© Springer India 2015
Savita Malhotra and Subho Chakrabarti (eds.)Developments in Psychiatry in India10.1007/978-81-322-1674-2_22. Development of Psychological Testing in India
(1)
Clinical Psychology, Department of Psychiatry, Postgraduate Institute of Medical Education and Research, Chandigarh, India
Keywords
Psychological testsConstructionStandardisationMeasurementS.K. Verma, Former Additional Professor of Clinical Psychology
1 Introduction
When I joined the Department of Psychiatry at the Postgraduate Institute of Medical Education and Research (PGIMER), Chandigarh, many of the psychological tests used in most parts of the country suffered from several deficiencies, known and unknown till then. In fact, one cannot work in an area where the tools in use are deficient in many ways. If one has to work honestly in this area, one cannot hide behind these imperfect tools, without doing anything about it. Blaming the tools alone cannot help one for long, particularly, when this happens to be one’s chosen field of work and worship. The God is everywhere, seeing everything and forgiving nothing.
But, before I name these deficiencies in tests in use then, I may be allowed to define an ideal test and mention its necessary characteristics. In all fairness, I must accept that no single psychological test can be considered as a perfect test for all types of populations, for all subjects and for all times to come. But, realising our own limitations in knowledge, wisdom and availability of time and resources, a psychological test has to have most of these desirable characteristics, while leaving scope for improvement in the future.
2 Definitions
A psychological test is a standardised method of (a) comparing the behaviour of two or more persons at the same time and/or (b) comparing the behaviour of the same person in different periods of time.
A standard is that in terms of which we measure something, e.g. a gram, a litre, a mile, a rupee or an inch; measurement is attributing a number according to well-defined rules.
Behaviour includes both overt and covert behaviour, observed and expressed verbally and/or measured.
Standardisation is a process of exactly fixing the stimulus variables, the exact situation of administration and of observing the behaviour, in terms of well-defined units of measurement and the exact method of interpretation, in the light of relevant norms, having high reliability and validity.
Objectivity refers to freedom from subjective factors in administration, measurement and interpretation.
Reliability means consistency of scores over a period of time and consistency in administrative methods, administration by self or by others, the method used by scorers as well as the test’s internal consistency. It makes one confident that every time a test is used, the same test is used and similar methods are used for administration, scoring and interpretation. Reliability can be of the total score, of various parts of the test or of every individual item. It is mostly measured in terms of coefficients of correlation, but can also be estimated in terms of percentage of agreement or disagreement.
Validity of a test shows whether a test measures what it purports to measure, e.g. a test may measure more than one aspects of behaviour, but one of them may be more specific.
Intelligence tests, for example, must measure intelligence in particular, but may also show the approach in solving a problem, a method of solving problems and how a person reacts to frustrations when difficulties arise while solving a problem. A personality test may also measure one or more aspects of personality, i.e. interest, attitude, preferences and aptitudes. A test is a valid measure of behaviour and may be valid for one aspect of behaviour only. The validities are also of different kinds: face validity, content validity, concurrent validity, construct validity, factorial validity and predictive validity.
Other features of validity of a test are as follows:
A test is valid for something with which it correlates.
Validity depends upon the reliability of the test. In fact, validity cannot exceed the under root of its reliability.
A test may have different validities for different aspects of behaviours.
Norms have to be relevant and meaningful. Local norms have to be used for interpretation. Population, reliabilities and validities are defined by the standardisation process and must be given in the test manuals. Norms have to be specific with regard to location, age, sex, education and all other significant variables likely to influence test scores, reported or not reported so far.
Test manuals should ideally provide all the above information. No test can be culture free and culturally fair unless it is proved, reported and defined. Manuals can also provide cross-validation data, if and when available. That is why a test can be revised and has to be revised after sometime, in order to remain a valid and reliable measure over time.
3 Drawbacks or Deficiencies of Psychological Tests
The following drawbacks were most prominent when I joined the department in 1968:
Most tests used were in English language only, and on-the-spot translations were being done on the bedside of patients in the general hospital setting. Uniform Hindi translations were needed, with local norms for use with them.
Patients were illiterate or barely literate and unsophisticated in the use of tests. Some of them were even so frustrated that they were ready to give their thumb impressions, in place of responding to test items.
Foreign norms were being used while scoring and interpreting the test scores, whereas there was the need to prepare local and relevant norms for use with our patient population.
The tests used were too time consuming in terms of length of items, e.g. nearly 200 in the General Medical Index Health Questionnaire or the CMI Health Questionnaire (Erdmann Jr. et al. 1952) and 60 in Standard Progressive Matrices (Raven et al. 1998). The language used in Hindi versions of personality tests was complicated, with difficult words, e.g. “ashrupat” for tears in the eyes, as in the Hindi Personality Trait Inventory (Verma et al. 1990).
Tests in simple Hindi were not available, which were constructed using patients’ own symptoms with their own expressions, and for which, local norms were available or which could be used at all educational levels.
A need for modifications of some tests was felt, but no constructive and useful work was done, and at best, only lip service was paid. The problems were noted, but nothing was done to address them. There was a scope for construction of tests in many areas, but much needed work was missing in the absence of a team of clinical psychologists devoted to this work wholeheartedly.
In a few areas, e.g. Bhatia’s battery of performance tests of intelligence in India (Bhatia 1955), where useful, constructive work was done with separate scoring systems for literate and illiterate subjects, there was lack of follow-up work with revision of the scoring system and revision of norms from different parts of the country.
In addition to the above drawbacks, the attitude of the so-called experts from Indian universities was deplorable, to say the least, because test construction was looked down upon even for a PhD work. In my personal experience, while doing my PhD, the “experts” made one excuse after another for it—such as “It is not sufficient work for PhD” to even saying “One PhD is not sufficient for test construction, or “Test construction in India is not accepted for PhD,” or “Only in North India some universities have accepted a few PhDs in test construction,” or “There is nothing original in test construction.” Even when evidence to the contrary was available, such excuses continued to be made.
I was lucky enough to have Prof. Wig (who accepted my arguments and allowed me to continue my PhD work) who accepted to be my supervisor and encouraged me to continue my experiments. (That is why I call him my godfather.)
These “experts”, when I approached them for guidance or advice, were of the opinion that we should adopt standardised tests from the West and use them with our population to solve any research problem, a piece of advice, which I could not and did not follow, for obvious reasons. I was too junior a researcher, so I bowed my head in silence, but continued my work, the only way open to me at that time. I do not know what kept me going in those early years, despite the continued opposition by my seniors, colleagues and juniors, both psychiatrists and clinical psychologists, in this Department of Psychiatry (not of Clinical Psychology). But looking back over the years, I can name a few possible “reasons” for it.
I saw construction of psychological test as writing beautiful poems as a hobby, enjoying the process, as my way of worship to the “God of Measurement”, if there is any. (If others can worship other Gods or Goddesses such as Brahma, Vishnu, Mahesh, Lakshmi or Saraswati, why cannot I have my own God and worship Him.)
I had read somewhere the definition of morale (given by a veteran soldier) as “Morale is what keeps your feet going while your head says it can’t”.
So I bowed my head and like a good student, started doing my work, as a worship in order to please him (i.e. the God of Measurement), to shower His invisible blessings (no matter if it is seen or appreciated by others or not). To be able to do one’s preferred work, is its own reward and can never be replaced by any other award or reward.
Appeal of my heart—“Do not give up before doing your best, no matter what the consequences”. So I did continue my work as worship, on a war footing, if I can call it that.
My father’s advice—“Whenever in difficult circumstances, bow your head in silent prayer and start doing what you think is right, without caring what others say or do”. My late father often referred to his own example—when he started his own work as a dentist, people made fun of a dentist as a mere tooth puller, as one who is destined to end up on the road calling people for repairing their broken teeth. But, he never gave up and ended as one of the leading dentists in the city of Varanasi at the time.
Never-ending faith of my late wife in my ability to overcome my shortcomings, my difficulties and be successful in the end. This moral support proved to be of great help, whenever I happened to be in low spirits and seeing faces of my own children, as if asking me “Water water everywhere nor any drop to drink” (Coleridge in his “Rime of the Ancient Mariner”), as to what messages would my life and efforts leave behind, for them to follow. (“Keep your head above the sea level, in order to survive and win”.)
It is rightly said that “All is well that ends well”. At the start of my professional career, opposition to my views was the rule; this was gradually converted into active cooperation by all concerned, from all parts of the country. My prayers were being answered after all and in my life time.
When there were other PhDs, not only by clinical psychologists but also by psychiatrists of the department, it was more than I could ever expect or imagine or live to see.
In this regard, I must gratefully acknowledge the significant contribution of other fellow clinical psychologists, particularly of Dr. Dwarka Pershad, Ex. Additional Professor of Clinical Psychology, Department of Psychiatry. Indirect encouragement to our efforts was soon available in the form of a few national awards for such tests from professional bodies, including one by the Indian Psychiatric Society, and my own nomination to advisory bodies and editorial boards of various journals in the country. I admit here once again the full and constant support of Professor Wig, without which I would have left this great department long ago. Last but not the least was indirect factors such as constructive criticism by eminent professionals, foremost among them were Prof. N.N. Sen, Prof. G.G. Prabhu and Prof. J.S. Neki, and international authorities on tests such as Prof. H.J. Eysenck, Prof. R.B. Cattell and Prof. Guilford.
4 Our Contributions
Our contribution in this area has to be seen as an insignificant and tiny effort, in the universe of a vast area of research and possibilities. Doing something (however small it might be) is better than doing nothing in the area of construction of psychological tests. I had read somewhere (in a different context though) that:
A piece of bread is better than nothing.
Nothing is better than God.
So, a piece of bread is better than God.
If I may be forgiven, I would like to replace “a piece of bread” by a psychological test in this statement.
Similarly, in another context I have read:
O God, if there is a God.
Save my soul, if there is a soul.
If I may be forgiven, I would like to put it as:
O God of Measurement, if there is a God of Measurement. Accept our small offerings (tests as flowers); if they can be offered and be accepted.
We are not sure (at least I am not sure) that what we have attempted to do in the field of psychological tests would make a difference to the world of measurement and be accepted as such. I have read somewhere—“A difference that makes no difference is no difference”.
To put it a little differently, I would like to make the following modified statement as—If a psychological test constructed by us in simple Hindi, for use with our population, majority of which consists of illiterates, barely literates, rural and unsophisticated people is better than a well-standardised Western test for use with their populations when used in India, without doing any modifications, it would make a difference to be called a real difference, in the end. On the other hand, if it does not make any difference, I would most humbly be ready to bow my head to be sacrificed.
5 A Start to Meet the Problems Faced
In order to meet the problems faced by us, we started with assessing the extent of the problems one by one and finding solutions that were available, practical and within our resources.
The first problem I faced was the length of the items in the test, particularly in the CMI Questionnaire (Erdmann Jr. et al. 1952). It was used in English with spot translation in Hindi. It contained nearly 200 items referring to different bodily systems and psychological complaints (A–L for physical section and M–R for psychological symptoms, with A–R referring to total complaints). We attempted to translate it in Hindi, so that the same language could be used with every patient. This took care of uniformity of the items, from tester to tester, and over time.
Another related difficulty arose with regard to the total scores. With the spot translations of English version, perhaps our patients could not relate to their problem correctly and did not endorse many items.
Patients’ relative positions with regard to the scores remained the same (as shown by the coefficients of correlation between English and Hindi versions remaining high), but the average score significantly increased with the Hindi version. This justified our attempt at developing a Hindi version, as patients could actually feel that the item really described their own difficulties or problems, versus the exact, literal, on the spot translation as attempted earlier.
Another issue was whether so many items were really required or whether the same results could be achieved by fewer items, which were more reliable and valid. So, item analysis was begun, to find out which items were readily accepted (frequently endorsed) by many patients, as opposed to others, which were rarely (it at all) endorsed by non-patients and patients (patients with both medical and psychiatric problems).
Those items that had zero or nearly zero percentage of endorsement were excluded in order to prepare a shorter version of the scale in simple Hindi.
When literature search for item construction was done, it was found that illiterate or barely literate people related better to items when expressed in first person (I, me, my versus you, your). In the English version, as well as in spot-translated Hindi version, the items were asked in the second person only, e.g. every time the items were “do you have” this or that problem. A need was felt for using first-person expressions in these questions. This shortened and simplified the statements (in place of questions), with the common question at the top asking them to tick the items applicable to them. Thus, the question form was converted into statement form to make it easier, simpler, shorter and quicker, taking less time to complete the test, while it still remained questionnaire (i.e. tick those items that are correct for you).
Related posts:
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
