3. Suppose that a school district conducts a randomized control trial of a reading intervention for

low-performing 8th-grade students. To determine whether the intervention improved reading test

scores, the district's data analysts estimated a regression using ordinary least squares (OLS).

The results are summarized in the table below:

Reading Test Score

Treatment

5.231

(2.015)

Intercept

50.812

(12.887)

Observations

956

Note:

Standard errors in parentheses.

(a) Write down the regression model the analysts estimated.

(b) What is the average reading test score of students in the control group after the intervention?

(c) What is the estimated treatment effect?

(d) Using the rule of thumb discussed in class, determine whether the treatment effect is sta-

tistically distinguishable from zero. Explain.

(e) What is the average reading test score of students in the treatment group after the inter-

vention?

(f) Under what conditions would the randomized control trial isolate the causal effect of the

reading intervention on reading test scores?

(g) Suppose that the randomly selected control group happened to have twice as many non-

native English speakers as the randomly selected treatment group. Non-native speakers

score lower on standardized (English) reading tests, on average. In light of this information,

is it likely that the estimated treatment effect isolates the effect of the intervention? Why

or why not? If not, what is the sign of the bias?