Segmentation analysis based on Likert I would like to do a segmentation analysis based on a Likert scale survey. I want to ask respondents to rate certain attributes on a scale of 5 points (not valuable at all, not very valuable, neutral, relatively valuable, very valuable), probably having 8-10 attributes in total.
Once I get all the answers, I would like to run an analysis to identify the most common types of respondents (price seekers, innovators...).
What is the exact analysis I would need to run to do this?
 A: Factor analysis can be an option.
Let's say you already have in mind of two common factors F1 and F2 as: price seeking and innovation. Then you can look at the sample correlation matrix to 
confirm your intuitions, that there are two groups of attributes, some high 
correlations exist within each group, and the collective meanings of attributes in each group match the corresponding factor in your mind.
I guess you can try different sets of attributes to use, and different number of common factors. In case of 2 factors and 8 attributes, you have the following representation:
$Z_1 = l_{11}F_1 + l_{12}F_2 + \epsilon_1$
$Z_1 = l_{21}F_1 + l_{22}F_2 + \epsilon_1$
...
$Z_8 = l_{p1}F_1 + l_{p2}F_2 + \epsilon_1$
Estimate the parameters.
Make appropriate rotations to make things more interpretable, ideally each attribute have one dominant loading $l_{ij}$ favoring one factor.
Estimate factor score for each person:
$(f_{11}, f_{21}), (f_{12}, f_{22}),...,(f_{1n}, f_{2n})$
People with high score on factor 1 can be grouped to the 1st group: price seeking group.
(SPSS, R and SAS should all be able to do it)
