"be consistent" PerlMonks

Re^2: Perl Developer Survey 2017 results are available (emacs less popular)

by tye (Sage)
 on Apr 28, 2017 at 14:12 UTC Need Help??

Come on, the math is not hard here. The question allowed multiple responses and in one case the division was done using the number of responders instead of using the number of responses.

And you got your conspiracy exactly backward. The result was to artificially inflate the percentage reported for emacs.

``` \$ say '85+65+78+519+355'
1102
\$ say '85/11.02,65/11.02,78/11.02,519/11.02,355/11.02'
7.71324863883848
5.8983666061706
7.07803992740472
47.0961887477314
32.2141560798548
\$ say 138/11.02
12.5226860254083

So just drop emacs' popularity from the reported 16.25% to the correct 12.5%.

All that is left is to figure out if the mistake was motivated by the preparer being an emacs fan or because they felt sorry for emacs fans and whether it was intentional or accidental. q-:

Update: Oh, and the percentages reported for Notepad++, Eclipse, and VisualStudio might well also be inflated by x1.3 (hard to say for sure since the counts were not reported for them).

Update: Though, I kind of find the 30% higher percentages more accurate. The problem was probably that they put the counts into the graph software and it is what provided the percentages. Makes me wish for a "Venn diagram" chart where the area of each shape is proportional to the number of people who chose that option and the areas of the overlaps between shapes is also proportional (to the number of people who chose all of the corresponding options).

- tye

• Comment on Re^2: Perl Developer Survey 2017 results are available (emacs less popular)

Replies are listed 'Best First'.
Re^3: Perl Developer Survey 2017 results are available (not about editors)
by LanX (Saint) on Apr 28, 2017 at 15:07 UTC
Hi Tye

welcome back to the monastery! :)

I figured that out already, and I'm not such a fanboy that I care about the results of a bad survey.

My theory is that the questions were single choice, but the other option allowed free text were multiple editors were inserted.

Anyway a decent survey would at least

• avoid opinionated choices
• publish the original data
• keep a snap shot of the original survey online
Otherwise we'll never know, that's why I give up.

Regarding venn diagram, this wouldn't be possible in many cases.

Imho the best representation is a hasse diagram.

Cheers Rolf
(addicted to the Perl Programming Language and ☆☆☆☆ :)
Je suis Charlie!

"I'm not such a fanboy that I care about the results of a bad survey."

Yet here you are ... dong that very thing as that very role.

Re^3: Perl Developer Survey 2017 results are available (emacs less popular)
by LanX (Saint) on Apr 28, 2017 at 17:43 UTC

I don't know how good this survey site can defy bots only responding for 3 questions.

Cheers Rolf
(addicted to the Perl Programming Language and ☆☆☆☆ :)
Je suis Charlie!

Keep scratching that itch until it becomes reality ...

Create A New User
Domain Nodelet?
Node Status?
node history
Node Type: note [id://1189112]
help
Chatterbox?
and the web crawler heard nothing...

How do I use this?Last hourOther CB clients
Other Users?
Others contemplating the Monastery: (1)
As of 2024-02-22 03:14 GMT
Voting Booth?
My favourite way to spend a leap day ...

Results (22 votes). Check out past polls.