10) In the first division of the sample, students who actually experienced FBTA are separated from those who did not.
FBTA clearly increase the probability of making a lower grade for the American students.
A significant finding here is that, in the math-track classes (column 2), FBTA does not seem to affect the probability of a grade significantly, perhaps due to the diminished importance of English language skills in mathematics classes.
The result for the FBTA confirms that FBTA adversely affect student grades in regular classes.
11) The focus here is on the FBTA variable because it turned out to be the only statistically significant explanatory variable related to the FBTA and the perception of their class performance.
Borjas' paper, although not the first, attempts to investigate the effect of FBTA on student learning, which is a sensitive and controversial issue in academia.
foreign-born students shows that the negative effect of FBTA is more pronounced on the American students than on the foreign-born students.
Nevertheless, for FBTA, cultural differences and lack of familiarity with American institutions remain barriers to more effective teaching in undergraduate principles classes.
For the FBTA in particular, familiarity with the culture and with private as well as government institutions requires time and personal experience.
If FBTA tend to be tough graders, and indeed poor students seek scapegoats, then grade, communication and preparation variables are endogenously determined.
To explore the possibility that multicollinearity is masking the statistical significance of COM2, additional models are run without the FBTA for the full sample, U.
Similar to Borjas' Column vi model, an interaction variable between FBTA and COM2 is added, and as a result, the coefficient of the interaction variable and FBTA become statistically insignificant, which could be due to multicollinearity observed between the two variables based on the values of the simple correlation coefficient and the VIF.