exams
General Test Information
Opscan Forms | Multiple Form Exams
Additional Test-Related Reports, by Request
To assure proper processing
of opscan forms, please ask students to:
- Use a Number 2 Pencil only. Do not use pens; the scanner cannot read ink.
- Do not fold, roll, staple, punch holes or otherwise damage the opscans - they must remain as flat as possible. (Thus, it is not advisable to have students take opscans home.)
- When filling in circles, make marks heavy and black, filling in the entire circle.
- Multiple marks are not allowed - Only one choice should be marked per item and if an item is to be omitted, leave all choices blank.
- Use all 9 spaces for the ID number, adding zeros if necessary. (Forms without a complete ID are rejected and scores are assigned to a dummy number.)
- Be sure students fill in the corresponding circle to which form they have, if there is more than one version of the test.
- Seat numbers, if used, must be right-aligned.
You will need to fill out a submission form for each exam.
We have them outside our door (the Pink form), in our office, or you can fill one in online, print it and bring it
with your exams.
Instructors must submit at least one form (answer key)
showing correct responses. Use an opscan sheet, marking responses as
though taking the test but leaving the ID area blank. If a question is
to be omitted, leave all item choices blank. Do not put X's or lines
through the blank answer choices on the key and warn students not to do
so as well.
It is not necessary to sort student opscan sheets by form before submission to Test Scoring. Please do not use paper clips or staples on forms. Photocopies of forms
cannot be processed. Forms must all be face up and aligned in the same
direction.
It is preferable that your answer keys be in alphabetic order,
beginning with "A", with no letter omitted.
Results are returned to instructors electronically as
an email attachment. For each student, output shows ID and name, number
of correct responses, percent correct, t-score, number of questions omitted, seat number, which form the student used, and a processing sequence number to help you locate a specific test sheet after processing. Also included are:
Histogram showing a graphical distribution of scores, with one asterisk for each student
Item Analysis, which can be used to determine which test questions were weak or represented unsatisfactory levels of achievement
Statistics, including number of students, means, and standard deviations
Information on students omitting a form letter, marking more than one answer for a question, and forms with a substantial number of light marks that may not have been processed correctly. These messages can be found near the top of the report.
For more information, see Understanding Test Output.
Test Scoring offers several choices in opscan forms
for student answer sheets. Standard forms are available free of charge
to faculty and staff. Most commonly used are "orange" sheets for up to
60 questions and "purple" sheets for up to 160 questions. A "green"
sheet is also available on which up to 48 questions may be printed
directly on the form. (See opscan forms for more information.) All sheets allow up to 10 choices for each question and no question may have more than one correct answer.
Space is also included for:
- Student ID number
- Form (or key) number (used when more than one version of a test is administered to a class)
- Seat number (optional, must be right-justified)
- Group number (optional, used to designate up to 10 different groups of the instructor's choosing. When used, results are tallied by group along with totals.)
Multiple keys or forms may be used to discourage cheating; the order of
test questions may be scrambled for different students or tests may
have different questions altogether.
For tests with scrambled
questions, output can be provided showing combined statistics for each
test question. To obtain this, instructors must provide an unscrambling
key showing the position on subsequent forms of each question on the
first form.
Should a student forget to indicate a form letter, the
responses are scored against all keys. If the highest resulting score
is at least 1.5 times as high as the next highest, it is assumed that
the form yielding the highest score is what the student used.
Regardless, the scores from all forms are printed out in descending
order in an error message near the beginning of the output.
Additional Test-Related Reports,
by Request
^ Back to Top of Page
Rescores
If you accidently miscode an answer on an exam key or want to
delete a poor question we can regrade your exam.
Either call us at 1-5413 or reply to the email we sent. Just tell us
the change(s) you want to make. We do not need to reprocess the opscans.
Occasionally, due to a typo on your exam you may want to
have two correct answers. Please call us if this is the case. It is much easier
to delete the question but we can produce a supplementary report with
the optional choice as correct.
Error Listings (Wrongs)
Upon request, Test Scoring will provide an additional
listing of each examinee's incorrectly answered items (including
multiple marks). This report can eliminate the need to return opscan sheets to
examinees. The listing also indicates the number of omissions and the
percentage of light marks.
Unscrambling Analysis
If you have multiple forms and simply rearranged the questions,
we can provide you with a combined Item Analysis. You will need to give us a
listing of the question locations on Form B, etc. that match the questions on Form A.
Below is an example of the kind of information we need.
Form A question numbers are in sequence in the leftmost column.
This sample exam has 3 forms - Forms B and C are scrambled versions of Form A.
Key
(Form) A
|
Key
(Form) B
|
Key
(Form) C
|
|
1
|
10
|
4
|
|
2
|
6
|
2
|
|
3
|
9
|
7
|
|
4
|
4
|
3
|
|
5
|
8
|
1
|
|
6
|
1
|
5
|
|
7
|
5
|
8
|
|
8
|
2
|
10
|
|
9
|
7
|
6
|
|
10
|
3
|
9
|
This means that Form A Question 1 is also located on Form B as Question number 6 and Form C as Question number 5.
The choices for each question must remain in the exact order on all forms.
Educational Technologies offers assistance in assessment design. For more information, see www.edtech.vt.edu/edtech/id/index.html.
Although these memos are offered for your information, Test Scoring no longer has the expertise to support them.
During the 1978-79 academic year, a Testing Advisory Committee served the
Research and Measurement Division of the Learning Resources Center. In response
to faculty inquiries about various aspects of testing, the Committee prepared a
series of seven Testing Memos, which were circulated every few weeks to users of
Research and Measurement's test scoring and analysis service.
Since 1991, four additional Memos have been circulated. Below are the
complete set with slight editing changes.
Members of the Committee were Robert B. Frary (Chairman),
Lawrence H. Cross, Jerard F. Kehoe, and Larry J. Weber.
Frary wrote MEMOS 1 and 2, Cross wrote MEMOS 3 and 6.
Kehoe wrote MEMOS 4 and 5, and Weber wrote MEMO 7.
Subsequent MEMOS 8-10 were written by Frary, and MEMO 11 is by
Cross.
Page last updated
August 7, 2008
.