Likert scale data is ordinal data. You cannot perform arithmetic operations on ordinal data and thus cannot calculate a mean or standard deviation. However, some people will sum together the data from multiple Likert scale questions and then treat that sum as interval data. You can calculate a mean and standard deviation for interval data.

Dear Mamoona,

The phrase Likert scale is ambiguous. It could refer to a single ordinal question (in which case Gordon is correct) or it could refer to a group of questions on a single construct, in which case the goal is to create a scale variable from the group of ordinal variables. Whilst it is common practice to simply add them all together (and hope for the best), this is not statistically valid. Please refer to my guide on scale reliability.

Hi Mamoona,

There are different views on Likert type data. Statisticians usually treat it as ordinal data (which is perhaps correct and I agree with Peter and Gordon) but there are others, specially researchers in social sciences who defy this. There are numerous examples available in literature coming from some respected elite journals. Their argument is if you have big sample size, Likert type data can be considered for statistical operations e.g. mean and standard deviation can be calculated.

For beginners, I would suggest you use "SPSS Survival Manual" by Julie Pallant. It is simple and easy to follow guide.

In my opinion, it would be best if you take help of a statistician or person having knowledge of statistics. Because even if you calculate mean and sd, interpretation of teh results and presenting (writing) it correctly in paper would be another challenge.

I was in the same situation but the Pallant book has really being of great help thanks

I shared the same challenge in analysing Likert scale data but the book of Pallant seemed the best help to solve that challenge.