I conducted a survey using a questionnaire. My target population is online mobile gamers in a certain area. The population is unknown, so I do the sampling. My research is analyzed using logistic regression. I was worried that I would not be enough data for each binary sample; for instance, in my case, the dependent variable is the "intention to play (or re-play) the game," which has a value of 0 or 1. Because I worried if the sample for 0 will not be enough, so I designed a questionnaire that allows the respondent to fill in more than one response, which means more than one sample. I'm confused because my target population is gamers, but my questionnaire captures the "game" as a sample. I'm new, please help! I also attached the example of my questionnaire. Thank you!
Edit: what I’m trying to do is identifying which variable is the best predictor of gamers’ intention to play (or re-play) the game
Hello Reza,
The second portion of the survey doesn't make much sense (to me) to be included, as you're asking potential respondents to make judgments about their perceptions during play of game they indicate they would not choose to play. As they may never have played the game, they might not be able to make accurate judgments. A good way to determine whether this concern might represent a negligible vs. major threat to your study would have been to run a small pilot sample and debrief the participants immediately after.
What you could do is, treat persons who selected a specific game as one they would play (e.g., Fortnite) as a group, and compare the groups on your response variables (for the first part of the survey only: the chosen game and the responses related to that game). This could be analyzed as a multinomial logistic regression model (or, if you believed the responses were sufficiently well-behaved, as a discriminant analysis or manova).
Good luck with your work.
I am sure your study design is faulted. I am sure your are not running an instrumentation study where you may want to see the Goodness-of-fit of variables or questionnaire or where you may want to test for the reliability and validity of the instrument or Measurement Error.
Actually, nothing is impossible but the results my not be far-fetched. Whichever questionnaire you administer first, the latter will always almost get you wrong response and might even affect the former. Respondents here are humans.
- Badges
- Science topic