e student question

Indicate whether they are true or false, and briefly explain WHY

1. If an observation is more than three standard deviations from the mean, it is an outlier and must therefore be eliminated from the sample.

2. If the only results that get reported are those that pass a P < .05 filter, there will be a tendency for them to overestimate the size of the true effect.

3. When a researcher finds that several models can plausibly estimate an effect size of interest, best practice is to select the model thought to be superior overall and report only its results.

4. A study is performed to determine whether a new food additive is carcinogenic. The finding is that the estimated effect of ingesting this additive on the likelihood of developing cancer is not statistically significant. This is evidence that the effect is nonexistent.

5. Suppose a statistical model can account for 80% of the variation in the outcome variable of interest; this does not mean that the model necessarily "explains" most of this variation in the normal, everyday sense of "explain".