Hey there,
for my bachelor thesis I made a survey using 5- point Likert Items from strongly disagree (1) to strongly agree (5). Is there any consensus on how to interpret using quartils and the IQR, I thought of something like:
Q3 equals 2,00: overall strong disagree
Q2 equals 2,00: overall disagree
Q2 equals 3,00: no statement or tendency
Q2 equals 4,00: overall agree
Q3 equals 4,00: overall strong agree
or is there any literature about how to interpret 5 point likert using Quartiles ?
and about the consens, is there any literature how to interpret the IQR as a tool of consensus ?
Thanks in advance and best wishes
Daniel
Daniel, I'm afraid your proposal makes no sense as far as I can see. Refer to the definition of quartile. D. Booth
Daniel, as these are the qualitative data you can't do mathematical computation on the same. Hence, please avoid it.
You can classify the responses by three categories, High (4,5), medium(3) and low(1,2) and calculate percentage of each category. This will help you.
Also, highest response category under a question gives the concensus.
There cannot be much of a consensus, because it is not well defined how, for instance, "agree" differs from "strongly agree" etc. The best you can do is to describe the distribution of answers you got. How you interpret it, depends on the subject, the context and on you. You can always write what's your interpretation so that any reader recognized this as your interpretation, and your thesis can (and should) leave it open to have other interpretations (and possibly come to different conclusions). You can (and should) use the discussion to place different possible interpretations side by side and, well, discuss them.
Your data has another problem, if I see that correctly: the center value (you call it "3" I think) seem to be defined as "no statement or tendency". This means, a person who does not like to state his/her degree of agreement (no matter what it is!) can answer "3". This is actually a missing value and does not indicate that the person is undecided or deos neither agree nor disagree. So "3" as a respnse can stand for three entirely different things, and in your analysis you mixe these all together, waht can severely bias your interpretation.
- Badges
- Science topic
- Similar topics
- Philosophy
- Esthetics
- Interpretation