Over a week ago, discussing on my local board on the matter of merits per week, and merely out of statistical curiosity, I took a look at what numbers would look like if we did a retrospective exercise. What I did is take a look at the number of potential forum candidates that had earned at least 1 Merit over the previous week of my mini study, and backtracked if they repeated merit earning over the previous individual weeks, over a 4 week period in total:
rank nUsers 1W nUsers 2W nUsers 3W nUsers 4W % Initial
Legendary 281 185 168 148 52,67%
Hero Member 112 67 54 44 39,29%
Sr. Member 84 45 41 35 41,67%
Full Member 79 38 31 25 31,65%
Member 61 29 17 13 21,31%
Note that the above is done from a global perspective, not delimiting the candidates to those actually participating in a given campaign.
Observations:- I did not consider Jr. Members.
- nUsers 1W represents the number of profiles that earn at least 1 Merit over the previous week (back when I did the exercise 10 days or so ago).
- nUsers 2W represents those that, also earned at least 1 Merit the prior week to week 1, and so forth. nUsers 4W would represent those that earned Merits over 4 consecutive weeks.
- I did the exercise backwards in time, from 1W to 4W going back in time. Theoretically, to be more precise (from a modelling perspective), I should have selected the candidates that got merited 4 weeks ago, and move forward week by week to see the results. It don’t think it would change the numbers much, but the better approach would have been the latter (I stated off only doing the exercise just covering a single week, but then extended it further back in reverse order for the sake of not redoing stuff).
- A given campaign may have selected some of the best candidates, and thus have better numbers than the overall numbers shown above. For example, I believe everybody complied with the 1 Merit requirement in the campaign cited here this past week, most without any issue, with a few struggling perhaps a bit.
- The 1 Merit requirement may push people to improve their posting, and thus the above numbers for a given set of selected people can be bettered due to this factor.
- The lower the rank though, in general terms, the more difficult it is to reiterate over a 4 week period measuring the 1 Merit requirement weekly. Again, I’m pretty sure that once subject to the rule within a campaign, participants will try to better themselves, thus improving the overall numbers. To what degree I can't tell.
