SAS, EViews, Excel, MATLAB, JASP & jamovi Analysis Support

Not every analysis happens in SPSS, Stata, R or Python. Clinical and pharmaceutical research leans on SAS; macroeconomics and finance courses often teach EViews; business and accounting programs expect Excel; engineering and signal-processing projects use MATLAB; and many psychology departments now teach with JASP or jamovi, free programs built on R. Each has its own workflow, file formats and typical mistakes.

This page covers support in all six. The service works the same way as our other software support: we clarify your research question, work in the required tool with a saved record of every step (a program, a workfile with a command log, a documented spreadsheet or a saved analysis file), and explain the output so you can report and defend it.

If your question outgrows the tool — for instance, a multilevel model in Excel, or panel GMM in JASP — we explain the limits and suggest the smallest step up that will do the job properly.

In SAS, the log is as important as the output. A program can finish and still carry a note such as 'MERGE statement has more than one data set with repeats of BY values', which usually means rows were matched incorrectly. We treat a clean log as part of the deliverable. For academic users, SAS offers free cloud options: SAS OnDemand for Academics for code-based work in SAS Studio, SAS Viya for Learners for low-code analytics, and the newer SAS Viya Workbench for Learners, which combines SAS with Python and R. Your course will usually specify one.

EViews is built around the workfile, and many errors come from setting the wrong frequency or sample range. We set these explicitly, create transformed series (logs, differences, growth rates) with clear names and save commands in a program file rather than relying only on menus. EViews 14 is the current release, available commercially, in a University Edition and as a free but restricted Student Version Lite; check the limits of your version before starting a large project.

Excel is fine for descriptive work, simple tests and transparent financial or operational models, and it is what many managers expect to see. It becomes risky when analyses depend on copied formulas, hidden rows or manually typed results. We separate raw data, calculations and outputs on different sheets, avoid hard-coded numbers inside formulas, and audit references so each result can be traced. If you need multilevel models, robust standard errors or large-sample survey weighting, we recommend moving that part to a statistical package.

MATLAB suits numerical, engineering and signal data. It is released twice a year (an 'a' release in spring and a 'b' release in autumn), and statistical functions depend on which toolboxes your license includes, so we check for the Statistics and Machine Learning Toolbox before relying on functions such as fitlme. We write functions rather than one long script so the same processing can run across hundreds of files.

Both programs are free and open source, update frequently, and run R underneath a point-and-click interface. Results update live as you change options, and the saved file keeps both the data and the analysis settings, which is a good basis for reproducibility if you share it. JASP is particularly strong for Bayesian statistics; jamovi's module library and syntax mode make it easy to see and export the R code behind an analysis.

The main risks are interpretive rather than technical: reporting a Bayes factor without saying which prior was used, or confusing BF10 with BF01, or relying on defaults for post hoc corrections. We explain these choices and help you report them clearly. Both projects release often — jamovi, for example, moved from 2.7 to a 28.x numbering series in 2026 — so we note the exact version in your methods.