Comp 333 Southwestern College Simple Ab Test for A Survey Questions

I’m working on a computer science multi-part question and need an explanation and answer to help me learn.

In this assignment, we will look into how to design a study that will help you to answer a research question. For this assignment, you will focus on designing a simple AB test for getting respondents to respond to your survey that you designed for the last assignment. For all of the tasks below, you will include your results in a single pdf document.

1. What is your research question? In other words, for your AB test, what are you interested in comparing? (E.g., Does sharing the survey over twitter or instagram lead to more answers?)
2. What is your hypothesis? Your hypothesis is what you think the answer to your research question will be. Explain why you think this, ideally backing it with evidence.
3. Based on your research question, what are your independent, dependent, and control variables? Make sure that what you list is concrete. The dependent variable cannot be “survey results” but it could be “number of people to complete the survey”. For the control variable, mention what factors you will make sure that you keep the same between your conditions so they do not bias the results.
4. For your study, are you going to use a within or between subject design and why? Have you considered all of the design factors you might need for your chosen design?
5. Now that you have designed your study, you can make your materials and procedure! For each condition that you want to test, what will it look like? This can be including the text that you will send out or an image of what it will look like. Basically, you need to describe each condition and explain how you are aligning with your control variables. Second, how will you share the two different conditions? Again, how will you take into account your control variables?
6. Description of how the dependent variable will be collected.
7. (Optional) Run your AB test. If your design requires you to manually share the survey, you will need to send it to at least 10 people in each condition. Share what the outcome is! Did it align with your hypothesis? Were you surprised?