Don

1
Dont Panic!

• A Hitchhikers Guide to Computerized Economic
  Teaching Experiments

2

• by
• Dieter Balkenborg
• Todd Kaplan
• Tim Miller

3
Why Experiments?

• They are a fun way to learn and to teach!
• Complements standard methods
• Helps students with low math skills
• Enhances credibility of our science
• Puts us on the test

4
Why Using Computerized Experiments

• Free ride on existing resources
• Little preparation
• speedy
• Automatic tabulation of results
• Some are difficult to do hand-run
• Limitations
• Class size
• Computing facilities
• Time constraints

5
FEELE veconlab, econport

• Three new names to learn
• Easy to google for
• We will spend some time on each site and learn
  how to use them
• We start with a visit to
• http//en.wikiversity.org/wiki/Economic_Classroom_
  Experiments
• Please add your experience!

6
On our wikiversity site

• 1 Why use experiments in teaching economics
• 2 List of Experiments
• 3 Hints for running economic classroom
  experiments
• 4 Useful Links and Related Literature
• Now, lets google for FEELE

7
FEELE Laboratory

• is mainly a research lab so look for
• Access our Website of Teaching Experiments
• View list of experiments (short descriptions)
• Experimenter access (thats soon you)
• Participant access (enter now)

8
You are invited to participate in my experiment!

• Experiments are an experience good!
• Enter participant access
• Register for my experiment (no login yet)
• You need the access code
• dgb1-dee
• which identifies my experiment
• Read the instructions
• Good luck!

9
Sample result Bertrand
Two Firms Fixed Partners
Five Firms Random Partners
Two Firms Random Partners
I learnt that collusion can take place in a
competitive market even without any actual
meeting taking place between the two parties.
Some people are undercutting bastards!!!
Seriously though, it was interesting to see how
the theory is shown in practise.
10
Now register as experimenter

• Code word deadsea
• Next time login with your assigned password (your
  initials number) and password (4-digit number
  e-mailed to you, KEEP ACCESSIBLE)

11
Set up experiment

• New, different suffix for each session you run,
  your username suffix access code for students
  
• Name Bertrand complements, DEE
• Select type of experiment from drop-down box
  (Bertrand)

12
Add session

• Press add session
• View instructions
• Configure initially make minimal adjustments
  complementary, number of subjects, number of
  rounds, number of players, fixed partners
• View instructions
• (Run session, View and save results)

13
Veconlab (Charlie Holt)

• FEELE copies this design
• Similar registration procedure and setting up of
  experiment
• One website for students, one for participants
• Dont miss the bubble experiment!!!!!

14
(No Transcript)
15
Testing a new experiment

• I demonstrate here how I can learn about a new
  experiment by
• Setting it up with a minimal number of
  participants
• Open a new window for each participant and
  register as participant
• Play against myself by switching between the open
  windows (use ALT-TAB to switch)
• See the results in the instructors window

16
Charlie HoltU. of Virginia

• Creator of Veconlab
• Has written textbook on teaching experiments.
• Book explains both handrun and computerized

17
EconPort (James Cox)

• The Best for
• Vernon Smiths double auction (graphs! Adding
  excise taxes!)
• Charlie Plotts asset market
• General normal form games
• Extensive games
• Also handbook, resources

18
How to use EconPort 1

• Java must be installed, testing necessary!
• First Configurations ADD A NEW GROUP (No
  changes in group public configurations)
• DO NOT CREATE NEW CONFIGURATION YOURSELF!!

19
How to use EconPort 2

1. Now go via tab to Experiments
2. Choose documentation, e.g. on marketlink or check
   the teaching module sections.
3. In the documentation you find standard
   configurations, e.g. for excise taxes in a
   competitive market, which you can add to your
   newly created configuration group. This
   configuration can then be edited.
4. Create a new group in your experiments
5. There you add the configuration as often as you
   want to run a new session. Per group you can only
   have one experiment at a time running.
6. It is a great software!!!

20
Thanks

• Thank you for your attendance, I hope we got you
  interested
• For short questions you can e-mail me on
  d.g.balkenborg_at_ex.ac.uk (have teaching experiment
  in the title!)
• Possibility of hands-on sessions in UK
  universities
• Thanks to the economic network for continuous
  support Charlie Holt James Cox