Disclaimer: I am a long-time SPSS user, but for that past few years have been using Stata too. My overall comment is that neither is perfect, and both have strengths & weaknesses. The choice may depend (to some degree) on what types of analyses you do.

SPSS certainly has a reputation for being one of the most user-friendly stats packages. But see Thom Baguley's first post in this thread for a different view on that--he calls it an "illusion of user-friendliness". ;-)

• https://www.researchgate.net/post/Which_one_is_the_best_for_statistical_analysis_and_modeling_R_or_SPSS

Consistency of command structure must be considered, I think, when judging ease of use. And on that point, I think Stata wins, hands down. Another academic I know who started life as an SPSS user but later switched to Stata summed it up by saying that in comparison to Stata, SPSS sometimes feels like 50 different programs written by 50 different people, each with their own ideas about how the syntax should work.

One other nice feature of Stata is that (unlike SPSS) it is not modular. When you buy a license, you get all of the features & commands.

HTH.

Take a look at the excellent and periodically updated article, "The Popularity of Data Science Software" by Robert Muenchen:

http://r4stats.com/articles/popularity/

You will see from several of the graphs that SPSS has enjoyed enormous popularity, but it has recently been in precipitous decline. Stata, on the other hand, appears to be steadily increasing in popularity by several metrics.

With its menu-driven interface, SPSS might be easier to use than Stata. On the other hand, it could be argued that Stata is more "powerful" because of its combination of menus and a command structure, along with its extensibility via "ado files".

Particularly among academics, and even more so among younger academics, "R" appears to be the clear winner among statistical software. Although R has a steep learning curve, it contains a massive and steadily growing library of packages, and given that it is a programming language, it can be tailored for individual data analysis needs.

Ultimately, the software one chooses for statistical analysis should depend on the user's data analysis needs and how well these are met by any given program.

Disclaimer: I am a long-time SPSS user, but for that past few years have been using Stata too. My overall comment is that neither is perfect, and both have strengths & weaknesses. The choice may depend (to some degree) on what types of analyses you do.

SPSS certainly has a reputation for being one of the most user-friendly stats packages. But see Thom Baguley's first post in this thread for a different view on that--he calls it an "illusion of user-friendliness". ;-)

• https://www.researchgate.net/post/Which_one_is_the_best_for_statistical_analysis_and_modeling_R_or_SPSS

Consistency of command structure must be considered, I think, when judging ease of use. And on that point, I think Stata wins, hands down. Another academic I know who started life as an SPSS user but later switched to Stata summed it up by saying that in comparison to Stata, SPSS sometimes feels like 50 different programs written by 50 different people, each with their own ideas about how the syntax should work.

One other nice feature of Stata is that (unlike SPSS) it is not modular. When you buy a license, you get all of the features & commands.

HTH.

Thanks to Bruce Weaver and his referral to a post on the "illusion of user-friendliness" applied to SPSS. I have had the same experience with SPSS and would agree that its apparent ease of use is somewhat an illusion. Consequently, I almost never use SPSS.

Indeed, for most of my data analysis and graphing needs, I turn to GraphPad Prism, which I find truly intuitive.

For more advanced needs, I use OriginPro, which has its own scripting language along with interfaces to Matlab, R, Python, and Mathematica.

When I need something even more specialized or customizable, I make use of various other software, such as JMP, KNIME, Matlab, Python, Stata, Statistica, Systat, or R.

There is no single tool that can handle every job. However, if I were just now starting out, I would choose to learn thoroughly two highly adaptable and complementary data analysis and programming languages: R and Python.

Frankly speaking R is the best software which provides every thing what we can think, new packages are continuously developing in it. Just a bit difficult as it involves programming but that we can get on R help menu or help on packages.

But if we compare SPSS and STATA , then no doubt STATA is much faster then SPSS, it has many more advance tools as compared to SPSS. STATA includes both programming and drop down menu that is also in SPSS but programmming in SPSS is too complex.

I thought STATA was the most expensive. Seems like that's wrong?? I've been a SPSS user for a long time and it covers most of my needs, except SEM (use Mplus). I've also heard that the STATA learning curve is too steep. Maybe that's also wrong??

best

John Kåre

The use of either one is based on choice, each package has its pros and cons.

Dear all,

Your input has been very important for a debate that has been going on and will impact not only specific economic / added value concerns by universities, but also to the researcher’s efforts and planning – specifically to learning curves (since the optimum working performance level takes some time to obtain and switch costs tend to be high) and tech proficiency (I myself gave up learning new programming languages as I have my hands overflowing with everything else – it’s a dilemma).

Thank you all very much for taking the time to answer this question.

Very best regards,

Mathias Carvalho