Hello, I am using SPSS to analyze my data but I am stucked . My independent variables are CEO Duality, Audit Committee and Board size and under each of these variables I have 10 questions. So for each variable there are 10 questions with a Likert scale of 1-5. I want to know how I can group these questions under one variable
Likert scales are based on taking the mean answer from several questions. So you can make a new variable based on those 10 questions by simply using the Compute Mean function. If you're using syntax you can copy and paste this, just change the names to whatever variable names you've used for your data.
Compute BoardSize = mean(v1 TO v10).
(Note: if your variables aren't listed in order you shouldn't use the 'TO' function, instead just type out all variable names.)
Edit: To clarify, if you want to create one likert scale based on 10 questions you should first run a factor analysis or a principal component analysis to verify that this is valid. For example you might need to not include one or two questions because they don't measure the same concept as the others.
Adding them up or taking the average would do fine. The former is called the summated scale score while the latter would be the mean scale score. Either ways, your results will not differ as long as you use them for regression or correlational analysis AND that you do NOT have missing ratings for every participant. If you have missing ratings for some participants and you still want to compute their scale scores, you need to apply the average.
FYI: https://stats.stackexchange.com/questions/148109/in-spss-should-i-calculate-the-mean-score-or-the-total-sum-score
If you know which questions respond to which variables, you can combine the Likert items by adding or averaging to form a scale. However, be sure to check the reliability of that scale before using it.
First, group your likert scale questions into different heading which you want to measure (re-code scale reversely if there is negative question) . Secondly, add all the ten questions and create scores.Finally, calculate the mean score from added output and divide the obtained variable into dichotomous form.
- Badges
- Science topic
- Similar topics
- Mathematical Sciences
- Statistics
- Statistical Software
- SPSS