CLOSED Poll: How to choose Contest Winners?

How to choose Contest Winners?


  • Total voters
    11
  • Poll closed .
Status
Not open for further replies.

ActiveAngel

Contest Committee Lead
Supporting
Contest Committee
Messages
2,307
Reaction score
2,724
Location
Under the sea
How to choose contest winners?

Notes:
We have outlined 4 options below based on user input from public discussion. But this doesn't mean the conversation is over. As always, we can update/amend these options, and/or add new ones as needed.
Follow-up question: Should only members get votes, or all ARC accounts?
You may vote for as many options as you wish, as the goal is to find the option that is least disagreeable.
You may choose to change choices at any time.
We now have additional volunteers to help with the Contest Committee; the result of this poll will be applied to the current contest.


Option A)
Use emojis/likes/loves to count popularity. Biggest number wins.
Almost instantly fast (5/5). Zero/Low effort (1/5).
  • Pros: 100% transparency for anyone that wants to count the likes at midnight.
  • Cons:Significant bias for first few people who post photos. Late submissions highly unlikely to win, regardless of photo.
    • Bias for people with followers.
    • High probability of repeat winners each month.
    • Question of how to weigh "loves" and other symbols, notably in the case of ties.
    • Easy potential for cheating if users create multiple accounts.
Option B)
Committee of volunteers chooses winner. Open to anybody that wants to have a vote.
Also fast (4/5). Low effort (2/5).
  • Pros: Discussion between voters can mitigate temporal/individual bias... and focus on merits of the contest parameters.
  • Cons: Appears non-transparent (users can always review photos independently, choose to be in voting conversation, and/or ask committee).
    • Conflict of Interest for voters not being able to be eligible to win... until committee size is large enough.
Option C)
Committee first chooses top 5*ish photos, then constructs blind poll of the top 5*. Popular vote on the poll is winner.
Slow (2/5). Moderate effort (3/5).
  • Pros: Poll results are transparent.
    • Polls can remove username to remove some bias.
    • Small number of photos are more likely to be reviewed "in full" by voters.
  • Cons: Still non-transparent for the committee (unless people want to participate or ask).
    • Still Conflict of Interest for voters.
    • Voting Bias during final selection (e.g., voters choosing pics of rare corals, rather than on photo characteristics or contest parameters)
    • Potential cheating with multiple accounts can determine winner, if not regulated.
Option D)
Committee member constructs blind poll of all photos. The Poll's top 5 photos go to committee, to discuss and choose the final winner.
Slow (2/5). High effort (5/5).
  • Pros: Poll results are transparent.
    • Poll can remove username to remove bias
    • Committee can discuss and judge final selection based on merit of the contest.
  • Cons: Still non-transparent for the committee (unless people want to participate or ask).
    • Potential 'first selection' bias during poll (Users may not be willing to spend time to view large number of photos.)
    • Still Conflict of Interest for voters.
    • May result in users disagreeing with committee selection.
 
A solution for preventing spam votes in scenarios C and D would be to ignore votes from accounts less than x days old with less than x amount of posts. That would at least make it more difficult to cheat, since you would have to make a fake account & also use it and probably ask fake questions and such in order to reach the requirements.
 
Status
Not open for further replies.
Back
Top