1
$\begingroup$

I have a voting system where projects are evaluated and given a score from 0 to 4.

I am calculating standard deviations for each but I now want to characterise standard deviations as something like 'low' or 'high' where:

'low' means a close grouping of scores (and thus concencus around the score)

'high' means a wider grouping of scores (and thus a lack of concencus around around the score)

'medium' might obviously be somewhere in the middle.

I think ultimately I want to be able to filter results something like all the project within 1 standard deviation of the mean, all projects within 2 standards deviations etc.

But how do I calculate 'within one standard deviation'?

Not a maths guys really so thanks for the help.

Danny.

$\endgroup$
  • $\begingroup$ So everyone voting on a project can give it a $0,1,2,3,$ or $4$? Is this a peer voting system (like students grading each others' projects)? About how many votes are you capturing? $10, 100, 1000, ...$? $\endgroup$ – John Aug 10 '16 at 22:09
  • $\begingroup$ Curatorial team evaluating submissions in a gallery. 4=yes, 3=yes/maybe, 2=maybe, 1=no/maybe, 0=no. Number of votes will probably vary a little depending on the size of the panel for each exhibition etc. but typically between 3 and 6 votes $\endgroup$ – Danny Browne Aug 10 '16 at 22:20
  • $\begingroup$ If you use a computer even for 6 votes all potential outcomes are 5^6 which is not a high number. Since you look at three qualifiers you may just group the total results into the three "low", "medium" and "high" options by deciding that in each group you keep 1/3 of the potential results and you order them based on the calculated standard deviation and have the cut off based on the split into 3. $\endgroup$ – Moti Aug 10 '16 at 23:31
  • $\begingroup$ I see what you are saying but that won't work. That would also assume that we only look at all the submissions we need to evaluate at end, but in reality evaluation will be going on while submissions are still coming in. I need someone to be able to see at a glance whether the scoring on each submission is mostly yes, mostly no or mostly maybe. $\endgroup$ – Danny Browne Aug 11 '16 at 13:58

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.