# Talk

## Need urgent help with statistics

(3 Posts)
FrazzledBrainedStudent Tue 24-Oct-17 16:36:53

I've been trying to suss out statistics for way too long and it's all gobbledegook to me still, so I need some guidance please as the longer I look at it the less I understand.

What I have are 3 questionnaires that I need to analyse in SPSS 24, but before I do that I need to have decided what analytical tests are best suited. The questionnaires are Likert scales; the first 2 are to test the participant's knowledge and have a choice of answers - strongly agree/somewhat agree/somewhat disagree/strongly disagree, and the 3rd one has a choice of frequency of different drinks consumed (never or less than once per month/1-3 times per month/1 per week/2-3 per week/5-6 per week/1 per day/2-3 per day/4 or more per day).

The questions I want answered are:
- Is there a difference in knowledge between males and females?
- Is there a difference in the type and frequency of drinks consumed between males and females?
- Is there a correllation between knowledge and frequency of drink consumed?
- Is there a correllation between knowledge and type of drink consumed?
- Is there a difference in the knowledge between the ages (14, 15, 16, 17, 18 year olds)?
- Is there a difference in the frequency between the ages?

Am I right in thinking I should do a Spearman’s rank test for the associations/correlations? Or an ANOVA?

Should I do a T-test to see if there's a difference between males and females?

Sorry, I really don't feel at all confident with it despite reading different websites and YouTube tutorials and having lectures in stats last year. I've spent so long trying to figure it out and I can't waste any more time; I have dyslexia and dyspraxia and am slow enough as it is! I really need help as nothing really makes a lot of sense to me, it's like I have a statistics blind spot in my brain! TIA.

ttlshiwwya Tue 24-Oct-17 17:00:59

Since your questionnaire data is Likert scale then you would be best to look at variations of the chi-squared test rather than ANOVA. I'm not familiar with SPSS but look at SPSS manual for chi-squared tests. Your LIKERT scale data are on ordinal scales and your groups (gender/type of drinks) are nominal. Using this info you can select the appropriate test for each question you want to answer e.g. Is there a correlation between knowledge and frequency of drink consumed? Here both measures are ordinal so you can pick up the chi-squared test for trend or correlation between increasing knowledge and frequency of drink consumed. Is there a difference in knowledge between males and females? Here knowledge is ordinal and gender is nominal so you can select the chi-squared test which compares mean responses amongst gender groups.

FrazzledBrainedStudent Tue 24-Oct-17 17:56:17

Thank you so much ttlshiwwya! I can't begin to tell you how helpful it is to have someone narrow things down so I know where to channel my searches

I'm not familiar with SPSS
No am I We used R Studio last year, but at least there seem to be tons of how to SPSS videos on YouTube.

Join the discussion

Registering is free, easy, and means you can join in the discussion, watch threads, get discounts, win prizes and lots more.

Register now »