Competition Entries

[Waaii]

If I collect tokens, or answers, to send in by post for a prize and I also answer by passwords, or whatever, online ,how are these and all the others amalgamated to choose a winner assuming that there is only one prize and all entries must legally have an equal chance of winning? How does this work?

Waaii Fri 18/08/06 09:19

[Buenchico]

Where competition entries come in from various sources, there are basically two ways of giving all of the entries an equal chance.

The first is to convert all entries to the same format (e.g. by printing out e-mail entries to postcard size so that they can be placed in the same sack as all of the entries on postcards).

The second is by assigning probabilities based upon the number of each type of entry. Let's suppose that there are 1000 entries for a competition, 200 come through method A, 650 come through method B and 150 come through method C.

It's a very simple matter to program a computer to select a random number between 1 and 1000. If the number chosen is between 1 and 200 a winner is then selected, at random, from the 'A' group of answers. If the number selected is between 201 and 850, a random selection is made from group 'B'. If the number is in the range 851 to 1000, an entry is selected at random from group 'C'.

That ensures that all entries have an equal probability of being selected.

Chris

[Waaii]

I have to discount the first suggestion as being too time consuming.I had thought of this but the envelopes with the tokens would be too different I feel, might work for purely postcard entries. Would incorrect entries be screened out?
I guess not as they could not be similarly discarded in the sealed envelope entries.
Yet they are in some competitions as you can not submit wrong choices of daily password online, but you can send an envelope with an incorrect number of tokens,duplicated dates or not sufficient 'special'ones.
The second suggestion seems kinda logical, tho' I feel uneasy about it without really knowing why!
I think it has something to do with the fact that my chances must increase if my entry is sent in by the most popular format as then I must have a greater breadth of numbers to be randomly chosen from. (Don't I)
What happens if the one entry from the one chosen group is incorrect, does a second one come from the same group or does the whole process start at the beginning again. I am not totally convinced!

[Buenchico]

Your quote:
"I think it has something to do with the fact that my chances must increase if my entry is sent in by the most popular format as then I must have a greater breadth of numbers to be randomly chosen from. (Don't I)".

That's exactly what I was trying to illustrate. In my example, Method B was the most popular format so, when the computer selects a random number, there are 650 chances out of 1000 that the an entry from the 'B' group wiil be selected.

You ask about the problem of dealing with incorrect entries. There are two possible solutions. The first is that only correct entries are 'put into the pot'. The second is that, if an incorrect entry is drawn, the process is started again. (i.e. the computer is instructed to select another random number).

Having written that (and my previous reply), I ought to state this: My degree is in 'mathematics as a double subject' and one of my specialist fields was 'probability and statistics'. (So I ought to know the 'theory'). However, I live in the real world and know that most competition organisers aren't interested in 'fairness' or 'equal probabilities'. They probably select competition winners simply on the basis on the easiest way of finding a winner. For example, you might have several sacks containing thousands of postcard entries and a computer system holding details of thousands of e-mail entries. To get your company's software boffins to find a random way of selecting an e-mail entry might take time and money. To shove your hand into a sack will take hardly any time at all and cost nothing. So what do you do?

As I've said, I live in the real world. If you believe that all competitions are fair then, with no offence intended, I have to suggest that you don't.

Chris

[Waaii]

Yes I do follow the reasoning. I think to work out a fair ratio of entries it must be correct to include all wrong AND all correct entries which can not be done by the present methods. Of course I am a realist too, so I just have to figure out whether more people will go online with a daily password than will bother to collect coupons! In fact I do both to 'hedge my bets.' Thanks for an interesting informed discussion. (Makes a change from some of the sillies!)
Chris from Chris.