Spss

[Susanbwh]

I would like some help on the following:

1) I am having trouble with finding the way to incorporate SD line (vertical) in the line graph in SPSS. Could you guide me to the right place?

2) I have ordinal variables (items), but if I added the scores of a few or several items to make up a factor or domains for example, the nature of the ordinal variables become more continous and can be treated as such in SPSS?

3) When I compare two populations (one with 150 people, and more normal and the other with 40 people with some illness) using t test (I guess I should use t test rather than z), how can I get the effect size (how much they differ) for certain variable? Should I enter into SPSS the Mean and SD from each population and then compare?

Please help me on these three things. Thank you.

[JohnM]

(1) I'm not sure what you're referring to here....

(2) That should be OK - just clearly note in your report what you did.

(3) For either a t-test or z-test, I believe SPSS should provide the effect size within the output under the column called "Mean Difference"

http://academic.uofs.edu/department/...9/level2b.html

[Susanbwh]

Thank you,
1) I am talking about the standard deviation (SD) line. I see some people present beautiful graphs (line) with vertical line for each X value. I just wonder if SPSS has the function.
3) Please forgive me for asking this again, but how do I enter and compare the data that are in separate file?

[JohnM]

Those are called "error bars" - I'm pretty sure SPSS has that feature, because I've used that graph before.

To compare data from another file, I would just copy/paste the data into the same file - make new variables or case descriptions if necessary.

[Susanbwh]

Thank you, Joh, I am impressed! I will look for the error bars.