February 2003 Issue
Do College Reading Placement Tests Measure College Reading Skills?
By Gary K. Probst
College reading placement tests play an important role in placement of students in most colleges. These tests can determine if a person is accepted into a college, not permitted to take certain courses, or is required to take a reading and study skills class. The terms used to describe the reading skills measured by these placement tests are very general impressive sounding cognitive terms. However, the test manual that comes with these tests contains a large amount of statistical information that tells very little about the exact reading skills measured by these tests.
What is the purpose of this article?
It is the purpose of this article to describe the exact reading skills measured by the following tests:
From the analysis of the skills on college reading placement tests the following will be shown:
This article will be in two parts. Part 1 will describe two reading placement
tests that do not measure college reading skills. Part 2 will
Before selecting a test to determine if a student has college reading skills, it is necessary to determine what are the reading requirements of a college course. The reading requirements of a college course I gave in an earlier article titled "What is the Omphalos of Your College Reading and Study Skills Course?" This article can be found at:
Keeping these reading requirements in mind, the question to ask is, "Are these reading requirements measured by college reading placement tests?"
Taxonomy of Cognitive Skills
In order to perform an assessment of what skills are measured by a reading
test, there must be a taxonomy of cognitive/reading skills. There are several
taxonomies of cognitive skills published and given as solutions to writing
objectives and test questions. Probably the most famous is Bloom's taxonomy
developed in 1956. However, this taxonomy as well as the others does not
describe the specific cognitive skills that are required on each level. While
these taxonomies are often used to describe learning objectives, they are so
general that they are of no use.
The taxonomy's four cognitive levels are facts, concepts, principles and problem solving. As one moves to the higher cognitive level, it becomes increasingly more difficult to perform the cognitive skills. This is because of the following two reasons:
By using this taxonomy, it is possible to identify the specific reading skills measured by a reading test.
Accuplacer Reading Test.
The Accuplacer tests are widely used in colleges for student placement. The
Accuplacer reading tests purports to measure two reading skills. Narrative
questions are given that required developing a topic sentence for a passage of
three sentences. Sentence relationship questions are given that have two
underlined sentences are followed by a question or statement about them.
The question for each underlined sentence pair asks how the second sentence is
connected or related to the first sentence as an example, cause, effect,
comparison or contrast.
Example 1: http://www.frc.mass.edu/casa/accuread.htm
After reading this passage, I question the following information:
1. Does the suggestion to read newspapers, etc in your home and then asking yourself, "What is the point? Really prepare your to take this test or prepare a person to take a test that measures the reading skills required in college courses?
2. Is knowing there is snow on the ground when shoveling a driveway an
inference required of a three year old or a college student?
The correct answers given are C and E.
For college placement the Accuplacer reading test has no passages to read. The test mainly asks a student to tell the connection or relationship between the second sentence to the first sentence. A few questions are given that require recognizing what would be a topic sentence for three sentences. The questions on the Accuplacer are not the type of questions one normally finds on a reading test or a college test. Therefore, the questions' specific cognitive level and cognitive skill cannot be identified. It appears the questions range from an easy to difficult principle level.
The Nelson-Denny Reading Test, Forms E and F, published in 1981
So as not to violate test security the number of the specific question on each cognitive level is not given. An analysis of this test reveals number of questions on the following cognitive skills:
Fifty percent of the questions on the Nelson-Denny test measure the ability to
read to locate a fact. Only twenty percent of the questions measure reading on
the principle level. The reading selections are short passages consisting of one
or two paragraphs. Therefore, the reading skill measured by these tests is
scanning for a fact.
Questions or comments? Discuss this article with the author or with other readers by clicking here.