additional time to be allocated when it's needed, without a significantly higher cost than the status quo.

Some schools offer special arrangements such as dictation for students who have difficulty writing for some reason (hand injury, mental disability etc.), so this additional time allocation would fit nicely alongside that sort of system. Actually, I'd be surprised if there doesn't exist a school or university such that a student may apply for extra time there.

 

[Duxwing]

As mentioned above, one reason is that some tests can be solved by brute force trial and error or deduction as opposed to knowledge; however, another reason is that an easy way to produce a normal distribution of scores is to simply give average students too little time to finish, and test makers want things to be as easy as possible.

-Duxwing

 

[Ex-User (5841)]

Should exams be scored by the simple formula: % correct / time taken to complete?

 

[Hadoblado]

What about just giving an extra hour to everyone?

It leaves time for a little bit of hard cracking, but honestly, is that such a bad thing? "Oh no, this person can use logic and reasoning to cover his limitations given a little bit more time, he should fail immediately." That sort of ability is useful in the real world, and speaks of scope in a career rather than entry level requirements. A person who isn't 100% able to remember every detail but can cover with reasoning can still pick up those details over the course of his career. If you can't reason your way out of a paper bag you'll likely never make it to the frontier.

 

[NullPointer]

Should exams be scored by the simple formula: % correct / time taken to complete?

Nah, the optimal strategy in that case would be to fill out the easiest question as quickly as possible, and then leave after 2 minutes.

What about just giving an extra hour to everyone?

It leaves time for a little bit of hard cracking, but honestly, is that such a bad thing? "Oh no, this person can use logic and reasoning to cover his limitations given a little bit more time, he should fail immediately." That sort of ability is useful in the real world, and speaks of scope in a career rather than entry level requirements. A person who isn't 100% able to remember every detail but can cover with reasoning can still pick up those details over the course of his career. If you can't reason your way out of a paper bag you'll likely never make it to the frontier.

Maybe, but that's not what they're trying to assess.

 

[Hadoblado]

It's not what they're trying to assess, but (presuming I'm correct in thinking it's more predictive of future success) should it be what they're trying to assess? Given the occupational function of the piece of paper you receive at the end, isn't limiting examination to rote learning a little short-sighted?

 

[Ex-User (5841)]

Just to clarify, I'm thinking a composite grading system that actually reports 2 grades: Overall percentage correct and correctness in time. Alpha and beta.

This doesn't preclude extra time, and it's not much more work for graders. It also identifies who raced through it to score higher, and who knew their stuff balls on.

 

[Hadoblado]

It seems like a useful method if you've reason to test these two types of understanding.

 

[Montresor]

Should exams be scored by the simple formula: % correct / time taken to complete?

Interesting thought but I think that it should be a ratio of z scores, not what you have described.


There is a time distribution and a % correct distribution and each will produce a unique z-score for every individual taking the test.



The final "ratio" will simply be another z score to be plotted on a new distribution which will produce grades (A,B,C,D) based on the previous criteria.


Then again, I'm not sure this will work in practice because I am wrong. Please fix my mistake lol.

 

[Ex-User (5841)]

Then again, I'm not sure this will work in practice because I am wrong. Please fix my mistake lol.

It's actually exactly the same thing except that a z score is reductionist in that one can't tell whether an individual's correctness was skewed toward knowledge or speed just by looking at the z score. (I assume this differentiation is important regarding division of labor, i.e. certain jobs require speed while others require accuracy).

 

[walfin]

You can give any bonus and any score you want but as a practical matter most employers just want to know the class of honours, so it'd have to be a unified score in the end.

 

[Ex-User (5841)]

You can give any bonus and any score you want but as a practical matter most employers just want to know the class of honours, so it'd have to be a unified score in the end.

Do you think this highlights a flaw in general employment protocol in that hires are made under the guise of replacement in the event of failure, as opposed to engineering a work environment through careful selection of the employees that compose the social system?

I view it analogously as for some reason choosing cheap plywood furniture over that made of hardwood dovetailed joinery; profit over enduring strength and reliability.

 

[Perfectly Normal Beast]

"Oh no, this person can use logic and reasoning to cover his limitations given a little bit more time, he should fail immediately." That sort of ability is useful in the real world, and speaks of scope in a career rather than entry level requirements. A person who isn't 100% able to remember every detail but can cover with reasoning can still pick up those details over the course of his career. If you can't reason your way out of a paper bag you'll likely never make it to the frontier.

It's not what they're trying to assess, but (presuming I'm correct in thinking it's more predictive of future success) should it be what they're trying to assess? Given the occupational function of the piece of paper you receive at the end, isn't limiting examination to rote learning a little short-sighted?

Perhaps it is another of the mechanisms by which society marks potentially 'dangerous' thinkers as failures. Much as those who question (often arbitrary, nonsensical or illegitimate) authority are expelled from the system and are thus less likely to acquire the power which might be used to change/overturn the status quo.
The capacity for rote learning which requires zero use of logic or reason is exactly what the system rewards in most disciplines. The unquestioning and incurious obedient are successful within this framework.
I have never actually considered the issue of exam time limits before but have often thought about how the education system (by design) makes learning a tedious, uninspiring chore for exactly those who would most benefit from it if it was not a tool for crushing individuality, independence and the joy of learning for curiosity's sake and churning out subservient worker drones who transfer their blind obedience to the state upon leaving education and who leave any capacity for questioning and thinking (about how society could be more fairly organised, for example) which they may have had behind with their childhood.
Thinkers are dangerous! (and more anxiety prone?)

 

[Montresor]

Do you think this highlights a flaw in general employment protocol in that hires are made under the guise of replacement in the event of failure, as opposed to engineering a work environment through careful selection of the employees that compose the social system?

I view it analogously as for some reason choosing cheap plywood furniture over that made of hardwood dovetailed joinery; profit over enduring strength and reliability.

Are you saying that the industrial world is always playing "catch-up" and it is the driving force for progress?

Sort of how evolution is supposed to be lagging by 10 000 years or something?

Anyways the immaculate dovetail joinery from imported hardwood might outlast a plywood table in Grandma's dining room, but I do believe that one (such as myself) would get more use (thus profit) from a properly made plywood table with 2" nails and strictly end-joints.

Shoot ... do I have a point?


mmm I think what I am saying is that the criteria Wally outlined for employers to make hiring decisions might actually be counter-productive, as the plywood table is solid and will withstand much abuse.