--Welcome to this sample experiment! Here you will learn how to randomly assign users to different conditions, how to keep words on the page for multiple screens, how to assign point values to questions, and other techniques useful in experiments. To get a more in-depth understanding of different types of questions, see also GuidedTrack Sample Survey.
--To see what this program looks live, visit this link: http://www.guidedtrack.com/programs/iwox5ds/run
--To start, we’ll quickly remind you of the 3 most basic rules of GuidedTrack language:
--1) Two dashes precede anything that is a comment. Only the people editing the survey can see comments. You’re reading a comment right now.
--2) Asterisks* precede GuidedTrack keywords. Keywords are used to build the design and functionality of the program.
--3) Indentations are important in GuidedTrack. They are used to indicate relationships between two or more lines of keywords or text.
*header: Does being “green” increase well-being?
--A *header keyword enlarges the text. In this example, users will see "Does being ‘green’ increase well-being?" in large text.
Welcome to the experiment.
--This is a standard text sentence. It looks just like normal text.
By clicking the button below, you are agreeing to our [Terms of Service|http://www.gimbeltech.com/documents/terms_of_service.html]
--A link has been inserted into the end of this sentence. The linked text and the URL are surrounded by [brackets]. The linked text and the URL are separated by a pipe: |
--In this example, users will be able to click the phrase "Terms of Service". Clicking "Terms of Service" opens the URL given. URLs always open in a new browser tab or window.
*button: Begin
--You can have a button that says anything you want on it. This example shows the user a button with the word “Begin” on it. When pressed, users move to the next screen. Buttons are useful for ending a page and requiring a click to go to the next page.
*question: Where do you live?
--This is a standard text question. The users see the question, a small text box in which to write their answer, and a “Submit” button, which they press after writing their answer to move to the next screen.
*question: What’s your age?
*type: number
--This question requires that the user enters a number only (e.g. “28”). The "*type: number" keyword ensures that any text response (e.g. “no”) will not be accepted. Note how an indentation is used here. This indentation shows that "*type: number" is a characteristic of this question (because it’s indented beneath the question).
*question: What’s your gender?
Male
Female
Prefer not to say
--This is a standard multiple choice question. The multiple choice answers must always be indented beneath the question they relate to. In this example, the user sees the question and three different possible answer choices. After the user selects an answer, they are taken to the next screen.
*randomize
All done with this section.
This section is now complete.
--Randomize is a keyword that randomizes what happens next. The possible paths users could go on must be indented one indentation beneath the *randomize keyword.
--In this example users will either see “All done with this section” or “This section is now complete,” but not both. There’s no good reason to use randomize here, we’ve included it only to get you ready for a trickier example later down the road.
Now you’ll take the /Elitepsych brief mood questionnaire/.
--This shows how to italicize text. The text you want italicized must begin and end with /forward slashes/.
*maintain: Please answer honestly
--Maintain will keep specified text (e.g. “Please answer honestly”) at the top of the page for as long as you want. The *maintain keyword will first present the user with a “Next” button. After clicking “Next,” the user will be taken to a new screen. The maintained text will appear at the top of the screen within a shaded box.
--In this example, the user will see the prior sentence that begins with “Now you’ll take...” followed by a button that says “Next.” Once clicked, they will see “Please answer honestly” at the top of the page in a shaded box. The question that follows will be below the maintained text.
*question: How happy are you right now?
I’m a miserable wretch
*points: 1 preIntervention
I’m kind of sad
*points: 2 preIntervention
I’m so-so
*points: 3 preIntervention
I’m pretty happy
*points: 4 preIntervention
I’m so happy I might explode
*points: 5 preIntervention
--You can assign point values to each possible answer choice. The *point keyword must always be indented beneath the multiple choice answer it refers to. You can include a unique name (e.g. "preIntervention") to help yourself differentiate what type of point it is. Your unique name should not include spaces and should consist only of letters, numbers, or underscores.
--You can have as many point names in your quiz as you like and can assign the same point name to as many questions as you like. Users will not know how many points of a particular type they got unless you tell them explicitly (only points not given a name are shown to the screen automatically).
--If you do not add a unique name to the points (e.g. you do not write “preIntervention”), then the points will be visibly tallied on the user's screen in the corner as they progress in the quiz.
--In this example, users who select “I’m so happy I might explode” will have 5 preIntervention points that will show up in a datasheet visible to the experiment creator.
Excellent.
*clear
--The *clear keyword will clear the maintained text (e.g. “Please answer honestly”). Users will be presented with a “Next” button that will take them to a new screen.
--In this example, users will see the word “Excellent” followed by a button that says “Next.” Once clicked, they will be taken to a new screen that begins with the text that follows (“You’ll now be assigned…”).
_You’ll now be assigned to one of two interventions_.
--This shows how to underline text. The text you want to underline must begin and end with _underscores_.
--Experiment is a keyword that randomizes what happens next, while keeping the number of participants in each group equal. With *randomize, you could have 15 participants that get one of your two conditions, while 5 get the other. The *experiment keyword keeps things even, ensuring if you have 20 participants total, there are 10 participants in each group.
--The experiment keyword must have a name, so you can find it easily in your data. We gave it the name “interventionGroups”
--In the example below, users will be randomized into one of two groups. Do you see the two *group keywords, indented one line beneath the *experiment keyword?
*experiment: InterventionGroups
*group
--The *group keyword is a special keyword that can be used with the *randomize keyword or the *experiment keyword. It keeps multiple lines of text or keywords organized within one group. All lines of text or keywords must be indented one extra indentation below the *group keyword they relate to. Without the *group keyword, you would have to keep the entire block of information you wanted to randomize on just one line, as we did in the first randomize example.
--you can also name the group if you wanted to, as in *group: reflectiveExercise
--In this example, users who are randomly assigned to this intervention will see the line of text below that begins with “Think about a time” and they will encounter the *wait keyword.
Think about a time when you did something good for the environment. Maybe you cleaned up a piece of litter on the ground, or high-fived a tree, or tried to reduce the amount of trash you were producing. I’ll give you a moment to think about it.
*wait: 10 seconds
--The *wait keyword pauses the program for the specified amount of time. Users will not be able to see what happens next until the allotted time has passed.
--In this example, users in this randomized group will wait 10 seconds before they see the next line of the program, which for them is the line that begins “Thanks!”
*group
Recall a time when you felt absolutely horribly sad and destitute and alone. Maybe your boss told you she or he didn’t like you, or you sat on a piece of used gum, or your trousers accidentally fell down in public.
*question: Describe your misery in your own words.
*type: paragraph
--This is a special type of text question with a larger paragraph-sized box in which to write an answer.
--In this example, the user sees the question, a large box in which to write their answer, and a “Submit” button, which they press after writing their answer to move to the next screen.
Thanks! Now we’ll give you the brief mood questionnaire again.
*maintain: Please answer honestly
*question: How happy are you right now?
I’m a miserable wretch
*points: 1 postIntervention
I’m kind of sad
*points: 2 postIntervention
I’m so-so
*points: 3 postIntervention
I’m pretty happy
*points: 4 postIntervention
I’m so happy I might explode
*points: 5 postIntervention
--This is exactly the same code that was written the first time the user received the brief mood questionnaire, only this time the name of the point has been changed to “postIntervention.” Researchers will be able to easily compare the number points users got before the intervention with the number of points they got after the intervention.
*clear
Thank you. That’s the end of this *extremely scientific* experiment.
--This shows how to bold text. The text you want bolded must begin and end with *asterisks*.