This experiment studies voting in an election with a top-two primary and multiple candidates. You will be paid for your participation on the basis of the decisions you make. If you are careful and make good decisions, you can make a substantial amount of money
In this experiment, there are multiple candidates (between 2 and 10), and you are a voter. You will participate in a number of periods, each consisting of an information buying phase, a straw vote, a primary election, a second information buying phase, and a general election.
In each period the candidates will take positions on a number line that goes from 0 to 100. Before each primary election all voters will have the opportunity to buy information about the candidates. Voters will then vote. Then, the top two primary outcomes will be announced to the group, where only the two candidates who receive the most votes advance. Voters will then have another opportunity to buy information about the candidates. Then, all voters will vote for a single candidate in the election and the candidate who gets the most votes will be considered the winner. At this point the candidates may move to new locations on the number line and the process will repeat.
Voters are paid for their participation on the basis of their payoff chart. Please look at the page that displays the payoff chart. The line is simply the set of all numbers between 0 and 100. The experimenters will select candidate positions on the line for each period. Candidates are equally likely to be at the end of the line as they are to be at the middle of the line. Each voter will be paid based on the position of the winning candidate on her payoff chart. For example, suppose candidate A is located at position 20, candidate B is located at position 25, and candidate B wins the election. Then you would earn 400 francs for that period. Note that on the sample payoff the maximum payoff is at position 45.
In the actual experiment, the payoff charts for each voter may be different. Each voter will have a maximum payoff and payoffs decrease symmetrically as candidate positions move away from the maximum in either direction as in the sample chart. However, different voters' maximums may be at different points on the number line, and their payoffs may decrease at different rates.
In the experiment the positions of the candidates will not be made public. You will be able to purchase a limited amount of information about positions of the candidates and the other voters based upon your budget. If you choose not to purchase information, your remaining budget will be converted into francs and increase your actual payment from the experiment. On your screen you will know the distribution from which all voters have drawn their maximum payoff. This line is in red. Higher lines indicate more voters will have their maximum payoff drawn from this area of the line, while lower lines indicate that fewer voters will have their maximum payoff drawn from this area of the line. Each experimental round the distribution of voters may change.
Each candidate line describes where the candidate is most likely to appear on the number line, with the height of the line describing where they are more likely to be. Initially, all of the candidate lines are evenly distributed because there is an equally likely chance that they are at any position. As you purchase more information about the candidate, their line gets updated to reflect a more accurate picture of where the candidate lies on the line. You should note that the lines only reflect the probability of where the candidate is with limited information and thus the highest point on the line is not necessarily where the candidate's true point is.
The sequence of events will be as follows: At the start of the experiment you may buy information about any candidate. Information does not always cost the same: candidates who have the same party as you will have cheaper information. You are not required to buy any information. You will then participate in a straw poll where all subjects are asked to identify for which candidate they are most likely to vote. After these results are revealed, you may then buy more information. You will then vote in a top two primary and the results will be announced. Yet again, you may then buy information but this time solely about either of the top two candidates. You will then vote in a general election. You will be paid based upon your remaining budget (if any) and the distance between the candidate wins the final election and your own maximum. After the last period the experiment will end. At this point, voters will be paid the sum of their payoffs. Your monetary payoffs are increasing in the number of francs you earn.
Please do not talk with other participants during the experiment. Are there any questions?