Design contest results

The deadline for voting has passed. The Candidates | The Contest Details

The table holding the results is: \d design_results
                            Table "public.design_results"
 Column  |     Type     |                          Modifiers
 id      | integer      | not null default nextval('design_results_id_seq'::regclass)
 user_id | integer      | not null
 choice1 | character(1) | not null
 choice2 | character(1) | not null
 choice3 | character(1) | not null
    "user_id" UNIQUE, btree (user_id)
    "fki_design_results_user_id" btree (user_id)
Check constraints:
    "choice_all_different" CHECK (choice1 <> choice2 AND choice1 <> choice3 AND 
                      choice2 <> choice3)
    "valid_choices" CHECK ((choice1 = ANY (ARRAY['A'::bpchar, 'B'::bpchar, 'C'::bpchar])) AND
 (choice2 = ANY (ARRAY['A'::bpchar, 'B'::bpchar, 'C'::bpchar])) AND (choice3 = ANY
 (ARRAY['A'::bpchar, 'B'::bpchar, 'C'::bpchar])))
Foreign-key constraints:
    "design_results_user_id" FOREIGN KEY (user_id) REFERENCES users(id)

The voting tally strategy involves ranking. First choice gets 4 points. Second choice get 2 points. Third choice gets 1 point.

Here are the results: select sum(rank) as rank, choice1 from ( select count(choice1) * 4 as rank, choice1 FROM design_results as first group by choice1 UNION select count(choice2) * 2 as rank, choice2 FROM design_results as first group by choice2 UNION select count(choice3) * 1 as rank, choice3 FROM design_results as first group by choice3 ) as tmp group by choice1 order by rank desc ;
 rank | choice1
  306 | A
  185 | C
  181 | B
(3 rows)

So yes, more people wanted option C (status quo) and option B than option A, but option A wins. By a significant margin.

Time to notify the winner.

Website Pin Facebook Twitter Myspace Friendfeed Technorati Digg Google StumbleUpon Premium Responsive

8 thoughts on “Design contest results”

  1. hmmm, as far as i can see the .png files aren’t loaded correctly. And no matter what i tried i found no way to fix it. But as i can not reproduce the “error” on an other PC i won’t bother you with this. Thanks anyway!

  2. By the way, did you randomize the order of the options when they were presented for voting? Otherwise I would expect option A to get the most votes, even if all three options were identical.

  3. Well, option A was visibly slower and more resource intensive (scrolling etc) in Opera 9.2 on WXP than the others. :-((

    Besides that I think option B is visually better and more accessible, that is to say you find what you look for more easily, information just stand out more. IMHO.

Leave a Comment

Scroll to Top