yeas

3%. INNOVATION ABSTRACTS xe:5"

4 j
BW tA Published by the National Institute for Staff and Organizational Development
ny With support from the W. K. Kellogg Foundation and Sid W. Richardson Foundation

 

 

 

 

 

TANDEM TESTING

The great American myth has it that success comes to those singular-minded, hard-driving individuals
who force their way to the front of the pack. Students who survive a competitive educational system, who
surpass all the barriers from admission to exams to graduation requirements, must be presumed to have the
advantage over the less competitive (though perhaps no less able) peers over whom they leapfrog. But for
some time social psychologists have maintained that cooperation is the key to success in any society.
Recently the business world has beheld with admiration the success of the Japanese "Theory Z" methods,
the essence of which is cooperation. James Coleman, in an article entitled "The Children Have Outgrown
the Schools" (Psychology Today, February, 1972), has emphasized the importance of education’s task of
turning out responsible, productive persons who can effectively participate in cooperative situations.

Providing opportunities for cooperation in the classroom is not a new idea. Group projects are a common
instructional approach, and there is a body of literature emphasizing the benefits. Peer instruction and
study groups provide evidence of the gains to be made. But what about student cooperation in testing?

In most classes, that would be called cheating, but in my classes, we refer to it as collaborating. The
procedure is simple: two students are given one copy of the test. They hand in one answer sheet with both
their names on it, thus getting the same grade. Scoring is done immediately following the test to allow the
pairs to examine any missed items. The pairs are formed by the students. Some pairs "bond" and test
together through a three-quarter General Psych sequence. Other pairs are less stable. I encourage students,
if they feel a partner is not pulling his/her weight, to find a new partner; but pair stability is the norm.

I began using tandem testing with one class in the summer of 1980, but it has been so well-received by
students and so effective that I now use it in all my classes. Classroom atmosphere on test day is relaxed.
A murmuring hum fills the air, punctuated by an occasional laugh. I can see the students’ involvement with
the material; in an attempt to persuade the partner of the correctness of an answer, one student may refer to
something "she wrote on the board"; another student may cite a certain section of the text—they are
actually teaching each other while taking the test! One student catches another's misinterpretation; one team
arrives at a compromise: "We'll put my answer for number 6 and yours for number 24, okay?"

Each test provides evidence that my aim of fostering cooperative skills is yielding results. Communication
skills are honed through struggles to express one’s ideas to a partner. Efforts to influence one another range
from "I don’t know why, I just think my answer is right" to detailed explanations and examples.
Development of trust follows the experience of going along with the partner, in doubt or in confidence, and
getting the item right. Conflict management styles emerge through practice, as both partners must agree on a
single answer.

Predictably, the students love it! In response to a brief questionnaire given to 103 students in my spring
1984 classes, 67 percent reported lower test anxiety with tandem testing while 15 percent said their anxiety
level was higher. Seventy-eight percent reported the same amount of study time, 12 percent said they
studied more, and 13 percent reported studying less. (In discussing the questionnaire, one student
volunteered that she studied less because the support of her partner allowed her to abandon her compulsive
over-studying.) Effect on grades: 53 percent believe their grades are higher with this system, 40 percent see
no effect, and nine percent felt their grade was lower. I plan an analysis to check statistically the accuracy of
the students’ perceptions.

 

Ce, Community College Leadership Program, The University of Texas at Austin, EDB 348, Austin, Texas 78712