Risk Ratio: 0.999 (95% CI: 0.993, 1.005)
0.996(95% CI: 0.989, 1.004)
0.985(95% CI: 0.886, 1.095)
0.953(95% CI: 0.828, 1.097)
1.059(95% CI: 0.910, 1.232)
1.054(95% CI: 0.863, 1.287)
Would you consider any of these above mentioned as borderline significant? and why?
BTW borderline significant is like beauty it's in the eye of the beholder. D. Booth
No (assuming the hypothesis is that the risk ratio is 1.0). Conventionally, 0.05 is just the borderline significance. It's that borderline, it's the worst possible outcome, that is conventionally considered statistically significant. All teh 95% CIs include the risk ratio of 1.0, so in all cases p > 0.05, so it's not even "borderline significant".
These ratios are incredibly close to 1, suggesting that whether they are significant or not, they are not important.
What are the predictor variables? The RR is the effect of a unit increase in the predictor variable, and some of these look like the measurement unit is pretty small (blood pressure, age maybe).
It is better to increase the size of your sample otherwise it is already done. The significance level set at 5% is conventional and the best threshold for p-value depends on the field and objective of your research. This means that if the threshold is set differently, everything changes in terms of significance.
- Badges
- Science topic