How people feel, or what they believe, is their attitude. But it is difficult to describe and measure attitude. Researchers must depend upon what people say are their beliefs and feelings. Through the use of questions or by getting people's expressed reaction to statements. A sample of their opinions is obtained. From this statement of opinion, one may infer or estimate their attitude- what they really believe.
However, inferring attitude from expressed opinion has many limitations. People may conceal their attitudes and express socially acceptable opinions. They may not really know how they feel about a social issue. People may be unaware of their attitude about a situation and may be unable to predict their reaction or behavior.
Even behavior itself is not always a true indication of attitude. When politicians kiss babies, their behavior may not be a true expression of affection toward infants.
Though there are several procedures for eliciting opinions and attitudes, the most widely use are Likert method, Thurstone technique and semantic-differential scale.
We want to know not only how each person feels about an object but also how each respondent's attitude compares with the attitudes of others. The Likerts scale, a technique based on summated ratings, provides such information (Likert, 1932).
The Likert scale consists of a series of statements about the object of interest. The statements may be positive or negative. The statements may be positive or negative. The respondent indicates how much he or she agrees with each statement. For example:
1) I think people should wait until they are married to have sex.
• Strongly agree ( +2)
• Agree (+1)
• Undecided (0)
• Disagree ( -1)
• Strongly disagree ( -2)
2) I think having sex before marriage strengthens the Inarriage.
• Strongly agree (-2)
• Agree (-1)
• Undecided (0)
• Disagree ( + 1)
• Strongly Disagree( +2)
Thus each item in the test is rating device designed to reveal both the direction of the individual's stand on the issue and the intensity with which he holds it. On this scale the higher value indicates a pro stand and the low values indicate an anti stand. These score values are assigned by the investigator. For instance, if you strongly agree with the above mentioned item 1(+2) and strongly disagree with the item 2 (+2). The score would be +4, indicating strong opposition to premarital intercourse. If your friend strongly disagrees with the statement that people should wait until they marry (-2) and might also disagree that premarital sex strengthens a marriage (+ 1). The resulting score of -1 indicates a slightly positive view of premarital intercourse.
An overall test score ts obtained by finding the sum of the numerical scores for the alternatives an individual checks on the various items. This overall individual" score can be interpreted only in relation to the distribution of scores made by other persons.
When a researcher wants to measure the attitudes of the people towards the women's reservation bill using the Likert scale, the following steps will be involved in the construction of the scale:
• The statements to be included in the test are chosen from a large number of current periodicals , books and some other tests and are formulated on the basis of empirical observation of different view points on the issue in question.
• Items should be clear-cut and unambiguous regarding the issue and should represent definite favorable or unfavorable stands.
• Preferably, an equal number of pro and con statements have to be included.
• Statements for any one scale should pertain to only a single issue but should not include statements on which all persons in a population will agree or disagree.
• All items are statements pertaining to desirable or undesirable behavior or courses of action and are not statements of fact.
Final statements are selected after the preliminary test has been administered to a large number of subjects. The results of this preliminary testing are analysed to pull out the items which do not differentiate between subjects who have high and low total scores on the test and are not highly correlated with total test scores.
Likert scales are advantageous as they can be devised for a large number of issues other than social distance. The total score represents a sum of values assigned so that alternatives on one side of the issue (eg-pro) carry relatively more weight than alternatives on the other side ( eg-con).
Another method of attitude assessment is known as the Method of equal appearing intervals. Following the procedures devised by Thurstone, many scales have come in to use to measure the attitudes of the individual towards various issues like war, privatization, crime etc.
A number of statements ranging from the most favorable to the most unfavorable stand on the issue are presented to the subjects on a form. The scale values of the statements are established before hand These scale values range from 0.0 (which represents one extreme) to 11 (which represents the opposite extreme). The statements are presented in the mixed order to avoid directional set.
Before the statements , there are certain instructions which include, "put a check mark if you agree with the statement". The stand of the individual on the issue is taken to be average of all the scale values for the statements with which he agrees.
Out of several hundred statements expressing an opinion on the issue, brief and relevant ones would be retained and edited in accordance with clearly stated criteria. The relevant statements cover a wide range of pro to con gradations on the issue. The statements would be mimeographed on small slips with a code number on each. Then, judges would be asked to sort them in to 11 piles. The statements expressing highest appreciation of the issue were to be put in the first pile (designated as A) and those expressing the strongest depreciation of the issue in the 11th pile( designated as K) other statements are to be sorted in the appropriate piles in between. The 6th pile (F) was to contain statements that expressed a neutral stand towards the issue.
The sorting by any judge who put 30 or more of the 130 statements into any one of the 11 piles was excluded from computation. The sorting of the each statement were then tabulated in the form of cumulative frequencies. The cumulative frequency, scale value and Q of a particular statement can be tabulated as :
The scale value: Q A B F k piles.
The scale value of a statement is the mid point of the cumulative frequency, i.e, half of the sorting lie above it and half below it.
In addition to securing statements representing different scale values, two major criteria used for inclusion in the final scale are -the criterion of ambiguity and the criterion of irrelevance.
"The criterion of Ambiguity" concerns the relative spread of judgments made over so many categories (piles). If the judges spread their placements of a statement over a wide range covering practically all of the categories, that statement was considered ambiguous and undifferentiated. Such statements were excluded from the final list. The measure used for this criterion was Q. Statements with large Q values were discarded and those with small Q values were retained for consideration.
"The criterion of irrelevance" concerns with the consistency of endorsement by individuals. To check the consistency, the 130 original statements would be given to each subject with the instructions to check the items with which he agreed.
The main assumption in the construction of a Thurston type scale is that, when the individuals are faced with the task of sorting a large number of statements ranging from very favorable through neutral to unfavorable, they will be able to sort them in to a number of equal-appearing categories unaffected by their own attitudes on the issue.
The semantic differential scale (Osgood, Suci, and Tannenbaum, 1957) is a technique for measuring connotative meaning of an object (i.e, a set of psychological meaning that vary from one respondent from another). For instance, one person may have had very positive experiences with the sexual intercourse, whereas another person's experience would have been frustrating.
An investigator presents the respondents with a series of bipolar adjective scales. Each of these is a scale whose ends are two adjectives having opposite meanings. The respondent rates the attitude object on each scale . After the data are collected , the researcher can analyze them by various statistical techniques. Analyses of such ratings frequently identify three aspects of connotative meaning : Evaluation, potency and activity.
Evaluation is measured by adjective pairs such as good-bad and positive-negative, potency is measured by weak~strong and light and heavy, and activity, by fast-slow and exciting-boring.
Example of semantic differential scale.
The semantic differential scale consists of a number of dimensions on which the respondent rates the attitude object.
Rate how you feel about premarital sexual intercourse on each of the following dimensions.
In the above example , two bipolar scales measure each of the three dimensions. Scores are assigned to each scale from +3 to -3, they are then summed across scales of each type to arrive at evaluation , potency and activity scores. In the above example, scores on each dimension could range from -6 (bad, weak and slow) to +6(good, strong, and fast).
One advantage of the semantic differential technique is that researchers can compare an individual's attitudes on three dimensions, allowing more complex differentiation among persons. Another advantage is that because the meaning it measures is connative, it can.be used with any object, from a specific person to an entire Nation, but a major disadvantage of this technique is that it requires more time to administer and to score.
The measurement of attitudes had been affected by a development which occurred outside of academic and scientific circles. This was the appearance of the "public opinion" polls on surveys which were used in attempting to tap attitudes of groups. Likert developed procedures for constructing attitude tests which were applicable to a wide variety of issues without some of the more difficult assumptions and procedures of the Thurstone technique. Thurstone proposed a technique for devising scales to measure attitudes on a variety of issues through the method of"equal appearing intervals". The third method of attitude assessment was developed by Osgood and Tennenbaum in which the respondent indicates an attitude or opinion between the two extreme choices. This method provides the individual with a seven-point scale with two adjectives at either end of the scale such as god-bad, healthy-unhealthy etc; the respondent is asked to rate a group, individual or object on each of this bipolar scales.
However, inferring attitude from expressed opinion has many limitations. People may conceal their attitudes and express socially acceptable opinions. They may not really know how they feel about a social issue. People may be unaware of their attitude about a situation and may be unable to predict their reaction or behavior.
Even behavior itself is not always a true indication of attitude. When politicians kiss babies, their behavior may not be a true expression of affection toward infants.
Though there are several procedures for eliciting opinions and attitudes, the most widely use are Likert method, Thurstone technique and semantic-differential scale.
LIKERT SCALE
We want to know not only how each person feels about an object but also how each respondent's attitude compares with the attitudes of others. The Likerts scale, a technique based on summated ratings, provides such information (Likert, 1932).
The Likert scale consists of a series of statements about the object of interest. The statements may be positive or negative. The statements may be positive or negative. The respondent indicates how much he or she agrees with each statement. For example:
1) I think people should wait until they are married to have sex.
• Strongly agree ( +2)
• Agree (+1)
• Undecided (0)
• Disagree ( -1)
• Strongly disagree ( -2)
2) I think having sex before marriage strengthens the Inarriage.
• Strongly agree (-2)
• Agree (-1)
• Undecided (0)
• Disagree ( + 1)
• Strongly Disagree( +2)
Thus each item in the test is rating device designed to reveal both the direction of the individual's stand on the issue and the intensity with which he holds it. On this scale the higher value indicates a pro stand and the low values indicate an anti stand. These score values are assigned by the investigator. For instance, if you strongly agree with the above mentioned item 1(+2) and strongly disagree with the item 2 (+2). The score would be +4, indicating strong opposition to premarital intercourse. If your friend strongly disagrees with the statement that people should wait until they marry (-2) and might also disagree that premarital sex strengthens a marriage (+ 1). The resulting score of -1 indicates a slightly positive view of premarital intercourse.
An overall test score ts obtained by finding the sum of the numerical scores for the alternatives an individual checks on the various items. This overall individual" score can be interpreted only in relation to the distribution of scores made by other persons.
Construction of a Likert scale :
When a researcher wants to measure the attitudes of the people towards the women's reservation bill using the Likert scale, the following steps will be involved in the construction of the scale:
• The statements to be included in the test are chosen from a large number of current periodicals , books and some other tests and are formulated on the basis of empirical observation of different view points on the issue in question.
• Items should be clear-cut and unambiguous regarding the issue and should represent definite favorable or unfavorable stands.
• Preferably, an equal number of pro and con statements have to be included.
• Statements for any one scale should pertain to only a single issue but should not include statements on which all persons in a population will agree or disagree.
• All items are statements pertaining to desirable or undesirable behavior or courses of action and are not statements of fact.
Final statements are selected after the preliminary test has been administered to a large number of subjects. The results of this preliminary testing are analysed to pull out the items which do not differentiate between subjects who have high and low total scores on the test and are not highly correlated with total test scores.
Likert scales are advantageous as they can be devised for a large number of issues other than social distance. The total score represents a sum of values assigned so that alternatives on one side of the issue (eg-pro) carry relatively more weight than alternatives on the other side ( eg-con).
THURSTONE SCALE
Another method of attitude assessment is known as the Method of equal appearing intervals. Following the procedures devised by Thurstone, many scales have come in to use to measure the attitudes of the individual towards various issues like war, privatization, crime etc.
A number of statements ranging from the most favorable to the most unfavorable stand on the issue are presented to the subjects on a form. The scale values of the statements are established before hand These scale values range from 0.0 (which represents one extreme) to 11 (which represents the opposite extreme). The statements are presented in the mixed order to avoid directional set.
Before the statements , there are certain instructions which include, "put a check mark if you agree with the statement". The stand of the individual on the issue is taken to be average of all the scale values for the statements with which he agrees.
Construction of the Thurstone scale :
Out of several hundred statements expressing an opinion on the issue, brief and relevant ones would be retained and edited in accordance with clearly stated criteria. The relevant statements cover a wide range of pro to con gradations on the issue. The statements would be mimeographed on small slips with a code number on each. Then, judges would be asked to sort them in to 11 piles. The statements expressing highest appreciation of the issue were to be put in the first pile (designated as A) and those expressing the strongest depreciation of the issue in the 11th pile( designated as K) other statements are to be sorted in the appropriate piles in between. The 6th pile (F) was to contain statements that expressed a neutral stand towards the issue.
The sorting by any judge who put 30 or more of the 130 statements into any one of the 11 piles was excluded from computation. The sorting of the each statement were then tabulated in the form of cumulative frequencies. The cumulative frequency, scale value and Q of a particular statement can be tabulated as :
The scale value: Q A B F k piles.
The scale value of a statement is the mid point of the cumulative frequency, i.e, half of the sorting lie above it and half below it.
In addition to securing statements representing different scale values, two major criteria used for inclusion in the final scale are -the criterion of ambiguity and the criterion of irrelevance.
"The criterion of Ambiguity" concerns the relative spread of judgments made over so many categories (piles). If the judges spread their placements of a statement over a wide range covering practically all of the categories, that statement was considered ambiguous and undifferentiated. Such statements were excluded from the final list. The measure used for this criterion was Q. Statements with large Q values were discarded and those with small Q values were retained for consideration.
"The criterion of irrelevance" concerns with the consistency of endorsement by individuals. To check the consistency, the 130 original statements would be given to each subject with the instructions to check the items with which he agreed.
The main assumption in the construction of a Thurston type scale is that, when the individuals are faced with the task of sorting a large number of statements ranging from very favorable through neutral to unfavorable, they will be able to sort them in to a number of equal-appearing categories unaffected by their own attitudes on the issue.
SEMANTIC DIFFERENTIAL SCALE
The semantic differential scale (Osgood, Suci, and Tannenbaum, 1957) is a technique for measuring connotative meaning of an object (i.e, a set of psychological meaning that vary from one respondent from another). For instance, one person may have had very positive experiences with the sexual intercourse, whereas another person's experience would have been frustrating.
An investigator presents the respondents with a series of bipolar adjective scales. Each of these is a scale whose ends are two adjectives having opposite meanings. The respondent rates the attitude object on each scale . After the data are collected , the researcher can analyze them by various statistical techniques. Analyses of such ratings frequently identify three aspects of connotative meaning : Evaluation, potency and activity.
Evaluation is measured by adjective pairs such as good-bad and positive-negative, potency is measured by weak~strong and light and heavy, and activity, by fast-slow and exciting-boring.
Example of semantic differential scale.
The semantic differential scale consists of a number of dimensions on which the respondent rates the attitude object.
Rate how you feel about premarital sexual intercourse on each of the following dimensions.
In the above example , two bipolar scales measure each of the three dimensions. Scores are assigned to each scale from +3 to -3, they are then summed across scales of each type to arrive at evaluation , potency and activity scores. In the above example, scores on each dimension could range from -6 (bad, weak and slow) to +6(good, strong, and fast).
One advantage of the semantic differential technique is that researchers can compare an individual's attitudes on three dimensions, allowing more complex differentiation among persons. Another advantage is that because the meaning it measures is connative, it can.be used with any object, from a specific person to an entire Nation, but a major disadvantage of this technique is that it requires more time to administer and to score.
SUMMARY
The measurement of attitudes had been affected by a development which occurred outside of academic and scientific circles. This was the appearance of the "public opinion" polls on surveys which were used in attempting to tap attitudes of groups. Likert developed procedures for constructing attitude tests which were applicable to a wide variety of issues without some of the more difficult assumptions and procedures of the Thurstone technique. Thurstone proposed a technique for devising scales to measure attitudes on a variety of issues through the method of"equal appearing intervals". The third method of attitude assessment was developed by Osgood and Tennenbaum in which the respondent indicates an attitude or opinion between the two extreme choices. This method provides the individual with a seven-point scale with two adjectives at either end of the scale such as god-bad, healthy-unhealthy etc; the respondent is asked to rate a group, individual or object on each of this bipolar scales.