I trust SPSS for my analytic needs.
Updated July 30, 2019
I trust SPSS for my analytic needs.
Score 8 out of 10
- IBM SPSS Statistics
Overall Satisfaction with IBM SPSS
SPSS is used only within our Member Insights Department and is used as our primary quantitative analysis tool. We use SPSS to handle all of our survey data needs, except for data collection. We use it to extract data out of our membership SQL database, pull a random sample of relevant members, match up to our collected survey data with relevant stored member data, and analyze the survey results to include in our reports and presentations. Our analyses always include descriptive measures (frequencies or means/medians), often include inferential measures (e.g., chi-square, ANOVA), and less frequently predictive measures as well (e.g., logistic regression).
- Biggest advantage: allows us to save all of our data manipulation and analysis instructions (SPSS calls it "Syntax") so that we can understand what we did, say, one year from now. This can be critical, especially if you are conducting year-over-year comparisons, or re-analyzing data in a new way after the primary reporting has been completed. Pretty tough to understand what was all done, say, in an Excel spreadsheet.
- Has all of the analytic capabilities we have needed so far (well, except for Random Forest).
- We use the Tables module a ton to create most of our crosstab kind of results, and SPSS gives you a lot of flexibility to create different kinds of tables with the same set of data.
- I learned SPSS when only Syntax was available, but I really appreciate being able to browse through the menus, and then just paste the Syntax you need. It also comes in handy when you're just playing, trying out different ways to look at the data.
- SPSS has for a long time touted it's "new and improved" graphing capabilities, and while they are that, they still have flexibility issues (or I just haven't found how to correctly use the software). For instance, why are data labels in a bar chart restricted to appearing only within the bar itself? Sure, you can have them at the bottom, middle, or near the top of the bar, but why not, say, just above the bar?
- When dealing with qualitative data, say comments from survey respondents, occasionally we run into characters likely typed by the respondent, which SPSS doesn't know how to handle, so it converts it to a space. It took me a while to figure out that it wasn't a space at all, and was wondering why my Syntax wasn't working (still assuming it was a space). It would help if SPSS could convert unknown characters to something more obvious.
- Biggest headache using SPSS is dealing with the IBM hierarchy when needing customer support or even upgrading to the new version (which comes out annually). In order to upgrade to v26 in April 2019, I eventually found out I needed not just an Authorization Code, nor a Full License Code, but also a Lock Code and spent way too many hours attempting to straighten this out with IBM customer support. I handle purchasing the upgrades too, and that task is not straightforward either. I miss the SPSS support before they were acquired by IBM.
We are able to efficiently conduct new or recurring projects by reusing stored Syntax.
SPSS is well suited to our survey data and analysis needs (but not data collection). It also works well for our data mining needs. Most of the staff I hire were exposed to SPSS in college, but either way, there is a learning period to understand how to document your work using Syntax (even if pasted from the menus) and I utilize mostly online resources to help them get up to speed.