Dear Respectful Listers,
We are working on an analysis with ordinal data and are seeking your kind
advice and help.
We have a questionnaire with 30 questions. Ten of them are designed and put
under one section to assess the level of knowledge of farmers on an
agricultural pest management technique. Each question under this section
measure on 3 level (Yes, Maybe, No) and we plan to treat this as ordinal
data corresponding to score of 3, 2, 1, respectively). With ten questions
assessing preliminary knowledge of 100 farmers, we plan to test the
correlation between their level of confidence on a Likert scale of 5
(ranging from 1 to 5) and the total scores under the section to assess
knowledge (total maximun score for this ten-item section is 30 and minimum
score is 10).
We are planning to use Goodman Kruskal Gamma test to test the trend of
association between farmers knowledge (in terms of the knowledge level)and
their confidence (in terms of ability to apply the knowledge). We also plan
to user Spearman test to test the correlation.
Would it be possible for us to apply the above techniques? And there any
better techniques that you would recommend?
Thank you so much for your kind help and support. And We are looking
forwards to your help.
With our best regards,
Tu
--
View this message in context:
http://www.nabble.com/Relationship-on-Ordinal-Data-tf3546120.html#a9899202Sent from the SPSSX Discussion mailing list archive at Nabble.com.
Locked
