# Measures of internal consistency of survey items

I have a question on a survey that asks students how useful a feature of the software was that they used for learning. I then have three questions that assess in what way the feature was useful. The measures use a 7-point Likert scale from Very Strongly Disagree to Very Strongly Agree:

The [[feature]]
...was useful for learning overall
...helped me to mentally organize the structure of the content.
...enabled me to control my learning pace
...made it easy for me to find the content I needed


The question I'm trying to answer is "If the feature was useful, in what way was it useful?"

I'm wondering what kind of statistical tools can be used to do this. Could a covariance matrix determine this sufficiently? Could I compute Cronbach's alpha for each question paired with the first question? I have also come across Confirmatory Factor Analysis, but I don't yet understand it well enough to know if it is an applicable tool.

I'm new to stats so I'm just looking for a pointer in the right direction.

Edit: If it makes a difference, it is a within-subjects experiment where the student has used the software both with and without the feature.

-
I think for each question you can look for the most frequently occurring score and perhaps the mean score to get a sense of how the respondents felt about each question. Also the Kappa statistic can be used to measure the level of agreement between students to the answers to the questions. Also you can look to see if any particular question stood out as having a high frequency of agree and strongly agree answers. I think the analysis should be mostly exploratory rather than confirmatory. – Michael Chernick Sep 11 '12 at 15:46
Michael's keep it simple approach makes sense to me. I am not sure I understand your within subject design, but it might be that you can also calculate a difference score for each person for each feature. The average difference score (for each feature, across people) would tell you if and how much the feature improved the usefulness of the software. In any case, you should be able to get some useful information from the within subject feature of your design. – Joel W. Sep 11 '12 at 19:16