The bias caused by order in this specific case is minor enough that it most likely will not be statistically significant due to the low sample size. There are some biases that this sample suffers from, but they make sense. It suffers from voluntary response bias, however we would only want those that care enough to vote to respond anyways. It will suffer from nonresponse bias, but that is unavoidable unless you want to track people down and show up at their house threatening them. This poll does not have response bias except for its ordering, and its ordering should only be of minimal influence.
so are you saying that to prevent bias in this poll we must go to every voter's house and threaten them with a weapon to have a full and complete sample?
The problem is that all of our voters are male, with the exception of one or two. We need more females. This leads me to believe that half of us should and must get a sex change so that the voting will be unbias.
Another problem is the age groups! Does the e-drama ever stop!? we need to wait another 50 years and hopefully get more younger members in PROS so that we can later divide the sample into male/female, and age groups. This will give us a more detailed analysis of who is voting and wherer their votes are going, thus we will plesase the majority.
I have also found another solution to the order of appearence bias. We must make people type the beginning of any name they want, and see if it is on the voting list. Hopefully if someone searches people, they are interested in people, so the People's Republic of Shenanigans will pop up first, followed by the second related search, followed by the third.
See....If only they let me and you run the poll, things would be a lot better.
And this is why, I conclude, that the loch-nedd monster is not real, for the tiny swedish town in which it was "pictured" was probably suffering from post world war economy crisis, and needed a touristic attraction to restore its shattered economy.
Poll