TOEFL Writing Work Sample
• Description of the TOEFL Writing task
• Examples of writing items
• Description of the TOEFL Writing Work Sample
THE TOEFL WRITING TASK
One part of the Writing section of TOEFL assesses the examinee’s ability to write in
response to a two-part stimulus on an academic subject. The stimulus consists of a
reading passage followed by a recorded lecture on the same topic. After reading the
passage and listening to the lecture, the examinee is directed by a prompt for a written
response. A typical prompt is “Summarize the points made in the lecture, being sure to
explain how they cast doubt on the specific points made in the reading passage.” The
stimulus and prompt together constitute the writing item.
The reading passage part of the stimulus is between 250 and 300 words long. It begins
with a brief introduction that explains an issue, practice, position, theory or explanation.
The reading states a thesis, that is, a position in favor of or against this practice, issue,
explanation, etc., and then presents three clear and distinct arguments that support its
thesis.
The lecture is on the same topic as the reading, and rebuts each of reading passage’s three
arguments. The lecture does not restate (though it may allude to) the arguments of the
reading, but it presents arguments of its own that “line up” with and counter the specific
arguments presented in the Reading. The lecture script is no more than 300 words.
The sample items that follow will interpret these abstractions. Note that after each of the
items there is a table of the “key points.” This is not seen by the examinee. The key
points are written by the item writer as a guide for the raters who will score examinees’
written responses.
THREE EXAMPLES OF TOEFL WRITING ITEMS
EXAMPLE 1
Were Dinosaurs Endotherms?
Reading:
Endotherms are animals such as modern birds and mammals that keep their body
temperatures constant. For instance, humans are endotherms and maintain an internal
temperature of 37°C, no matter whether the environment is warm or cold. Because
dinosaurs were reptiles, and modern reptiles are not endotherms, it was long assumed that
dinosaurs were not endotherms. However, dinosaurs differ in many ways from modern
reptiles, and there is now considerable evidence that dinosaurs were, in fact, endotherms.
Polar dinosaurs
One reason for believing that dinosaurs were endotherms is that dinosaur fossils have
been discovered in polar regions. Only animals that can maintain a temperature well
above that of the surrounding environment could be active in such cold climates.
Leg position and movement
There is a connection between endothermy and the position and movement of the legs.
The physiology of endothermy allows sustained physical activity, such as running. But
running is efficient only if an animal’s legs are positioned underneath its body, not at the
body’s side, as they are for crocodiles and many lizards. The legs of all modern
endotherms are underneath the body, and so were the legs of dinosaurs. This strongly
suggests that dinosaurs were endotherms.
Haversian canals
There is also a connection between endothermy and bone structure. The bones of
endotherms usually include structures called Haversian canals. These canals house nerves
and blood vessels that allow the living animal to grow quickly, and rapid body growth is
in fact a characteristic of endothermy. The presence of Haversian canals in bone is a
strong indicator that the animal is an endotherm, and fossilized bones of dinosaurs are
usually dense with Haversian canals.
(N) Now listen to part of a lecture on the topic you just read about.
Lecture:
(WA) Many scientists have problems with the arguments you read in the passage. They
don’t think those arguments prove that dinosaurs were endotherms
Take the polar dinosaur argument. When dinosaurs lived, even the polar regions where
dinosaur fossils have been found were much warmer than today—warm enough during
part of the year for animals that were not endotherms to live. And during the months
when the polar regions were cold, the so-called polar dinosaurs could have migrated to
warmer areas or hibernated like many modern reptiles do. So the presence of dinosaur
fossils in polar regions doesn’t prove the dinosaurs were endotherms.
Well, what about the fact that dinosaurs had their legs placed under their bodies, not out
to the side, like a crocodile’s? That doesn’t necessarily mean dinosaurs were high-energy
endotherms built for running. There’s another explanation for having legs under the
body: this body structure supports more weight. So with legs under their bodies,
dinosaurs could grow to a very large size. Being large had advantages for dinosaurs, so
we don’t need the idea of endothermy and running to explain why dinosaurs evolved to
have their legs under their bodies.
OK, so how about bone structure? Many dinosaur bones do have Haversian canals, that’s
true, but dinosaur bones also have growth rings. Growth rings are a thickening of the
bone that indicate periods of time when the dinosaurs weren’t rapidly growing. These
growth rings are evidence that dinosaurs stopped growing or grew more slowly during
cooler periods. This pattern of periodic growth—you know, rapid growth followed by no
growth or slow growth and then rapid growth again—is characteristic of animals that are
not endotherms. Animals that maintain a constant body temperature year round, as true
endotherms do, grow rapidly even when the environment becomes cool.
Prompt: Summarize the points made in the lecture, being sure to explain how they
challenge the specific points made in the reading passage.
Dinosaur Key Points:
Reading:
Thesis: There is strong evidence for
believing that dinosaurs were endotherms
like modern mammals and birds.
Polar Dinosaurs
The presence of dinosaur fossils at the
poles suggests that dinosaurs were
endotherms because only endotherms can
remain active under conditions of cold.
Leg Position and Movement
In general reptiles have legs out to the side.
In contrast endotherms have the legs placed
under the body to allow the sustained
running than an endothermic metabolism
makes possible. Dinosaurs had their legs
under their body, so they were probably
running endotherms.
Haversian Canals
Endotherms grow relatively rapidly and the
bones of rapidly growing animals typically
have Haversian canals. Dinosaur bones are
generally dense with Haversian canals. It
seems likely, therefore, that dinosaurs were
endotherms.
Lecture:
Antithesis The arguments that dinosaurs
were endotherms all have weaknesses.
The polar regions were warmer during the
dinosaur age, and during the cold season,
dinosaurs may have migrated or hibernated.
Having legs under the body allowed
dinosaurs to grow to large size, which was
advantageous but which has nothing to do
with endothermy. So it cannot be
concluded from the placement and
movement of the legs that dinosaurs were
endotherms.
Besides Haversian canals, dinosaur bones
often show growth rings that suggest that
dinosaurs stopped growing when
conditions became cold. True endotherms
continue to grow rapidly even during the
cold periods, so dinosaurs were probably
not true endotherms.
Example 2
Life on Mars
Reading Passage
For years, scientists have been trying to discover whether life exists on Mars. Most have
concluded that the Martian environment today is too harsh to support living organisms.
However, three recent discoveries lend support to the idea that there may very well be life
on Mars after all.
First, large amounts of methane have been detected in the Martian atmosphere. This
suggests the presence of living methanogens, tiny organisms that release methane into the
atmosphere as a by-product of their life processes. Methanogens get their energy from
hydrogen in rocks and do not need oxygen or sunlight to live, so they can survive in harsh
environments like Mars. Since methane can remain in Mars’ atmosphere only for a few
centuries after it's created, these methanogens must be present and producing methane
today.
Second, the existence of life on Mars is also strongly supported by a meteorite discovered
in Antarctica in 1996. The particular chemical composition of this meteorite indicates
that it is a rock from the surface of Mars. The meteorite also contains chemically pure
and flawless crystals of the compound magnetite. Such magnetite crystals are produced in
nature only by bacteria.
Third, observations of hydrogen on Mars made in 2001 by NASA’s Mars Odyssey
spacecraft suggest the presence of liquid water there that could support organisms. The
spacecraft detected the hydrogen in the top layer of Martian soil. This hydrogen is just
what one would expect if there were liquid water under the surface. A subterranean
reservoir of liquid water is an extremely suitable environment for microscopic life.
[Narrator] Now listen to part of a lecture on the topic you just read about.
Lecture
[Male Professor] Mars is a very unfriendly place for life. The evidence that has been put
forward as supporting there being life on Mars is actually very ambiguous. The facts as
we know them can easily be accounted for without supposing life on Mars.
First, methanogens are not necessarily the source of the methane. Methane can be
produced by nonbiological processes like volcanic eruptions. The same experiments that
detected the atmospheric methane also indicated recent volcano activity. Moreover,
unlike methanogens, volcanoes have actually been found on Mars. So since no
methanogens have been found yet, it’s more likely that volcanoes are the source of
Mars’s methane.
Second, while the magnetite crystals found in the meteorite from Mars were probably
made by bacteria, there is no indication that the bacteria themselves originated on Mars.
The meteorite arrived on Earth about 13,000 years ago, which means it has had thousands
of years to become exposed to bacteria on Earth. So it could have been bacteria on Earth
that formed the crystals we now observe. Indeed, recent research performed on the
meteorite found that, in fact, it has become contaminated by Earth bacteria. Since the
sample has been contaminated, the magnetite crystals on the meteorite cannot be used as
evidence for life on Mars.
Third, what about the supposed existence of liquid water near enough to the surface of
Mars to support life? The recent investigations of Mars detected hydrogen; they didn’t
detect liquid water per se. It’s actually much more likely that if there’s any water close to
the Martian surface, it’s frozen; in other words, it’s ice, not liquid water. And that’s not a
great environment for life.
Prompt: Summarize the points made in the lecture, being sure to explain how they
cast doubt on the specific points made in the reading passage.
Life on Mars Key Points:
Reading
Thesis: The evidence supporting the theory
that there are living biological organisms
on Mars is convincing.
Supporting Arguments
The large amount of methane detected in
the atmosphere suggests the presence of
living methanogens.
A meteorite from Mars contains magnetite
crystals, which can be produced only by
living bacteria.
Recent data indicate that there may be
liquid water under an ice layer, and thus
there may be living organisms.
Lecture
Antithesis: Though the evidence for life on
Mars is interesting and may warrant further
investigation, but at present it is
questionable.
Rebutting Arguments
Methanogens have not yet been found on
Mars, while volcanoes, which also produce
methane, have been.
The magnetite crystals on the meteorite
from Mars may have been caused by
bacteria form Earth; after all, the meteorite
has been on Earth for a long time and
studies show that the meteorite has been
contaminated by Earthly bacteria. Because
of the contamination, it cannot be
concluded that the crystals were formed by
Martian bacteria.
The existence of the ice layer is only an
inference from the presence of hydrogen in
the soil, and there are other equally
plausible, non-biological explanations for
that soil hydrogen.
Example 3
VOTING MACHINES
Reading Passage:
Critics say that current voting systems used in the United States are inefficient and often
lead to the inaccurate counting of votes. Miscounts can be especially damaging if an
election is closely contested. Those critics would like the traditional systems to be
replaced with far more efficient and trustworthy computerized voting systems. The
advocates of replacing the old system of voting with computerized voting machines
support their position with several arguments.
In traditional voting, one major source of inaccuracy is that people accidentally vote for
the wrong candidate. Voters usually have to find the name of their candidate on a large
sheet of paper containing many names—the ballot—and make a small mark next to that
name. People with poor eyesight can easily mark the wrong name. The computerized
voting machines have an easy-to-use touch-screen technology: to cast a vote, a voter
needs only to touch the candidate’s name on the screen to record a vote for that
candidate; voters can even have the computer magnify the name for easier viewing.
Another major problem with old voting systems is that they rely heavily on people to
count the votes. Officials must often count up the votes one by one, going through every
ballot and recording the vote. Since they have to deal with thousands of ballots, it is
almost inevitable that they will make mistakes. If an error is detected, a long and
expensive recount has to take place. In contrast, computerized systems remove the
possibility of human error, since all the vote counting is done quickly and automatically
by the computers.
Third, some people say it is too risky to implement complicated voting technology
nationwide. But without giving it a thought, governments and individuals alike trust other
complex computer technology every day to be perfectly accurate in banking transactions
as well as in the communication of highly sensitive information.
[Narrator] Now listen to part of a lecture on the topic you just read about.
Lecture:
[Professor] While traditional voting systems have some problems, it’s doubtful that
computerized voting will make the situation any better.
Computerized voting may seem easy for people who are used to computers. But what
about people who aren’t? People who can’t afford computers, people who don’t use them
on a regular basis—these people will have trouble using computerized voting machines.
These voters can easily cast the wrong vote or be discouraged from voting altogether
because of fear of technology.
Furthermore, it’s true that humans make mistakes when they count up ballots by hand.
But are we sure that computers will do a better job? After all, computers are
programmed by humans, so “human error” can show up in mistakes in their programs.
And the errors caused by these defective programs may be far more serious. The worst a
human official can do is miss a few ballots. But an error in a computer program can result
in thousands of votes being miscounted or even permanently removed from the record.
And in many voting systems, there is no physical record of the votes, so a computer
recount in the case of a suspected error is impossible!
As for our trust of computer technology for banking and communications, remember one
thing: these systems are used daily and they are used heavily. They didn’t work
flawlessly when they were first introduced. They had to be improved on and improved on
until they got as reliable as they are today. But voting happens only once every two years
nationally in the United States and not much more than twice a year in many local areas.
This is hardly sufficient for us to develop confidence that computerized voting can be
fully trusted.
Prompt: Summarize the points made in the lecture, being sure to explain how they
oppose specific points made in the reading passage.
Voting Machine: key points
Listening
Thesis: Traditional methods of recording
votes should be replaced with
computerized voting machines.
Reading
Antithesis: Introducing computerized
voting machines will not solve the
problems critics have of traditional voting
methods.
People unfamiliar with computers will find Computerized voting is user-friendly and
computerized voting difficult, and may
will make casting the correct vote easier for
miscast their votes or even not participate
people who occasionally miscast their vote
in elections.
using ballots.
Programs for computerized voting
Counting ballots by hand is tedious and
machines may contain errors, which may
results almost inevitably in errors; recounts
result in more serious miscount than errors would also be expensive; these errors will
due to human counting and no possibility
be eliminated by the computerized voting
of recount for some machines.
machines, because computers do not make
counting errors.
Financial and communication technology
Governments and individuals trust complex
are used daily and have been made reliable; technology to be fully reliable for financial
computerized voting takes place so seldom, transactions and communications.
it doesn’t have the chance to be fully
Computerized voting technology should be
tested.
no different.
TOEFL WRITING WORK SAMPLE
Each applicant for an internship in the TOEFL Writing summer institute is required to
submit a Writing Work Sample. The work sample has two parts.
Part I: Write an item
Write a reading + lecture stimulus with three point/counter point pairs based on one of
the two sources provided at the end of this document.
Part II: Produce a table of key points for a possible item based on the remaining source.
In this part, produce the point table only. Don’t write the item itself. (You do not need a
key points table for the item you produce for Part I.)
Both the item and the key points table should be put in a single Word document with a
title in the form of “last name first initial TOEFL Writing Work Sample” and submitted
along with the rest of your summer institute application.
General advice on writing your item.
Clarity, simplicity and directness: These are primary virtues that can take precedence
over completeness of coverage or nuances of a position or argument. Sometimes there is
an interesting or important argument that must be foregone because it cannot be delivered
with sufficient simplicity or because the reply to it cannot be made simply and directly.
Distinct: The distinctness of points/arguments should be maximized. Of course, each of
the three arguments supports the same thesis, but consistent with that, each point should
be as distinct as possible. [E.g. in an item about global warming, three different
arguments about carbon dioxide concentrations will all blur together in the examinee’s
mind. So one argument about carbon dioxide, one about retreating glaciers, and one
about the geological record.] In practice, the desideratum of distinctness may force not
using a good point/counterpoint because it would infringe on the idea of another point
pair.
Arguments: An argument is not just an opinion. For each individual point there has to
be a basis, grounds, evidence for the conclusion (the thesis). The rebuttal of the lecture is
not just a contrary opinion, and the rebuttal cannot ignore the reading argument. Each of
the three lecture rebuttals has to engage the reading argument of the corresponding point.
It has to undermine the evidence, or provide an alternative explanation of the evidence, or
take issue with an assumption explicitly or tacitly made in using the evidence to support
the Reading thesis. One can call the accuracy of the Reading evidence into question,
“new data suggests that...”, but one cannot rebut the reading argument just by disagreeing
with the argument’s conclusion.
Take into account the fact that the examinee will hear the lecture only once. So the
lecture in particular needs to be clear, simple and straightforward. So do not use
difficult vocabulary and be sure as part of the exposition to provide a brief gloss of a
technical term that cannot be avoided.
Each of the three points and their counters should (respectively) make a single reasoned
point, not a family of related points and counter points.
Do not try to cover all the material in the source or even on a given point. It is generally
necessary to simplify the issues. The point of these items is to provide a basis for
assessing the examinee’s ability to write, and we can’t do that if the response limiting
factor is the complexity of the reading passage and lecture.
Source material for your work sample:
Source 1
On the controversy of what caused the megafaunal extinctions at the end of the
Pleistocene
http://faculty.washington.edu/grayson/jas30req.pdf
Source 2 (presented in full below).
“The Case of the Servant with the Fur Collar”
Copyright 2005 The New York Times Company
The New York Times
September 22, 2005 Thursday
By CAROL VOGEL
Why was she wearing fur?
That was one of the first questions experts asked when they began studying a 17thcentury portrait of a woman who had the unmistakably stolid face of a servant but was
decked out in a sumptuous fur collar. And why did the light on her face appear to be
reflected off the dark surface of that collar when it should be absorbed by it?
These were puzzling questions, since the woman, whose head is covered in a plain
white bonnet, certainly did not seem to belong to the class of 17th-century Dutch society
that had its portraits painted. Some experts would have taken one look at the canvas and
immediately dismissed it as the work of a minor artist.
But when Ernst van der Wetering, the head of the Rembrandt Research Project, saw
the painting, he recognized something far more important than her dress.
For more than two years now, the painting on wood panel has been undergoing slow
but extensive restoration and study under the care of Mr. van der Wetering and Martin
Bijl, a former head of conservation for the Rijksmuseum in Amsterdam. With scientists’
help, they have concluded that the painting is a Rembrandt from about 1640 that someone
tried, a century later, to transform into a formal portrait.
No other experts have seen the painting—until today, when it goes on view at the
Rembrandt House in Amsterdam in anticipation of a celebration of the 400th anniversary
of the artist's birth in 2006. Now, the theory will be posed to the world’s Rembrandt
scholars. The painting is then headed for the auction block, at Sotheby's January sale of
old master paintings in New York.
George Wachter, director of old master paintings for Sotheby’s worldwide, has put
an estimate of $3 million to $4 million on Portrait of an Elderly Woman in a White
Bonnet a conservative figure, since it is not signed and a study. Mr. Wachter said he
wanted to let the market decide how much it is worth. (The last Rembrandt to come up at
auction was a self-portrait sold to the casino owner Steven A. Wynn two years ago at
Sotheby's in London for $11.3 million.)
How Old Woman came to Sotheby’s is a familiar tale in the world of old master
paintings. More than two years ago Howard Walsh Jr., a collector from Fort Worth,
approached Mr. Wachter to ask his opinion of the painting. He had inherited it from his
parents, Howard and Mary D. Fleming Walsh, who bought it from the Newhouse Gallery
in Manhattan in 1971. At the time they were told it was a Rembrandt.
Scholars were not so sure. It had been included in an 1836 catalogue raisonne of
Rembrandt's work and in a 1915 catalog about the artist. The painting was last published
in 1931. After that it was rejected from further Rembrandt catalogs.
“We couldn't sell it as a Rembrandt until we knew more,” Mr. Wachter said. “When
I first saw it, I thought it was a beautiful picture, but there was a white haze over the face
because the varnish had changed over time. Still, I could see there was a lot of quality to
it.”
He promptly sent a photograph of the painting to Mr. van der Wetering, who said he
was interested in seeing it firsthand—an encouraging sign, Mr. Wachter said. So with Mr.
Walsh's blessing he sent the painting to Amsterdam.
As head of the Amsterdam-based group of experts who have the last word on
whether a painting is by Rembrandt, Mr. van der Wetering has examined more than his
fair share of questionable old masters.
“We get piles of photographs, but when I first looked at this painting, I couldn’t help
but get excited, which is rare,” he said last month as he was standing in front of Portrait
of an Elderly Woman in a White Bonnet, which rested on an easel by the window of a
conference room at Sotheby’s in Amsterdam. “The execution of the head has a kind of
spontaneity.”
He then pointed to some delicate wisps of hair naturally spilling out of the bonnet
and the kind of intricate rendering of the bonnet, the woman's ear and her neck, which he
said he thought were unmistakably Rembrandt-like. “The execution of the head has that
combination of free draftsmanship and painterly execution.” But, he added, the way the
light reflected off the collar simply “didn’t make sense.” He was also intrigued by how
the artist captured a sense of light falling obliquely from behind the sitter in a way that
left her face largely in shadows.
He still could not confirm its authenticity until it had been restored. That’s when Mr.
Bijl, now an independent curator, was enlisted. He, too, saw the genius of Rembrandt in
the painting. “The first thing I always do is look,” he said in a telephone interview. “Then
I start talking with van der Wetering to see if our ideas are the same.”
Peter Klein, a wood biologist from Hamburg, Germany, was brought in to test the
oak panel. He discovered that it had come from the same tree as Rembrandt’s “SelfPortrait With a Hat” (1633), which is in the Louvre, as well as “Portrait of Willem
Burggraaff,” also for 1633, in the Staatliche Kunstsammlungen in Dresden, Germany,
which has been attributed to the artist's workshop, and a landscape in the Wallace
Collection in London that is thought to be from 1640 and was painted by a student of
Rembrandt.
By examining the work carefully, Mr. van der Wetering said he and Mr. Bijl could
see seams where wood panel had been added to the original painting. The dark
background had been painted over so old and new would look as one.
X-rays and pigment tests confirmed their suspicions that in an attempt to make it
more salable, someone had transformed the painting into a formal portrait, changing its
shape from arched to rectangular and adding a fur collar to make the sitter look more like
a lady.
In addition to removing decades of varnish and over-painting, Mr. Bijl also used a
Japanese saw (because it is both thin and delicate) to remove the portions of panel that
had been added, bringing the portrait back to its original shape.
X-rays showed many layers to the painting. Around the woman’s neck was the fur
collar, but under it there was a black layer of paint, and under that what Rembrandt had
originally intended: a whitish collar. That explained the reflections, since light could well
reflect off a white collar, but not off dark, fur.
“Thinking is the longest part,” Mr. Bijl said. “We had to be sure of each step we
took.” After two years of debate, he removed the fur, which revealed the yellowish-white
collar. The collar, unlike the whiter bonnet, has a yellowish cast because, a textile expert
confirmed, that “poorer people used a kind of starch for their collars that had a tendency
to turn yellow, but they used a better type of starch for their caps,” Mr. van der Wetering
said. “And Rembrandt was very aware of the difference of these tones.”
The bonnet’s point is folded back in such a way that the ear and the metal structure
supporting the bonnet are partially visible. “In our eyes, even though the cap is unusual in
the oeuvre of Rembrandt, it could only be by Rembrandt,” Mr. Bijl said.
Only after the fur collar was removed, he said, could he see the reflections of light
on the jaw, the cheek and the chin. Another Rembrandt touch: a small spot on the
woman’s cheek. In a number of self-portraits the artist added a blemish to his own skin.
Both men were so intrigued by Rembrandt's lighting that they tried to create a room
with it. Mr. Bijl then asked his wife to pose to try to determine Rembrandt's complex
rendering of reflections and shadows.
In the end Mr. van der Wetering concluded that the sitter was one of Rembrandt’s
household servants and that the painting is an oil study he did to experiment with the
issue of light.
“Rembrandt was called the painter of light, and rightly so,” Mr. van der Wetering
said. “He sometimes used himself as a model to solve particular pictorial problems in
front of the mirror. In this case, the problem of the figure lit obliquely from behind was a
problem that, in one form or another, occupied him throughout his life.”
[end of source article]