I am trying to determine the statistical analysis and sample size of my dissertation research. I plan to examine the relationship between race-based stress and cannabis use. The measure for race-based stress has three subscales that are Likert scales ranging from 1 (Never) to 6 (Almost all the time). I have two cannabis use questions and am trying to decide which question to use to run the analysis. The first cannabis question is a dichotomous question that asks if you have used cannabis in the past 30 days, with options "yes" or "no." The second cannabis use question asks about frequency of use over the past month and is asked as follows: How often did you use cannabis in the past month? 1 (Less than once a month) t0 4 (Almost every day).
Based on this information, which cannabis use question should I use, what statistical analysis should I use, and what would my sample size be for said analysis?
Thanks in advance for your help with these questions!
Dear, I hope you’re doing well.
First of all I'm confused about your question that is the question "What is a good sample size for a dissertation?" or "What is the analysis and sample size needed for dissertation study?
By the way I'm answering first one. As seen in the arithmetic below, a sample size is adequate to make assumptions about almost any population size at a 95% confidence level with a 5% margin of error. For example, if you took a 100 sample size, your average should be 95:5. Where 95 is your actual process and 5 is errors. If you think you need to get a better result, you may take 1500–2000 samples. Actually, it depends on your research subject. Some research is short-length, using 80-100 samples. Some research is broad, using 10000+ samples. Usually, samples of 400–500 are commonly used in research.
Omar Faruq hello. Thank you for your response. The question “What is the analysis and sample size needed for dissertation study?” was used because I couldn’t fit the whole question into the space. I gave more description below it. I appreciate your help with this question.
Whitney Contreras. the answer I provided, I hope it will help you. I read your details. You can use either or a combined question set. For example, some questions may use true/false or Yes/No, and some questions may use multiple choice or something like this. You may create a Google Form or any other platform for the survey. then you should collect minimum 100 response. More response means more strong research, more precise, and more valuable. You can take a target of 1000 responses. then go for analysis. Thank you.
Whitney Contreras Also, you can find other related articles for better understanding. Search "Cannabis – Science topic" for more results.
Omar Faruq that was very helpful. I was not sure about the sample size. Do you think a regression analysis would work to analyze the data?
- Badges
- Science method