## Tuesday, July 7, 2015

### All Data Are Not Created Equal

Weighting the Educator Student Growth Performance Score

Purpose: To adjust the Student Growth based on the number of students.

By analyzing multiple evaluation systems, we discovered that the evaluation method used to calculate teacher performance is not adequate. The intent behind this article is to offer a different   method to deal with data when there is a combination of performance factors. This process will equalize the contribution that students’ growth has on the final results when calculating the total student growth associated with the determination of the educator’s final student growth performance. This situation happens when an educator has mixed sections with different numbers of students or when a teacher is teaching multiple subjects that have various numbers of students in each class. This method will give the right weight that the amount of students has when calculating the final results of the educator’s student growth performance.  The data is corrupted if the evaluation system does not take into consideration this adjustment. Such corruption is a threat that adds a fatal error to the calculation of the educator’s student growth performance and needs to be corrected before the final student growth performance is reported.

Case scenario—Mr. Thomas’ mixed classes:

Mr. Thomas is teaching the same subject (reading), but he has two classes, a 4th grade class and a 5th grade class. The classes have a different number of students. After calculating the student growth using the Value Added Model (VAM) data, each group had a different student growth performance as follows:

 Grade Subject Number of Students in Class VAM Student Growth Measure (SGM) Performance 4 Reading 4 60 5 Reading 14 98

Threat to the final performance evaluation

The common error happens when a mixed SGM result of either data sets (or multiple data sets) is treated as if they contributed equally to the final average results.  As an example,  using the previous SGM to calculate the final student growth performance, and treating the data as if both SGM contribute equally, the result is (60 + 98) / 2 = 79. Based on this calculation, Mr. Thomas’ final student growth performance score is 79. This calculation has a fatal error that requires weighting the data according to the number of students in each class.

Weighted data procedure

The student growth of 14 students does not contribute equally to the final performance when it is compared with the student growth of 4 students.  It’s not the influence of the teacher that’s the concern here, but rather the contributory value given to the two sets of data. The data must be weighted in order to obtain the correct final performance value. In order to obtain the final performance value, the following steps should be used to correctly evaluate Mr. Thomas’ student growth performance.

I- Determining the contribution factor

1) Add the total students in each class. (4 + 14 = 18) This number represents 100% of the students or total students.

2) Divide the number of students in each class by the total number of students as a method to determine the contribution factor for each group of students:

(a) 4th-Grade Class (4/18 = .22)
(b) 5th-Grade Class (14/18 = .78)

II- Implementing the contribution factor

3) Now take the SGM Performance for each grade level (or subject) and multiply it by the contribution factor in order to obtain the real contribution that each group of students has on the educator’s final performance:

(a) 4th-grade class contribution: (60 X .22 = 13.2)
(b) 5th-grade class contribution: (98 X .78 = 76.44)

4) The addition of both (or multiple) contributions is the correct final student growth performance for Mr. Thomas. This method will eliminate the un-weighted error.

(a) (13.2 + 76. 4 = 90): Now based on this calculation,
Mr. Thomas’ final student growth performance score is 90

Contribution Factor for Final Performance Summary
 SGM Contribution for each grade or subject (Multiply by the) Contribution Factor Total student growth contribution per class 60 .22 13.2 98 .78 76.44 Weighted Performance Score 89.6

Citation required: Raul Baez Hernandez (2014). Weighting the educator student growth performance score. Coach One Evaluation System