CSPR Tables:

Students served (long-term student count; tables 2.4.1.2/2.4.2.2)

Academic performance in reading and mathematics (tables 2.4.1.6.1/2.4.2.6.1 and 2.4.1.6.2/2.4.2.6.2)
EDFacts File Specifications:

C113 & C125: Neglected or delinquent (N or D) academic achievement

C135: N or D long-term
Reporting high-quality and accurate data for the number of students with complete pre- and posttest results, as well as the number of students demonstrating improvement, is a critical component of the Title I, Part D, program performance. This tool provides sample calculations that can be used to avoid common data collection errors and increase the likelihood that the academic achievement data reported are accurate and consistent.
Reporting pre- and posttest results in C135 are optional for at-risk programs.

Two of the improvement levels for reading and mathematics—“up to one ½ grade level” and “1/2 to one full grade level”—have been merged into a single level, “up to one full grade level.”
Results for academic achievement in reading and mathematics are based on students who have complete pre- and posttest data available. Consequently, academic achievement data should meet the following two criteria:

The number of students with complete pre-post test data should not exceed the number of long-term students.

The number of students with complete pre- and posttest results should always equal the sum of long-term students who had any pre- and posttest results (e.g., results across negative change, no change, or improvement) reported for mathematics and reading.
Performing a few simple calculations before submitting academic achievement data will help to identify any inconsistencies to make sure that the number of students who have complete pre- and posttest data aligns with the number of students demonstrating results. Exhibit 1 provides sample calculations to help guide technical assistance around these data.
Last Updated July 2013
This NDTAC tool is intended as a quick reference and does not supersede or replace any Federal forms or information about the collection and reporting of data pertaining to Title I, Part D.
1

Using the data in Exhibit 1, the example below provides step-by-step information for data providers on how to avoid common errors when reporting pre- and posttest data.
Exhibit 1: Sample Information for Calculating the Percentage of Students Who Are
Improving in Adult Corrections Programs (Subpart 1 Only)
CSPR Measure Number of Students
1. Number of long-term students who tested below grade level upon entry 100
2. Number of long-term students with completed pre- and posttest results (data)
3. Number of long-term students who demonstrated negative grade-level change from the pre- to posttest exams
4. Number of long-term students who demonstrated no change in grade level from the pre- to posttest exams
75
25
0
5. Number of long-term students who demonstrated improvement up to one full grade level from the pretest
33
6. Number of long-term students who demonstrated improvement of more than one full grade level from the pre- to posttest exams
35
The calculation below results in a reasonable outcome percentage for the number of students who are demonstrating improvement for academic achievement:
(rows 5 and 6) = 33 + 35 = 68
Number of long-term students who have complete pre-and posttest data (row 2) = 75
Nonetheless, data reported in Exhibit 1 presents a data quality issue because the number of long-term students with data is lower than the total number of students demonstrating results , as shown in the calculations below:
Number long-term students who have complete pre- and posttest data (row 2) = 75
Number of long-term students with reported results (rows 3–6) = 25 + 0 + 33+ 35 = 93
If the results reported in rows 3–6 are correct, then the percentage of students demonstrating improvement should be:
Inconsistencies between the number of students with complete results and the actual results provided (as demonstrated in
Exhibits 1 and 2) can call into question the accuracy and quality of the data reported.
The number of students with complete pre- and posttest data should always equal the sum of all the result counts provided for long-term students (e.g., results across negative change, no change, or improvement, or rows 3–6) in mathematics and reading.
Data providers should review their data for consistency. If the misalignment cannot be resolved, a comment should be provided in the CSPR that explains the discrepancy and what can be done to resolve the data quality issue.
Last Updated July 2013
This NDTAC tool is intended as a quick reference and does not supersede or replace any Federal forms or information about the collection and reporting of data pertaining to Title I, Part D.
2