Overall Satisfaction with IBM SPSS
SPSS is used in various pockets throughout the institution by both researchers and administration. It is far from the only statistical tool in use, but because many of the administrators were trained in colleges of education it is the most commonly used tool. Further, as a menu driven tool, the learning curve is easier for new users who enter administrative departments. Note that in MOST cases SPSS is being used as a desktop product.
- Menu-driven interface allows for new, non-programming users to become productive quickly. This capability is significantly stronger than competing products, whose menu-driven interfaces often lack many important analytical features.
- Syntax (code-based) interface allows for coding of repetitive analysis and automation. While most tools have this, it is an invaluable piece of the software that allows for access to less common options and provides a clear way of documenting any process. Further, SPSS puts the syntax for each command into the output file, making it easy to see exactly what command resulted in what output.
- The output from SPSS commands has been enhanced to provide most of what an analyst needs in order to make a decision. This is helpful for new analysts, as it can guide them in the direction of a more complete analysis.
- SPSS is rock solid. This is a well-established product with 24 versions under it's belt.
- While substantial improvements have been made, the graphic capabilities remain opaque syntax-wise and relatively rigid through the menu driven interface. You need to know what you are doing, and even then the graphs created are not necessarily the best looking. I tend to extract the data and build presentation graphs in other tools.
- Data mining capabilities are provided, but at a HUGE cost in a different tool. As a non-profit, SPSS Modeler is out of our price range.
- Picky thing - SPSS does not compute marginal effects. Tools like SAS, STATA and R either provide those calculations automatically or at the very least make them trivial to get at. SPSS can't do it at all. From an econometric perspective, this is a HUGE drawback.
- I'm not sure we need a new version with new licensing every single year. This is especially the case if you aren't pushing the limits of the feature set.
SPSS provides the ability to get new analysts up and running more quickly than programming-oriented language, while still providing the flexibility you would expect in a code-able tool.
Depends on the department - Improved decision-making throughout the organization, but also student retention and progression improvements, financial savings in procurement and other tasks.
- SPSS has made repeatable (yearly) analysis a reality, with minimal maintenance and therefore substantially less analyst time after the first year. It's hard to put time saved into $$ but the increase in time available for other tasks has been statistically significant.
I use Stata for tasks that SPSS cannot support, but ultimately SPSS has a short learning curve, strong statistical processing, and a mature tool set. SAS is also mature, but more programming based. JMP tries to 2nd guess what I need. NOTE: R (open source) is a great option as well, BUT also is programming only. SPSS hands down has the best menu-driven options.
Using IBM SPSS
500 - Research and development, analytics and certain functional areas that require survey analysis.
All users are using the desktop version of SPSS, therefore the only support required is installation and re-licensing on a regular basis.
- Statistical analysis of research projects
- Predictive models using logistic regression
- none that I can think of.
- None for base SPSS - new ways to use it will involve purchasing SPSS modeler.
Evaluating IBM SPSS and Competitors
- Product Features
- Prior Experience with the Product
Experience with the product in university. I already knew it, it had a short learning curve for even advanced features and it could do almost everything I needed. Reducing startup time was invaluable.
No change, since it was skills I learned prior to this position that drove the selection. If I didn't know ANY tools, however, I don't know whether SPSS would have won out or not. Never underestimate the power of teaching your tool to college students.
IBM SPSS Implementation
- Implemented in-house
Change management was minimal
- A new version is released every year and keeping up with those versions and their licensing has proven a challenge in a large organization.
IBM SPSS Support
Pros | Cons |
---|---|
Quick Resolution Good followup Knowledgeable team Problems get solved Kept well informed Support understands my problem Support cares about my success Quick Initial Response | Difficult to get immediate help |
No - Too many individuals would require access, making it not an effective expense for the university
Using IBM SPSS
Pros | Cons |
---|---|
Like to use Relatively simple Easy to use Technical support not required Well integrated Consistent Quick to learn Convenient Feel confident using Familiar | None |
- Menu-driven interface makes nearly everything easy to do
- Built in identification of duplicate records based on one or more criteria
- You cannot get marginal effects