Results of 4/30/98 WebCT Survey III
  1. How useful would would the following be: an online "practice gateway exam" which gave you random questions and an instant score upon completion, and which you could take up to (say) three times before taking the real thing? 
    (not at all)   (a bit)   (somewhat)   (pretty well)   (excellent)
     2            4          5             12              2
  2. How useful would would the following be: online practice quizzes which gave you random questions and an instant score upon completion, and which you could take up to (say) three times before taking the real thing? 
    (not at all)   (a bit)   (somewhat)   (pretty well)   (excellent)
     2            4          8              7              4
    How often would be best? 
    (once per week)(every 2 weeks)(once per month)(before each exam)
       5              3              3               14


  3. How useful would would the following be: actual online quizzes which you could take at any convenient time (within a start and end dates), which gave you random questions and an instant score upon completion? 
    (not at all)   (a bit)   (somewhat)   (pretty well)   (excellent)
     7            4          8              4              2
    How often would be best? 
    (once per week)(every 2 weeks)(once per month)(before each exam)
       4              1              7               9
(Plus 1 write-in for "never" and several left blank)
  4. How useful would would the following be: actual online quizzes which you could take at any convenient time (within "start" and "end" dates), which gave you identical questions (every student has same questions)and an instant score upon completion? 
    (not at all)   (a bit)   (somewhat)   (pretty well)   (excellent)
     3            6          13             2              2
    How often would be best? 
    (once per week)(every 2 weeks)(once per month)(before each exam)
       3             3             5                 10
(Plus 1 write-in for "never" and several left blank.
Note that 50% of respondents (11) wanted more frequently
than once per exam.)
  5. How big a problem were technical difficulties, such as the server being down, or your not being able to come online? Not an annoyance here and there, but a real difficulty? 
    (not at all)(a bit)(somewhat)  (real problem) (major problem)
     8         4       7.5         0.5              2*
(*: One of the two wrote in: "but it was my system problems")

  (never)  (rarely)(sometimes)  (>3 times)(never able to login)
     2         5       5.5         2.5              1
(Half-votes represent students who split their votes between two choices).
  6. In general how do you feel about human versus computer grading? One might argue that the computer grading is (a) faster, (b) more accurate, (c) can still be double-checked by a human in case error is suspected, and (d) gives student instant feedback, (e) easily generates random collections of questions when that is desired.

    On the other hand, it (a) takes work to put in all the questions into the computer's database, (b) the computer can't (presently) grade much more than multiple-choice, or Column-A/Column-B matchup type questions, and (c) issues of access, or fair/equal access, to the computer-testing module.
    How do you feel about these issues? Anything else you'd like to tell us?
    Yep. I like both. I like the site especially the [Homework] hints.
    I think that, in the end computer are still much more efficient.  
    Computer quizzes would be ok but not tests. They are quick and can help you measure how much you know  
    I would prefer human grading so I could prove things and not only have to do exact multiple choice questions.  
    S U C K S B A D
    Good for math-type classes, bad for essays No
    Whatever nope
    These don't matter: problems MAY be ambiguous. If the test is on the computer, we may not be fairly graded "Lived" backwards is "Devil"!
    (Live & evil, too!)
    Do both (1/2) + (1/2) -- computer + human.  
    Computer practice quizzes are okay, but real tests should be human graded.  
    Whatever works  
    It depends on the type of quiz/test. Some things are better if the grader can see the work + thought process I think it would be better to have smaller quizzes + tests than have big tests + a final
    I'd rather have humans grading tests for things like partial credit.  No
    It's excellent for practice and review work  
    The biggest problem is the limitation of multiple choice question only, because our actual exams are not multiple choice  
    It's hard to visualize from computer symbols to mathematical symbols [Ie, "integral" symbols etc -HB] , and partial credit can't be given, plus time pressure is [in?] all [up?] comptuters nope
    Although computer grading is nice because of speed, it doesn't always show that we know the material + it can't give us partial credit  
    I like personal feedback but quizzes would be ok  
    Access really matters. Everyone must have an equal choice and conditions when taking a quiz -- using a computer will be different for everybody and would introduce unfair uncertainties (Also, the [mathematical] notation is confusing)
    Required online quiezzes are not a good idea. Some people have problems with this.  
    I like online quizzes. All tests + quizes + classes should be done on the internet This class now is made for math geeks -- make it so people who have other things in their life other than math can be in this class [Referring to overall HW/time workload in UMTYMP? -HB]
    Computers are good for practice  
    Computers should grade multiple choice No
    Human is better Nope
    Use computers for (multiple-choice) and (matchup), human for the rest. Love is a narcotic, injected with sin
    The fire of passion is the same hell I'm in
    It numbs you to evil, you fee no pain
    Now I'm in withdrawl & I'm going insane!
    Thank you for your time and for sharing your thoughts and ideas!