What would be the most appropriate method to analyze different variable's variation across 3 periods ? Some variables are numeric and some are factors (Frequency of usage such as "once a month", but can be converted into factors)
I have some issues :
- Assumption of normality is not met (cannot run an ANOVA)
- Not all sample was measured across 3 periods (over 154 members of the sample, only 54 were measured for all variables across 3 periods since some did not use the product during one of the periods), but some variables (frequency of usage, motives of usage...) can be just registered as zero for the periods in which the product was not used (I am not sure how to distinguish the effect of the variation of variables and the effect of stopping sage of the product altogether)
What would be the best way to make the most out of my data?
The analysis should only be performed on the same 54 people.
Analysis of variance or even covariance for data abnormalities is Robust.
I suggest you good with the paired-samples t-test in SPSS. SPSS allows you to have multiple time horizon for your independent variable. ANOVA is of no use here. Lavene's test of homogeneity does not require the assumption of normality.
In the last table in output window, look for variance and p value. If you observe p is significant than you can interpret that there is a difference in usage or motivation in 154/54 customers etc.
Also, zero values or blanks create false-positive/negative results, use 'MEAN' of the column instead of zero.
I suggest you good with the paired-samples t-test in SPSS. SPSS allows you to have multiple time horizon for your independent variable. ANOVA is of no use here. Lavene's test of homogeneity does not require the assumption of normality.
In the last table in output window, look for variance and p value. If you observe p is significant than you can interpret that there is a difference in usage or motivation in 154/54 customers etc.
Also, zero values or blanks create false-positive/negative results, use 'MEAN' of the column instead of zero.
- Badges
- Science method
- Similar topics
- Mathematics
- Statistics
- Statistical Modeling
- Analyses of Variance
- ANOVA