RESEARCH DIARIES: ENTRY 5: Scaling Techniques: Likert Scale

Scaling techniques is a big chapter. So, rather than writing a big blogpost covering many of them in one go, let us get started with the first and the most popular one (rating scale / Likert Scale)

 Rating Scale: A type of question to learn about how respondents feel about the item being surveyed. The respondents are asked to rate something from "Very Good" to "Very Bad". It is a variant of MCQs (Multiple Choice Questions). These are also called "Categorial Scales". One can use two to seven points scales (or even more), as, there is no rule about how many points scale can be used. But generally, three to seven points scales are used. It is advised by some that the number of point scales should not be odd, but even, as even number of points scales will not allow the respondents to display the "central tendency ticking" behaviour.

For example, if the scale were:

[ ]Very Good    [ ] Good     [ ] Neutral       [ ] Bad       [ ] Very Bad

A person might choose Neutral in all the answers and that response will not be of much use.

But, if the scale were: 

[ ]Very Good    [ ] Good       [ ] Bad       [ ] Very Bad

it would not be possible to choose neutral in all the answers

1.1 Graphic Rating Scales: (Most popular example of this is the Likert Scale): Here the respondents can select an option on a scale. One extreme is presented on one side, while other extreme is presented on the other side. It has applications in Marketing, HR etc. An example:

[ ]Very Good    [ ] Good     [ ] Neutral  [ ] Bad  [ ] Very Bad

The problem is that the definition of "Very Good" or "Good" is left to the discretion of the respondents. 

Likert Scale is generally the one with five response items / categories. For example, it can start with "Strongly Disagree" and end with "Strongly Disagree". Named after its inventor Rensis Likert, it is generally used to measure the attitude of people towards any event, object etc.

Here, sometimes, values might be assigned to certain type of experience. Like, 1 = strongly disagree, 2 = disagree, 3 = neutral / neither agree/disagree, 4 = agree, 5 = strongly agree). Then sentences are offered to them. For example,

1. My teacher teaches well

2. My employers pays me well

3. I feel delighted with my "X" brand care

Then, these scores can be calculated. The main advantage of the Likert scale that it is easy to construct and administer. At the same time, it has a limitation that the respondent might misinterpret a statement.

Please note that (as per URL, Are Likert scales ordinal or interval scales? (scribbr.com)) "Individual Likert-type questions are generally considered ordinal data, because the items have clear rank order, but don’t have an even distribution.

Overall Likert scale scores are sometimes treated as interval data. These scores are considered to have directionality and even spacing between them

The type of data determines what statistical tests you should use to analyze your data."

Further, according to website, Likert Scale Definition, Examples and Analysis | Simply Psychology

How can you analyze data from a Likert scale?

The response categories in Likert scales have a rank order, but the intervals between values cannot be presumed equal.

Therefore, the mean (and standard deviation) are inappropriate for ordinal data (Jamieson, 2004)

Statistics you can use are:

• Summarize using a median or a mode (not a mean as it is ordinal scale data ); the mode is probably the most suitable for easy interpretation.

• Display the distribution of observations in a bar chart (it can’t be a histogram, because the data is not continuous).

Please note that above views are also challenged in many places (Is a Likert-type scale ordinal or interval data? (researchgate.net)) but we will stick to the above conclusions.