Baldwin, Doug. A Guide to Standardized Writing Assessment. Educational Leadership, Oct 2004, Vol.
62 Issue 2, p.72-77.
Emphasis on “writing across the curriculum” and “writing assessment” are currently hot topics
in education. In Doug Baldwin’s 2004 article, “A Guide to Standardized Writing Assessment”, he states, “Strong
writing skills have become an increasingly important commodity in the 21st century. Students must be able to create
and polish pieces of writing on demand in preparation for the real world where they must be able to write quickly and clearly”
(Baldwin, 72). With more and more “paperless” workplaces using e-mail,
Blackberry, instant messaging and electronic papers and portfolios, writing has also taken a new shape. Students who will soon be entering the workplace must be familiar with this new world.
Technology is changing assessment .Not only how we teach writing, but also how we score and assess writing in the classroom. The new SAT is forcing all educators to take a look at assessment and how they can
prepare students for the new test. Many teachers do not like the idea of “teaching to the test”, however, we can
help students hone their writing skills by practicing the same types of assessments in the classroom. We must become familiar
with what standardized test scores are looking for. In the past holistic scoring, which is basically a score based on a first
impression of the overall quality of a piece of writing, was popular. The challenge
with the holistic model “has been scoring the increasingly large number or responses accurately and efficiently”
(Baldwin, 73).
Standardized paper and pencil scoring is slowly changing over to computer scoring.
In 1997, ETS (Educational Testing Service) began using the Online Scoring Network to score pieces for the (GMAT) Graduate
Management Assessment Test. Using this system, test takers’ responses are
sent via email to trained scorers who read the essays online. These scores are certified and take an online course that determines
whether they qualify to score or not. If qualified, like calibrating a scale, they need to successfully score a minimum of
10 papers before scoring actual test taker essays. In 1999, ETS took it even
further- with very little human scoring, and scored the test using e-rater software.
This software was developed by computational linguistics and “trained on hundreds of human-scored responses.
Currently the GMAT receives one score from e-rater and one from a human.
As a language arts teacher, this scares me. I feel it may lead to less
stylistic and creative writing. It will take the innocence and joy out of writing. Students will become hyper-focused on mechanics and writing to the test. This is similar
to the great rubric debate. Rubrics, when done correctly, can foster creativity, but they can also stifle. I am not completely
opposed to e-rater technology as long as it is only one piece of the assessment puzzle and there is always human scoring involved.
As Standard IV. states, technology facilitators should “Apply technology in assessing student learning of a subject
matter using a variety of assessment techniques”. It does not state
that it should be the only way.
