Page 72 - DEDU504_EDUCATIONAL_MEASUREMENT_AND_EVALUATION_ENGLISH
P. 72

Educational Measurement and Evaluation


                   Notes              domain to be measured.” Explaining the meaning of representative sampling, Frederick G.
                                      Brown has said,  “Representative sampling means selecting items in proportion to their
                                      emphasis or importance.”
                                      The tests constructed for evaluating content knowledge, skill or other teaching objectives should
                                      be established with content validity, and this is a very important task, The chief purpose of
                                      achievement tests is to evaluate students’ knowledge and skills etc. relating to the content.
                                      “Content validity is the chief characteristic of this type of tests. Content validity indicates how
                                      far a test evaluates that knowledge, that is, how far a student has acquired learning in that
                                      subject, how much his understanding is, what skills he has acquired, etc.
                                      By content is meant both subject matter and instructional objectives. By subject matter is meant
                                      behavioural changes which are produced in students by different units of knowledge and
                                      instructional objectives. How far a test evaluates these two, can be found out from its content
                                      validity. Subject matter can be used to find out validity only when it is helpful in realizing
                                      teaching objectives, and not merely for teaching in the classroom. According to the National
                                      Committee on Test Standards (1966), “In the case of an educational achievement test, the
                                      content of the test may be regarded as definition of the one or more educational objectives.”
                                      (a)  To list out all units of the whole knowledge under that subject matter and to select those
                                          important units on which the test has to be based.
                                      (b)  To list all those important and desirable behaviours which are necessary to be produced
                                          in students through instruction, and to select only those of them on which the test will be
                                          based.
                                      (c)  Of the above two lists, to determine how much importance to be given to each list.
                                      (d)  To prepare table of specification.
                                      (e)  To determine the number of questions or problems or items of the test on the basis of
                                          table of specification.

                                      Despite the fulfillment of the above characteristics, content validity has its certain limitations.
                                      According to Frederick G. Brown, its chief limitation is the absence of quantitative index which
                                      can summarize the degree of validity. In his words, “The major limitation of content validity
                                      is the lack of quantitative indices that summarize the degree of validity.” Secondly, this type
                                      of validity cannot be used in finding out the validity of ability tests and personality tests. Also,
                                      it is not a simple task to select the representative samples.
                                  2.  Criterion-related Validity : When we desire to predict about future working efficiency or
                                      performance of some individuals on the basis of some test, or we want to estimate the
                                      performance of some individuals in a certain field, then the test has to be based on criterion-
                                      related validity. For example, if we want to know, how far the intelligence level of students can
                                      correctly predict their success in some future examination, then we will have to study criterion-
                                      related validity. This type of validity is called predictive validity, because it indicates to the fact
                                      about what can be said about the achievement or success of some individuals in a future test.
                                      In the second case, the purpose of a test may be to measure the present ability, skill or behaviours
                                      in a certain field today. How far a test measures these present characteristics of individuals
                                      correctly, is called its concurrent validity. These two types of validity come under criterion-
                                      related validity.
                                      According to Grounlund, “Criterion-related validity may be defined as the extent to which
                                      test performance is related to some other valued measure of performance.”
                                      In predictive validity, this ‘other measure of performance’ is available in the future, and it
                                      indicates to the fact how for the success in a future task their success can be predicted.
                                      In the words of Frederick G. Brown, “Predictive validity refers to situations in which  criterion
                                      data are collected at some future time.”



        66                                 LOVELY PROFESSIONAL UNIVERSITY
   67   68   69   70   71   72   73   74   75   76   77