Many educators advocate the use of
authentic assessment, which refers to
measurement of student performance on
tasks that would be expected of a person functioning in society as
a citizen or employee.
Authentic assessment has several possible benefits: (1) It focuses educational
measurement upon relevant,
worthy educational outcomes, (2) it implicitly conveys to students that they
are learning valued skills rather than trivia, and (3) the assessment
process is value-added, as the assessment can be educative.
One method of improving student assessment in biology is to employ
case problems featuring authentic scenarios and data. The use of case problems in
test questions forces the instructor to focus on realistic tasks.
Realistic Test Questions
Authentic assessment requires educators to create
real rather than contrived activities and to create real contexts for student performance.
Written examinations do not strictly qualify as authentic assessment,
but in budget-limited freshman college courses they may be
a necessary component.
Benefits attributable to authentic assessment may be approached
by striving for realism in test questions.
Realism of Multiple-choice Questions
The authenticity of multiple-choice questions can be maximized by framing
each question stem as a case example.
A weakness of multiple-choice questions is their failure to emulate the demands of a
workplace in which employees lack access to a fixed list of choices each time a
decision is required. Free-response questions, by contrast, require the student
to construct
a response in the absence of hints or suggestions.
A strength of multiple-choice questions is that they do emulate a
workplace in which colleagues or clients put forth an number of
hypotheses which must each be understood, considered, and weighed in arriving at a decision.
Validity of Multiple-choice Questions
To what extent could a student
who chooses a suitable answer
from a list also produce a response of equivalent quality in the absence of choices?
Multiple-choice questions do not allow direct measurement of the latter, but they may be
valid indices of competency.
Research indicates that a given student is likely to
have a similar performance ranking on both multiple-choice
tests and tests employing free-response open-ended questions.
This may tell us only that intelligent
hard-working students do well on exams. Student achievement may be
driven by underlying variables such as reading-comprehension and self-discipline. For example, the Whimbey Analytical Skills Inventory is a powerful independent predictor of success in biology.
But there is some direct evidence that multiple-choice items can predict
the quality of answers that would be generated in the absence
of choices, although this relationship requires validation with each
learning outcome. A multiple-choice
item might identify with great confidence
an ability to perform a complex multi-step calculation in the absence of choices,
but caution would be required in
determining to what extent multiple choice questions could identify scientific
innovators, great public speakers, or superb tennis players.
If multiple-choice questions are the only cost-effective assessment option for instructors with budget-limited marking time, the questions should be employed with these limitations in mind.
Concisely-worded examples, such as those shown below,
allow testing of basic scientific
literacy and
fluency with a large number of core concepts and skills.
Subject matter can include
any concept deemed worthy
of testing, but note that (1) questions are cast in the context of a situation that a real
person could encounter, (2) a diversity of learning outcomes is covered, and (3)
some questions require that the student
invoke a number of skills, concepts, and algorithms in order to construct a suitable answer.
The following examples illustrate a range of responses that might be expected from students
given brief statements of realistic case scenarios.
Interpret Meaning:
The sexually-transmitted disease gonorrhea is becoming difficult
to treat because the causative bacteria are evolving resistance to antibiotics.
For example, in Hawaii between 1997 and 1999 resistance to fluoroquinolones
increased from 1.4 percent to 9.5 percent. Scientists attribute this to
natural selection. What does natural selection mean in this context?
Interpret Data:
An investigator monitoring physiological parameters in an Olympic athlete in training
subjected the athlete to an episode of extreme exercise and
found a plasma anion gap of 20 (reference value =12). What is the most likely explanation for this?
Weigh Hypotheses:
An 18-year-old woman with healthy parents showed muscle
weakness. Her brother had been diagnosed with an X-linked recessive muscular dystrophy
caused by an allele that impairs production of
dystrophin, a component of the sarcolemmal
cytoskeleton. About one third of such cases are new mutations. What is the
LEAST likely cause of the woman's muscle weakness?
Analyze Systems:
You are involved in a marine habitat conservation project and find that
that crab (Carcinus maenas) predation by herring gulls in the intertidal zone is
20x greater on sites lacking concealment cover
of fronds of the rockweed Ascophyllum nodosum. How
might you best characterize the ecological status of Ascophyllum in this situation?
Recognize Patterns:
A habitat conservancy group became alarmed when they performed
a survey within a group of islands, and found only four species of fern
on a small island whereas there were typically 16 species of fern on a
larger islands. If this were a natural pattern of distribution, how might it best be explained?
Infer Causes:
A dog-owner noticed that her pet had an enlarged liver and shrunken muscles.
If this were an osmotic problem, which explanation would be most appropriate?
Predict Effects:
Domoic acid, isolated from a diatom, has been found
experimentally to bind to hippocampal glutamate receptors. If a person were to accidently
consume shellfish contaminated with this organism, what effect might be expected?
Calculate Values:
You are working for a biotech company that is attempting to
develop a blood screening test for ovarian cancer. A preliminary
assessment on a population in which the confirmed ovarian cancer
rate is five per thousand shows that the test correctly diagnosed four
out of every five cases of ovarian cancer but showed a false positive in
36 cases per thousand. If a woman tests positive, what is the probability that she has ovarian cancer?
Calculate Outcomes:
Lannie and Phil both have healthy parents but each has a sister with
autosomal recessive cystic fibrosis. If Lannie and Phil have a child, what is the
probability that it will be born with cystic fibrosis?
Critique Investigations:
A biologist studied the environmental impact of an herbicide on squirrel reproduction. He
selected two small islands with similar vegetation and similar-sized squirrel populations
(about 200 squirrels) with equal reproductive rates. One island, selected at random,
was sprayed with herbicide, and
the other was used as a control. Squirrel reproduction rates were measured on both islands
before and after the treatment. What is the greatest flaw in this investigation?
Weigh Arguments:
A man diagnosed with Parkinsonism suspected that pollution at his workplace (a mill)
may be to blame. Which of the following statements would lend greatest support to his suspicion?
Manipulate Data:
A coach is concerned about the electrolyte balance of an athlete performing
in a warm climate. She estimates that ambient temperature and
humidity conditions will result in
a perspiration loss of about two liters of NaCl solution from the athlete during an
Olympic event. The athlete weighs 50 kg and begins the event with a plasma
sodium concentration of 140 mmol/L. Assume average sodium sweat-loss of 70 mmol/L, assume no
water intake,
and ignore kidney function. Approximately what would be the plasma sodium concentration
at the end of the event?
Critique Claims:
George claimed that a tropical animal protein supplied by a local
naturopath was effective in lowering blood pressure, and George was planning to
invest in the product. As evidence of his claim,
George, who had hypertension, said that he felt much better after the
treatment and had much more
energy. In comparison to a credible scientific investigation, how is
George's inference flawed?
Prioritize Questions:
Commercial fishers are concerned that sea lions are reducing the number of
salmon available to the fishery.
If the size of the salmon run were known, and if limited additional
research funds were available and you were ranking proposals
to investigate the salmon predation concern, which of
the following investigations would you deem LEAST urgent?
To contact, click HERE
Case Problems in Authentic Assessment
CONSTRUCTING QUESTIONS EMPLOYING CASE PROBLEMS
Copyright 1999 Peter Ommundsen