Page 219 - DEDU504_EDUCATIONAL_MEASUREMENT_AND_EVALUATION_ENGLISH

Unit 18 : Marking System : Need, Problems, Components

outcomes are clearly listed in term of competencies implied in various instructional Notes
objectives. However, its utility is limited to a small fraction of school learning.
Moreover, the task is to be specified to such an extent that it leads to trial outcomes
that become unwieldy.
(b) Marks in terms of students ‘potential to perform : It raises the issue of whether what
one can do, does he really do ? How best can we measure the potential that is not
contaminated by the present achievement ? Can we really make the needed judgment
of achievement in relation to potential, segregating it from simple achievement ? If
so, how valid and reliable such a judgment can be made by teachers. Moreover, how
marks can then be used in different contexts, e.g. for admissions to higher courses,
employments, selections etc., if assigned in terms of potential of the performer. What
is important is that what John can in fact do or not, whether he does the best he can.

(c) Performance in relation to peer groups : It means that an individual’s performance
can be judged in relation to some reference group. This is the type of referent that can
be used in relation to a class, school, district, state or the national group (as used in
standardised test) norm. This is indeed norm-referenced judgment. Problem here is
the nature of the group in relation to which a student’s performance is judged, whether
random or stratified. Government schools, central schools, public schools etc all make
a difference in judging performance of a student, with reference to a brighter, poorer
or similar group performance and the size of the group.
(d) How standards to be equated : How best to anchor any specific group to that frame of
reference ? For example, common syllabus and course objectives taught by teachers
of all sections of a school or teachers of different schools following the same course
provide reasonable anchoring device. In groups that are more diverse, one may think
of the previous grade-point average or scholastic aptitude test as a rough anchor
(Thorndike). If common test is given to three sections and frequency is worked out,
we can calculate the number getting As, Bs, Cs and so on, depending upon the policy
of grading in an institution or an examining agency, e.g. 15%As, 25%Bs, 40%Cs,
15%Ds, and 5%Fs etc. Sometimes a student getting B grade may finally get A grade if
his previous grade-point average is taken into account by the institution or another
indirect indicator like S.A.T. is considered besides the anchoring test. Percentage of
each grade awarded corresponds, step by step, with the percentage in that category
on the common test. This may lead to more than 15% As in a section and less than
25% Bs or less than 15% As in another section with more than 25% Bs etc.

18.4 Derived Scores

Test scores can be expressed in a variety of different units in relation to different scales of
measurement. A score may represent very good performance on a test, but the same score on
another test may represent very poor score. Therefore some common basis has to be established
to make comparison of scores on those widely different tests possible. This may reflect relation
of derived scores to norms. These are of three types.
(a) Derived Scores Based on Median Performance
These are the grade scores or grade equivalents and the age scores Both depend on norm
tables with raw scores that grade equivalent or age scores can be determined.
(i) Grade-equivalent scores : Indicate the position on a grade scale at which a students’
test performance places him. A child scores on a test that is equal to the median or
average score of pupils, of say 5th grade students. His grade equivalent on the subject-
matter of the test is 4.0 regardless of whether he is in class V or III. Sometimes these

LOVELY PROFESSIONAL UNIVERSITY 213