Latin American Economic Review

Table 4 Summary statistics on average scores of 9th graders in math

From: Closing the achievement gap in mathematics: evidence from a remedial program in Mexico City

	Within treated schools			Between schools			All the schools in the sample
	Treatment	Control	Simple differences (1–2)	All in treated schools	All in non-treated schools	Simple differences (1–5)	All the schools in the sample
	(1)	(2)	(3)	(4)	(5)	(6)	(7)
Pre-treatment
Partial 1	6.296 [1.273]	6.856 [1.424]	−0.560*** [0.057]	6.797 [1.420]	6.923 [1.533]	−0.619*** [0.058]	6.904 [1.505]
Partial 2	6.261 [1.190]	6.918 [1.746]	−0.6572*** [0.069]	6.866 [1.671]	6.932 [1.756]	−0.647*** [0.068]	6.911 [1.733]
Partial 3	6.222 [1.213]	6.635 [2.108]	−0.413*** [0.082]	6.588 [2.029]	6.807 [2.032]	−0.556*** [0.079]	6.761 [2.034]
Average pre-treatment	6.260 [1.047]	6.803 [1.480]	−0.544** [0.058]	6.758 [1.427]	6.908 [1.550]	−0.607*** [0.059]	6.868 [1.506]
Post-treatment
Partial 4	6.511 [1.270]	6.645 [2.426]	−0.134 [0.094]	6.630 [2.322]	6.751 [2.322]	−0.204** [0.090]	6.728 [2.322]
Partial 5	6.718 [1.560]	6.695 [2.630]	0.024 [0.103]	6.698 [2.530]	6.897 [2.516]	−0.139 [0.098]	6.863 [2.522]
Average post-treatment	6.615 [1.209]	6.670 [2.436]	−0.056 [0.094]	6.694 [2.296]	6.823 [2.304]	−0.171* [0.089]	6.817 [2.311]
Observations	689	5,258	–	5,947	16,278	–	22,225

This table gives the mean scores for the five partial tests (and the average of the test taken before and after the intervention) for treatment and comparison students. Columns (1)–(3) show the average scores in the treatment group, the group of non-treated students in participating schools, and the difference between the two grade averages. Columns (4)–(6) show the average scores for all students in treated schools, and all non-treated students in all schools, and the simple difference between the last group and the treatment group. Column (7) shows mean scores for all students, treated and non-treated, from all the schools in the sample. Standard errors are in brackets
* Significant at 10 %; ** significant at 5 %; *** significant at 1 %

Back to article page