This prevented cognitive factors as those mentioned above to bias our results, but importantly, it also prevented subjects of using these landmarks to estimate moving speed.
The scene we used provided rich visual flow information but was clear of any landmarks to prevent subjects using cognitive strategies to estimate moving speed.
(1)This value was first calculated per marker and then used to estimate moving averages over a window of consecutive markers, which is the measure used to infer selective sweeps in our analysis.
We must also assume that the risk ratio estimates the effect in the target population, that there is no bias, and that there is no change in the effect estimate moving from the study to the target population (Greenland 2004).
This suggests, in the absence of confounding amplification, that the tight control of this covariate in Model 2 would ordinarily result in a less biased treatment effect estimate moving towards, not away from, the null.
If it had begun 2 weeks earlier than the actual date, the number of cases prevented would have been ≈59% greater than the base estimate; moving the program ahead by 8 weeks would have resulted in a ≈306% increase in cases prevented compared with the base estimate.
