Part I: True or False (25 points) For each of the following statement, report whether the statement is true or false. If it is true, you do not need to explain your answer. If it is false, give a counterexample to the statement or briefly explain your reasoning. For example, the statement "The sky is always blue" is false, and a counterexample is "At night it is black, not blue" or "It is not blue when it is cloudy" Each question is worth 5 points 1. All else equal, if you have a smaller sample size it will be easier to reject a null hypothesis that a regression coefficient is equal to zero. 2. If you omit a variable from a regression that is correlated with your outcome variable, it will always bias treatment effect regression coefficients. 3. In a regression discontinuity model, increasing the bandwidth will lower standard errors, however it could also increase bias. 4. All else equal, the larger the effect that you expect an RCT will have, the larger sample size that you will need to identify that effect in the data. 5. You can only estimate treatment effect coefficients in a fixed effect model if the treatment changes across different observation with the same fixed effect