1
$\begingroup$

I am using SPSS for the first time and just trying to make sense of my findings.

I had two IV (nationality, age group), each with 2 levels (English-non English; young-old), which is why I ran a 2x2 ANOVA on SPSS. All my hypotheses were null, as no significance was found between the variables(nationality; age group; age group*nationality). However, the intercept has a p-value of p<.001

So, I'm a bit confused on how to interpret my data...

Improve this question
$\endgroup$
3
  • $\begingroup$ You might find some useful information on this other post there: enter link description here $\endgroup$
    – Timelate
    Commented Apr 20, 2022 at 7:52
  • $\begingroup$ As a side comment, you might check to see that your model is appropriate for the data you are studying. That is, think about the assumptions of homoscedasticity and normal distribution of errors. ... If you expect "significant" effects for these factors, and you don't find them, there may be several explanations, but sometimes it's because the model or model type isn't the best for the situation. $\endgroup$
    – Sal Mangiafico
    Commented Apr 20, 2022 at 14:26
  • 1
    $\begingroup$ You mention that nationality and age are independent variables, but that use of 'independent' is more a custom in controlled studies. Nationality and age are not controlled and in observational studies (which I believe you are dealing with) it might be that your 'dependent variable' has a causal effect on the observations of age and nationality. $\endgroup$
    – Sextus Empiricus
    Commented Apr 20, 2022 at 17:30

1 Answer 1

Reset to default
-2
$\begingroup$

It means the slope of your intercept was greater than 0 (and outside the confidence intervals too). Depending on your software your using, the intercept usually just happens to be the first variable in your dataset (this is how it works in R) unless you specify it otherwise.

Improve this answer
$\endgroup$
4
  • 2
    $\begingroup$ Intercept could well be smaller than zero. What is tested as a standard is an "equal to zero" null hypothesis. Also the term "slope" is normally used for the non-intercept terms in a regression. $\endgroup$
    – Christian Hennig
    Commented Apr 20, 2022 at 9:16
  • 1
    $\begingroup$ Furthermore there is the possibility of type I errors in tests, meaning that we do not know for sure that the slope is not zero, rather only that there is evidence in the data that it is. $\endgroup$
    – Christian Hennig
    Commented Apr 20, 2022 at 9:17
  • $\begingroup$ Please see stats.meta.stackexchange.com/questions/6304/my-upvoting-policy, when you find a question sufficiently clear to write an answer, consider to upvote the question! $\endgroup$
    – kjetil b halvorsen ♦
    Commented Apr 22, 2022 at 16:32
  • $\begingroup$ If you found this answer helpful, then please consider upvoting and/or accepting it. $\endgroup$
    – kjetil b halvorsen ♦
    Commented Apr 25, 2022 at 13:50

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.