ADVERTISEMENTS:
This article throws light upon the three types of scales that are used for measuring the attitude of individuals. The types are: 1. Differential Scales 2. Summated Scales 3. Cumulative Scales.
Type # 1. Differential Scales:
These scales used for the measurement of attitudes are closely associated with the name of L.L. Thurstone, hence are often called Thurstone- type scales.
Thurstone-type scale consists of a number of statements whose position on the scale has been determined by a ranking operation performed by judges who are persons whose judgements about the relative rank of different statements along a dimension can be relied on.
ADVERTISEMENTS:
Various method of securing judgments of scales-position have been, viz., the method of paired comparisons, the method of equal-appearing intervals, etc. The method of equal-appearing intervals is most commonly used in the construction of this scale.
This procedure is as under:
In selecting the statement for the scale and assigning scores to them, the following procedure is used:
(a) The researcher gathers a large number of statements (several hundred) conceived as related to the attitude being investigated,
ADVERTISEMENTS:
(b) A large number of judges (up to 300) working independently, are requested to classify these statements into eleven categories or piles.
Each of the judges is requested to place in the first pile the statements which he thinks are most unfavorable to the issue (or, most progressive or most permissive, depending upon the dimension along which the statements are to be placed) in the second pile, those he considers next most unfavorable and proceeding thus, in eleventh pile, the statements he considers most favourable.
The sixth position on this continuum is defined as the point at which the attitude is ‘neutral.’ The first pile (i.e., statements rated as most unfavorable) carries the score of 11 while the statements assigned the last pile carry a value of 1 (each),
(c) The scale value of any one statement is computed as the mean or median position to which it is assigned by the group of judges. Statements that have too broad a scatter, that is, whose evaluation by different judges varies very widely, are discarded as ambiguous or irrelevant,
(d) A final selection of statements is made, taking evaluated items or statements that spread out evenly along the scale from one extreme position to the other (scale values like 10.3, 9.4, 8.4, 7.5, 6.4, 5.3, 4.5, 3.4, 2.6 and 1.6) may be included in the scale to be administered.
The resulting scale is thus a series of statements, usually about twenty, the position of each statements on the scale having been determined by the judges’ classification. The subjects are asked during the administration of the scale questionnaire to check mark the statement or statements with which they agree or to check two or three statements that are closest to their position.
In MacCrone’s study of attitudes toward ‘natives’ in South Africa using the Thurstone- type scale, the statement which was judged as one with extremely unfavorable import for the natives and having the scale value 10.3 was, “I consider that the native is only fit to do the ‘dirty’ work of the white community.”
The most favourable statement having a scale value of 0.8 was “I would rather see the white people lose their position in this country than keep it up at the expense of injustice to the native.” Between these two opposite extreme statements were a number of statements arranged according to their scale-values (from higher of lower).
Of course, the scale values are not shown on the questionnaire and the statements are usually arranged in a random order. The mean (more preferably the median) of the scale values of the items the individual checks is interpreted as indicating his position on the scale of favourable-unfavorable attitude (or any other dimension) toward a given issue.
ADVERTISEMENTS:
The Thurstone method of equal-appearing intervals has been widely used. Such scales have been utilized to measure attitudes toward varied issues such as war, religion, and various ethnic groups. The Thurstone-type, scale has also been adopted for the analysis of documentary data such as newspaper editorials.
The Thurstone scale is most appropriate and reliable if the scale is measuring just a single attitude and not a complex of attitudes. If the responses of an individual scatter widely over statements having scale values differing widely, his attitude score is not likely to have the same meaning as a score with little scatter.
The scattered responses may also be interpreted to mean that the subject has no clear attitude toward the issue or that his attitude is not organized in a manner assumed by the scale.
Several objections have been raised against the Thurstone-type scale. The major ones may be listed as under:
(a) The procedure involved in constructing such a scale is very cumbersome. This involves, as we have seen, a lot of work such as preparing several hundred statements relevant to the issue taking care that different shades of attitudes are represented, getting these statements rated by a large number of judges and then choosing the most reliable statements constituting a smooth continuum, etc.
(b) Another justified criticism of this scale is that, since a subject’s score is the mean or median of the scale-values of several statements he checks, there is the likelihood that the same total score obtained by different persons may express different attitudinal patterns. For example, an individual checks two items (statements) have scale-values 8.6 and 7.6 respectively.
His score will be 8.6 + 7.6 ÷2 = 8.1 (the mean of scale-values). Now another individual may check three statements having scale-values 10.4 + 5.5 ÷3 = 8.1 (approx.). Thus, two individuals get rated as being equivalent in regard to their attitude even though the latter’s choice of statements and the scatter of his responses is appreciably different from that of the former.
(c) The statements comprising the scale have a reference to varied dimensions of an issue and hence cannot really occupy different positions on a single continuum.
(d) Yet another objection concerns the extent to which the scale values assigned to statements are influenced by the attitude of judges themselves.
Although some earlier studies, notably by Hinckley and MacCrone, had come to the conclusion that the attitudes and backgrounds of the judges do not affect the position of various statements on the scale, more recent researches notably by Hovland and Sherif (1952) and by Kelly and associates (1955) have come out with findings suggesting that the attitudes and backgrounds and even the intelligence of the judges do affect the scale-values they assign to the statements.
Such findings cast serious doubts on the meaning of the scale positions and the distances between them. It should be noted, however, that even these studies found a high agreement in the rank-order in which the different groups of judges arranged statements along the continuum.
Thus, according to most writers on research methodology, the Thurstone-type scale despite this weakness may be said to constitute a reasonably satisfactory ordinal scale. In practice, if individuals agreed with only a few contiguous items on the scale, the Thurstone scale would provide highly satisfactory ordinal scales since in this case the score would have a clearer meaning.
Type # 2. Summated Scales:
A summated scale like the differential scale just discussed, consists of a series of statements to which the subject is asked to react. The main difference between the two is that unlike the differential scale or favourable statements definitely favourable or unfavorable toward a given issue are used in the scale (the intermediate shades being excluded).
The respondent/subject indicates his agreement or disagreement and degree thereof with each item. Each response is given a numerical value corresponding to its favourableness or unfavourablenss. The summation of the scores of individual’s responses to all the separate statements gives his total score. This score represents his position on the continuum of favourable-unfavourableness toward an issue.
The type of summated scale most frequently used in the study of social attitudes follows the pattern devised by Likert. Therefore, it is customarily referred to as the Likert-type scale. In this scale, the subjects are asked to respond to each of the statements in terms of several degrees of agreement or disagreement; for example, (I) strongly approve, (II) approve, (III) undecided, (IV) disapprove, (V) strongly-disapprove.
We can see that these five points constitute the scale; at one end of the scale is strong approval, at the other end is strong disapproval and between them lie intermediate points. The subject indicates with reference to each statement where he stands on this scale.
Each point on the scale carries a score or a value. Response indicative of the most favourable attitude (strongly approve) is given the highest score of 5 or + 2, while the one conveying most unfavorable attitude (strongly disapprove) is given the least score, i.e., 1 or 2.
The following will make it clear:
Of course, the score values corresponding the scale positions do not appear on the questionnaire supplied to respondents.
The procedure for constructing a Likert-type scale is as follows:
(a) The investigator assembles a larger number of statements considered relevant to the attitude being investigated, which are either clearly favourable or clearly unfavorable.
(b) The statements are administered to a small sample of subjects, i.e., representation of those on whom the questionnaire is to be finally administered. The subjects indicate their response to each item by checking one of the categories of approval or disapproval on the scale below each statement.
(c) The responses to various items are scored in such a way that a response indicative of the most favourable attitude is given the highest score (or highest positive score). It is important that the responses are scored consistently in terms of the attitudinal direction they indicate.
Whether ‘approve’ or ‘disapprove’ is a favourable response vis-a-vis an issue depends on the content and wording of the statement. For example, with reference to the following statements, its disapproval by a subject will indicate favourable attitude toward the issue.
It may be noted that in the above illustration, the order or scale-values of the positions has been reversed. Comparing it with the scale used for statement No. I, we find that for statement No. I the position I (strong approval) had the scale-value of 5 (or + 2), the same position as used for statement No. II has the scale value of 5 (or + 2).
This is quite understandable, because the person who strongly disapproves statement No. II which in terms of content is the opposite of statement No. I, i.e., expresses unfavorable view about co-education.
(d) The next step in the procedure is to compute the total score of the individual by adding his item-scores (i.e., scores received for separate statements).
(e) Finally, the responses are analysed to determine which of the statements discriminate most clearly between the high scores and the low scores on the total scale.
In other words, the investigator exerts to identify the items which have a high discriminatory power. The items with low discriminatory power or those that do not show a substantial correlation with the total score are eliminated to ensure that the questionnaire is internally consistent, that is, every item or statement is related to the same general attitude.
The Likert-type scale has several advantages over the Thurstone scale.
(1) It permits the use of items that are not manifestly related to the attitude being studied. This is so because in the Likert method any item (statement) that is found empirically consistent with the total score can be included.
Unlike Thurstone type scale, there is no necessity of agreement among judges that restricts the items (statements) to content which is obviously related to the attitude being studied. It is a great advantage to be able to use items that do not, on the face of it, appear to have a direct relationship to the attitude being studied.
(2) The Likert type scale is generally considered simpler to construct. At least, the procedure of construction is less cumbersome.
(3) It is likely to be more reliable than a Thurstone type scale comprising the same items. The Likert type scale permits the expression of several degrees (usually five) on the continuum of agreement-disagreement, whereas the Thurstone type scale allows for a choice between only two alternative responses, i.e., acceptance or rejection.
(4) The range of responses permitted to a statement in the Likert type scale provides more precise information about the individual’s opinion on the issue.
However, that the Likert type scale does not provide a basis for saying how much more favourable one is, compared to another, or for measuring the amount of change after some experiment or exposure. Thus, in effect, the Likert scale does not rise to a stature higher than that of an ordinal scale.
One major weakness of the Likert scale is that often, the total score of an individual has little clear meaning in as much as many patterns of response to the various statements may produce the same score. But this lacuna applies with even greater force to the Likert scales, since they provide a greater number of response possibilities.
So in Likert scale as in the Thustone type scale, the meaning of identical total scores of two or more persons may be markedly different. Pragmatically viewed, it may appear that the scores on the Likert type questionnaire often provide a basis for a rough ordering of people on the characteristics being measured.
Type # 3. Cumulative Scales:
Cumulative scales like the earlier scales are made up of a series of .items with which the respondent indicates agreement or disagreement.
The distinctive feature of a cumulative scale is that the items therein are ordered or related to one another in such a way that an individual who replies favorably to item No. 3 also automatically replies favorably to item No. 2 and No. 1 and one who replies favorably to item No. 4 also reacts favorably to items Nos. 3, 2, and 1 and so on.
Thus, all individuals who react to a given item favourably have higher scores on the total value scale than the individuals who react to that items unfavourably. The individual’s score is computed by counting the number of items he answers favourably. This score places him on the scale of favourable-un-favourable attitude provided by the relationship of the items to one another.
One of the earliest scales for measurement of attitudes, the Bogardus social distance scale, was intended to be a scale of the cumulative type. The social distance scale which has become a classic technique in the measurement of attitudes toward ethnic or racial groups lists a number of relationships to which members of a given ethnic group might be admitted.
The respondent is asked to indicate for specified ethnic or racial groups the relationships to which he would be willing to admit members of each group. This attitude is measured by the closeness of relationship that he is willing to accept or the social distance that he would wish to maintain.
The Borgardus scale is illustrated below:
The respondent is directed to circle or tick off (✓) each of the classifications relationship categories) to which he would be willing to admit the average member of a particular ethnic, racial or nationality group (neither the best members nor the worst, he has known). The answers are supposed to reflect the respondent’s first feeling reactions.
In the above scale, it is reasonable to expect that a respondent who ticks off 3 in relation to French, i.e., indication of his willingness to have them as his neighbours, would ordinarily also tick off 4 and 5, i.e., accepting them as employees in his occupation and accepting them as citizens in his country.
But he would not tick off 6 and 7, because these are statements connoting exclusion. If an individual does not tick off 3, then it is reasonable to except that he will not tick off 1 and 2, since these are statements indicating even closer relationships (marriage and friendship) compared to 3 (neighbourhood).
The underlying assumption that these items form a cumulative scale (with reference to America) has, by and large, been borne out. In practice, however, some reversals may occur. For example, one who would object to living in the same neighbourhood with a particular group would not object to having these people in an informal social club (i.e., accepting 2, but rejecting 3).
The reason may be their dirty living habits or domestication of certain pets which are a nuisance to the neighbourhood. Experience has demonstrated, however, that such reversals, while possible in individual cases, do not obtain in regard to an entire group of respondents.
It should be noted that reversals can be interpreted by postulating the intrusion of some extraneous factors such as respondent’s apprehension that land values may fall if members of a particular group (of low social status) were to live in his neighbourhood.
After 1940, there has been a rapid development of techniques for determining whether the statements form or a scale in the ideal sense ideally have. This led to a methodological emphasis on the development of uni-dimensional scales, that is, scales consisting of statement that do no refer to issues extraneous to the characteristic being measured.
It had been pointed out by a number of investigators that the Thurstone or Likert scales contain statement about various aspects of the characteristic under consideration and thus these statements belong in fact, to two or more different scales. Thus, scales using statements are in fact ‘multi-dimensional’ scales.
For example, in a scale that was designed to measure attitude toward war (a Thurstone-type Scale), the most favourable statement was, “war is glorious” and the most unfavorable statement was “There is no conceivable justification for war” and its mid-point was “I never think about war and it does not interest me.”
Carter pointed out that these statements could hardly be considered as falling along a straight line, since they refer to different aspects of the phenomenon of war. It is understandable that combining statements referring to different aspects of the phenomenon, e.g., war, makes it impossible to specify exactly what the scale is measuring.
In such a scale characterized by multi-dimensionality, there is a very high possibility of the subject’s response being scattered along statements which differ widely in their scale-value; this makes it difficult to give any clear meaning to the score based on the average of the scale-values of statements checked by the individual.
A notable attempt to approach this problem has been made by Guttman. The technique developed by Guttman, known as ‘scale analysis’ or the ‘Scalogram method’ has for its main purpose, to ascertain whether the attitude or characteristic being studied actually involves only a single dimension.
In the Guttman procedure, a universe of content (the attitude or characteristic under study) is considered to be uni-dimensional only if it yields a perfect or almost perfect cumulative scale.
That is, it should be possible to arrange all the responses of any number of subjects into a pattern such as the one given below:
The important thing about this pattern is that if it holds good, i.e., the given score on a particular series of items (statements) always has the same meaning. That is, if one knows a subject’s score, he can tell, without consulting his questionnaire, exactly which items the subject must have endorsed.
Samuel Stouffer points out the characteristic feature of the Guttman technique thus:
“…it must be possible to order the items such that, persons who answer a given question favorably all have higher ranks than persons who answer the same question unfavourably. The response to any item provides a definition of the respondent’s attitude.”
This quality of being able to reproduce the responses to each item knowing only total score is called ‘reproducibility’, which is one of the major tests of whether a set of items constitutes a scale in Guttman’s sense. Let us take an illustration for added clarification.
Consider the following items, with which the respondents are asked to agree or disagree:
(1) Co-education in the long run helps subsequent marital adjustments. Agree disagree.
(2) Parents should send their sons and daughters to co-educational institutions. Agree disagree.
(3) Co-education should be made compulsory by an Act of law. Agree ________________ disagree.
If these items were found to form a perfect cumulative scale, then we would have all individuals with a score of 2 on the scale believe in the first statement that ‘co-education helps adjustment among sexes’ and believe, of course, in the second statement that ‘parents should send their children to co-educational institutions’ but not in the third statement that ‘co-education should be made compulsory by law.’
In practice, speaking of social research perfect cumulative or uni-dimensional scales are rarely if ever found. But approximations to them can be developed. Scalogram analysis uses several criteria for deciding whether or not a particular series of statements may be regarded as a uni-dimensional scale.
The most important of these is the reproducibility of the responses. The proportion of the subject’s responses which fall into the scale pattern presented above, provides a measure of the extent to which particular responses are reproducible from the total score. Guttman has set 0.9 as the level of minimum reproducibility.
The Guttman technique may rightly be regarded as a method of determining whether a set of statements forms a uni-dimensional scale. It affords no guidance, however, for selecting statements that are likely to form a uni-dimensional scale. The scale discrimination technique developed by Edwards and Kilpatrick is a method of selecting a set of items likely to form a uni-dimensional scale.
The procedure suggested is as under:
(a) A large number of statements dealing with the issue concerning the study is collected. Items that are ambiguous, irrelevant, neutral or too extreme are eliminated by inspection.
(b) As in the Thurstone method, a large number of judges are requested to place the remaining statements in eleven piles according to their degree of favourableness- unfavourableness toward the issue. The unreliable items are discarded and each of the remaining items is assigned a scale value (median position).
(c) These statements are then transformed into a Likert type scale by providing for expression of various degrees of agreement disagreement in response to each of the items. This scale is administered to a large group of subjects and their responses are analysed to determine which of the items discriminate most clearly between the high scorers and the low scorers on the total scale.
Such items as have the highest discriminatory coefficients in their scale interval (say, all items having scale value between 7.0 and 7.9 or 6.0 and so on) are selected in twice the number that is actually wanted for use in the final scale. For each scale interval, an equal number of items is selected.
(d) The statements or items in the resulting list are arranged in order in their scale value. The list is sub-divided into two counter-parts of questionnaire. The even numbered statements are assigned to one and the odd numbered ones are assigned to the other questionnaire.
But the uni-dimensional scales do suffer from certain limitations which we would do well to note:
(1) Uni-dimensional scales hardly constitute a reliable basis for assessing attitudes of persons towards complex objects or phenomena or for predicting the behavioural responses of individuals towards such objects or phenomena. For instance, ‘war’ or ‘modernity’ is a complex concept hence, the uni-dimensional scales do not quite help us to measure the attitudes of men toward these intricate concepts of war and modernity.
It is, of course, possible to construct and use uni-dimensional scales for the implications of war for economy, health, morality etc., in independent studies.
Although these dimensions of war do affect the final shape of persons’ attitudes towards war, independent enquiries basing their assessment on uni-dimensional scales would hardly afford us a total perspective about a person’s attitude toward ‘war’ in all its many-splendored implications.
Secondly, a scale may be uni-dimensional for some persons but not so for others. In our discussion on cumulative scales, we had shown how the items on the scale may not constitute a cumulative series for one and all.
Differences in educational levels and experience are reflected in a person’s subjective evaluation of the scale items and as such a person may indicate his disagreement to item No. 2 but not to item No. 3 on the cumulative scale. Harding and Hogerfe have in their study demonstrated how a single scale did not, in effect, function as a uni-dimensional scale on the three different categories of workers.