Overall Satisfaction with IBM SPSS
Researchers on the Market Intelligence team use SPSS regularly to analyze survey data. Typically, we are correlating or cross-tabulating audiences and insights to make recommendations to the business. Some researchers do more advanced statistics such as clustering analyses for segmentation, regressions, and factor analysis. The most frequently used functions are cross-tabs and descriptive tools.
- Has a GUI, as opposed to other analysis tools like R or Python
- Accepts and organizes data relatively well
- Makes statistics accessible to non-statisticians
- Lots of finicky work to do simple tasks
- Usability is atrocious [in my opinion]. No ability to customize.
- Would love to see product enhanced with interpretation features or citation tools (e.g., report results APA style)
- Bulk editing variables would be an improvement
- UI looks like its straight out of AOL days
SPSS lets us do analyses quickly without having to do any programming. Everyone uses it differently, but it is an important component for us making sense of our data and returning insights to the company.
I have used R when I didn't have access to SPSS. It takes me longer because I'm terrible at syntax but it is powerful and it can be enjoyable to only have to wrestle with syntax and not a difficult UI.
If you aren't comfortable programming and you need to do statistics, SPSS is a great choice. Psychology undergrads, for example, have to take statistics. Their difficulty with the programming portion cuts into their ability to learn the actual analyses, so SPSS is good for that. I would love to see IBM SPSS invest more in their UI/UX and increase the usability of this tool because it could be so much better. Their iconography is a mess, there's very little feedback, and don't even get me started on that last "big update" that basically truncated all functionality. The biggest thing I would like to see is to have the "Analyze" function broken into categories, or other buttons, or be able to customize it in some way. Having to repeatedly go through drop down menus to run analysis after analysis is a huge time suck, and very error prone (easy to click on the wrong thing). Lastly, one of the biggest challenges for beginning (and even experienced!) statisticians is interpreting output. More tool tips, feedback, integrated coaching, or something like that to help people understand more about what they are seeing could be so beneficial.