Karl-Franzens-University Graz

A Comprehensive Introduction to z-Tree

Stefan Palan
stefan.palan@uni-graz.at
http://www.palan.biz/academic

References

Fischbacher, U., z-Tree Tutorial, www.iew.uzh.ch/ztree/

ztree21tutorial.pdf, 3rd of January, 2002. (Tutorial)

Fischbacher, U., z-Tree Zurich Toolbox for Readymade Economic

Experiments - Experimenters Manual, Institute for Empirical
Research in Economics, University of Zurich, Working Paper No. 21,
ISSN 1424-0459, 1999. (Manual)

Fischbacher, U., z-Tree Reference Manual,

http://www.iew.uzh.ch/ztree/ztree21ref.pdf, 6th of January, 2006.
(Reference Manual)

Z-Tree Wiki, https://www.uzh.ch/iew/ztree/ssl-dir/wiki/, 26th of

March, 2012. (Wiki)

Lecture ressources: www.palan.biz/academic - select Downloads

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

2
Graz, SS 2012

A Comprehensive Introduction to z-Tree

Introduction

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

3
Graz, SS 2012

1. Introduction to z-Tree

Structuring a typical session 1/3

1. Subject arrival

Subjects checked on list

Excess subjects are sent home with show-up fee
Other subjects are (randomly) assigned to workstations
2. Instructions on mechanism

Hand out printed instructions, read them out loud

Control questions
3. Training round(s)

4. Bathroom break

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

4
Graz, SS 2012

1. Introduction to z-Tree

Structuring a typical session 2/3

5. Instructions on parameters

Hand out printed instructions, read them out loud

Control questions
6. Run treatment

Repeat 5 & 6 for multiple parameter constellations

7. Questionnaires

Solicit questionnaire responses

Solicit risk-aversion, etc.

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

5
Graz, SS 2012

1. Introduction to z-Tree

Structuring a typical session 3/3

8. Payment

Random number generation

As transparent as possible (e.g. real dice)

Random numbers for individual subjects reduce cost variance
but increase required time

Pay subjects individually and anonymously

Have subjects sign receipt
Ask subjects not to talk about experiment with others

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

6
Graz, SS 2012

1. Introduction to z-Tree

General rules

Follow session structure cheat sheet diligently

Contains procedural steps

Contains program settings
Contains required material (dice, cards, etc.)

Note special occurrences in experimenter diary

Backup everything

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

7
Graz, SS 2012

1. Introduction to z-Tree

Characteristics and features

Zurich Toolbox for Readymade Economic Experiments

Simple, flexible programming language for economic

experiments

Client-server architecture

Automatic networking, data collection and payoff

calculation

Crash recovery capability

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

8
Graz, SS 2012

1. Introduction to z-Tree

Network topology

Source: Tutorial, p. 6.

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

9
Graz, SS 2012

1. Introduction to z-Tree

Network connections: Critical issues

Setup with file server

Ensure server has read/write access
Ensure client has read access
Put all files into one directory
Clients can be restarted by Run-Restart All Clients

Setup without file server

Ensure server has read/write access
Point client to server IP
Clients can only be restarted manually

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

10
Graz, SS 2012

1. Introduction to z-Tree

Network connections: Critical issues

Server IP (in server.eec)

Command line parameter, e.g.: zleaf /server 10.0.0.1
IP file is in the zleaf.exe-directory
IP file is in the directory c:\expecon\conf
IP equals local machines IP

Client name
Command line parameter, e.g.: zleaf /name pc1

File name.eec in zleaf.exe-directory

Name equals local machines TCP/IP hostname
A Comprehensive Introduction to z-Tree
Stefan Palan, Karl-Franzens-University Graz

11
Graz, SS 2012

1. Introduction to z-Tree

Other important command line options

zleaf.exe
Language: /language
e.g. /language en, or /language english
Screen resolution: /size
e.g. /size 1024x768

ztree.exe
Change default directories

/xlsdir, /sbjdir, /adrdir, /paydir, /gsfdir, /tempdir, /datadir

(autosaves of .ztt and .ztq), /leafdir (server.eec)

e.g. /adrdir c:\data

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

12
Graz, SS 2012

1. Introduction to z-Tree

Files used by z-Tree

ztree.exe

Server program file

zleaf.exe

Client program file

name.ztt

Treatment code file

name.ztq

Questionnaire code file

name.txt

Parameter input file

@1.ztt, @2.ztt,
@1.ztq, @2.ztq,

Backups of treatments and

questionnaires

@db.txt, @lastclt.txt, @prevdb.txt Temporary files

090330_0804.adr

Subject address file

090330_0804.gsf

Crash recovery file (binary)

090330_0804.pay

Payout information file

090330_0804.sbj

Questionnaire response file

090330_0804.xls

Main data file (~ MS Excel compatible)

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

13
Graz, SS 2012

1. Introduction to z-Tree

Example: Public goods experiment*

Groups of n = 4 subjects

Initial endowment of subject i: wi 20

Contribution of subject i: ci 0, wi

1 4
Profit of subject i: i wi ci 1.6 c j
n j 1
New z-Tree treatment:

* Taken from the z-Tree Tutorial, sections 2.2-2.3.

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

14
Graz, SS 2012

1. Introduction to z-Tree

Define parameters

Define general parameters in

Background:

Create new program

Select logfile in stage tree

Click Treatment-New Program

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

15
Graz, SS 2012

1. Introduction to z-Tree

The first stage

Add the first stage

Select the Background
Click Treatment-New Stage

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

16
Graz, SS 2012

1. Introduction to z-Tree

Designing the screen 1/2

Add the first box

Select Active Screen in the Contribution Entry stage
Click Treatment-New Box>Standard Box
Click OK

Add an output item

Select the new
Standard Box
Click TreatmentNew Item

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

17
Graz, SS 2012

1. Introduction to z-Tree

Designing the screen 2/2

Add an input item

Select the first item
Click TreatmentNew Item

Add a button
Select the input item
Click TreatmentNew Button
Enter OK as name and click OK

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

18
Graz, SS 2012

1. Introduction to z-Tree

Profit calculation

Add new stage Profit Display

Add profit calculation

Select Profit Display stage
Click Treatment-New Program

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

19
Graz, SS 2012

1. Introduction to z-Tree

Profit display

Add new box to Active screen

of Profit Display stage

Insert display items

Sum of all contributions

Subjects income for the
period

Insert Continue button

Save the treatment (.ztt file)

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

Order of execution

Own contribution

20
Graz, SS 2012

1. Introduction to z-Tree

Starting multiple z-leafs for testing

Either create shortcuts:

or use the following batch file:

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

21
Graz, SS 2012

1. Introduction to z-Tree

Advice for testing 1/2

Switch between leafs using <ALT>-<TAB>

Kill leafs using <ALT>-<F4>

Display connected clients by clicking Run-Clients Table

Pause time using <F12>

Resume time using <Shift>-<F12>

Stop testing by clicking Run-Stop after this period

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

22
Graz, SS 2012

1. Introduction to z-Tree

Advice for testing 2/2

To test at your leisure:

Make periods very short (e.g. 2 sec)

Start program (<F5>)

then immediately pause the time (<F12>)

Test at will

Resume/Pause to move to next period (<Shift>-<F12>)

Click Run-Stop after this period to initiate end of testing

End testing by clicking OK on all waiting screens

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

23
Graz, SS 2012

1. Introduction to z-Tree

Resulting screen output

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

24
Graz, SS 2012

1. Introduction to z-Tree

Resulting data output

Output in YYMMDD_hhmm.xls:

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

25
Graz, SS 2012

1. Introduction to z-Tree

Creating new questionnaire

Click File-New Questionnaire

Click QuestionnaireNew Address Form

Questions left empty will
not be asked
May be entirely empty
Address form plus empty
question form suffice to
complete questionnaire
Run questionnaire (may
be empty) at least once
to write payment file

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

26
Graz, SS 2012

1. Introduction to z-Tree

Payment file

Created after treatment and questionnaire have been run

Contains information from address form

Contains information from session table

Subject
1
2
3
4
Experiment

Computer
Interested
Name
Player1
OK
Jane Parker
Player2
OK
Jim Smith
Player3
OK
John Doe
Player4
OK
Bill Farmer
C:\Institut\zTree\120328_1127.pay

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

Profit
21.52
15.43
20.00
17.23
74.18

Signature

27
Graz, SS 2012

A Comprehensive Introduction to z-Tree

The Basics of z-Tree

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

28
Graz, SS 2012

2. Basics of z-Tree

Structure of a treatment

Background
Contains general settings
Defines the tables used in the treatment
Defines default Active Screen and Waiting Screen
Contains programs which run at the beginning of a period

Stages
Correspond to the screens displayed to subjects
Contain programs which run at the beginning of a stage

* Adapted from a z-Tree lecture by Maria Bigoni: http://www2.dse.unibo.it/bigoni/courses/ztree/Lecture_notes_handout.pdf, 31.08.2012,

accessed on 19.12.2012.

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

29
Graz, SS 2012

2. Basics of z-Tree

Creating a stage

Name is for documentation only; must be unique

Start

Timeout

Active screen/Waitingscreen/Use background/Header

Screen layout
Containers
Boxes
Absolute/relative size/position
Adjustment of the remaining box

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

30
Graz, SS 2012

2. Basics of z-Tree

Items: Text output

Result:

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

31
Graz, SS 2012

2. Basics of z-Tree

Items: Variable output

Result:

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

32
Graz, SS 2012

2. Basics of z-Tree

Items: Variable input

Result:

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

33
Graz, SS 2012

2. Basics of z-Tree

Screen layout

Screen layout can be controlled in box dialog:

See Demo_ScreenLayout.ztt for example layouts

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

34
Graz, SS 2012

2. Basics of z-Tree

Screen layout
1
4

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

35
Graz, SS 2012

Program a treatment which displays (approximately) the

following:

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

* Inspired by lecture slides by Verena Utikal, Friedrich-Alexander-University Erlangen-Nrnberg.

2. Basics of z-Tree

Exercise Screen layout*

36
Graz, SS 2012

A Comprehensive Introduction to z-Tree

Advanced Item Output

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

37
Graz, SS 2012

3. Advanced item output

Conditional output
Layout:
!text: 2="Less than the third.; 3="The
third.; 4="More than the third.;

Results:
Period 1+2:

Attention:
The first text version that is being
displayed sets the size of the item. If
the variable value (and accordingly
the text) changes, longer text may
not display correctly. Therefore, add
enough blank space after each
option to display even the longest.
Period 3:

Period 4+:
A Comprehensive Introduction to z-Tree
Stefan Palan, Karl-Franzens-University Graz

38
Graz, SS 2012

3. Advanced item output

Inserting variables into text

Layout:
<> !text: 1="This item displays the
content of the variable Period: <Period |
1> and Subject: <Subject | 1>.

Result:

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

39
Graz, SS 2012

3. Advanced item output

Formatted text
Layout:
!text: 1="{\rtf \fs22\b\i This item displays
formatted text.}"

Result:

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

40
Graz, SS 2012

3. Advanced item output

Variables embedded in colored text

Layout:
<> !text: 1="{\rtf {\colortbl;
\red255\green0\blue0;
\red0\green0\blue255;\red0\green0\blue
0;} \fs24 This is a \cf2 colorfully
formatted \cf3 output box displaying the
variables votes: \cf1 <\votes | 1>\cf3 ;
and NumSubjects: \cf1 <\NumSubjects |
1>\cf3 .}"

Result:

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

41
Graz, SS 2012

3. Advanced item output

Conditional formatting
Layout:
<> !text: 1="{\rtf \fs24 <\Period | !text:
1="\i " 2="\b ">Test}"

Result:
Period 1:

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

Period 2+:

42
Graz, SS 2012

3. Advanced item output

Rich text formatting

Switch
\tab
\par
\line
\bullet
\ql
\qr
\qc
\b
\b0
\i
\i0
\sub
\super
\strike
\ul
\ul0
\cfn
\fsn

Effect
tabulator
new paragraph
new line
thick dot
aligned to left
aligned to right
centered
bold
not bold
italic
not italic
small and inferior numbers (index)
small and superior numbers (exponent)
crossed through
underline
do not underline
Text color. n is the index of the color table which is defined by colortbl. See example.
Font size in units of half a dot. The font size must be explicitly given, otherwise it is larger
(24) than usual in z-Leaf.

Source: Reference manual, p. 56-57.

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

43
Graz, SS 2012

3. Advanced item output

Alternative formats

Source: Reference manual, p. 30.

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

44
Graz, SS 2012

4 Programming: Basics

General notes on formatting

Theoretically, any RTF code (e.g. saved in Word) should

display correctly

Variable refresh
Put text in layout (as opposed to label) section to
allow variables to refresh
Screen space is reserved for initial value of variables

Change font size generally through z-leaf command line

switch /fontsize

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

45
Graz, SS 2012

3. Advanced item output

Checking entries for correctness

Select a button
Click Treatment-New Checker

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

46
Graz, SS 2012

A Comprehensive Introduction to z-Tree

Programming basics

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

47
Graz, SS 2012

4 Programming: Basics

Examples of simple functions

result = if ( k < 5 | k >= 10, 1, 10);

result = abs ( c + pi() );

result = round ( a, 0.2);

result = roundup ( random() * 5, 1 );

result = exp ( randomgauss() );

result = sqrt (b ^ 2);

result = max ( ln ( x ), log ( y ) );

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

48
Graz, SS 2012

4 Programming: Basics

Variable types

Floating point variables

Standard variable type
Saved in binary format (minimal deviations from
decimal format)
Array variables
Used to save (infrequently) recurring values
Array size must be fixed in background (cannot be
calculated)
array a[3];
array a[1,5,2];
a[2]=10;

String variables
Do not exist
(There is a limited possible workaround)

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

49
Graz, SS 2012

4 Programming: Basics

Conditional execution in z-Tree

If-statement for value assignment:

result = if ( k < 5 | k >= 10, 1, 10);

If statement for code execution:

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

50
Graz, SS 2012

4 Programming: Basics

Controlling stage flow

Participate variable
Variable in the subjects table
Set to 0 to make a subject skip this stage (must be
done in the stage background)

LeaveStage variable
Variable in the subjects table
Set to 1 to make a subject move immediately to the
waiting screen of the current stage

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

51
Graz, SS 2012

4 Programming: Basics

Defining parameters in the Background

Put general parameters into program in globals table in

Background, e.g.:

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

52
Graz, SS 2012

4 Programming: Basics

Defining parameters in the Background

Put general parameters into program in globals table in

Background, e.g.:

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

53
Graz, SS 2012

4 Programming: Basics

Defining switches in the Background

Use variables to switch between test modes, and

between different treatments within one .ztt file:

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

54
Graz, SS 2012

4 Programming: Basics

Defining switches in the Background

Use variables to switch between test modes, and

between different treatments within one .ztt file:

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

55
Graz, SS 2012

4 Programming: Basics

Calculation of payouts: Background

# practice periods

# paying periods

Value in real currency of 1 unit of

the experimental currency (ECU)

Lump sum payment

Periods with profit

Exchange rate

Periods without effect on profit

Value in ECU added to TotalProfit at the beginning of period 1 of

the treatment

Show up fee

Value in real currency added to subject payout at the beginning

of period 1 of the session

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

56
Graz, SS 2012

4 Programming: Basics

Calculation of payouts: Bankruptcy

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

57
Graz, SS 2012

4 Programming: Basics

Calculation of payouts: Profit/TotalProfit

Table: subjects
Profit

Contains profit in a period, needs to be calculated

TotalProfit

Contains total profit up to period t, calculated automatically as the sum

of the variables Profit of periods 1 to (t 1).

Table: session
FinalProfit

MoneyAdded

Subjects profit excluding the show-up fee

Money added to a subject who faced bankruptcy but was allowed to

continue

MoneyToPay

Equals FinalProfit plus ShowUpFee plus MoneyAdded

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

58
Graz, SS 2012

4 Programming: Basics

Calculation of payouts: Good advice

Set exchange rate in Background to 1

Create variables in table globals.

exchangerate: The exchange rate in real CU/ECU

currency: A dummy variable for currency, e.g. 1 for EUR, 2 for USD

Only one place for

changing these
parameters (in the
program in the
globals table)!

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

59
Graz, SS 2012

4 Programming: Basics

Programs, tables and scope operators

Programs run in a
table
Programs can contain
commands running in
different table
Structure here is:
globals

subjects
contracts

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

Scope operators:
: one step up
\ all steps up to
globals table

60
Graz, SS 2012

4 Programming: Basics

Table functions

Can be preceded by table name, e.g. subjects.sum()

Examples:
sum ( [condition] , variable );
average ( [condition] , variable );
product ( [condition] , variable );
minimum / median / maximum ( [condition] , variable );
find ( [condition] , variable );
count ( [condition] );
regressionslope ( [condition] , x , y );
stddev ( [condition] , variable );

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

61
Graz, SS 2012

4 Programming: Basics

Programs, tables and scope operators

Programs run for all
records in a table
(unless restricted
by conditions)

Variables in
table globals

Command run in
table subjects

Variable in
table subjects

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

62
Graz, SS 2012

4 Programming: Basics

Programs, tables and scope operators

Example: (Program runs in table A)

Table A

Table B

Table C

x = v + B.sum ( v * :v - C.product ( v - :v - ::v) )

A A
B A
C B A

Function same() as a special case

If variable only defined in one table, scope

operator may be omitted (dangerous!)

Last periods tables: OLDsubjects, etc.

Source of the form of presentation: Esarey, J., zTree Workshop: Fundamentals of zTree, http://userwww.service.emory.edu/~jesarey/
ztree.ppt, 2005, p. 81.

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

63
Graz, SS 2012

Source: Reference manual, p. 59.

4 Programming: Basics

The complete scope reference

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

64
Graz, SS 2012

4 Programming: Basics

Loops in z-Tree

Loops through the records of a table:

contracts.do { player = :Subject; }

Loops using while ( ):

while ( i <= 5 ) { i = i + 1; }

Loops using repeat { } while ( ):

repeat { i = i + 1; } while ( i <= 5 );

Loops using iterator( ).do:

iterator( i, 8, 12, 2 ).do { :player = i; }

Loops in time using later ( ) do/repeat { }:

later ( 15 ) do OR repeat { }

While and repeat loops can be left with <CTRL>-<ALT>-<F5>

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

65
Graz, SS 2012

A Comprehensive Introduction to z-Tree

Box types

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

66
Graz, SS 2012

5. Box types

Box types
Container box:
Used to organize other boxes on the screen

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

67
Graz, SS 2012

5. Box types

Box types
Contract list box:
Used to display records from a z-Tree table:
Condition for
displaying box
Makes own
records font
color blue
Condition for
displaying
individual
records
Variables to
sort by

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

68
Graz, SS 2012

5. Box types

Box types
Contract list box:
Code:

Output:

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

69
Graz, SS 2012

5. Box types

Box types
Contract list box:
Contracts can be selected
Button in contract list box:
Program is executed for selected contract
Error message if no contract selected
Current subjects contract in subjects table
can be accessed using scope operator, e.g.
Creator = :Subject;

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

70
Graz, SS 2012

5. Box types

Box types
Contract creation box:
Used to create new records in a z-Tree table:
Condition for
displaying box
Table wherein new
records should be
created
Additional settings for
contract creation
Number of contracts
which can be entered
simultaneously

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

71
Graz, SS 2012

5. Box types

Box types
Contract creation box:
Code:

Output:

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

72
Graz, SS 2012

5. Box types

Box types
Contract creation box:
Item entry limitations determine what values are
permissible
Program in button
Used to write additional variables
Automatically limits its operation to new record

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

73
Graz, SS 2012

5. Box types

Box types
Grid box:
Used to display records from a z-Tree table:
Condition for
displaying box
Defines size of
the box
Determines
how the
ordering of the
items is
interpreted
Formatting
options

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

74
Graz, SS 2012

5. Box types

Box types
Grid box:
Code:

Output:

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

Attention:
Drawing a grid box with
many items takes z-Tree
a long time. Whenever
possible, use e.g. a
contract list box instead.

75
Graz, SS 2012

5. Box types

Calculator button box

Calculator buttonbox:
Gives subjects access to the Windows calculator

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

76
Graz, SS 2012

5. Box types

Box types
Chat box:
Used to display/exchange chat messages:
Condition for
displaying box
Variable name
and entry
limitations
Condition for
displaying
individual
records
Output text
format

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

77
Graz, SS 2012

5. Box types

Box types
Chat box:
Code:

Output:

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

78
Graz, SS 2012

A Comprehensive Introduction to z-Tree

Programming

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

79
Graz, SS 2012

6.1 Programming: Program execution

Running programs

Program in the background

Runs after the setting of standard variables

Program in a stage
Runs at beginning of stage, before checking Participate

Program in a button
Program in subjects table runs only for the subject
pressing the button
Runs after checkers

Program started by later ( ) do/repeat { } command

(Re-)Runs after pre-specified time

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

80
Graz, SS 2012

6.1 Programming: Program execution

Parameter table

1. Specific parameters
(after programs in the background)

2. Role parameters

3. Period parameters & prompt

A Comprehensive Introduction to z-Tree
Stefan Palan, Karl-Franzens-University Graz

81
Graz, SS 2012

6.1 Programming: Program execution

Parameter table variable import

Variables can be imported from ASCII file (one-by-one)

Allows parameter customization before session

Tab-separated values
A Comprehensive Introduction to z-Tree
Stefan Palan, Karl-Franzens-University Graz

82
Graz, SS 2012

6.1 Programming: Program execution

Period structure
1. Setting of standard variables

2. Programs in the Background

3. Specific parameter programs
4. Role parameter programs
5. Period parameter programs
6. Programs at the beginning of a stage

7. Programs in buttons (when clicked)

Delayed sub-programs: later ( ) do/repeat { }

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

83
Graz, SS 2012

6.2 Programming: Advanced programming techniques

Data import and export 1/2

Tables can be imported and exported from/to ASCII file

New (Version 3.3.0) stage tree elements

Table dumper:
Allows exporting
table to ASCII
file
Table loader:
Allows importing
table from ASCII
file (append or
replace)

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

84
Graz, SS 2012

6.2 Programming: Advanced programming techniques

Data import and export 2/2

Demo_DataFileRW.ztt

Table dumper:
parameters
parameters
parameters
parameters

Period
1
1
1

x
68
3
94

Table loader (replace):

Replaces table records from
first downwards with data in
ASCII file

Table loader (append):

Adds new table records after last one

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

85
Graz, SS 2012

6.2 Programming: Advanced programming techniques

z-Tree programming with Excel

Use MS Excel to generate repetitive code:

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

86
Graz, SS 2012

6.2 Programming: Advanced programming techniques

z-Tree programming with Excel

Complex code generation:

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

87
Graz, SS 2012

6.3 Programming: Miscellaneous

Groups

Groups are defined in Specific parameters in Parameter

table

Groups can be modified using variable Group in subjects

table

Automatic group matching:

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

88
Graz, SS 2012

6.3 Programming: Miscellaneous

Group matching

Select desired periods in

parameter table:

Choose matching procedure from Treatment menu:

Partner:

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

Absolute stranger:

89
Graz, SS 2012

6.3 Programming: Miscellaneous

Calculating rank in z-Tree (from the Wiki)

Simple rank calculation (grouping is optional):

rank_low = count (same (Group) & value > :value) + 1;
rank_high = count (same (Group) & value >= :value);

Rank calculation excluding

ties (assuming value is
integer):

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

90
Graz, SS 2012

6.3 Programming: Miscellaneous

Sorting values in z-Tree (from the Wiki)

Assume there are

integers v1 to v5 in the
globals table

Strategy:
Put data into a table
Calculate the rank
Transfer the data
back

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

91
Graz, SS 2012

6.3 Programming: Miscellaneous

Sorting values in z-Tree

Example: Call Auction

Program in globals table aggregates offers in contracts table into
price/volume list - sorted ascending in price - in pricelist table:
if (contracts.count(p>0)>0) {
//Writes minimum and maximum price into variables in the globals table
minprice=contracts.minimum(same(Period),p);
maxprice=contracts.maximum(same(Period),p);
//Sets variable price equal to the lowest price in this period
price=minprice;
//Loops through all prices offered in this period and writes possible purchase and sales volume
//into pricelist table
n=1;
repeat {
//Creates new entry in pricelist table
pricelist.new {
//Entry consists of price, volume that could be sold at this price, volume that could be bought
//at this price, the total possible transaction volume at this price, and a counter variable
p=:price;
sellvol=contracts.sum(same(Period)&p<=:p&q<0,-q);
buyvol=contracts.sum(same(Period)&p>=:p&q>0,q);
vol=min(sellvol,buyvol);
n=:n;}
n=n+1;
//Increments the price to the next offered price or to a price>maxprice if maxprice has been reached
if (price<maxprice) {price=contracts.minimum(same(Period)&p >\price,p);} else {price=maxprice+1;} } while (price<=maxprice);
//Sets price equal to the mean of the prices which produce maximum volume
price=pricelist.average(same(Period)&vol==pricelist.maximum(same(Period),vol),p);}

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

92
Graz, SS 2012

6.3 Programming: Miscellaneous

Markets and auctions: contracts table*

Flexible number of records

Adding records
Contract creation boxes
Command: contracts.new { x=1; }

Changing records
In contract list boxes and contract grid boxes
Command: contracts.do { x=2; }

Deleting records
Marking a contract as deleted (e.g. status = 3;)
* Adapted from lecture slides by Verena Utikal, Friedrich-Alexander-University Erlangen-Nrnberg.

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

93
Graz, SS 2012

6.3 Programming: Miscellaneous

Markets and auctions: contracts table*

Buttons in contract creation box and contract list box

Can contain checkers (e.g. price * volume < cash)
Operate only on new/selected record
Access to subject who pressed the button using scope
operator

As if in contracts.new/do {}

E.g. set subject using: CreatorOrSelector = :Subject;

* Adapted from lecture slides by Verena Utikal, Friedrich-Alexander-University Erlangen-Nrnberg.

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

94
Graz, SS 2012

6.3 Programming: Miscellaneous

Markets and auctions: Full execution

Goal: Market for buyers and sellers of a uniform good

Every trader can buy and sell

Orders must be fully executed

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

95
Graz, SS 2012

6.3 Programming: Miscellaneous

Markets and auctions: Full execution

Example record
ID = 1;
price = 23;
volume = 5;
status = 0;
type = 0;

Variable status
0 open offer
1 traded
2 cancelled

3 invalidated
4 expired

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

buyer = 1;
seller = 0;
offertime = 2.381;
statustime = 18.221;

96
Graz, SS 2012

6.3 Programming: Miscellaneous

Markets and auctions: Full execution

Variable type
1 buy offer
1 sell offer

Open buy offer characterized by:

type == 1 & status == 0

Open sell offer characterized by:

type == -1 & status == 0

Transaction characterized by:

status == 1

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

97
Graz, SS 2012

6.3 Programming: Miscellaneous

Markets and auctions: Full execution

Example records:
ID

price

volume

status

type

buyer

seller

offertime

statustime

23

-1

5.304

-77777

25

7.221

18.475

30

13

12.100

16.224

15

-1

25.132

33.251

Record 1: Outstanding sales offer from subject 3

Record 2: Subject 1 sold after a buy offer from subject 2

Record 3: Cancelled buy offer from subject 1

Record 4: Invalidated sell offer from subject 5

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

98
Graz, SS 2012

6.3 Programming: Miscellaneous

Markets and auctions: Full execution

Variable intialization:

Record starting time for later calculation of time elapsed

Set dummy number for variable initialization
Set endowment
Initialize variable for convenient calculation of trade number

Initialize variables

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

99
Graz, SS 2012

6.3 Programming: Miscellaneous

Markets and auctions: Full execution

Order input

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

100
Graz, SS 2012

6.3 Programming: Miscellaneous

Markets and auctions: Full execution

Order input

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

101
Graz, SS 2012

6.3 Programming: Miscellaneous

Markets and auctions: Full execution

Order output

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

102
Graz, SS 2012

6.3 Programming: Miscellaneous

Markets and auctions: Full execution

Order output
//Processes trade
//Updates contract information
seller = :Subject;
status = 1;
statustime = gettime() - \Starttime;
//Updates cash and share holdings
subjects.do {
//Updates seller
if ( Subject == :seller ) {
NumTrades = NumTrades + 1;
Cash = Cash + :price;
Shares = Shares - :volume;
}
//Updates buyer
if ( Subject == :buyer ) {
NumTrades = NumTrades + 1;
Cash = Cash - :price;
Shares = Shares + :volume;
}
}

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

103
Graz, SS 2012

6.3 Programming: Miscellaneous

Markets and auctions: Full execution

Order output

//Invalidates illegal offers

//Loops over all contracts
contracts.do {
//Looks only at outstanding offers
if ( status == 0 ) {
//Only buy offers
if ( type == 1 ) {
//Invalidates buy offers if proposer has insufficient money
if ( price * volume > subjects.find ( Subject == :buyer , Cash ) ) {
status = 3;
statustime = gettime() - \Starttime;
}
}
//Only sell offers
if ( type == -1 ) {
//Invalidates sell offers if proposer has insufficient shares
if ( volume > subjects.find ( Subject == :seller , Shares ) ) {
status = 3;
statustime = gettime() - \Starttime;
}

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

104
Graz, SS 2012

6.3 Programming: Miscellaneous

Markets and auctions: Full execution

Order output

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

//Marks offer as cancelled

status = 2;
statustime = gettime() - \Starttime;

105
Graz, SS 2012

6.3 Programming: Miscellaneous

Markets and auctions: Full execution

Order output

Buy orders

Sell orders

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

106
Graz, SS 2012

11. Exercises

Joint exercise: Limiting order legality

Write down the condition for a checker which ensures
that the price of a new buy (sell) offer is lower (higher)
than the lowest existing sell (highest buy) offer price.

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

107
Graz, SS 2012

11. Exercises

Joint exercise: Limiting order legality*

Write down the code for a bid-ask improvement rule
(i.e. a new bid must be higher than the current highest
bid price and a new ask price must be lower than the
current lowest ask price.)

* Inspired by lecture slides by Verena Utikal, Friedrich-Alexander-University Erlangen-Nrnberg.

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

108
Graz, SS 2012

11. Exercises

Exercise Market with partial order execution

Write down how you would structure the tables and
variables of a market to fulfill the following criteria:

Subjects can offer to buy and sell multiple units of a

uniform good, entering a price and volume

Partial execution is possible

Only one Buy (Sell) button (with transaction warning)

No information is lost:
Who made the first offer, who accepted
Time where the offer and acceptance were made
Initial volume, how much was exchanged per transaction

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

109
Graz, SS 2012

6.3 Programming: Miscellaneous

Markets and auctions: Partial execution

Variables (partial execution possible)
Table offers
ID
execution (0,volume)
price
offerer
volume
offertime
type (-1,1)
Statustime
status (0,1,2,3,4)

Table transactions
offerID (ID)
acceptanceID (ID)
accepterID

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

volume
time
(price)
110
Graz, SS 2012

6.3 Programming: Miscellaneous

Markets and auctions: Partial execution

Create record in table offers for every order

Upon new order entry, check transactability

Create new record in table transactions for every

transaction induced by new order

Variable offerID in transactions contains ID of first

order

Variable acceptanceID in transactions contains ID of

order which led to transaction with previously submitted
order offerID

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

111
Graz, SS 2012

A Comprehensive Introduction to z-Tree

Questionnaires

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

112
Graz, SS 2012

7. Questionnaires

Questionnaires: Overview

Run after 1 treatment (sets number of subjects)

Adress form needed to write payment file (delete

components by deleting caption in adress form)

Question forms used for:

Solicitation of feedback
Display of results

Questionnaire may be empty

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

113
Graz, SS 2012

7. Questionnaires

Questionnaires: Overview

Rules set the regions where

labels and questions are
positioned

Maximum one button per

question form

Last question form may not contain a button

* Adapted from a z-Tree lecture by Maria Bigoni: http://www2.dse.unibo.it/bigoni/courses/ztree/Lecture_notes_handout.pdf, 31.08.2012,

accessed on 19.12.2012.

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

114
Graz, SS 2012

7. Questionnaires

Questionnaires: Using the session table

Conditioning on variables in session table (Participate)

Run treatment writing variables into session table

(FinalProfit, MoneyAdded, ShowUpFee, MoneyToPay and
MoneyEarned are always available)

Access variables in session table in question form

definition in questionnaire

Use Participate variable to control which subjects see

which parts of the questionnaire

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

115
Graz, SS 2012

7. Questionnaires

Questionnaires: Using the session table

Example: Use a variable x from the subjects table.

In the treatment create a program in the session table:

x = : x;

In the questionnaire, use a text like:

<> Displays x: < x | 1 >

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

116
Graz, SS 2012

7. Questionnaires

Questionnaires
Automatic checking of control questions

Use normal treatment (i.e. not a questionnaire)

Use buttons with checkers

Only possible for numerical entries

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

117
Graz, SS 2012

A Comprehensive Introduction to z-Tree

Helpful Advice

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

118
Graz, SS 2012

8. Helpful advice

Helpful advice

Program all treatments in one treatment file

Use speaking variable names

Define all variables in the Background

Prepare session structure help sheet (script):

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

119
Graz, SS 2012

8. Helpful advice

Commenting
Comment your code extensively:

Single-line commenting:
// until the end of the line

Inline commenting:
x /* comment */ =5;

Multi-line
commenting:
/* multi
line
comment */

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

120
Graz, SS 2012

8. Helpful advice

Helpful advice

Test extensively

Attempt analysis based on data generated by test run

Keep a documentation of tables and variables:

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

121
Graz, SS 2012

8. Helpful advice

Helpful advice*

Implementing observer or experimenter subject (e.g. for

dice-throw)
Variable that is 1 for observer and 0 for others
Create specific observer stage displaying relevant
variables or permitting input of a variable
Use Participate to exclude observer from normal
stages and exclude subjects from observer stage
Use checkers to let experimenter allow subjects to
continue step-by-step in training phase
* Source: z-Tree manual.

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

122
Graz, SS 2012

8. Helpful advice

Helpful advice: Tables

Use instead of arrays for better output format

Use table loader to quickly import new parameters

Use to transfer data between treatments (instead of

OLDsubjects, etc.)

Use table functions to quickly calculate summary

information

Table lifetime
Period (contracts, globals, subjects, )

Treatment (summary, )
Session (session, )
A Comprehensive Introduction to z-Tree
Stefan Palan, Karl-Franzens-University Graz

123
Graz, SS 2012

8. Helpful advice

Helpful advice

Search and replace using export function

Use File-Export-Treatment to export code to a .txt-file.
Use search/replace in Notepad/Word
Re-import treatment file into z-Tree

Calculate exact time using gettime ()

gettime() returns time since computer was started
Set variable equal to gettime() at beginning of period

Calculate difference between time of interest and time

calculated at the beginning
A Comprehensive Introduction to z-Tree
Stefan Palan, Karl-Franzens-University Graz

124
Graz, SS 2012

8. Helpful advice

Timelog

Create timelog table:

Allows custom logging of
important events

Record starting time and create first record:

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

125
Graz, SS 2012

8. Helpful advice

Timelog
Create entries for all events of interest:

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

126
Graz, SS 2012

8. Helpful advice

Timelog
Create entries for all events of interest:

Period

ID
0
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1

Time
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17

Subject
0
0.328
0.421
0.546
3.916
6.755
7.457
7.987
13.588
16.505
16.833
22.714
25.257
25.881
31.746
31.762
34.133

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

Event
0
0
0
0
1
1
1
1
1
2
2
2
3
3
3
0
0

1
2
4
6
7
8
7
9
18
7
9
18
7
9
18
19
20

127
Graz, SS 2012

8. Helpful advice

Notes on data processing

Main data is saved in YYMMDD_hhmm.xls

Reformatting possibilities:
Manually
Using z-Trees Tools - Separate tables command
(see following slides)

Using my Excel macro (see following slides)

Using Kan Takeuchis Stata import procedure
Using Oliver Kirchkamps R import procedure
(all included in the resource pack at www.palan.biz/academic select Downloads)

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

128
Graz, SS 2012

8. Helpful advice

Notes on data processing

Z-Trees Tools - Separate tables command

Select .xls-file to process
Output:

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

129
Graz, SS 2012

in advice
Part III Advanced8.Topics
Helpful
z-Tree

Notes on data processing

Using my Excel macro

Start macro
Answer dialog questions
Output:

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

130
Graz, SS 2012

8. Helpful advice

Topics in data processing

Opening .xls output file in German Excel creates problems

with data interpretation (decimals interpreted as dates)

Time data
Recorded for all button clicks (if stage has a timeout)
Times are recorded as time remaining in period:

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

131
Graz, SS 2012

A Comprehensive Introduction to z-Tree

Troubleshooting

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

132
Graz, SS 2012

9. Troubleshooting

Helpful advice from the manual

Change timeout during the experiment:

Use a variable to set the timeout for a stage
Modify the variable in the period parameters in the
parameter table (for periods yet to be played)

Stop after this period option

Use both in testing and, if necessary, in an
experiment

Period runs its course, but treatment then stops

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

133
Graz, SS 2012

9. Troubleshooting

Debugging*

Errors when closing a program

Read the error message
z-Tree usually puts the cursor into the line causing the
error
Use comments to locate the error

Errors while the program is running

Test different combinations of parameters
Consult the tables to check intermediate results

Do so within z-Tree (stop the clock)

Use the .xls output file

* Adapted from lecture slides by Verena Utikal, Friedrich-Alexander-University Erlangen-Nrnberg.

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

134
Graz, SS 2012

9. Troubleshooting

Crash handling
Client Crash

If the PC is still working, simply restart the z-leaf

(<Alt>-<F4> to shut it down)

The z-leaf will then replay the treatment up to the

present situation

* Adapted from lecture slides by Verena Utikal, Friedrich-Alexander-University Erlangen-Nrnberg.

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

135
Graz, SS 2012

9. Troubleshooting

Crash handling
Client Crash

If the PC is not working, start

a new PC

Start z-leaf on new PC

Open clients table on z-tree
Disconnected client appears
in parentheses
Move new client over field of
old client

* Adapted from lecture slides by Verena Utikal, Friedrich-Alexander-University Erlangen-Nrnberg.

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

136
Graz, SS 2012

9. Troubleshooting

Crash handling
Server Crash

Restart z-Tree

Open the clients table

Restart all clients with menu

command Restart all Clients

If some clients do not

reconnect, manually restart
their z-leafs
* Adapted from lecture slides by Verena Utikal, Friedrich-Alexander-University Erlangen-Nrnberg.

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

137
Graz, SS 2012

9. Troubleshooting

Crash handling
Server Crash

If no client connects, you can:

Restart the clients manually

Wait (up to 4 minutes) and restart the clients later
Start z-Tree on another computer
Shut down and restart the server PC

* Adapted from lecture slides by Verena Utikal, Friedrich-Alexander-University Erlangen-Nrnberg.

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

138
Graz, SS 2012

9. Troubleshooting

Crash handling
Server Crash

Use Run/Restore Client Order to sort the clients into the

same order as before the crash

Restore all tables with Run/Reload database (tables are

stored after each period)

Check the summary or subjects table to see how many

periods have been played (n)

Open the treatment that ran before the crash (e.g.

@1.ztt)

Set the number of practice periods to n

Restart treatment
Run/Start
treatment
(<F5>)Erlangen-Nrnberg.
* Adapted fromwith
lecture slides
by Verena Utikal,
Friedrich-Alexander-University

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

139
Graz, SS 2012

A Comprehensive Introduction to z-Tree

Graphics Teaser

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

140
Graz, SS 2012

Part III Advanced Topics in z-Tree

Graphics teaser 1/3

z-Tree offers graphics capabilities since version 3.0

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

141
Graz, SS 2012

Part III Advanced Topics in z-Tree

Graphics teaser 2/3

Selection of plot items:

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

142
Graz, SS 2012

Part III Advanced Topics in z-Tree

Graphics teaser 3/3

Demo treatments demonstrate graphics capabilities:

colortriangle.ztt
movepointdemo.ztt
selectlinkdemo.ztt
z-draw.ztt
animatetreedemo.ztt

More information on graphics in z-Tree:

z-Tree lecture by Maria Bigoni:
http://www2.dse.unibo.it/bigoni/courses/ztree/Lecture_notes_handout.pdf

z-Tree Wiki:
https://www.uzh.ch/iew/ztree/ssl-dir/wiki/
A Comprehensive Introduction to z-Tree
Stefan Palan, Karl-Franzens-University Graz

143
Graz, SS 2012

A Comprehensive Introduction to z-Tree

Exercises

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

144
Graz, SS 2012

11. Exercises

Exercise A simple guessing game*

Program the following treatment:

The computer draws a random integer between 1

and 10 (y)

Each subject enters an integer between 1 and 10 (x)

If x equals y, the subject receives 100 ECU

Display the outcome of the treatment

Hints:
y=roundup(random()*10,1));
Profit=if(x==y,100,0);
* Adapted from lecture slides by Verena Utikal, Friedrich-Alexander-University Erlangen-Nrnberg.

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

145
Graz, SS 2012

11. Exercises

Exercise Ultimatum Game 1/2

Program the following treatment:

2 subjects, 1 period

Subject A receives 10 euros and decides how much

to transfer to subject B

Subject B can either accept or reject the offer

If B accepts, A gets what he kept and B gets what
he was given

If B rejects, both players receive 0

Display the outcome of the treatment

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

146
Graz, SS 2012

11. Exercises

Exercise Ultimatum Game 2/2

Subject A

Structure

Subject B

Program start

Proposer decision stage

Responder decision stage

Profit display stage

Hint: Participate = 1 - Proposer;

A Comprehensive Introduction to z-Tree
Stefan Palan, Karl-Franzens-University Graz

147
Graz, SS 2012

11. Exercises

Exercise Trust Game

Program the following treatment:

2 subjects, 1 period

Subject A receives 10 euros and decides how much to

transfer to subject B

The amount transferred is multiplied by 3 when it arrives

in subject Bs account

Subject B decides how much to transfer back to subject

A

The amount transferred is multiplied by 3 when it arrives

in subject As account

Display the outcome of the treatment

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

148
Graz, SS 2012

11. Exercises

Exercise Keynes beauty contest 1/2

Program the following treatment:

9 subjects participate in an experiment with 3

periods

Subjects are randomly matched and rematched into

groups of 3 (absolute stranger matching)

Each subject enters an integer x between 0 and 100

The subject whose x is closest to 0.5 times the

group average of x receives 100 ECU

Display the outcome of the treatment

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

149
Graz, SS 2012

11. Exercises

Exercise Keynes beauty contest 2/2

Hints:

distance=abs(x-0.5*subjects.average(same(Group),x));

Winner=if(distance==subjects.minimum(same(Group), /*
*/ distance),1,0);

Extra tasks:

In case there would be more than one winner, make

sure that one of them is randomly chosen to receive
the 100 ECU

Implement an experimenter subject which sees the

entries made by the other subjects (contract list)

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

150
Graz, SS 2012

11. Exercises

Exercise Dutch auction 1/2*

Program the following treatment:

There are 15 subjects, in 3 groups of 5 subjects

Each subject is endowed with 100 ECU

Each subject can buy a maximum of 1 unit of a good

The good has different values for different subjects

A subjects valuation is randomly determined between
10 and 90
Each subject is informed only of their own valuation
* Adapted from lecture slides by Verena Utikal, Friedrich-Alexander-University Erlangen-Nrnberg.

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

151
Graz, SS 2012

11. Exercises

Exercise Dutch auction 2/2*

The computer conducts a Dutch auction for 3 units of the

good in each group

The computer starts the auction at a price of 100 and

counts down to 0, reducing the price by 1 every 3 seconds

If a subject buys the good for the current price, the auction
for the next good starts, until all goods are sold

At the end, subjects see their profit, which is:

If they bought a good: The difference between their
private valuation of the good and their purchase price
If they bought no good: Zero

Subjects may not buy at a price higher than their valuation

* Adapted from lecture slides by Verena Utikal, Friedrich-Alexander-University Erlangen-Nrnberg.

A Comprehensive Introduction to z-Tree

Stefan Palan, Karl-Franzens-University Graz

152
Graz, SS 2012

Karl-Franzens-University Graz

A Comprehensive Introduction to z-Tree

Stefan Palan
stefan.palan@uni-graz.at
http://www.palan.biz/academic