0% found this document useful (0 votes)
355 views

Writing Effective Tests

The document provides guidance on writing effective tests from a book on developing and using tests. It discusses the importance of tests having clear purposes and benefits, and outlines problems with many college tests. It then offers tips for constructing different types of test items and scoring essays effectively. The overall goal is to help faculty design tests that reliably and validly measure student learning.

Uploaded by

Terry Doyle
Copyright
© Attribution Non-Commercial (BY-NC)
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPT or read online on Scribd
0% found this document useful (0 votes)
355 views

Writing Effective Tests

The document provides guidance on writing effective tests from a book on developing and using tests. It discusses the importance of tests having clear purposes and benefits, and outlines problems with many college tests. It then offers tips for constructing different types of test items and scoring essays effectively. The overall goal is to help faculty design tests that reliably and validly measure student learning.

Uploaded by

Terry Doyle
Copyright
© Attribution Non-Commercial (BY-NC)
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPT or read online on Scribd
You are on page 1/ 28

Writing Effective Tests

 The following slides summarize the work


of Lucy Jacobs and Clinton Chase from
their book Developing and Using Tests
Effectively: A Guide for Faculty

 The book is a compilation of years of


research on how to develop effective
test instruments.
Writing Tests
 Most higher education faculty never took a
course in how to write tests. Their experience
comes from the many tests they have taken over
their many years in school. This personal
experience is not a reliable way to learn to
develop tests that will determine so many
important outcomes for our students.

 Even with diligence, time and practice writing


tests, most tests given in college have
weaknesses and are not highly reliable.

 However, the following presentation can


significant help to improve the quality of the tests
Writing Fair and Effective
Tests
 Purpose of  Benefits to
Tests Students
 Grades  Learn how they
 Planning content are tested
presentations  Engage in deep or
 Facilitate learning surface learning
 Motivate students based on tests
 Feedback on
learning
Test Benefits to Instructors
 Focus on essential  Feedback if course
course content objectives were
 Message to achieved
students on what’s  Feedback on
important effectiveness of
 Valuable feedback instruction
on learning  A motivational
strategy
Problems with College Tests
 Don’t focus on  Too short to
what’s important sample content
 Too little feedback fairly
given to students  Purpose of test not
 Questions made clear-so
ambiguous or students see it as
unclear penal
 Too few tests-
don’t have a broad
view of students
knowledge
How Long Should a Test Be?
 True-False 30  Matching 30
sec. sec.
 Multiple choice 1  Short Essay
min.  10-15 minutes.
 Completion 1  Extended essay
min.  30 minutes
 Short answers 2
min.
 Multiple choice
90sec. higher
level thinking
Number of Test Items
 Fifty Minute Tests
 40-45 multiple
choice
 60-80 true-False
 Fastest students
finish in half the
time of the slowest
 Tests should allow
almost every
student to attempt
every item
Test Item Difficulty
 Tests should  Difficulty can be:
produce a spread  A. content it asks
of scores
 Test items should  B. cognitive skill
be of average level being
difficulty measured
 Items that 50-70%
of the students
can answer
 C. Construction of
correctly the
 If they didn’t study item
Test Reliability
 Long enough to  Directions should
sample content be clear.
well  Scoring procedure
 Most student needs to be free of
should be able to errors.
finish  All students should
 Score range take the same test
should be wide
 Items free of
ambiguity and
tricks
Test Validity
 Samples the  What limits
course objectives validity?
in proportion to  A. Directions not
their importance clear
 Use a test matrix  B. Tests skills not
to check part of the
distribution of course objectives
questions by  C. Too short
content and
cognitive level.
 D. Subjectivity of
scoring
Multiple Choice Tests
 Write a stem in the  Do not pad the
form on an stem-write them
incomplete simple and clear.
statement or  Distracters must
question that be plausible
could be answered  Avoid using humor
without look at the
four choices
 Make certain there
is only one best
 Always put a verb answer
in the stem
Multiple Choice Tests
 Make distracters  Avoid giving
fairly grammar clues
homogeneous-this  careful with
will force students specific modifiers--
to be ie. Never, always
discriminating  Limit use of all of
 Don’t give the above and
irrelevant clues or none of the above
tricks-these
reduce validity
 Make answers
Multiple Choice Tests
 Arrange options in  Avoid asking trivia
logical(alpha )  Ask students to
order choose the best
 Keep correct answer not the
response letter correct answer.
( number)
balanced
 Don’t lift stems
verbatim from text
 Arrange answers
in vertical columns
True-False Tests
 Advantages  Limitations
 A. Sample many  A. Difficult to write
more bits of so there not
information ambiguous or
 B. Ask more obvious
question in 50  B. Difficult to make
minute time clearly true or
period helps clearly false
reliability statements
 C. Research  C. 50-50 chance
affirms True-False  D. Educated guess
True-False Tests
 Avoid use of  Write questions
specific that are clear and
determiners ( all, simple to read.
never)  Make it clear
 Avoid use of where the answers
qualifiers(sometim are to be placed--
es, usually) don’t use + or -
 Avoid use of
indefinite terms
(long time ago)
True-False Higher Level
Thinking
 Use propositional  Setting up a
logic--if then problem that
approach. requires analysis.
 Example-If the  Example --John
gross national bought a car…it
product goes up won’t start. Which
and inflation is are possible
3.4%…then the reasons?
stock market will
likely go up.
Short Answer or Completion
Items
 Requires recall  Questions should
rather than have a single word
recognition of answer.
information.  Make certain there
 Doesn’t test is only one right
higher level answer
thinking  Delete only key
 Best used to see words
what information  Make blanks same
students have length
collected.  Don’t lift
Matching Tests
 Stimulus column  Put stimuli in
on the left alphabetical order
numbered--  Use 10-15 items
response column  Keep entire test on
on right lettered. one page.
 Provide space to  Have at least five
right answer to the more stimuli than
left of stimuli. responses
 Use only
homogeneous
material
Essay Tests
 Advantages  4. Can’t simply
 1. Can assess recognize an
complex learning answer
 2. Easy to write  5. Doesn’t allow
 3. Emphasize for guessing--
communication except bluffing
skills as  6. Can see how
fundamental students select,
organize and
evaluate ideas and
apply them
Essay Tests
 Limitations  4. Score
 1. Difficult to score influenced by
 2. Scores are less readers overall
reliable impression of the
student
 3. Sample only a
limited amount of
 5. Not favorable to
content developing good
writing skills
Essay Reliability Concerns
 Less reliable than  A second reading
objective tests often results in a
 Environmental different grade.
factors influence  Expectations of
scoring ie. Time of students’
day, mood of performance
reader where  Physical elements
paper is in the of the paper--
stack handwriting
 Previous paper can  Limited number of
influence current questions luck of
paper the draw
Making Essay Tests Better
 Restrict essays to  Keep question
assess only focused on a
outcomes that single objective
require complex  All students should
higher level respond to the
cognitive same questions-
functions. don’t give choices
Examples  Grammar and
 1. Compare spelling count only
-Contrast if taught as an
 2.Present objective in the
Making Essay Tests Better
 Directions need to
be very clear and
include type of
response
expected--Example
outline, list,
discussion
 List the point value
of each question
Scoring Essay Tests
 Conceal students  Use a prescribed
names reading procedure
 Skim a few before  A. Key procedure
starting to grade  lays out ideas that
to get a sense of should have been
responses in the essay
 Read only one  B. Ranking
item across all Procedure
papers  Makes piles based
 Reshuffle the stack on quality
Open-Book Tests
 Advantages  Limitations
 Used in classes  Students tend not
where formulas, to prepare well
tables or graphs  No evidence to
are needed to find suggest students
solutions do any better
 Student
demonstrates they
know how to use
references
 Reflect real life
Take Home Tests
 Advantages  Limitations
 1. More thorough  1. Question rarely
exploration of sample adequately
ideas can take course material
place  2. Much more
 2. No time grading time
restrictions  3. Students often
 3. Usually better unsure of how
written thorough to be.
Re-takes of Tests
 Advantages  Limitations
 Retakes do result  A large test bank
in better learning is needed
 Lower test anxiety  Grading time
 Students learn increases
from their  Students don’t
mistakes often prepare well
 It eliminates the for the first test
need for make-up
tests
Helping Students Prepare for
the Tests
 Review the scope  Review make-up
of the test policy
 Use a practice test  Provide a review
 Clarify time limits session
 Clarify what  Make provisions
material will be for last minute
needed and questions
allowed  Allow for breaks
 Review grading during long exams
procedure

You might also like