Talk

Advanced search

Threads started in this topic after 9th November 2018 will no longer be removed after 90 days. A new topic called 90 Days Only can be found in the Other Stuff category of Talk.

Anyone know anything about statistics (ANOVA) and spss

(7 Posts)
trexforever Sat 11-Aug-18 14:51:20

Help!

I have run an ANOVA and my results have revealed no significant difference between groups. However the estimated marginal means has revealed a significant difference between 2 of my subgroups although this isn't revealed in the standard post-hoc analysis.
Can anyone tell me what this means as I am none the wiser after Google.

TIA!

seeingdots Sat 11-Aug-18 16:37:56

I'm a frequent user of SPSS but it's been a long time since I've done an ANOVA so not sure if I can be any help but would you not normally run post hoc tests only if your ANOVA result indicated a significant difference? I don't know about marginal means but I'd think if you can't reject the null hypothesis from either the ANOVA or post hoc tests you don't have a significant result to speak of. What's your research on, out of interest?

DomesticAnarchist Sat 11-Aug-18 16:40:48

Ugh, this needs a trigger warning.

(I can't decide whether I'm being lighthearted or not, actually)

seeingdots Sat 11-Aug-18 18:30:06

@DomesticAnarchist grin

ItsTheGloaming Sat 11-Aug-18 18:32:36

I actually shuddered reading the thread title. Which is rediculus as I can't actually remember enough from my degrees to be of any useful help to you OP!

MedSchoolRat Sat 11-Aug-18 18:37:01

Does that mean you should run a t-test between the 2 subgroups that had the interesting MM difference? Or is t-test what you mean by "standard post hoc analysis"?

-Stata girl myself, but I respect SPSS for being able to handle files nothing else can

MedSchoolRat Sat 11-Aug-18 18:44:10

It seems like all these links are saying the EMMS tells you that some variable has been left out, but OP said ANOVA not GLM...

www.theanalysisfactor.com/why-report-estimated-marginal-means-in-spss-glm/

I hate stackexchange... any website where your every post gets graded by Internet Randoms is Total Yick. That said, it's a good place to pick up info.

This thread any help?

or... stats.stackexchange.com/questions/124674/spss-calculation-of-confidence-intervals-in-estimated-marginal-means-of-repeat

Join the discussion

Registering is free, easy, and means you can join in the discussion, watch threads, get discounts, win prizes and lots more.

Register now »

Already registered? Log in with: