• Matchings: The experiment consists of a series of rounds. In each round, you will be matched with the same group of 3 other people. The decisions that you and the other people in your group make will determine the amounts earned by each of you.
  
  
• Investments: You begin each round with a number of "tokens," which may either be kept or invested. At the same time, the 3 people you are matched with will decide how many of their tokens to keep, and how many to invest. Neither of you will be able to see the other's decision until after your decision is submitted.
  
  
• Earnings: The payoff to you will equal:
  $0.10 for each token you keep,
  $0.08 for each token you invest, and
  $0.08 for each token invested by the 3 other people who you are matched with.
  
  
• Subsequent Matchings: The groups of 4 people will be the same in all subsequent rounds, so the 3 other people you are matched with in one round are the same people that you are matched with in the next round.

  ---

  Vecon Lab - August 26, 2005

• There will be 5 rounds, and in all rounds you will begin with a new endowment of 25 tokens, each of which can either be kept or invested. The 3 other people in your group will also have 25 tokens.
  
  
• Everybody earns money in the same manner: $0.10 for each token kept, $0.08 for each token invested, and $0.08 for each token invested by the 3 other people.
  
  
• At the start of a new round, you will be given a new endowment of 25 tokens. Your are free to change the numbers of tokens kept and invested from round to round.
  
  
• Note: You will be matched with the same people in all rounds.

• You will be matched with the same group of 3 other people in each round. There will be a total of 5 rounds in this part of the experiment.
  
  
• All people will begin with 25 tokens which they may keep (and earn $0.10 each) or invest (and earn $0.08 each), knowing that they will also earn $0.08 for each token invested by other people in the group. You will begin each round with a new endowment of 25 tokens, irrespective of how many tokens you may have kept or invested in previous rounds.
  
  
• There will be a total of 5 rounds in this part of the experiment. Your earnings for each round will be calculated for you and added to previous earnings, as will be shown in the total earnings column of the record form that you will see next.
  
  

*************************** NEXT SCREEN ********************************

*************************** NEXT SCREEN ********************************

Overview: This program sets up an experiment in which each person is given an endowment of tokens that may either be kept or contributed to a public good. The value of a token kept is typically specified to be greater than the "internal return" that the individual receives from making the contribution, so there is a private incentive not to contribute, the strength of which depends on the size of the internal return. However, each token contributed produces an "external return" to the others in the group, and contributions are socially optimal in the sense that the sum of the external returns to others and one's own internal return is greater than the private value.

The public goods game has fascinated social scientists because the conflict between social and private incentives provides a platform for studying other-regarding preferences (e.g. altruism) or reciprocity, and how motives for contributions are sensitive to demographic and treatment variables. The fear that "free riding" behavior may generate inefficiently low levels of public good provision has been on economists' minds at least since Adam Smith's famous observations about street lights.

For a summary of experimental economics research, see Ledyard's survey in the Kagel and Roth (1995) Handbook of Experimental Economics, Princeton University Press. Experiments with differing internal and external returns are reported in Goeree, Holt, and Laury (2001) "Altruism and Noisy Behavior in One Shot Public Goods Experiments," forthcoming in the Journal of Public Economics.

Group Size, Matchings, Rounds, and Parameters: The total number of participants in the session is divided into groups of a size that you specify, and each group represents a separate game. Grouping may be fixed (same in all periods) or randomly reconfigured after each round. You also choose the number of rounds, the token endowment, and the internal and external returns for contributions to the public good. You also have the option of switching to a new set of payoff parameters after a specified number of rounds, so that a cohort of participants goes through two treatments. The instructions are automatically configured to conform to the parameters you select, and your instructor results page calculates data averages by round and by treatment. The best way to become acquainted with the various features of this program is to test it (see instructions on the next page).

Discussion: The discussion can focus on extent and trends in free riding behavior. Speculation about effects of changing group size and the economic incentives (the returns) can be answered by doing two treatments. It is also interesting to try to discern gender or other demographic effects, but experimental evidence on these things is mixed (see Ledyard, 1995). Finally, discussion can be directed to more practical matters, such as factors like "seed money gifts" that might increase alumni donations, etc.

Session Name: This is one of the Veconlab web-based experiment programs, and to use them you will need your personal "session name." If you do not already have one, it can be obtained quickly online from the Guide to Experimenters link on:

*************************** NEXT SCREEN ********************************

Experiment Setup: The next page in this sequence is the Set Parameters screen that lets you alter the default settings if you wish. Then you go to a Show Matchings screen that shows how players are matched with each other, by round. Note: You should never reload the Show Matchings screen during an experiment, since this will reconfigure the random matchings and erase existing data.

Tracking Results: After looking at the matchings, continue to the Admin Results page where you can watch subjects as they log in. When the Results page is refreshed, you will see any new results that have come in since the last time the page was loaded. This page can be saved, printed, or projected during the subsequent class discussion. You can also read instructions and send messages to participants from links on this page. At the bottom of the results page, there is a Create Data Table link that produces an unformatted block of data, with column headings, which can be copied and pasted into a spreadsheet program like Excel. If your connection is lost or you wish to watch results on a different computer from the one that was used to set up the experiment, you can use the PG View program to go directly to the results page for an ongoing experiment without erasing data.

Subject Login: The participants can begin the login phase after you finish the setup. They will need to use your session name when they log in from:

Recommended Testing: A simple (5-10 minute) test is essential to develop the confidence and experience needed for a smooth classroom implementation. To test this program, go ahead to the next pages to set up an experiment for only two participants (not necessary, but it speeds up the testing). Continue until you reach the results page with a message that "Only 0 of two participants have logged in." Then, open two new browser windows (on the same computer or on different computers), and go to the participant login page at the above address. After logging in each of the players using your session name, you can go through the instructions pages and make decisions for them, and watch the decisions and earnings show up once the Admin Results page is refreshed. From this page, you can also practice reading instructions, sending messages, and creating a data table that is copied and pasted into a spreadsheet.

• Please enter your preassigned Session Name below, and tell participants to enter the same Session Name on their login screens. Your Password, consisting of 1-4 letters or numbers (no symbols), will be required if you wish to use the PG View program (on the Admin menu) to view the results without clearing out the database. Your password can also be used in the PG Restart program for participants who have forgotten their passwords. To disable all password checks, just leave both password boxes empty. The Description box lets you enter a short text summary of the experimental conditions (university, date, class, nature of payoffs, etc.).
  
  

  Session Name:
  Admin Password (no symbols): (e.g. last four numbers of your SSN)
  Please reenter password:
  Optional Description:
  Number of Treatments:	1 2

  
  
  Note: The default two-treatment setup implements a decrease in the Marginal Per Capita Return from 0.8 to 0.3 with groups of size 4. (The internal and external returns are both equal to 0.8 in one case and 0.3 in the other, so group payoffs are maximized by full contribution in both cases.) An alternative would be to hold one of these returns constant and alter the other, or to hold all returns constant and alter the size of the group. With these defaults, you will have to set the number of participants at the top of the next page to be a multiple of 4, or of the other group size(s) that you select.
  

  ---

  Vecon Lab - August 25, 2005

*************************** NEXT SCREEN ********************************

*************************** NEXT SCREEN ********************************

*************************** NEXT SCREEN ********************************

*************************** NEXT SCREEN ********************************

*************************** NEXT SCREEN ********************************

• Session Name: You will need a "Session Name," which also serves as a personal data storage area. The Session Name is obtained by registering online:
  Obtain Session Name and Register
  or by emailing me at holt@virginia.edu Please include three initials, your full name, and your school affiliation with the request, so that we can provide a session name that is easy for you to remember.
  
  
• Session Name Use: When you set up an experiment, you must give your session name to the participants, so that their decisions go into the correct database for your experiment. The next time you use your session name, the data from the previous experiment will be erased. Thus you should not let other instructors use your session name. Naturally, you tell the students your session name, but if you don't call it a password there should be no problem with misuse. If you encounter problems, please notify me at holt@virginia.edu and I may add an additional security check. When you set up an experiment, you also have the option of choosing a short password that will prevent participants from viewing the results from the View programs on the Admin Menu. This password will also be needed if you wish to use the View programs to look at the results later from a different computer. I have a "master key," so email me if you forget the password you used on the first setup page.
  
  
• Data: Each experiment has an "admin" sequence of linked web pages for the experimenter (you). The final page in this sequence is a results page, which can be saved, projected, or printed like any HTML file. The results page will contain a Create Data Table button that produces a block of unformatted data, which can be imported into Excel or other programs for further analysis and storage. The results pages also have links to graphics displays that are configured to show results and theoretical predictions (in most cases). The data graph is labeled with the description (e.g. date, course, university) that you enter from the first setup page. Please remember to save or print data and graphs, since the next use of your session name will refresh the database and remove old data.

  Introduction, Session Names, Instructions, Data Displays, Demonstration, Large Classes, Microeconomic Principles, Acknowledgements, Main Menu

  ---

  Vecon Lab

*************************** NEXT SCREEN ********************************

*************************** NEXT SCREEN ********************************