The semantic differential technique was developed by Osgood, Suci, and Taneabaum (1957).
Its main objective is to examine the meaning of certain concepts, e.g., church, wife, socialism, and so on. In a sense, it is an attitude scale. The groups of subjects may be asked to rate a given concept on a series of bipolar rating scales.
Generally, a seven-point scale is utilized, whose two ends will use adjectives, which are opposite (called bipolar terms), e.g., good-bad, complex-simple, passive-active, friendly- unfriendly, and so on.
The response categories consist of seven categories ranging from one extreme to the other with the middle category representing neutral. The seven positions for each item are assigned a score from 1 through 7.
Each respondent is then asked to indicate his or her perception about the object by placing a checkmark, that is, by choosing a score on each item that describes the object with respect to the bipolar terms for that item.
The individual scores for the items are then summed, and his total score is the measure of the respondent’s perception of the object.
The semantic differential is similar to the Likert scaling categories ranging from strongly disagree to strongly agree, except that in the semantic differential, only the two end categories have names.
The middle categories simply have a blank space, or sometimes a number. Also, the two end categories are not strongly agree-strongly disagree, but rather a pair of opposite adjectives (bipolar terms) thought to express the subject’s feelings about the concept.
For example, opinion on the issue of abortion or early marriage may be expressed as good or bad as extremes.
The format for response categories may be displayed as below:
Osgood has investigated the correlations between the scores given to a set concept on different bipolar scales by conducting a series of factor analyses.
The three main factors are;
- Individual evaluation (e.g., social- unsocial, successful-unsuccessful),
- Potency (e.g., weak-strong, heavylight).
- The individual’s activity of the object or concept. (e.g., active-passive, slow-fast)
In other words, people tend to react to various concepts largely (but not entirely) in terms of these dimensions.
The semantic differential makes the measurement and comparison of various objects or concepts possible.
In order to form an attitude scale, what we need is to decide the description of the issue to be studied, and to choose suitable objective pairs for it.
A respondent’s total score is the measure of his attitude.