How do I analyze perception on Likert scale questionnaire (that was poorly designed)?

So, I have a somewhat difficult task at hand right now, I was contacted to help on a research that aims to identify the perception of teachers regarding their pedagogical practices, if they tend to eurocentric pedagogical practices or non eurocentric pedagogical practices (scenario: a South America country).

This was conducted through a questionnaire with 2 sessions, one with participant's overall traits (like age and education level, all in options format, thus qualitative data), and the other with Likert like itens on questions like "Do you often teach non European authors?"

The research hypothesis is that teachers do have non eurocentric pedagogical practices despite having an eurocentric pedagogical formation, but I'm pretty sure the questionaire can't answer this (this whole research was very poorly designed, if you want my opinion)

So, how can I work this research? Work the Likert like session as a Likert scale, return the descriptive statistics like mean and standard deviation? But I was contacted for doing inferential statistics. Also, does that answer the "what's the teachers perceptions" question?

Can I test if the Likert scale scores seen in the research is different than an expected in the population? Should I test if the scores differ from group to group (like, by age or level of education or years as a teacher)? Should I use the scores as the dependent variable?

TLDR: How do I analyze the perception on a poorly designed Likert-like questionnaire? Does mean and SD answer my research question?

  • Mathe Mathe

    I believe this question is far from obvious and warrants a bounty.

1 Answer

Analyzing a poorly designed Likert-like questionnaire can be challenging. You can start by cleaning and organizing the data. Calculate descriptive statistics like mean and standard deviation for each Likert item. Consider using inferential statistics to test for differences between groups or compare responses. Be cautious when interpreting the results due to questionnaire limitations. Complement with qualitative methods if available. Clearly state the study's limitations in the research report.

Join Matchmaticians Affiliate Marketing Program to earn up to a 50% commission on every question that your affiliated users ask or answer.