Key Concepts in Regression Discontinuity

The key assumption is that subjects cannot manipulate the running variable that determines treatment assignment.

Researchers can perform balance tests on covariates and check the number of observations around the cut-off.

Fuzzy RD involves a discrete change in treatment intensity that affects the probability of receiving treatment, rather than a strict on/off eligibility at the cut-off.

<p>Researchers can compare the outcomes of subjects just below and above the cut-off using a scatter plot.</p> Signup and view all the answers

<p>Clear policy rules ensure that the cut-off is consistently applied, preventing manipulation and ensuring valid comparison of treatment effects.</p> Signup and view all the answers

<p>RD eliminates selection bias by observing treatment effects on groups just above and below a cutoff, controlling for unobserved variables.</p> Signup and view all the answers

<p>Graphical evidence in RD analysis visually illustrates the treatment effects and confirms the validity of the estimated impacts across the cutoff.</p> Signup and view all the answers

<p>A fuzzy RD design involves situations where treatment assignment is not strictly enforced at the cutoff, affecting the treatment probabilities.</p> Signup and view all the answers

<p>Credible RD estimates require a clear policy rule, no manipulation of the running variable, and a suitable specification and bandwidth.</p> Signup and view all the answers

<p>In the first stage, eligibility is regressed on treatment receipt; in the second stage, the treatment is regressed on the outcomes derived from the first stage.</p> Signup and view all the answers

<p>Simple OLS may overestimate the treatment effect because better scientists are more likely to be eligible and therefore receive the grant, skewing the results.</p> Signup and view all the answers

<p>Clarity in the policy rule ensures that the treatment assignment is well-defined, allowing for reliable comparisons around the cutoff.</p> Signup and view all the answers

<p>The objective is to determine how receiving a grant influences the quantity and quality of future scholarly outputs.</p> Signup and view all the answers

<p>RD eliminates selection bias by comparing subjects around the cutoff, ensuring that treatment and control groups are similar in characteristics.</p> Signup and view all the answers

<p>The balancing test checks if subjects on either side of the cutoff are similar in characteristics, confirming the validity of the RD design.</p> Signup and view all the answers

<p>LATE refers to the treatment effect estimated only for those individuals near the cutoff, who are most likely influenced by the treatment.</p> Signup and view all the answers

<p>Covariates should change smoothly around the cutoff to avoid omitted variable bias, ensuring they do not affect the treatment and outcome relationship.</p> Signup and view all the answers

<p>Clear policy rules ensure that the treatment assignment is transparent and consistent, which is critical for the validity of the RD analysis.</p> Signup and view all the answers

<p>Both methods aim to isolate the effect of the treatment, but RD uses a cutoff for treatment assignment instead of randomization principles.</p> Signup and view all the answers

<p>In an RD design, the graphical representation typically shows outcome values plotted against the running variable, with a noticeable jump at the cutoff indicating treatment effects.</p> Signup and view all the answers

<p>It means that treatment effects are estimated based solely on individuals whose values for the running variable are near the cutoff, rather than the broader population.</p> Signup and view all the answers