That's a strange question. It's a bit like if someone asks: "Which kinds of typewriters must I learn to be an excellent book writer?"

If you want to be an excellent biostatistician, you need a good understanding of statistics (and of biology, of course). Important skills besides this are the abilities of abstract thinking and problem solving. The latter requires a lot of creativity, which is thud another very important skill. Rather than knowing any particular program (software) you should be able to find a tool appropriate for the task and you should be able to learn (quickly) how to use the tool, where better not to use it, and what are all its assumptions and specialities. What counts is that you understand the problem to be solved, to judge the traps in the experimental design, to judge the quality and the problems with the data, and to understand what a particular manuipulation and analysis does, so that you solve the problem without making a mess.

It is understood that an excellent biostatistician can use any available tool after a short time. The tools will change over time, there might be restrictions given by the organisation,. and no tool is the best solution to all problems.

So it should be clear that no knowledge of specific programs can substitute for the main required skills I mentioned above. But if you want to work now as a biostatistician, it is surely not a disadvantage to gave good knowledge of R, Python and Julia, and to know how to use at leas one of Matlab, SPSS, SAS or Statistica. And it may be further helpful to have some skills in some other general-purpose programming languages like C++, Fortran, or Java, especially if you head towards bioinformatics as well. If you will work together with people mainly using Excel, it's not harmful to have goos skills in Excel (including VB programming).But at the end, thats just asking for the kind of typewriter. The book won't be good because some "high-end typewrite" was used!

That's a strange question. It's a bit like if someone asks: "Which kinds of typewriters must I learn to be an excellent book writer?"

If you want to be an excellent biostatistician, you need a good understanding of statistics (and of biology, of course). Important skills besides this are the abilities of abstract thinking and problem solving. The latter requires a lot of creativity, which is thud another very important skill. Rather than knowing any particular program (software) you should be able to find a tool appropriate for the task and you should be able to learn (quickly) how to use the tool, where better not to use it, and what are all its assumptions and specialities. What counts is that you understand the problem to be solved, to judge the traps in the experimental design, to judge the quality and the problems with the data, and to understand what a particular manuipulation and analysis does, so that you solve the problem without making a mess.

It is understood that an excellent biostatistician can use any available tool after a short time. The tools will change over time, there might be restrictions given by the organisation,. and no tool is the best solution to all problems.

So it should be clear that no knowledge of specific programs can substitute for the main required skills I mentioned above. But if you want to work now as a biostatistician, it is surely not a disadvantage to gave good knowledge of R, Python and Julia, and to know how to use at leas one of Matlab, SPSS, SAS or Statistica. And it may be further helpful to have some skills in some other general-purpose programming languages like C++, Fortran, or Java, especially if you head towards bioinformatics as well. If you will work together with people mainly using Excel, it's not harmful to have goos skills in Excel (including VB programming).But at the end, thats just asking for the kind of typewriter. The book won't be good because some "high-end typewrite" was used!

Agreed with Joachen. There are many software and packages for statistical analysis, e.g. SPSS, MATLAB, SAS, Statistica, and Maple, as well as R, Python, and Julia which are freeware.

R is currently the best choice as far as I know for its extensive libraries and functions particularly made for statistics.

Statistic drives to the software no backwards...

Hey, check out these two instructors and their courses at coursera.org:

https://www.coursera.org/instructor/rdpeng

https://www.coursera.org/instructor/~694443

They both teach Biostatistics at the Johns Hopkins Bloomberg School of Public Health.

I concur with Jochen, there is no program that will make you an excellent statistician. Never the less I myself use the program R which seems to used for the most part so it will be a good starting point.

The answer hides itself in the keywords of your question :) : R + Python.

I agree with everyone's response. Your question is ill-posed. There is no correct answer. A good starting point is whatever software has good support (online and from colleagues in-person), your colleagues/ collaborators use it, and does what you need and want it to do, and is easy for you to learn and use and easy to communicate the results, and time efficient to use. In my extensive experience, R programming, SPSS, Design Expert, NCSS-PASS, and MS Excell all meet this criteria. An example of software that doesn't meet this criteria *for my needs* (your needs may differ) from my experience is SAS which lacks user friendliness, support, and is hard to communicate the results, and seems to be more time-consuming to use than other softwares *for my needs*. I have not extensively used other software such as Python so I can't comment on that one but its popularity implies that it is a good candidate statistical programming language/software to consider learning.

Thank you so much!