0
$\begingroup$

I've used a prop.test to compare proportions. I understand the results but I don't know how to reports the statistical results in my thesis:

The results show a significant difference, the proportion in group 2 is greater than in group 1 (χ2 (1)= 38.764, p < 0.001).

(I've turned off the Yates' continuity correction, since the numbers are above 10. I hope this was correct)

enter image description here

$\endgroup$

2 Answers 2

1
$\begingroup$

I'm a big fan of reporting

a) The estimate of the difference, and b) A confidence interval for that difference.

So you could say something to the effect of

I used a two sample test of proportions to estimate the difference between groups. I estimate that the difference in betweeen and name is 0.082 (95% confidence interval: 0.056 - 0.100).

I'm not a big fan of reporting p values anymore, though some journals may require them. Whatever you do, just make sure you use appropriate significant figures (I am, and always have been, bad at that).

$\endgroup$
2
  • $\begingroup$ They should also list the distribution and degrees of freedom / test used to find the confidence interval. $\endgroup$
    – M Waz
    Jul 18, 2019 at 16:04
  • $\begingroup$ Oh my bad, yea TIL prop.test uses a chi-squared test. Yea, those things should be listed then. $\endgroup$ Jul 18, 2019 at 16:19
0
$\begingroup$

You can simply say:

The estimated difference in proportions between group 2 and group 1 is 0.082 (95% confidence interval of 0.056 and 0.100, $\chi^2$ = 38.784 with one degree of freedom from Pearson's Chi-Squared test).

$\endgroup$

Your Answer

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

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