Prompt: Your CEO believes that your customers are reacting badly to messaging about your latest product and he'd like to test updated messaging. He wants you to set up an experiment testing two versions of messages about your product and test them on sample customers.
Instructions: You will design an experiment in Qualtrics to test your CEO's hypothesis. Your experiment will include a control and an experimental group of participants based on two versions of your product messaging (your experimental condition). For each group, you will create an alternate version of messaging around a product/service of your choice. For example, if you were testing advertising copy, you would create two versions of the advertising text. This is a between-subjects design, so each participant will see only one version of the messaging. You can use whatever text you like, but I recommend something related to your real-world job or, if possible, to your dissertation. After creating your survey in Qualtrics, you will use it to generate random responses and test those responses in Jamovi or another analysis program. To complete the assignment, you will turn in a copy of your survey (exported from Qualtrics to Word) and a copy of your analyses.
To create and test your survey in Qualtrics, follow these steps:
Create a new project/survey in Qualtrics.
In Block 1, explain any instructions and any warnings to participants. A sample consent form is at the bottom of this assignment.
In Block 2, add any demographic questions. At minimum, include gender, age, and ethnicity. Think carefully about how you'd like to ask these questions and what sort of responses would be meaningful for you.
In blocks 3 & 4, create each of your two experimental conditions. This will be asking participants to read one of the two messages carefully. You may want to add comprehension check questions or timers to ensure that participants read the messages carefully.
In Block 5, create your survey questions for participants. These are the DVs of your study. These could include questions like, "How positive do you feel towards this product after reading this message?" You must include at least two Likert-type questions (i.e., scale from 1-5) and at least one open ended essay response.
In Block 6, write your back matter including any instructions or debrief. Include your contact information in case they have questions.
To set your experimental condition, go to the Survey Flow section and create a randomizer after your Block 2. Under the randomizer, send participants to one of either Block 3 or Block 4.
If you've set any other fail conditions for the survey (for example, if they do not consent to your Block 1), then also create those end of survey conditions in Survey Flow.
Optional: Explore some of the settings in QUaltrics such as response controls, recoding, and backgrounds.
Once your survey is ready and you've previewed it, use Qualtrics to generate 200 responses (under Tools). Export these responses into a .csv file for analysis. Be sure to set to export numeric responses.
Load the exported responses into Jamovi or another analysis program and run the following analyses:
Descriptive statistics table including the mean and standard deviation of each variable.
Correlation between your demographics, experimental condition, and all DVs.
2-way ANOVA between your experimental condition and gender on each of your DVs.
Deliverables:
Print-out of survey: You can export your survey as as Word document that you will submit with this assignment. Be sure to include your survey flow in your export.
Analysis output: As with past assignments, send me your analysis in whatever format you use. For example, if using Jamovi, attach the .omv file in your assignment submission. I prefer to see the analysis than just the results (i.e., the .omv rather than the results copy/pasted into a Word document).
Sample Solution