
| Resource Home| Documents Home |
November 30, 1998
Executive Summary
The " Accelerated Schools Program " was instituted at Tully Elementary School at the beginning of the 1994-95 school year. This inquiry was intended to investigate the effect that the Accelerated Schools Program has had on student achievement beyond what would be expected from more traditional (previous) forms of instruction. Inferring that the program has had a positive impact on student achievement would not only require observing an increase in student performance following institution of the program, but also superior student performance at Tully compared to other Elementary schools with similar student populations. Tully Elementary students were compared to the average of three matched control schools 1 on a standardized test of achievement over a period of six school years. Schools were matched according to their Stress Index 2 . The Stress Index for Tully was (-.308) indicating a moderate degree of stress. The average Stress Index for the three comparison schools was (-.312). Tully and the control schools were compared on their average Reading scores (Figure 1), Language scores (Figure 2) and Mathematics scores (Figure 3). There are several salient points to be noted in all three figures. First, Tully's performance on the Reading and Language components of the Stanford 9 increased following institution of the Accelerated Schools program in the 1994-95 school year. Second, academic performance of the control schools also increased during the same time. Finally, after adjusting 3 for the larger percentage of gifted students at Tully, performance was no longer superior for Tully students compared to the control schools. Results of this comparison found no evidence to suggest that the Accelerated Schools program impacted student achievement beyond what would be expected from more traditional approaches to instruction.
1
The three Elementary Schools that made a reasonably good match with Tully Elementary were: Myers/Ganoung, Howell, and Warren Elementary.2
The Stress Index indicates degree of stress in terms of SES, %SDPE, %Minority, %Mobility.3
The impact of the percentage of gifted students was determined and adjusted down to a level equivalent to that of the control schools. The impact was determined to be 9.4 points for Language, 12 points for Reading and 11 points for Mathematics based on 1997-98 Stanford 9 data.
A Six Year Comparison on Reading Scores
Between Tully Students and Students from Three
Matched Control Schools

Mean NCE
(n)|
Schools |
1992-93 |
1993-94 |
1994-95 |
1995-96 |
1996-97 |
1997-98 |
|
Tully Elementary |
48.17 (124) |
48.86 (81) |
47.25 (72) |
58.19 (54) |
49.05 (227) |
51.08 (247) |
|
Tully Adjusted 4 |
36.17 |
36.86 |
35.25 |
46.19 |
37.05 |
39.08 |
|
Matched Controls |
39.44 (174) |
41.11 (189) |
41.38 (185) |
44.87 (167) |
44.51 (557) |
45.56 (552) |
|
Mean Difference* |
8.73 |
7.75 |
5.87 |
13.99 |
4.54 |
5.52 |
|
Mean Difference** |
-3.27 |
-4.25 |
-6.13 |
1.32 |
-7.46 |
-6.48 |
*
Mean Difference = (Tully Elementary - Matched Controls)**
Mean Difference = (Tully Adjusted - Matched Controls)4
Reading means were adjusted by the formula: ReadAdjusted = ReadNCE + [(5.07 - 30.36) x .48]where 5.07 is the percentage of gifted students in the control schools, 30.36 is the percentage of gifted students in Tully and .48 is the regression coefficient representing the effect of gifted students on Reading scores.
A Six Year Comparison on Language Scores
Between Tully Students and Students from Three
Matched Control Schools

Mean NCE
(n)|
Schools |
1992-93 |
1993-94 |
1994-95 |
1995-96 |
1996-97 |
1997-98 |
|
Tully Elementary |
51.39 (124) |
49.70 (81) |
57.93 (72) |
60.89 (54) |
48.14 (227) |
47.99 (247) |
|
Tully Adjusted 4 |
41.99 |
40.30 |
48.53 |
51.49 |
38.74 |
38.59 |
|
Control Schools |
42.83 (174) |
42.35 (189) |
47.91 (185) |
48.17 (167) |
42.51 (557) |
44.74 (552) |
|
Mean Difference* |
8.56 |
7.35 |
9.39 |
12.72 |
5.63 |
3.25 |
|
Mean Difference** |
-0.84 |
-2.05 |
0.62 |
3.32 |
-3.77 |
-6.15 |
*
Mean Difference = (Tully Elementary - Matched Controls)
4
Language means were adjusted by the formula: LangAdjusted = LangNCE + [(5.07 - 30.36) x .37]where 5.07 is the percentage of gifted students in the control schools, 30.36 is the percentage of gifted students in Tully and .37 is the regression coefficient representing the effect of gifted students on Language scores.
A Six Year Comparison on Mathematics Scores
Between Tully Students and Students from Three
Matched Control Schools

Mean NCE
(n)|
Schools |
1992-93 |
1993-94 |
1994-95 |
1995-96 |
1996-97 |
1997-98 |
|
Tully Elementary |
53.60 (124) |
51.07 (81) |
60.19 (72) |
59.35 (54) |
47.89 (227) |
49.84 (247) |
|
Tully Adjusted 4 |
42.60 |
40.07 |
49.19 |
48.35 |
36.89 |
38.84 |
|
Control Schools |
44.57 (174) |
44.85 (189) |
48.87 (185) |
52.07 (167) |
40.44 (557) |
43.70 (552) |
|
Mean Difference* |
9.03 |
6.22 |
11.32 |
7.28 |
7.45 |
6.14 |
|
Mean Difference** |
-1.97 |
-4.78 |
0.32 |
-3.72 |
-3.55 |
-4.86 |
*
Mean Difference = (Tully Elementary - Matched Controls)
4
Mathematics means were adjusted by the formula: MathAdjusted = MathNCE + [(5.07 - 30.36) x .44]where 5.07 is the percentage of gifted students in the control schools, 30.36 is the percentage of gifted students in Tully and .48 is the regression coefficient representing the effect of gifted students on Mathematics scores.