1
$\begingroup$

Technically, endogeneity occurs when a predictor variable (x) in a regression model is correlated with the error term (e) in the model. This can occur under a variety of conditions, but two cases are especially common in inequality research: (1) when important variables are omitted from the model (called “omitted variable bias”) and (2) when the outcome variable is a predictor of x and not simply a response to x (called “simultaneity bias”). At least part of the latter problem is often called “selection.”

Hello, With reference to the above, can I ignore simultaneity bias if my key independent variable occurs before the dependent variable? As such, if my independent variable is a requirement that remains fixed, and I want to study the impact of such requirement on the performance or outcome. Also, there are no empirical studies that determines the level of requirement. With the case I am working on, the requirement varies at a group level. Some groups have higher requirement and some lower and I want to test the level of requirement on their performance. There is no disclosure as to how this requirement is determined for different groups. From private sources I know that the groups can negotiate by demonstrating why the requirement must be lowered or can willingly accept a higher requirement.

$\endgroup$

1 Answer 1

0
$\begingroup$

You should ask yourself: "who decided to impose such requirements? what was the ratio behind them? is the reason related to variables that I'm not including in my model (among the $X$s) and to the outcome variable?". If the requirements happen to be motivated by previous outcomes, then you have omitted variable bias. If the requirements are decided exogenously/independently of your outcome measure, this would not be a potential channel of bias (there could be others though, as usual).

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Not the answer you're looking for? Browse other questions tagged or ask your own question.