Write short note : Measurement of Attitudes.

“Write short note : Measurement of Attitudes.” (1989)

  • An attitude is a relatively stable predisposition to respond favourably or unfavourably toward an object, person, group, or issue — it is not the behaviour itself but the underlying orientation presumed to guide it.
  • L.L. Thurstone’s claim that attitudes “can be measured” launched the systematic quantification of attitudes within sociology and social psychology, treating a mental orientation as something that could be placed on a numerical continuum much like height or income.
  • The core measurement problem follows directly from this: an attitude is internal and unobservable, so it must always be inferred from an expressed response — a verbal statement, a questionnaire tick, or a behavioural cue — rather than observed directly.
  • Because of this inferential gap, sociologists developed a family of scaling techniques, each offering a different trade-off between construction rigour, ease of use, and the kind of attitude object it suits.
  • This note surveys that field at the level of technique-types; the mechanics of how each scale is actually built belong to a related, narrower question.

Thurstone’s Equal-Appearing Interval Scale

  • Thurstone pioneered the first rigorous psychometric approach: a large pool of statements about the attitude object is generated, spanning strongly unfavourable to strongly favourable positions.
  • A panel of expert judges sorts these statements into eleven ordered categories purely by how favourable each statement appears, not by whether the judges personally agree with it — this separates the statement’s position on the continuum from any single judge’s opinion.
  • The final scale retains statements on which judges showed reasonable agreement, each carrying a scale value derived from the judges’ placements; a respondent’s score is the value of the statements they endorse.
    • “Attitudes can be measured.” — L.L. Thurstone
    • The claim was radical in its time precisely because it treated subjective orientation as amenable to the same interval-level measurement natural science applied to physical quantities.

Likert’s Summated Rating Scale

  • Rensis Likert offered a far simpler alternative: respondents indicate their degree of agreement with a set of statements (typically on a five-point strongly-agree-to-strongly-disagree continuum), and the item scores are summed into a single total.
  • Its comparative ease of construction — no expert judge panel is required — is a large part of why it became the most widely used attitude-measurement format across sociology, market research, and public opinion polling.

Bogardus’s Social Distance Scale

  • Emory Bogardus designed a scale specifically for inter-group attitudes, asking respondents how close a relationship they would accept with members of another group — from marriage into the group, down to exclusion from one’s country.
  • The scale is cumulative in structure: accepting a closer degree of relationship is taken to imply willingness to accept the more distant ones as well, making it a natural tool for studying caste, communal, and ethnic distance in Indian contexts.

Guttman’s Scalogram

  • Louis Guttman formalised the cumulative logic into a general technique: a genuinely unidimensional scale is one where agreement with an extreme item logically implies agreement with all the milder items on the same scale.
  • Because real responses rarely form a perfectly cumulative pattern, Guttman scaling requires an empirical check of how well actual answers fit the expected cumulative structure before the scale can be trusted.

The Semantic Differential

  • Charles Osgood and his collaborators developed the semantic differential, which asks respondents to rate a concept along a series of bipolar adjective pairs — good–bad, strong–weak, active–passive — usually on a seven-point continuum.
  • Because the same grid of adjective pairs can be reused across many different concepts, the technique allows quick, standardised administration and easy comparison of attitudes toward several objects at once.

Projective and Indirect Techniques

  • Where respondents are unlikely to state an attitude directly — on prejudice, sexuality, or political dissent — researchers turn to projective techniques: sentence-completion tasks, picture-based prompts, or word-association exercises that infer attitude indirectly from what the respondent projects onto ambiguous material.
  • These sit closer to qualitative method than to formal scaling, trading precision for access to attitudes respondents would otherwise conceal or be unaware of holding.
  • Contemporary survey methodology continues to pursue this same problem through implicit and indirect measurement designs intended to work around socially desirable responding, extending the logic of projective technique into more formally validated instruments.

The Persistent Limitation: Attitude Does Not Predict Behaviour

  • Richard LaPiere’s classic demonstration — travelling across the United States with a Chinese couple and being refused service at only one of over two hundred establishments, yet later receiving overwhelmingly negative responses when the same establishments were mailed a questionnaire asking whether they would serve Chinese guests — remains the starkest illustration that a measured attitude and actual conduct can diverge sharply.
  • No scaling technique, however well constructed, escapes this gap entirely, because expressed attitude and enacted behaviour are shaped by different situational pressures.
  • Attitude-measurement techniques range along a continuum from the laborious rigour of Thurstone’s judge-based construction to the administrative simplicity of Likert’s summated ratings, with Bogardus and Guttman contributing cumulative logics and Osgood contributing rapid multi-concept comparison.
  • Projective and indirect techniques extend this toolkit to attitudes respondents are unwilling or unable to state directly.
  • None of these techniques resolves the underlying attitude–behaviour gap, which means measured attitudes are best read as indicators of orientation rather than as predictions of actual conduct.
  • Their continuing relevance in Indian sociological research — on caste distance, communal perception, or gender attitudes — lies precisely in making otherwise invisible orientations comparable and trackable over time, even while their behavioural translation remains an open empirical question.