Compare means of different variables

Question

I've got more than 20 (10 point likert scale) variables with more than 1000 entries each. What I want to do is compare the means of the answers on the questions.

A one-way anova seems suitable for this, but you can only categorize by the values of a variabele. I want to categorize by question, the variable itself.

Is there any way to do this without having to put all answer beneath each other and having another variable saying which question it is.

Just to be clear, I've got:

variable: question1 --> 1000 entries ranging from 1 to 10

variable: question2 --> 1000 entries ranging from 1 to 10

...

variable: question20 --> 1000 entries ranging from 1 to 10

I want to compare the means of the different questions with one-way anova but I can't choose to factor by question.

Answer 1

• If the same 1000 participants answered each question:
  • consider using paired samples t-tests and repeated measures ANOVAs
• If a different 1000 participants answered each question:
  • then it sounds like you need to restructure your data file to conform to the expectations of the statistical package you are using.