How to analyse 5 point Likert scale data in SPSS

How do you analyze Likert scale data? Likert scales are the most broadly used method for scaling responses in survey studies. Survey questions that ask you to indicate your level of agreement, from strongly agree to strongly disagree, use the Likert scale. The data in the worksheet are five-point Likert scale data for two groups

Read my post that compares parametric and nonparametric hypothesis tests.

Most people are more familiar with using parametric tests. Unfortunately, Likert data are ordinal, discrete, and have a limited range. These properties violate the assumptions of most parametric tests. The highlights of the debate over using each type of test with Likert data are as follows:

• Parametric tests assume that the data are continuous and follow a normal distribution. Although, with a large enough sample, parametric tests are valid with nonnormal data. The 2-sample t-test is a parametric test.
• Nonparametric tests are accurate with ordinal data and do not assume a normal distribution. However, there is a concern that nonparametric tests have a lower probability of detecting an effect that actually exists. The Mann-Whitney test is an example of a nonparametric test.

What is the best way to analyze Likert scale data? This choice can be a tough one for survey researchers to make.

If you want to find correlations between Likert items, be sure to read my post about Spearman’s correlation because that analysis is designed for ordinal data.

Learn more about Ordinal Data: Definition, Examples & Analysis.

Which Test is Better for Analyzing Likert Scale Data

Studies have attempted to resolve this debate once and for all. Unfortunately, many of these studies assessed a small number of Likert distributions, which limits the generalizability of the results. Recently, more powerful computers have allowed simulation studies to meticulously analyze a broad spectrum of distributions.

In this post, I highlight a study by de Winter and Dodou*. Their study is a simulation study that assesses the capabilities of the Mann-Whitney test and the 2-sample t-test to analyze five-point Likert scale data for two groups. Let’s find out if one of these statistical tests is better to use!

The investigators assessed a group of 14 distributions of Likert data that cover the gamut. The computer simulation generated independent pairs of random samples that contained all possible combinations of the 14 distributions. The study produced 10,000 random samples for each of the 98 combinations of distributions. Whew! That’s a lot of data!

The study statistically analyzed each pair of samples with both the 2-sample t-test and the Mann-Whitney test. Their goal is to calculate the error rates and statistical power of both tests to determine whether one of the analyses is better for Likert data. The project also looked at different sample sizes to see if that made a difference.

Comparing Error Rates and Power When Analyzing Likert Scale Data

After analyzing all pairs of distributions, the results indicate that both types of analyses produce type I error rates that are nearly equal to the target value. A type I error rate is essentially a false positive. The test results are statistically significant but unbeknownst to the investigator, the null hypothesis is actually true. This error rate should equal the significance level.

The 2-sample t-test and Mann-Whitney test produce nearly equal false positive rates for Likert scale data. Further, the error rates for both analyses are close to the significance level target. Excessive false positives are not a concern for either hypothesis test.

Regarding statistical power, the simulation study shows that there is a minute difference between these two tests. Apprehensions about the Mann-Whitney test being underpowered were unsubstantiated. In most cases, if there is an actual difference between populations, the two tests have an equal probability of detecting it.

There is one qualification. A power difference between the two tests exists for several specific combinations of distribution pairs. The difference in power affects only a small portion of the possible combinations of distributions. My suggestion is to perform both tests on your Likert data. If the test results disagree, look at the article to determine whether a difference in power might be the cause.

In most cases, it doesn’t matter which of the two statistical analyses you use to analyze your Likert data. If you have two groups and you’re analyzing five-point Likert data, both the 2-sample t-test and Mann-Whitney test have nearly equivalent type I error rates and power. These results are consistent across group sizes of 10, 30, and 200.

Sometimes it’s just nice to know when you don’t have to stress over something!

Reference

*de Winter, J.C.F. and D. Dodou (2010), Five-Point Likert Items: t test versus Mann-Whitney-Wilcoxon, Practical Assessment, Research and Evaluation, 15(11).

Two scientific pocket calculators on a desk. Photo by Michael Kwan [CC BY-NC-ND]

This post will give you some advice about using SPSS to summarise data that were generated with a Likert scale. If you want to read up on Likert scales before you go on, you can find some information in this post.

Before we start

Why should you summarise Likert scale data

Elsewhere in this blog, I have written that a Likert scale might consist of several items that measure a similar underlying construct (a latent variable). For instance, if I want to measure people’s attitudes towards sweets, I might ask them to record what they think about the following statements:

In order to interpret these data, we need to summarise the data in the scale. We can do this in two ways: adding the data or estimating the median. In this post, I will show you how to estimate the median, because this is slightly harder. The same steps can be modified to add up the data. 

Using the same example as above, I need to create a new ‘super-variable’, which shows the mean of items (1), (2) and (3) for each respondent.

My assumptions about you

I assume that you will already know how to define variables and values, how to toggle between the numerical expression and verbal descriptor of the values (i.e., you can make SPSS show responses as “strongly agree/agree/disagree/strongly disagree” or as “1/2/3/4”), and how to key in data. I will also assume that you have already established that your scale is internally consistent, so I will focus only on the technical aspects of merging the variables.

Here’s how to merge the Likert items

Starting out

Your starting point for summarising Likert scale data with SPSS will be a dataset similar to the one shown in Figure 1, below.

When you have created the dataset by typing your data into SPSS, and after you have tested for the internal consistency of the scale (use Cronbach’s α), it’s time to create a new variable.

Merging the variables

From the top menu bar in SPSS, select Transform -> Compute variable. You should now see the following dialogue box.

1. Assign a name to the new variable (e.g., Sweets);
2. Scroll down the Function Group, and select Statistical;
3. From the functions that appear select the Median. [ΝΒ it is possible to select the mean, but I don’t recommend it]. At this point, the following formula should appear in the numerical expression box: Median ( , )
4. Place the cursor in the brackets, select the variables you want to merge, and click on the arrow. Repeat with all the variables, separating them with comas.
5. Click on OK.

Your new Likert scale

SPSS will automatically generate a new variable, which will appear at the end of your dataset. This will be in numerical form (1, 2, 3, …), but you can change it to a verbal descriptor for consistency (Figure 3). You can use this variable for descriptive statistics (e.g., estimate the central tendency and dispersion), cross-tabulations, correlations and so on…

Now wasn’t that very easy?

Frequently Asked Questions

Over time, a lot of people have asked questions about Likert scales in the comments section of this post. I have collected the most usual things people ask in this section.

There are decimal points in the median I calculated. Is that a problem?

If your median falls between two values, it will have a ‘half’ (e.g., 2.5, 4.5 etc.). This is normal. You can report the median as you see it.

Why you do not recommend grouping the Likert scales as means and you recommend using medians?

The data produced by Likert type items are, strictly speaking, ordinal data. That means that they can tell us how to rank responses (strongly agree is ‘more’ agreement than agree) , but they do not give us information about the distance between them (strongly agree is not twice as much agreement as agree). Think of the medals in the Olympics: they can tell you if an athlete came first, second or third, but you cannot use them to calculate average speed. The median is a cruder statistic than the mean, because it does not take into account the ‘distance’ or ‘weighting’ of responses. In this case though, it is the best statistic we can legitimately use because this ‘distance’ is unknown.

OK, I did what you said, but what should I do next with my study?

It’s hard to answer such a question without knowing more about what you’re trying to find out (your research question) and your data. This is the kind of question that your advisor or mentor will be better qualified to answer.

Where can I find out more information about all this?

There are many statistics manuals you could read, if you want to follow up on the information in this post. My personal favourite is Andy Field’s Discovering Statistics with SPSS. I have also written some more posts about quantitative research below,, which you might find useful:

Before you go

If you landed on this page while preparing for one of your student projects, I wish you all the best with your work. There’s a range of social sharing buttons below in case you feel like sharing this information among fellow students who might also find it useful. Also feel free to ask any other questions you may have, using the contact form.

• Share on Facebook
• Share on Twitter
• Share on Linkedin
• Share via Email