factorWiz for a Factorial Design Radio Buttons   Help   Copyright

Experiment Name: Exp Short Name:
Row Factor Name: Col Factor Name:
No. levels of Row (9 or less): No. levels of Col (9 or less):
Row 1=Col 1=
Row 2=Col 2=
Row 3=Col 3=
Row 4=Col 4=
Row 5=Col 5=
Row 6=Col 6=
Row 7=Col 7=
Row 8=Col 8=
Row 9=Col 9=
Row 10=Col 10=
Row 11=Col 11=
Row 12=Col 12=
Separator between Factors
No. of Buttons: Left End Label: Right End Label:

HELP Return to factorWizRB

1. To use this program for first time, work through the following example with a factorial design of two sets of five adjectives. Type in the experiment name, Judgments of Likeableness, in the first row. For the shorter name, try like5x5. Don't enter quotes in any cells. This (shorter) name, like5x5, will be printed as the first variable (00exp) in the data file. The 5x5 will help to remember that this is a 5 x 5, Row by Col, factorial design.

2. The row and column factor names are not used in this program. Just enter Row and Col in the spaces for the names.

3. Enter 5 for the number of rows, and 5 for the number of columns.

4. For the row stimuli, enter the following adjectives in the first 5 rows. (Anything in later rows will not be used.)
phony, squeamish, blunt, informal, sincere
For the column stimuli, enter the following adjectives in the first 5 rows:
mean, listless, solemn, lighthearted, trustworthy

5. For the separator between factors, put in the word " and " (always without entering quotes, but notice that you should put in a leading and following space, i.e., space-and-space).

Change the number of Radio Buttons from 7 to the desired number (in this example, 9). Type in the enpoint labels of the response scale. In this example, enter Dislike Very Very Much for the Left End Label and Like Very Very Much for the Right End Label.

6. Press the Make the Form button. To view the survey in Netscape Navigator, push the Display button. The Display button does not work in Internet Explorer.

7. Copy the entire contents of the text box to your text editor (e.g., NotePad, SimpleText, or BBEdit), and save it as a text file, with the extension of .htm or .html. For example, you could save it with the name, likeExpt1.htm.

8. If you would like another random order of trials, push Make the Form again. Copy the text box to a new file, and save it under a different name, for example, likeExp2.htm. You can create different random orders this way.

9. Replace the words, "(put your instructions here)" with the instructions for the task. This editing can be done in the HTML window of factorWiz or in the Text Editor. For example, in this case, you could use the following instructions:

This is a study of how people form impressions of personality. Each trial below lists two personality trait adjectives. Imagine that each trial represents a different person. Each adjective has been contributed by a different acquaintance who knows that person well. Sometimes people don't agree in how they would describe a person. Your task is to imagine a person who would be described by both adjectives, and rate how much you would like that person. Make your ratings on the following scale:
1=dislike very very much
2=dislike very much
4=dislike slightly
5=neutral (neither like nor dislike)
6=like slightly
8=like very much
9=like very very much

10. You could assign different people to different random orders, or you could have the same person do the task twice, using different orders. The assignment to conditions of different random order can be done by CGI script, by JavaScript, or by creating a page in which the participant clicks on his or her birth month, and different birth months are assigned to different random orders of the stimuli. In judgment tasks, order effects are minimal after a 7 trial warmup that previews the range (the extremes) and distribution of the stimuli. To view a sample JavaScript page to assign people to different orders, click here; to try it out in your browser, click here.

11. Load the HTML file into your browser, and edit it in the text editor to improve its appearance or make any further changes.

12. The form created is programmed to send the data to a file, data.csv. You can change the URL in the ACTION=URL to the URL of your own CGI script. To test the form, you can also change the ACTION to mailto: your email address, as follows: ACTION=mailto:user@address.edu.

A script in PERL by William Schmidt that works with factorWiz can be used to save data to your local server. To see the script and instructions, click here. That page also includes a PERL script to randomly assign people to different random orders.

Unless you change the ACTION= in the FORM tag, the data are sent to a file from which you can FTP READ ONLY by typing the following into the URL box in your NETSCAPE or IE browser:

Note that the above address is FTP: not HTTP: --Also note the @ in the address. This FTP site is download only, and all files are Copyright Michael H. Birnbaum, all rights reserved.

You are welcome to use the default option under the following conditions. You may freely use the data generated from your own experiment without requesting permission; however, you may not use data from another person's experiment without permission from the author of that experiment, nor may you make any other use of the program or data that would be considered unethical according to the APA code of ethics.

Click here to start FTP session: Download data.csv

13. Study the data file created in the file, data.csv.
The first variables sent by the script will be as follows:

experiment short name
Computer address (this identifies the computer network that sent it. This may be useful if you see two successive entires from the same address, to help you catch multiple submissions by the same person)
Gender (M or F for male or female)
The data will be in the following order (from left to right):
(row 1, col 1), (row 1, col 2)...(row 1, col c)
(row 2, col 1), (row 2, col 2)...(row 2, col c)
(row r, col 1), (row r, col 2)...(row r, col c)
14. If you rearrange the random order of trials, keeping the variable names the same, then the script will continue to return the variables in the above order. This means you can switch the order of sections of the experiment without affecting the data file created.

Birnbaum, M. H. (2000). SurveyWiz and FactorWiz: JavaScript Web pages that make HTML forms for research on the Internet. Behavior Research Methods, Instruments, and Computers, 32, 339-346.

Copyright Michael H. Birnbaum, 1998, all rights reserved. This program may be used freely for educational and other noncommercial scholarly uses. Software is "as is," no guarantees or warranties can be made. You may modify the program, but retain this copyright notice.

This material is based upon work supported by the National Science Foundation under Grant No. SBR-9410572. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author and do not necessarily reflect the views of the National Science Foundation.

Send Comments, Corrections, Permission Requests

Return to factorWizRB

Return to the index