Creating and Interpreting Lickert Scale Data in Market Research

Case Study - Measuring Consumers' Preferences for Digital Devices

What scale? Lick-urt, s'il vous plait. Getty Images | Christoper Futcher | E+

The Likert Scale - A Reliable Standby for Surveys Research

As consumers adopt technological advances, the demand for digital content on various mobile devices,such as smartphones, tablets, and laptops, increases dramatically.  Consider that a market research client might want to better understand consumers’ preferences regarding different types of digital platforms and to explore the primary drivers of consumer video viewing for entertainment and for business needs.

The market research client has asked that a survey be developed to explore consumer attitudes about the use of technology platforms for content distribution.  The survey will be conducted over several months in order to collect data on how technological change and implementation influences the perceptions, attitudes, and behaviors of the survey participants. Both quantitative and qualitative data have been requested by the market research client.  Random sampling will be used to select the survey participants thereby establishing a probability sample, which will enable the application of inferential statistics to the data.  Random sampling helps to effectively reduce bias to acceptable levels

Examples of a 5-Point Likert Scale

A 5-point Likert scale can be used to record the responses of the survey participants.  (The name Likert is pronounced "Lick-urt" since it is a French surname.)  A Likert scale is a version of a summated rating scale, which is configured in a way that enables the conversion of text responses to quantifiable categories which can be summed to reflect the relative differences of the individual or aggregate responses.

  Even though there are no correct answers attached to the question items, a summated rating scale results in better reliability than a single rating scale tends to provide. 

Below are example questions that might be used in this survey.

Video content is sufficiently detailed such that I do not need to read web content.

__Strongly Agree   __Agree  __Neutral  __Disagree  __Strongly Disagree

After viewing a video, I usually go to the website for more in-depth information.

__Absolutely True   __Somewhat True  __Neutral  __Somewhat Untrue   __Absolutely Untrue

Consumers experience superior quality using UI / UX applications on business websites.

__Always   __Often  __Sometimes  __Seldom  __Never

The examples are formatted according to a 5-point Likert scale.  Since people tend to think in terms of a larger number indicating greater agreement or "truthfulness," the scale is configured so that when the scores are summed, a larger number is read as aligned or in agreement with the question item (which is really a statement, not a question).  

5 = Strongly Agree   4 = Agree  3 = Neutral  2 = Disagree  1 = Strongly Disagree

5 = Absolutely True   4 = Somewhat True  3 = Neutral  2 = Somewhat Untrue   1 = Absolutely Untrue

5 = Always   4 = Often  3 = Sometimes  2 = Seldom  1 = Never

How Can Likert Scale Data Be Interpreted?

However, it is important to recognize that a primary drawback of a summative score of the ordinal numbers from a Likert scale is that it the score imparts a sense of meaning that is not truly representative of any real magnitude.

 For the quantitative data that results from summing the points respondents record for each question item, statistical analyses will be utilized to determine relationships between the responses to the questions.  Accordingly, statistics may then be used to provide information about acceptable rates of reliability, validity, and sensitivity.  For instance, most market researchers insist that data from Likert scales pass the Cronbach’s alpha or the Kappa test of intercorrelation and validity.

Sources:

Jupp, V. (2006).  The SAGE dictionary of social research methods.

Likert, R. (1932).  A technique for the measurement of attitudes.  Archives of Psychology,140 (55).

Martinez-Martin, P.  (2010, February 15).  Composite rating scalesJournal of Neurological Science, 289 (1-2), 7-11.  doi: 10.1016/j.jns.2009.08.013.

Zikmund, W. G., Babin, B. J., Carr, J. C., & Griffin, M. (2013). Business research methods (9th ed.). Mason, OH:South-Western.