ztree2stata

ztree2stata is an ado file for STATA. It imports a z-Tree (Fischbacher, 2007) data file and converts it into Stata format. Specifically ztree2stata:

1. opens an ASCII text file saved by z-Tree,
2. keeps the data of the specified table,
3. renames variables using the variable names in z-Tree, and
4. converts the data into numerical type if possible.

Installation

copy ztree2stata.ado and ztree2stata.hlp into C:/Program Files/StataNN/ado/base/z. If you cannot find such a directory, then run sysdir command and see [U] 20.7 How do I add my own ado-files?.

Quick start

Import the subjects table of 040517_1230.xls

ztree2stata subjects using 040517_1230.xls

Import the globals table of 040517_1230.xls after clear the data in the memory

ztree2stata globals using 040517_1230.xls, clear

Import the data of from the 2nd and the 4th treatments and save the data in dta format

ztree2stata subjects using 040517_1230.xls, treatment(2 4) save

Import the subjects table of 040517_1230.xls and 221211_1749.xls and combine them

ztree2stata subjects using 040517_1230.xls, save
ztree2stata subjects using 221213_1749.xls, clear save
use 040517_1230-subjects.dta, clear
append using 221213_1749-subjects.dta

Syntax

ztree2stata table using filename [, options]

Options

clear permits the data to be loaded even if there is a dataset already in memory and even if that dataset has changed since the data were last saved.

save allows Stata to save the data in memory as filename-table.dta.

replace permits save to overwrite an existing dataset.

treatment(numlist) specifies treatments that will be imported into memory. If this option is omitted, then all treatments will be imported.

except(string) specifies strings that will be used to skip the automated renaming. If there is any trouble with variable names, then the except option may solve it. See the example below for more details.

string(string_varlist) specifies variables to which destring (the conversion process to numeric) is not applied. This option was necessary for the previous versions through 2014; it is no longer required to specify which variable is a string in the current version.

Remarks and examples

z-Tree records the various data during the experiment in multiple tabular formats, and after the experiment is completed, these tables are combined and output as a single tab-delimited ASCII file with xls extension. There are several issues when Stata imports the file. They include, for example,

1. The file includes several tables (e.g., subjects, globals, and clients), though they must be separated for analysis.
2. Because the file includes variable names as strings, every numeric data is interpreted as a string.
3. When z-Tree data includes several treatments, the width of the selected table varies with treatments; thus, variables may be recorded in different columns.

ztree2stata can handle these issues and conveniently convert the z-Tree data file into Stata’s dta format.

⊳ Example 1

ztree2stata subjects using 040517XY.xls

Stata reads 040517XY.xls and keeps the data of the subjects table.

• table = subjects
  Specify one of the tables which have been defined in z-Tree.

• filename = 040517XY.xls
  Stata looks for 040517XY.xls in the current directory. Therefore, you need to put your data file in the current directory. If the size of the file is too large, then you need to increase the memory size before you use ztree2stata.

⊳ Example 2

ztree2stata globals using 040517XY.xls, tr(2 4) save

Stata opens 040517XY.xls and keeps the data of the globals table in treatment 2 and 4.

• options = tr(2 4)
  Stata reads the data of treatment 2 and 4. If the data does not include some of the specified treatments, then it returns an error.

• options = save
  If 040517XY-globals.dta does not exist yet, this option allows Stata to save the data in memory as 040517XY-globals.dta; Otherwise, Stata returns an error message and does not save the data in memory. To overwrite the existing file, use save with replace option.

⊳ Example 3

ztree2stata subjects using 050301AB.xls, except(foo goo) 

• options = except(foo goo)
  If you have any trouble with variable names, then you may want to define an exception. In this example, ztree2stata renames a variable whose name includes “foo” or “goo” into foo# or goo#, where # is the original column number where the variable is located. For instance, suppose that you have 3 variables in a z-Tree table — “food,” “good?,” and “goodfood!” — and that they are located in the 6th, 7th and 10th column, respectively. Then, they will be renamed as “foo6”, “goo7,” and “goo10.”

Technical Notes

• The ado and hlp files, ztree2stata.ado and ztree2stata.hlp, are available through the author’s website.

• ztree2stata cannot open Excel files. A data file created by z-Tree is not an Excel file but a tab-delimited ASCII file, while its extension is xls. Therefore, once a data file is overwritten as an Excel file, ztree2stata can no longer open it.

• The following symbols will be deleted from the variable names: exclamation marks (!), colons (:), semi-colons (;), equal signs (=), double quotation marks (“), and spaces ( ).

• The except option can be used to eliminate any other symbols or letters from variable names.

• Note that when there is an unpaired double quotation mark in a data, Stata’s import results may not be as intended.

• This command is provided as it is, without any warranty.

Reference

Fischbacher, U. 2007. “z-Tree: Zurich Toolbox for Ready-made Economic Experiments,” Experimental Economics, 10(2), pp.171-178. (http://www.iew.unizh.ch/ztree/index.php)

Takeuchi, K. 2023. “ztree2stata: A data converter for z-Tree and Stata users,” Journal of the Economic Science Association, forthcoming.