PCC Project Report

Fall 2000

1. Current Project Status 
2. Suggestions for continuation of the project 
3. Related Links 
4. Other information

1.    Current Project Status

After recruiting at PCC for several times, we were able to get around 40+ people registered in our database. Our first experiment was conducted on Monday 9pm, 11/20/00. All of the communications, including our announcement of the experiment, were on email bases. Only one people from PCC successfully completed the whole experiment, although 3-4 PCC students did show up and attempt to get into our system. We had several discoveries after our first try:

• The show-up rate of PCC students tends to the very low (at that time, there were around 40 people in the database.)

• Our market experiment instruction did not appear to be intuitive enough for these PCC students. The behavior of the PCC student  participated has demonstrated it.

•   There are some technical difficulties involved. One PCC students wrote to us after the exam claiming s/he was not able to find the order form. Very likely, s/he was using a small screen.

To identify the problems, we conducted an email survey and gave a monetary offer based on how fast people relied to our survey. We had 28 PCC students replied to our email. This indicates around 70% of the PCC students in our database check their emails at least once a week (also indicates this group of people have a more reliable Internet access or they have more interests on online surfing, etc.)

We also did a major revision to the instruction page. Faria added some screen shots to make the explanation much more intuitive and shorter. We then conducted our second experiment on Wednesday 9pm, 11/29/00. This time, we announced out experiment around one week before the experiment date. In order to get more people show up, we also offered a 5-dollar show-up bonus. Within one hour before the experiment started, we had 6 PCC students showed up. However, due to server problem, the experiment was cancelled. We could expect 1-2 more participants showing up after 9pm.

During the preparation of the paychecks, we also found many PCC students did not enter their complete information (such as city and zip code).  We have notified them to complete their records.

2.    Suggestions for continuation of the project

We should be aware of the followings when conducting the next round experiments or recruiting for more PCC students:

• We should always ask them to enter their full information in our database.

• We should announce the experiment well ahead of the time to allow sufficient time for them to check their emails and remind them again 3-4 days before the experiment.

• PCC students seem unwilling to reply to our emails. It is in general very difficult to get responses without any special offers.

• Try not to use technical phrases when communicating with PCC students. And also try to make emails simple and short. 

• We should test our server ahead of the time. Probably we should avoid running several experiments at the same time.

• There is a possibility to include Whitman student and students from Germany into this project. But make sure you take time difference into account. Prof. Plott will have the information/contacts. 

Also, as last time we told them the 5$ bonus for showing up would only be offered for our second experiment, we need to come up with some other ideas to get more people show up. (We can try some drawings from all the participants and give that person a big prize, etc.)

We have established some good relationship with Econ and CS teachers at PCC. Since we only recruited one CS classes, we should consider get more subjects from other CS classes. 

In the experiments we had so far, we designed markets with basic demand-supply relationship. Since only one PCC student participated, we can reuse the previous design for the next experiment.

3. Related Links

1. The main Marketscape instructions and help site.

2. Instructions for how to set up a MarketScape Experiment

3. Instructions for how to run Marketscape Experiments

4. A modified version of the experiment Instructions

5. A map of the experiment directory from11/29/2000 experiment (The website for this experiment is http://eeps4.caltech.edu/market-001129/)
6. A description of one of our recruiting experiences 

4. Other Information

The below are some email communications/handouts we had with PCC students. 

1. A handout we used when doing recruiting in PCC classes. (in word format)

2. An email we sent out right after we did recruiting in PCC classes. 

Date: Mon, 13 Nov 100 10:41:50 -0800 (PST)
From: Laboratory for Experimental Economics and Political Sciences <eeps>
To: eeps@hss.caltech.edu
Subject: On-line Market Experiment

Hi:
This is a follow-up to the announcement that was given in your economics class this morning. Be sure to put your name in the database at http://eeps.caltech.edu/register  

We are thinking about running an experiment this Wednesday evening either from 7:30 to 9:00 or from 9:00 to 10:30 or so. Are those times good? Can you make other suggestions?

Thanks

P.S. When attempting to register you might receive an error message telling you that the JavaScripts on your computer are not enabled. This is easy to correct. Below we outline an example of how to correct this for one version of Netscape.

3. An email we sent out to these people unregistered and also to postpone our experiment:

Subject: market experiment

Hi,

Our system indicates that your registration is not completed as of this afternoon (11/14/00). Although money is involved, we are having trouble getting people registered. As a result we will not conduct an experiment tomorrow (Wednesday, November 15). This letter is to nudge people along and we will run experiments as soon as possible. Please register by going to: http://eeps.caltech.edu

Thanks
eeps

4. Our first experiment announcement email:

 
5. An email survey we designed after the first experiment and summary of the survey result:

The main purpose of the survey is to get a sense of how often people check their emails. A total of 28 PCC students replied to our email survey. A summary of the survey result is listed in the second column. (please note for Q3, some people made multiple selections.) We did not schedule our second experiment based on the survey result because the survey ended on 11/26/00 and we had to make an announcement well ahead of the experiment date (11/29/00). The result will certainly be useful for scheduling future experiment. 

Subject: Caltech On-line Market Experiment

We are attempting to find ways to make participation in the on-line market experiments easier. The following questions will help us schedule notices and the timing of the experiments. If you answer this by midnight Wednesday Nov. 22nd we will send you a check for $10. Later answers will be paid less according to the schedule at the bottom.

Simply reply to this email and mark an X beside the appropriate answers to the questions below (and include your name).

1. Approximately, how often do you check your email?

2. What time would be most convenient for you to participate in an hour-long on-line market experiment?

3. What day of the week is most convenient for you to participate in an on-line market experiment?

4.. Your name:

If you respond to this email you will be paid according to the following schedule:

$10 if your response is received by 11:59 p.m. Nov. 22nd
$8 if your response is received by 11:59 p.m. Nov. 24th
$4 if your response is received by 11:59 p.m. Nov. 26th

In order for us to mail you a check you need to be registered in our database. (We need your address and social security number for payment.)
If you have not already done so, please register at http://eeps.caltech.edu/register

Thank you for expressing your interest in our On-line Market Experiments.

Sincerely,
Market Experiment Team
Email address: eeps@hss.caltech.edu

Summary of Result from Q2 and Q3 if we combine them:

5. Our second experiment announcement email:

We are conducting an experiment (about 1 hour) on Wednesday, November 29 starting at 9 PM. This is an Internet experiment so you can participate from home. You will be paid based on the amount you earn during the experiment plus you will receive a $5 payment if you participate actively during the experiment. The website for this experiment is http://eeps4.caltech.edu/market-001129/

If you go to this website before Wednesday at 9 p.m. you can read the experiment instructions. To see what the experiment looks like you can log in using id = 1 and password = a (you cannot use this password for the actual experiment). The better you understand the instructions, the more money you can make!

You can get your personal id and password about an hour before the start (you should write these down). Do not get an id and password unless you can participate for the whole experiment.

6. An email we sent out to these who did not provide complete information in our database

Subject: We need your complete address to send check

Thank your for responding to our survey and registering in our database. We noticed that your city and postal code are missing in the address field. In order to send your check we will need this information.

Please go to http://eeps.caltech.edu/register today enter your social security number and then click change and add your city and postal code in the mailing address field.

If you have problems, please email your full address to eeps@hss.caltech.edu.

We would appreciate if you could provide this information before December 10th so that we can send your checks before Christmas. The sooner you provide this information the sooner you can get your check. 

Thanks!
Market Experiment Team.

Initial draft: 12/09/00
Revised: 12/10/00