Friday, March 29, 2024 | Ramadan 18, 1445 H
clear sky
weather
OMAN
25°C / 25°C
EDITOR IN CHIEF- ABDULLAH BIN SALIM AL SHUEILI

Standardisation vital for online tests

Untitled-1
Untitled-1
minus
plus

Dr Adil al Khafaji


Standardisation of education is an issue that calls for a level playing field for all the educational institutions be it private or government. Two standpoints are often discussed in education, one that calls for unifying instructions and procedures and the other that encourages diversity of modes of actions.


The first provides better control over relevant operations and eliminates biased results, the second allows for opinions coming from different backgrounds and diverse schools of thinking. For academics and students, the first standpoint holds importance.


The general requirements of a ‘good test’ can be adapted to provide a framework for unifying testing and measurement procedures. It is believed that a degree of standardisation of test preparation and administration is an important necessity in the case of formal academic online testing.


Preliminaries


A test is a tool that is used to measure skills and knowledge learned in a given level usually through planned instructions, such as classroom instruction, and/or through self-study. This is of course not to deny other types of tests like those designed to measure talent and inborn potentials.


We usually give tests to assess the students’ level of learning. It is aimed at assessing whether the students have learned what we intended to teach. It is to gather information for adapting or improving instruction and to increase the level of learning and motivation. The last point, however, should not be taken to mean we teach through tests. ‘When you are an examiner you are no more a teacher’.


Standardised and non-standardised testing


There are four major elements in a test – questions, administering the test, scoring the answers, and interpreting the results. These four elements are consistent in a standardised test. Test design, administration, and scoring, as well as interpretation of results, are performed in a predetermined standard fashion. In non-standardised testing, the students are not necessarily tested in the same manner. Different questions may be given to different groups of the same level. On another count, the same test may be used but the time allotted may not be the same. Non-standardised testing can also be noticed in employing different methods of evaluation of answers and interpretation of results.


Standardisation can be performed at different levels — at the group level, at the level of one institution, a network of institutions, or even the entire country as in pan-country government examinations. Decisions to provide a degree of standardisation can be relevant to any of the above four elements and may vary in nature, but better be directed at controversial issues like the use of cameras to invigilate online testing.


In certain contexts, test-takers, particularly females, may object to this but would show less resistance if the decision comes from high levels of the hierarchy. Other contentious issues that can be standardised in online testing are consideration of technical failures and the provision of grace time.


Achievement tests and proficiency tests


Tests can be written with specific content material in mind. This material would have been the subject of instruction during a course of study at a certain level. The test can be based on this kind of material. Test writers are interested in knowing how much knowledge or skill the test-takers have been able to achieve or obtain during the study. Such tests are called achievement tests. They are usually used in formal teaching to decide whether a learner may proceed to an upper level of study or not. Such tests may not necessarily measure abilities like reasoning, problem-solving, evaluation, and the use of existing knowledge to acquire new knowledge. The label ‘proficiency tests’ is sometimes, but not always, reserved to measure the ability to perform processes of the latter type.


Standardisation would work on a different count when achievement and proficiency are in the spotlight. It would set up common minimum requirements for progression to upper levels in the case of achievement tests and delineate the required target performance in proficiency tests. Such delineations can be crucial in disaggregating test-takers’ results to consider fitness for certain tasks.


Validity, reliability, practicality, coverage, and relevance are characteristics of a good test. Other elements include balance, objectivity, clarity, and marks distribution.


For examiners, it is important to maintain a unified philosophy in test design, test administration, scoring, and result interpretation. He should be able to test the learners’ abilities to make initiatives and use taught material to acquire new knowledge.


There should be a test for one thing at a time. Items should test what they are supposed to test. Better use one consistent criterion of scoring, and consider available resources when preparing your test. Neither neglect available resource potentials nor over-estimate them. A good examiner should try to cover all taught material and test all relevant skills and test items should be relevant to the field and type of knowledge at hand.


Ideally, test items should reflect different degrees of challenge to human minds. They should have discriminative power. Tests should have model answers either conceptualised or written down to ensure the objectivity of scoring.


(The writer is an Associate professor of Linguistics and Translation at Al Zahra College for Women, Muscat)


SHARE ARTICLE
arrow up
home icon