Hi Marta

what is/are your research question/s? Your statistical analyses depend on what you want to find out and on the actual questions/items in the questionnaire. The more information your items contain (continous vs. binary), the more options are available to you. Any handbook on descriptive and inferential statistics will teach you the basics, but it is up to you to pick the method that matches your research question/s.

Best

Marcel

Thank you very much Marcel Grieger ! I have binary data, we asked people how much they agree or disagree with statements about nature (e.g. what do you search for when you visit a natural place? We gave different options and they expressed their level of agreement or disagreement for each of those). We have different research questions, I think some of them could be answered with simple correlation analyses for the others I will follow your advice and search for a handbook about descriptive and inferential statistics.

Thanks, bests

Marta

Hi Marta

just quick follow-up: Binary or dichotomous data mean question to which the answer may be "yes" or "no". Levels of agreement or disagreement are more usually answered with a Likert-style response format (e.g. ranging from 1 "do not agree at all" to 5 "fully agree"). With the latter you can computer factor score if the items are meant to load onto one factor as opposed to binary options.

Best

Marcel

Thanks for the specification Marcel Grieger ! In fact, I have both binary and Likert scale responses. I will search for a handbook to better understand how to manage the latter data.

Thanks, bests

Marta

A quantitative research can be analyzed using the mean for the likert scale as the statistical tool

Thank you Leila Sylvia FILARCA Bautista !

Marta. The topic of your question is how to statistically handle ordered categorical data, since the so-called Likert scale is a numeric scale of the levels of agreement to statements. However the concept is used for numeric ordered categories as well. Ordered categorical data, e g Likert scales, have no standardized magnitude or equi-distance - only rank-invariant properties.

This means that if you treat data as if they have the same properties as quantitative data, you will get results that are not valid. Take care of your data properly - instead.

Svensson E. Guidelines to statistical evaluation of data from ratings scales and questionnaires. Journal of Rehabilitation Medicine 2001;33: 47-48.

Svensson E. Recent developments in analysis of paired ordinal data. Discussion of the paper Ivy Liu and Alan Agresti: The analysis of ordered categorical data: An overview and a survey of recent development. TEST 2005;14(1):1-73 pp 44-46.

Thank you very much Elisabeth Svensson for your precious advices!