Which software (e.g., MS Excel / Word, R, Matlab etc) do you find best to use for presenting statistics data visually for publication? @PhDForum @AcademicChatter @JASPStats #HE @OxExpPsy @RHULPsychology @VBourne_stats @ProfAndyField @MaartenSpeek @UCLPALS @o_guest @PaLSPPG
-
Show this thread
-
Replying to @VictorKovalets @PhDForum and
R is more satisfying to figure out, but I can make a plot indistinguishable from R using Excel literally 10 times faster.
2 replies 1 retweet 6 likes -
Replying to @LaurentWada @VictorKovalets and
I want to anonymous, but I also want to post some of my published figures and have people guess whether it was R or Excel.
1 reply 0 retweets 2 likes -
Replying to @LaurentWada @VictorKovalets and
People get way too snobby about software, it’s just a tool to do a job. If you’re happy with SPSS use it, if you prefer R use it, love Excel? No problem. Use what works for you (and your collaborators).
10 replies 15 retweets 73 likes -
Replying to @ProfAndyField @LaurentWada and
I mean sure, but the reproducibility of Excel is lower unless you always document the version and which buttons you pressed. With code the commands are the figure so less work to makes reproducible by default.
3 replies 4 retweets 20 likes -
Replying to @o_guest @ProfAndyField and
Sure. But this ignores that you need to know what you are doing, when you code your statistics in R. If that's not what you are good at, it is better to stick to what you know. IMHO (I am guessing this is what Andy means)
2 replies 0 retweets 1 like -
Replying to @GordonFeld @ProfAndyField and
You need to know what you're doing when you use Excel too. Just because the output looks good doesn't mean it's correct. You need to know how to do your job regardless.
1 reply 1 retweet 5 likes -
Replying to @o_guest @ProfAndyField and
Of course. But I would argue that in Excel and SPSS it is easier to know, (roughly) what you are doing. There are just less options and you can usually follow a recipe. In e.g. R there is much more that can go wrong.
3 replies 0 retweets 1 like -
Replying to @GordonFeld @ProfAndyField and
Given that I know of bugs in previous versions of SPSS where the p values output were mathematically wrong, I don't think it's a fair thing to say people know what they are doing if they are just mashing buttons into a GUI.
2 replies 0 retweets 2 likes -
Replying to @o_guest @ProfAndyField and
Ok. So if I understand you correctly, this means that a (quite large) group of people (i.e. most psychologists) who have been extensively trained in stat methods shouldn't be doing statistics. What about programming-pros, who have no formal training in stat methods?
2 replies 0 retweets 1 like
I have no idea how you concluded this from my remarks.
-
-
Replying to @o_guest @ProfAndyField and
Really the only thing I am saying is that "allowing" people to use SPSS or Excel is more inclusive. (And knowing what a stat method does theoretically may be more important than being able to implement it in a programming language...)
2 replies 0 retweets 1 like -
Replying to @GordonFeld @ProfAndyField and
I didn't disallow them. I can't disallow them. I merely pointed out it's less reproducible or requires extra effort to be reproducible. Brief discussion here on the basics of what that means: "Dialogue: What is Computational Reproducibility?" http://oliviaguest.com/doc/guest_rougier_2016.pdf …
0 replies 0 retweets 1 like
End of conversation
New conversation -
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.