Hello everyone,
I am writing my thesis right now and am researching a company's webshop to see whether improvements can be made. I have sent out a questionnaire to a population of almost 2200 people asking for their satisfaction regarding different factors of the webshop (such as navigation, design and information on the webshop) and their overall satisfaction. I have used a semantic differential scale to measure these factors. I want to analyze if there is a correlation between the different factors and the overall satisfaction of the customer.
The problem, for now, is that I have only received around 30 responses out of the total 2200. Is it still possible to conduct any form of analysis with this small data set? I was originally thinking about conducting a Chi-square analysis, but am not sure anymore. What can I do for now? Thanks in advance.
The type of analysis to use depends on the results you want. For example,
1- Use the Chi square test to test the independence between two category questions;
2- Use the correlation test to test the strength of the relationship between two numerical questions.
3- Use the ANOVA test to test the relationship between a category question and a numerical question.
and so on...
Dear Jos Oosting ,
If I understand your concerns correctly, you're asking more for an appropriate test to be used in connection to the sample?
One way or another, I'd suggest you another stance. Statistics is just a tool. It can help you get some more information from a seemingly blurred pool of answers. Nevertheess, Who are the 2200 people how did you coose them? Are they the whole population or gust a selection?
Out of them, you could count on 40% response rate. It's still round 800 participants.
With this, you'd be only fishing for something which is not even there. Various test can be done, but unless the sample is selected properly, it'd be just numbers without much reason ;-)
If you can share more about your sample selection, I guess we could help you.
Best woshes,
Martin
- Badges
- Science topic