Hi
For my research work I have a Teacher Credibility Questionnaire, which is semantic differential or bipolar adjective questionnaire. My question is how to enter the data in SPSS software and how to measure the reliability?.
In the attached section you can see the questionnaire.
Really thanks for your answers in advance.
Hi,
For such a scale (it seems to assume a Likert nature), you may have a look at the original published literature to see if the primary investigators have conducted psychometric analysis. If it's validated, try communicating with the investigators to get the original scoring methods.
For such a scale, data entry in SPSS should be as of an individual variable with the particular response score. For e.g, irritable, gloomy, etc..each will be a single variable and the response score are entered separately. You may then tabulate the score to check for normality.
For reliability, you may use Cronbach's alpha to determine the scale's consistency.
Good luck!
This type of data is called a "semantic differential."
AS Kurubaran Ganasegeran recommends, I would treat it as a set of LIkert-scored items. You could indeed look at a Cronbach's alpha for the entire data set, but I would also person am Exploratory Factor Analysis to determine where all the items measure the same underlying concept. If you do perform an EFA in SPSS, I recommend not using the defaults; instead, do a Principal Factors with oblique rotation.
Dear Farzad,
A few points to add to the previous answers:
Hi again.
Thanks for your answers.
Dear Peter about 2nd point of your reply to me I should mention that half of the positive items are designed in one side to encourage the respondents to read each item. I have translated that questionnaire into Persian in order to remove the linguistic barrier for respondents. This is the picture of the first version of the questionnaire.
I found someone had coded 1&7 as extremely and 2&6 as moderately and 3&5 as fairly and 4 as neutral for a similar semantic differential questionnaire. Is this type of scoring correct? shall I do the same?
I would be thankful if you answer me more.
I don't see any reason to "collapse" the categories, since this amounts to throwing away data.
Sorry Dear David L Morgan
I didn't get what you mean....
I need more ... .
thanks
This is similar to a typical theory of planned behavior questionnaire (TPB). Check out the use in on of the thousands of articles with that theory. You don't need to code them on the survey (neither in SPSS), the semantic differentials on each line is enough. See examples below.
Data Theory of Planned Behavior measure
Article Predicting patient post-detoxification engagement in 12-step...
Best
John Kåre
- Badges
- Science topic
- Similar topics
- Mathematical Sciences
- Statistics
- Statistical Software
- SPSS