The Testing Trap:
How State Writing Assessments Control Learning
by George Hillocks, Jr. (2002)
New York: Teachers College Press
What effect do high-stakes L1 writing tests have on writing ability?
This book attempts to answer that question in K-12 contexts by examining the mandated testing programs
of five U.S. states. It illustrates how government mandated tests can adversely impact writing instruction
and underscores some of the problems of relying on high-stakes tests to drive curricula. Hillocks, a professor
emeritus of writing at the University of Chicago, denounces the No Child Left Behind policies of Bush
for promoting the superficial performance of explicit test features at the expense of in-depth learning.
This book starts off by examining how political factors impact assessment policies. Outlining the rise of
standards-based education in the United States, the author questions the rhetoric espoused by politicians
seeking to initiate school reform through testing. Noting how state-mandated testing programs tend to increase
drill and memorization practice and the likelihood of non-performers dropping out, Hillocks rhetorically
. . . should we expect tests, in and of themselves, to bring about higher achievement in schools?
Do we believe testing will bring about changes in what happens in classrooms, change that will
in turn bring about different and perhaps higher levels of student achievement? Do we believe that
simply the threat of sanctions will encourage students and teachers to work harder? Why should
we believe that? (p. 12)
The first sixteen pages of this text offer a montage of what happens when political imperatives collide with
constructionist notions of learning. The aftermath is not only a compromise of test content, but also a
sort of educational McDonalization (Hayes and Wynyard, 2002). One thing that means is that extended
essay writing which is not rapidly tested gets sidelined.
Hillocks notes how testing pervades our culture and our general ambivalence towards testing. Though many
accept the myth that "tests indicate achievement, intelligence, or aptitude, or all of these" (p. 14), others
complain about the human toll involved in testing and our morbid obsession with achievement scores. Hillocks echoes
Jones (2000) in stating ". . . the prerequisites of high stakes have lead inexorably to standardization and disempowerment."
"Most writing tests, in Hillocks' view, amount to little more than superficial pre-writing exercises,
since examinees are not given enough time to critically develop their essays."
The author encourages people to consider whether tests measure what they purport to. Most writing tests,
in Hillocks' view, amount to little more than superficial pre-writing exercises, since examinees are not given
enough time to critically develop their essays. Schuster (2004) echoes this view, describing most standard
writing tests as "tests of drafting" rather than "tests of writing" due to time constraints and the pervasive
use of the five-paragraph essay format. Hillocks appraises most timed student compositions as nothing but
"organized blether" [i.e. blather] (p. 80) and questions whether the tendency to superficialize education
is purely accidental by suggesting:
Part of the problem in this country is that systematic thinking about difficult problems seems
to be confined to an elite group. Most Americans are not willing to think much beyond their
own desires and perspectives, which often determines what they think is right and appropriate.
Perhaps most people cannot think carefully about complex problems. (pp. 6,7)
Would these same comments apply to Japan's educational system? If Baker and LeTendre (2000, p. 356) are correct, the situation
is by no means exclusive to the USA.
Hillocks directs his rhetorical cannons at the writing assessment tests in five U.S. states.
Noting how "millions of dollars, thousands of teacher hours, and hundreds of thousands of classroom student hours" are
spent on mandated writing assessments, the author concludes it's mostly a wasted investment. He also explores the connections
between assessment methodology and current traditional rhetoric (CTR). Rooted in the notion of an objective, external truth,
CTR is juxtaposed to the New rhetoric of Perelman (1969) and Berlin (1982). The author's sympathy for the latter and disdain
of the former is not concealed. At times Hillocks' censure of opposing views is almost too doctrinaire, as in this example:
Contemporary educational theory and practice argue that effective learning must be constructivist
in nature, and that students learn best and perhaps only when they can construct knowledge for
themselves within the framework of their existing knowledge. (p. 22)
This type of argument presupposes that there is only one sort of learning. However theorists such as Bloom (1956) and Gardner (1983)
suggest that there are in fact multiple types of learning. Rote learning might be appropriate for certain tasks, but certainly
not for the full scope of education.
Differences in the assessment standards of several U.S. states are quickly overviewed. Whereas some states
have minimal writing competency requirements for high school graduation, others don't. Moreover, whereas some states include student
portfolios in their assessments, other states rely exclusively on timed in-class exercises. Instead of thinking of "writing tests"
as uniform, the author points out how writing is operationalized quite differently by various testing agencies.
Chapter 5 is a lengthy lambaste of the Texas Assessment of Academic Skills, which was in use from 1991 to 2003 (Wikipedia, 2006).
Hillocks contends that the narrow focus of this test fosters writing instruction on surface mechanics rather than deep content.
Noting how a composition ". . . may be grammatical, well-organized, and coherent without being well-reasoned, thoughtfully developed,
or effective in any way", (p. 35) he alludes to the need for teaching more critical thinking skills in composition classes. Hillocks
also criticizes the way the passing scores can be politically manipulated – instead of investing money to raise test standards,
it's often easier to simply lower test standards so that a greater number of students will pass.
The chapter that follows claims to examine teacher responses to the Texas test. Unfortunately, the author doesn't mention how
many teachers were interviewed or anything about his data collection process. In a book that is critiquing testing methodology,
this type of error is perplexing. All of Hillocks' information is reported in percentile figures without any reference to sample
sizes. One could, perhaps, regard this chapter as a qualitative case study. Even by those standards, however, not enough background
information about the study was reported to actually interpret it. In short, we are left with nothing but rhetoric and anecdotes.
The attack then shifts to Illinois. Hillocks contends that the benchmark papers and detailed prompts for the Illinois Goals Assessment
Program (IGAP) exam leads to a "homogenization of writing" (p. 110) which is predictably banal. Hillock ascribes this
condition to two factors: government restrictions and the fact that the tests aren't solidly grounded in what he believes to be sound
writing theory. Since only 15 school days per academic year in Illinois may be used for testing, the state writing exams are limited to
40 minutes. In Hillock's view, the theories of Kinneavy (1971) provide the best grounding for writing tests. Kinneavy creatively fused
classical rhetoric with communication theory and underscored the importance of textual modes (description, narration, evaluation, and
classification). His ideas have shaped the writing exams in many U.S. states, but apparently not Illinois. The Illinois test, according
to Hillocks, focuses too much on the traditional five-paragraph essay and not enough on rhetorical principles.
Teacher responses to this test were described as "ambivalent" (p. 124). Once again, not enough methodological information is provided to
critically interpret the data.
The next chapter starts out with high praise for the target objectives of the New York State Regents exams, then a close look at a
benchmark essay for that exam. Gradually, the author voices concerns about whether or not the test is being graded in a way that measures
its avowed standards. Hillocks contends that what is purportedly "in depth analysis" often amounts to little more than a blithe regurgitation
of previously stated facts.
Chapters 10 and 11
Shifting to Kentucky and Oregon, the author quickly delineates how writing is assessed in both states. Both states employ some type of
portfolio assessment as well as in class writing. Predictably, the author tears apart how those tests are graded,
contending that the benchmark sample papers often do not fit the standards specified in the test rubrics. This chapter concludes by offering
some praise for the Kentucky writing portfolio assessment system. Hillocks says it is one of the few writing tests he examined that strives
for some semblance of a real audience and a broader writing purpose (p. 46).
"the Kentucky writing portfolio assessment system . . .is one of the few writing tests . . . that strives for some
semblance of a real audience and a broader writing purpose."
The book concludes with a recap of the main themes of the text. In fact, readers could skip the first eleven chapters of this book and glean the pivotal points from this final chapter. After briefly overviewing the malaise afflicting all five states examined in this text, the
author turns his attention to possible solutions. The main need, in Hillocks view, is not more testing but better teacher training:
At the center of the K-12 testing fury is the myth that testing alone is able to raise standards and the rate of learning. Certainly,
testing assures what is tested is taught, but tests cannot assure that things are taught well. If states want teaching to improve they
will have to intervene at the level of teaching. Teachers need opportunities to learn more effective procedures for teaching writing.
Tests of writing cannot teach that. (p. 204)
In addition to providing more teacher training, Hillocks feels it is essential that teachers have adequate time to teach writing well. Devoting
classroom time to tests takes away from time that could be spent on writing. And since most American writing textbooks geared for the K-12
market are stuck in CTR-mode, Hillocks believes teachers also need time to develop their own writing materials. However, with an average
American K-12 writing teacher dealing with 130 students a day, can teachers actually develop their own materials and also give detailed
feedback on the essays they receive? The fact that formulaic CTR essays are much easier to grade results in a sort of "production line"
mentality among many public school teachers and there is an economics at work that favors McDonaldized instruction. Unless public pressure
is increased on law makers and state officials, we may well see a two-tiered educational system: with elite schools for the wealthy few and
mediocre schools for all others. Hillocks ends exhorting readers to get away from "vacuous thinking and writing" and make a commitment to
At times this book could be criticized for being too narrowly focused on the high school writing programs of merely five US states during the 1990s.
However, this work also raises perennial questions relevant to language teachers around the world.
"Writing is a skill which defies superficial, facile assessment: the communication of deep ideas is hard to translate readily into numerical variables."
One issue concerns content validity. Although many of the writing tests examined in this book have technical reliability, most lack content validity.
Writing is a skill which defies superficial, facile assessment: the communication of deep ideas is hard to translate readily into numerical variables.
Another issue concerns the optimal mode of instruction. In high stakes testing environments, Hillocks contends that "instruction" often degenerates
into mere examination-preparation drilling. He argues against the traditional lecture/recitation modes of learning and emphasizes the need for
more in-depth student interactions. He also complains about the excessive teacher talk in most classes, noting that when students do speak
it is often to answer teacher questions through a single word or phrase. In other words, classroom discourse too often is disconnected and fragmented –
ironically similar to the information on many tests. Hillocks echoes Hargreaves and Fink (2003) in criticizing the breadth-without-depth nature of
most school interactions, which in turn is mirrored in the shallowness of most tests. Stressful high-stakes testing environments, in Hillocks view,
effectively prevent deep interaction and augment the sense of alienation reverberating through many classrooms.
A third issue concerns test washback in general. Perhaps too simplistically, Hillocks views state-mandated exams as significantly determining
teacher behavior. Cheng (2004, p. 148) suggests washback is a more complex phenomena and Stoll (cited in Haney, 2000) adds, "It's immensely hard
to get a critical mass of teachers within a school, let alone a district, to significantly change their practice." Even as new laws and new tests
are enacted, teachers are often remarkably inure to change.
Are the 224 pages of this book actually worth reading? For American K-13 school administrators, perhaps. For teachers of EFL, or indeed any foreign
language with an interest in assessment issues, however, I can't help but wish this volume would be condensed into a single chapter of a larger work
which deals with test impact in terms of all four language skills from a wide range of international contexts. Such a book would begin to cover the
scope suggested by its title.
- Reviewed by Tim Newfields
Baker, D. P., & LeTendre, G. K. (2000). Comparative sociology of classroom processes, school organization, and achievement
In M. T. Hillinan (Ed.). Handbook of the sociology of education. Berlin: Springer. 345-364.
Bloom B. S. (1956). Taxonomy of educational objectives, Handbook I: The cognitive domain. New York: David McKay Company Inc. (republished in 1984 by Pearson Education).
Cheng, L. & Watanabe, Y. (2004). Washback in language testing: Research contexts and methods.
Mahwah, NJ: Laurence Erlbaum & Associates.
Gardner, H. (1983). Frames of mind. New York: Basic Book Inc.
Hargreaves, A. & Fink, D. (2003, September). Educational reform and school leadership in 3-D perspective.
Nottingham: National College for School Leadership. Accessed December 4, 2006 at
Haney, W. (2000, August 19). The myth of the Texas miracle in education, Part 8: Summary and lessons learned.
Education Policy Analysis Archives, 8 (41). Accessed December 4, 2006 at
Hayes, D. & Wynyard, R. (2002). The McDonaldization of higher education. Westport, CT & London: Bergin & Garvey.
Jones, K. (2000). High stakes vs. democracy. FairTest Examiner.
Accessed December 1, 2006 at http://www.fairtest.org/examarts/Fall%2000/Jones.html. [Expired Link].
Kinneavy, J. L. (1971). A theory of discourse. Englewood-Cliffs, NJ: Prentice-Hall.
Newkirk, T. (1999, December 1). Teaching to the test means 'dumbing down' the curriculum.
Accessed December 19, 2006 at http://www.nexthorizon.unh.edu/news/news_releases/1999/december/tm_19991201curriculum.html.
Schuster, E. H. (2004, January). National and state writing tests: The writing process betrayed.
Accessed December 1, 2006 at http://www.pdkintl.org/kappan/ k0401sch.htm. [Expired Link]
Wiggins, G. P. (1998). Educative assessment: Designing assessments to inform and improve student performance. San Francisco: Jossey-Bass.
Wikipedia. (2006). Texas Assessment of Academic Skills.
Accessed December 6, 2006 at http://en.wikipedia.org/wiki/Texas_Assessment_of_Academic_Skills.