Statistics Definitions > Semantic Differential Scale
The semantic differential scale measures the connotative meaning of things. For example, while the word “heart” is defined as the organ that pumps blood around the body, it’s connotative meaning is love or heartache. The scale is used in surveys to gauge people’s feelings towards a particular subject.
Denotation vs. Connotation
Denotation is the exact meaning of a word. It’s what you would find if you looked in a dictionary. A few examples of denotation:
- Sweater: a knitted garment worn to keep warm.
- Abyss: a deep or seemingly bottom chasm.
- Diamond: a precious, clear and colorless stone made from pure carbon.
- Lion: a large, fawn-colored cat that lives in prides.
Connotation is an idea or a feeling that the word invokes . The above words have many implied meanings in pop culture and literature, including:
- Sweater: friendship, fireplaces and hot cocoa.
- Abyss: a really bad situation.
- Diamond: anyone who stands out and “shines.”
- Lion: bravery.
The Semantic Differential Scale
A semantic differential scale measures attitudes towards something. For example, you could measure a person’s attitude to the word “Work” with the following scale:
The terms to the left and right are polar opposite adjectives. For example, “necessary” is the opposite of “unnecessary.” There are usually five intervals, although some scales have seven. Instead of blank spaces to mark, you could have radio buttons or boxes to check.
According to Ron Garland, three types of semantic differential scale exist:
- Scale points are unlabeled.
- Scale points are labeled.
- Scale points are numbered (i.e. from 1 to 5).
Garland found that respondents preferred labeled scale points as they are easier to comprehend, easier to complete, and are more useful for expressing opinion. Labeling the scales does not seem to introduce bias.
Semantic Scale vs. Likert Scale
With the Likert scale, people state how much they agree or disagree with a particular statement; with the semantic differential scale, people filling in the questionnaire decide how much of a trait or quality the item has.
A Likert scale measures agreement or disagreement to a particular statement. The scale ranges from “strongly agree” to “disagree” with neutral in the center. You can easily quantify the results. It’s usual for the highest agreement to be rated a 5, neutral as 3 and the lowest agreement or no agreement. This makes it easy to compare results. An example of a Likert scale question would be:
My job is drudgery.
Strongly disagree / Disagree / Neutral / Agree / Strongly Agree
On the other hand, the Semantic scale is very subjective on the user’s part. Plus, there’s no “neutral” answer, which makes it difficult to quantify. You can’t pinpoint direct metrics with this type of scale. For example, you couldn’t find out if “on hold wait times are less than five minutes.” You can only ask how the respondent feels about wait times (acceptable/not acceptable). A potential benefit of the semantic differential scale is that the user is really only posed with two options that are opposites, where the Likert scale has a range of intensities to choose from.
The two scales have some middle ground. Technically, the Likert Scale can be designed with opposites, like love/hate or happy/sad. This is similar to the idea of polar opposite adjectives for the semantic differential scale. Although “like/hate” may be polar opposites, they are not adjectives* and so cannot be included in a semantic differential questionnaire.
Which scale you choose largely depends on what information you want to know. As far as the three semantic differential scales are concerned, a rule of thumb is: choose the unlabeled scaled if you think your respondents are good with abstract thinking, and the labeled one if you are surveying the general public. Choose the numbered scale if you are sure your respondents have numerical aptitude.
*The word “like” could be an adjective, as in “the twins are very like.” But in this context of love/hate, it is a verb.