Hi, but most of time Likert scale answers are just for individual questions belonging to an instrument. It is usual to add up all the questions in order to calculate the instrument score. Some instruments have other scoring rules.

Thus, the idea is not to analyze an individual question, rather it is to analyze the whole score from those instruments.

Now, my questions:

What is the scale of the instrument score (from Likert-type questions)? Ordinal or Interval?

How should I analyze the instrument scales?

Thank you.

If these data fulfill the assumptions of parametric tests, yes you can. But it is preferred to use non-parametric statistics

I am just wondering that many scholars use parametric tests to analyze ordinal date (Likert-scale) and those papers get published. It has become a common norm!!!

Yes it is true especially if those articles are not reviewed by a statistician. 

Thanks prof. Mona for your contribution...

This issue has a big debate in educational statistics. Some researchers believe that the Likert Scale data comes between Ordinal and Interval measurement scale. Others believe that it is in the ordinal scale but for application purpose we can assume it is in the interval scale. This is why many researches in education ANOVA, T-Test, and others to analyze this type of data

And the debate continues....

Thanks Dr Mahmoud for your share of knowledge.

In my opinion it is appropriate to assume interval scale for an item if the response alternatives are formulated in a way that you can see them as equidistant.

Also I would like to remind you that if you state that the items of likert scales are just ordinal, you cannot construct a scale out of them because transformations like calculation of the sum or the mean make no sense with ordinal data.

Hi, but most of time Likert scale answers are just for individual questions belonging to an instrument. It is usual to add up all the questions in order to calculate the instrument score. Some instruments have other scoring rules.

Thus, the idea is not to analyze an individual question, rather it is to analyze the whole score from those instruments.

Now, my questions:

What is the scale of the instrument score (from Likert-type questions)? Ordinal or Interval?

How should I analyze the instrument scales?

Thank you.

A Likert-scale is a scale over several items and can be analysed using conventional test that are quite robust. Likert-type-items are a different matter. Here is a nice overview of the debate:

Carifio J, Perla RJ (2007) Ten Common Misunderstandings, Misconceptions, Persistent Myths and Urban Legends about Likert Scales and Likert Response Formats and their Antidotes. Journal of Social Sciences 3 (3):106-116

Thanks guys for this useful discussion.

We need to differential between Likert-type items which refer to individual items. This can only be analyzed using non-parametric tests.

Likert-scale, on the other hand, is the sum of a group of items that measure one construct. When we sum such items scores, we can use parametric tests give that the assumptions are met.

 Andres is correct where he outlines most of time Likert scale answers are just for individual questions belonging to an instrument. It is usual to add up all the questions in order to calculate the instrument score in social sciences. However on an individual question level you can indicate that you assume your likert scale approximates to an interval scale (especially if parametric conditions met).

Canany one share this article @ [email protected]   

Carifio J, Perla RJ (2007) Ten Common Misunderstandings, Misconceptions, Persistent Myths and Urban Legends about Likert Scales and Likert Response Formats and their Antidotes. Journal of Social Sciences 3 (3):106-116

Can we use inferential statistics tests to analyze five-point likert-scale data?. Available from: https://www.researchgate.net/post/Can_we_use_inferential_statistics_tests_to_analyze_five-point_likert-scale_data [accessed Apr 23, 2015].

Inferential statistics, unlike descriptive statistics, is a study to apply the conclusions that have been obtained from one experimental study to more general populations. This means inferential statistics tries to answer questions about populations and samples that have never been tested in the given experiment.

Inferential statistics infer from the sample to the population, they determine the probability of characteristics of a population based on the characteristics of your sample, they also help assess the strength of the relationship between your independent variables, and your dependent (effect) variables. With inferential statistics, you can take the data from any samples and make generalizations about a population. 

There are two main areas in case of inferential statistics, estimating parameters. This means taking a statistic from your sample data and using it to say something about a population parameter and hypothesis tests, this is where you can use sample data to answer research questions. 

Regards