Overview for Instructor:
        WEXTOR is a useful tool not just because of what it produces, but also its logical progression in developing the experimental materials.  It gives students an organized avenue for constructing an experiment.  One drawback is that after you specify the experimental procedure you have to finish constructing the appropriate webpages and then post them to your own website.
        One note about the example provided below.  In my research methods class, I'm not interested in my students learning to make webpages, but rather to think about what's involved in making the experiment.  To this end, the example below provides most of the finished (i.e., polished) materials at the end.
        This example was adapted from a classic demonstration from person perception research.

Objectives:

• Familiarize with WEXTOR
• Think about the layout and experimental procedures of a study
• Practice variable designation

• Finished experiment
• Modifiable handout for students (Includes URLs rather than active links)
• Ideas for course use and other projects

• Reips, U.-D. (2002). WEXTOR:  A Web-based tool for generating and visualizing experimental designs and procedure.  Behavior Research Methods, Instruments, and Computers, 34, 234-240.

        WEXTOR is a tool for constructing and organizing a research procedure.  You begin by defining any between-subjects, within-subjects, and quasi-experimental variables.  You then define the order of these variables, see a visual representation of the experimental order, and modify the webpages needed to conduct the study.  Ultimately, you can conduct the experiment over the Internet or a lab-based computer.
        Let's put together a simple research study in WEXTOR.  Research on impression formation shows that the order in which we learn information about others influences our overall evaluation of that person.  We're going to put together a conceptual replication a classic study on impression formation.  Here's the example we're going to use.  Participants will be given 6 trait terms that describe an individual.  However, the order in which the traits are presented will differ for difference participants.  Even with the same list of traits, the evaluations of the targets should differ.
        Here are the lists of traits we'll use:
 

        We're going to put together this brief research study with 2 Factors (there's no within-subjects factor in this example).  The first factor is a between-subjects variable representing which list order a participant gets.  As a quasi-experimental variable, we want to know if the participant is a native English speaker.

        Respondents will read either List A or List B and then, based on only this limited information, rate the target on several personality traits (sociability, happiness, aggressiveness, and rudeness) as the dependent measures.

Getting Started:

1) go to the WEXTOR website.  If you haven't created a login id, go ahead and do that.  Once you've done that, get logged in.

2) Click Create/modify an experimental design to get started.  We want to Create an Experimental Design.  First, you need to name your experiment.  Let's call it "Impression Formation".  Click Create when you've entered the text.

Specifying the factors:

3)  Next, we begin to define each of our factors.   WEXTOR accommodates between-subjects, within-subjects, and quasi-experimental factors, and it provides brief explanations if you need a refresher.
        We need to specify how many of each type of factor our experimental procedure will have.  Open the drop-down menu below the between-subjects factor and select "1".  Use the drop-down menu for within-subjects factors to select ")".  Use the bottom drop-down menu to indicate that "Yes", we have a quasi-experimental variable.  If you try to move on without at least specifying you don't have any variables of a certain type,you will receive a friendly error message and let you correct your mistake.
        When you're done, click the > arrow to continue.

4)  Now we need to name each of the factors and specify the number of levels for each.  Using the Factor Information List provided above, enter the name and number of levels for each factor and then click the > arrow to continue.  For the Quasi-experimental factor, we do not require different webpages for each level of the factor.

5) Now we have to specify the name for each level of each factor.  Use the information in the Factor Information List above to enter these labels and click the > arrow to continue.

Specifying the Experimental Procedure:

6) Now, in STEP 4 of WEXTOR, we define the experimental conditions.  Two important things to note here:

        a) The name you specify here essentially creates a folder in which materials for a specific experimental condition (e.g., List A) will reside. Since, the name you provide will be part of the URL used in the experimental procedure,  you may want to use a name that will not reveal too much information to the research participant.  To this effect, WEXTOR offers a codename of sorts by default, but you can change this if you want.
                Let's rename the conditions using something more descriptive to us.  We'll name them using the first letter of each trait.  So, for Experimental condition 1, erase the default name that's provided and enter "ESCIII".  For condition 2, enter "IIICSE".

        b) Another important thing to note about this screen is it provides you the opportunity to eliminate any conditions you don't need, if you have an incomplete (i.e., not completely crossed) design.

7) In Step 5 of WEXTOR, we have to specify how many pages (i.e., webpages) the experimental conditions will consist of.  First, you should note that WEXTOR automatically generates 5 pages.  Take a minute to review those.
        OK, now let's specify how many pages we'll need (not including the automatically included ones).  The procedure will consist of 1 instructions screen, 1 screen consistenting of the trait list, and 1 screen of trait ratings.  So, that's 3 screens total.

8) In Step 6 of WEXTOR, we specify the names of the webpages we'll need.  Names are already provided for the automatically generated webpages.  Let's keep the names simple, so use these:
 

        WEXTOR can help protect the integrity of the filenames by adding some random characters to your filenames (so they aren't so obvious).  In this case, let's not worry about doing that.

9)  We don't need to worry about Step 7 because we don't have an within-subjects factors.  So, let's move on.

10)  Now we need to specify the question addressed by our quasi-experimental variable.  In the top textbox, type the question "Are you a native English speaker?"  In the next textbox, type the variable name "Nat_Eng".  We've called the 2 levels of the variable "Yes" and "No", but WEXTOR knows we may need to call them something different for the respondents.  In this case, those names work fine, so in the next textbox type "Male" and in the final textbox type "Female".  Finally, click the button for Radio buttons.  Then, click the > arrow to move on.

11) In Step 8 of WEXTOR, we need to specify that both groups in our quasi-experimental variable can be randomly assigned to either level of our between-subjects variable.  So, click all 4 of the check boxes and then continue.