It just depends on what measures you have that you want to compare between the groups. So for the first question it could be a t test if you have an appropriate outcome (continuous ideally, but people often use t tests with discrete counts and ordinal outcomes). For the second question with 5 options as outcomes it sounds like unordered categories and so something like a Chi-square test might be appropriate.
Other factors can influence the choice further but these are both common options.
Thom
________________________________________
From: Research of postgraduate psychologists. <[log in to unmask]> on behalf of Debbie Grimmond <[log in to unmask]>
Sent: 19 April 2016 20:21
To: [log in to unmask]
Subject: Chi-squared or t -test
Dear All,
This is very embarrassing but I just want to double check something. I am more familiar with ratio or ordinal data but I have added another aspect to my research and it is some time since I did this type of stats, and at my very old age, I am looking for reassurance.
I have a questionnaire element to my study. I am looking at motivation in career path choice and how this may be influenced by various factors.
In the questionnaire the first question gives them 3 possible paths A, B and C. I want to check if there is any significant difference between groups ( male and female but also monolingual, bilingual and EAL).
I then ask a question about who has influenced them giving them 5 options. Again, I want to see if there is a significant difference between the groups.
I am confused because in my mind I should be using chi-squared as it it categorical/ nominal data but a number of papers use t-tests.
Can any body advise me?
Many thanks in anticipation of any words of wisdom.
Debbie
DISCLAIMER: This email is intended solely for the addressee. It may contain private and confidential information. If you are not the intended addressee, please take no action based on it nor show a copy to anyone. In this case, please reply to this email to highlight the error. Opinions and information in this email that do not relate to the official business of Nottingham Trent University shall be understood as neither given nor endorsed by the University. Nottingham Trent University has taken steps to ensure that this email and any attachments are virus-free, but we do advise that the recipient should check that the email and its attachments are actually virus free. This is in keeping with good computing practice.
|