I am studying the generational changes in the relationship between empathy and social media usage. I am using a Likert Scale to measure both empathy levels and social media usage; however, I do not know how to approach this data.
Any help or suggestions would be greatly appreciated.
Sophie Cutter
Likert scale would provide ordinal data which is analysed via nonparametric statistical tests. Please email or message your conceptual framework and data collected I would be happy to help you out.
This is a very recurrent question:
https://www.researchgate.net/post/What_statistical_analysis_should_I_use_for_Likert-Scale_data
https://www.researchgate.net/post/How_to_score_a_likert_Scale
https://www.researchgate.net/post/What_statistical_analysis_should_I_use_for_Likert_scale_data
My answer would be:
inspect the histograms. Do these distributions look quite normal? If so, use parametric tests. The F- or t-tests are known to be quite robust.
Good that you ask again. The answers in the previous thread all cling to non-parametric methods, some even claiming that this is the only way of doing it. Non-parametric methods are very limited and there exist parametric models for all kinds of DVs.
First, check the linear model distribution assumptions (Normality, constant variance.) Sometimes with Likert scales they are just fine and you can use linear regression or ANOVA.
If that is not the case, there are two parametric options left in the family of Generalized Linear Models:
1. Ordinal logistic regression
2. Beta regression
I describe them both here (chapter 7.6): https://www.researchgate.net/project/Book-New-statistics-for-the-design-researcher
kindly send your data set and i will analyse the same for the result of the dependent variable towards empathy and social media usage. there is relationship, difference, variance, impact, contributions and influences between variable to factor, and factor to factor.
- Badges
- Science topic
- Similar topics
- Social Science
- Media
- Social Media