THE STORY BEHIND THE SCALE – Holly Foster Media

THE STORY BEHIND THE SCALE

In 1932, Rensis Likert was pursuing his Ph.D. in Psychology at Columbia University. His dissertation research was part of a larger study, aimed at evaluating the technical problems involved with studying social attitudes.

Likert was seeking a better way to measure human attitudes (qualitative data) using statistical (quantitative) methods.

Although Likert was a pioneer, he was not the first to study attitude measurement.

Four years prior in 1928, a method of measuring attitudes had been developed by Louis Thurstone. His Thurstone scale was considered reliable and valid, but the procedure was time consuming and expensive (Prochaska, n.d.).

Measurements Require the Use of a Scale

The Thurstone scale requires gathering consensus from eleven outside people, known as “judges,” who determine the scale (range of intensity) of the survey questions to be used.

Although Thurstone’s method was highly-regarded, it was labor intensive. Could there be a better way?

Using questions from similar studies on attitudinal differences, Likert created the Survey of Opinions, which used three different question formats covering three public issues: internationalism, racism, and imperialism.

Three issues, three measurement scales.

“A measurement scale is used to qualify or quantify data variables in statistics. It determines the kind of techniques to be used for statistical analysis” (7 Types of Data Measurement Scales in Research, 2019).

Survey Question Types

The survey was administered to approximately 2,000 undergraduates (mostly males), with the final data culled from 650 randomly chosen surveys. Example questions from the survey are shown below (Likert, 1932).

TYPE 1 – Three-Point Statements

A statement followed by three choices:

Do you favor the early entrance of the United States into the League of Nations?


TYPE 2 – Five-Point Statements

A statement followed by five possible answers:

Using the term “armaments” to mean equipment devised for war rather than for police purposes, our policy should be to favor


TYPE 3 – Five-Point Continuum

A statement followed by a continuum of five attitudinal options:

All men who have the opportunity should enlist in the Citizens Military Training Camps.

As he studied the data, Likert noticed the answers to each of the five-point questions (both the statement type and the continuum type) “yielded a distribution resembling a normal distribution” (Likert, 1932), or bell curve. Based on his observation, and for the purposes of the study, he made the assumption that attitudes themselves, too, are distributed normally.

The Calculations

After the questions were sorted by issue, the question types had to be made comparable to one another for computation.

Likert converted each question type (the three-point, and each of the five-point types) using a normal distribution based on a sample of 100 responses. Without getting too deeply into statistics, he used sigma units, which helped to prove his calculations to be as reliable as those using the Thurstone scale.

” … statistical significance is usually expressed in units of the standard deviation, or σ (sigma), from the average value” (What Is a Sigma? | Perimeter Institute, n.d.). 

Going one step further, he tried an even simpler technique.

“The simpler technique involved the assigning of values from 1 to 5 to each of the five different positions on the five-point statements. The ONE end was always assigned to the negative end of the sigma scale, and the FIVE end to the positive end of the sigma scale” (Likert, 1932).

That left the three-point questions. Three different methods were applied to test consistency, each of which yielded similar results to those of the successful sigma method.

The Results

Likert was successful in discovering two additional methods of measuring attitudes:

  • the sigma method
  • assigning consecutive numerical values to different alternatives, known today as the Likert scale.

“The first published examination of Likert-type scales involved items written to address racist, internationalistic, and imperialistic attitudes. The scores had high reliability and correlated with scores from other measures, supporting the use of Likert-scaled items” (Horst & Pyburn, 2018). 

I believe Likert summed his results up best when he said:

“In this study, however, each statement becomes a scale in itself and a person’s reaction to each statement is given a score. These scores are then combined by using a median or mean” (Likert, 1932).

Likert’s simple, yet accurate approach has been widely used ever since.

” … it yields reliabilities as high as those obtained by other techniques, with fewer items” (Likert, 1932).”

Full details of the study can be found here.

“The use of Likert scaling is widespread throughout education, psychology, business, and other disciplines involving research examining people’s attitudes, values, beliefs, dispositions, or psychological states and traits” (Horst & Pyburn, 2018).


Four Tips

Looking to get feedback on your website? Interested in people’s opinions on something? The Likert Scale is a great way to get answers. Here are some tips on how to create questions that will provide valuable insight.

  1. Keep your questions focused on one topic only.
  2. Use an odd number of responses with “neutral” in the middle. Five is the most common number, but seven can also be used if you wish to granulate your choices.
  3. Use a consistent scale to make reporting and analysis easier.
  4. Keep the word count of each question to twenty or less. Make sure the questions you ask are easy to understand. Avoid jargon.
  5. Limit the number of questions. Less is more. Twenty is acceptable, but if you can get the information you need with fewer, do so. If your survey is too long people will avoid taking it.

What Do You Think?

I found this blog informative.

Use the comments section to let me know what you think!

until nxt time …

Many studies use the Likert scale based on it’s simplicity and accuracy. Here are a couple of examples if you would like to check them out:

Sauro, J., & Dumas, J. S. (2009). Comparison of three one-question, post-task usability questionnaires | Proceedings of the SIGCHI Conference on Human Factors in Computing Systems. 1599–1608. https://dl-acm-org.libraryproxy.quinnipiac.edu/doi/10.1145/1518701.1518946

Wetzlinger, W., Auinger, A., & Dörflinger, M. (2014). Comparing Effectiveness, Efficiency, Ease of Use, Usability and User Experience When Using Tablets and Laptops. In A. Marcus (Ed.), Design, User Experience, and Usability. Theories, Methods, and Tools for Designing the User Experience (pp. 402–412). Springer International Publishing.

References

7 Types of Data Measurement Scales in Research. (2019, December 28). Formplus. https//www.formpl.us/blog/measurement-scale-type

Boslaugh, S. (2008). Statistics in a Nutshell: A Desktop Quick Reference O’Reilly Media.

Crossman, A. (n.d.). Normal Distribution and Why It Matters. ThoughtCo. Retrieved from https://www.thoughtco.com/what-is-normal-distribution-3026707

Edwards, A. L., & Kenney, K. C. (1946). A comparison of the Thurstone and Likert techniques of attitude scale construction. Journal of Applied Psychology, 30(1), 72–83. Retrieved from: https://doi.org/10.1037/h0062418

Glen, S. (2015, August 16). Likert Scale Definition and Examples. Statistics How To. Retrieved from: https://www.statisticshowto.com/likert-scale-definition-and-examples/

Glen, S. (2016, June 29). Thurstone Scale: Definition, Examples—Statistics How To. Retrieved from: https://www.statisticshowto.com/thurstone-scale/

Horst, P. (1955). L. L. Thurstone and the Science of Human Behavior. Science, 122(3183), 1259-1260. Retrieved from: www.jstor.org/stable/1750776

Horst, S. & Pyburn, E. (2018). Likert scaling. In B. Frey (Ed.), The SAGE encyclopedia of educational research, measurement, and evaluation (Vol. 1, pp. 975-977). Thousand Oaks, CA: SAGE Publications, Inc. doi: 10.4135/9781506326139.n396

L. L. Thurstone | American psychologist | Britannica. (n.d.). Retrieved from: https://www.britannica.com/biography/L-L-Thurstone

Likert, R. (1932). A Technique for the Measurement of Attitudes. Archives of Psychology, 140, 1–55.

Prochaska, F. (n.d.). Thurston, Guttman, and Likert Scales. Retrieved from: http://www.sjsu.edu/people/fred.prochaska/courses/ScWk242/s0/Thurston-Guttman-and-Likert-scales.pdf

Rensis Likert | American social scientist | Britannica. (n.d.). Retrieved from: https://www.britannica.com/biography/Rensis-Likert

Sinaian, P. (2020, February 20). SurveyLegend ® | What are Likert-Type Scale Responses. SurveyLegend. Retrieved from: https://www.surveylegend.com/likert-type-scale-responses-examples-with-examples/

The Bell Curve (Normal/Gaussian Distribution) Explained in One Minute: From Definition to Examples. (n.d.). Retrieved from: https://www.youtube.com/watch?v=DJzmb7hGmeM

What Is a Sigma? | Perimeter Institute. (n.d.). Retrieved from: https://www.perimeterinstitute.ca/what-sigma

Written by

Leave a Reply

Your email address will not be published. Required fields are marked *