The following instructions are for the appropriate assembly of student information in a comma-separated values (CSV) file, termed the "information file".

The DSPP machine learning model uses the data within the uploaded information file to compute predictions and append them to the end of the file once complete. For the machine learning model to provide the most accurate predictions, it is necessary to use the provided template CSV as a guide to input your data.

Please ensure that your information CSV file contains the exact column headers as shown in the template CSV. Failure to do so will result in inaccurate predictions.

General Information

---

• Each row in the CSV file represents a student, with each of their information parameters in a unique column in that row.
• Your CSV information file should only contain information from one STAAR subject test area over one grade. For example, an information file can include third-grade students' information and their math benchmark test scores from the same academic year. An information file should NOT include varying subject areas and or grade levels.
• Please avoid leaving any blanks in the CSV file. Doing so will yield an inaccurate result for that student.

Step 1

---

As seen in the column headers 'A' to 'F' of the information file CSV template, the following data is required for each student. Details on how to convey each information parameter can be seen in the table below.

Number	Header	Name	Values
1	A	Numerical Identifier	A number that can be used to reidentify students after receiving results while masking student identities. It is recommended to simply use the row count as seen in the template. Do NOT use student school ID numbers due to FERPA. Do NOT include actual student names due to FERPA.
2	B	Grade	A number corresponding to the student's grade. For example, if the student is in third grade, put 3.
3	C	Ethnicity	AM - American Indian or Alaska Native A - Asian B - Black or African American H - Hispanic or Latino or Spanish Origin P - Native Hawaiian or Other Pacific Islander W - White T - Two or more ethnicities
4	D	ECD	Economically Disadvantaged 1 - Yes 0 - No
5	E	EB	Emergent Bilingual 1 - Yes 0 - No
6	F	SpEd	Special Education 1 - Yes 0 - No

Step 2

---

Headers 'G' and 'H' should be populated with student test scores. These scores can be from previous STAAR tests, school district benchmark tests, or a combination thereof. Details on how to convey each test score can be seen in the table below.

Number	Header	Name	Values
7	G	BM 1	Benchmark 1 Enter the second most recent STAAR or benchmark test score for the corresponding subject. This number should be a percentage. For example, if a student got 18 out of 36 questions correct, enter 50. Do not include the percent sign. This number can be a decimal.
8	H	BM 2	Benchmark 2 Enter the most recent STAAR or benchmark test score for the corresponding subject. This number should be a percentage. For example, if a student got 18 out of 36 questions correct, enter 50. Do not include the percent sign. This number can be a decimal.

Step 3

---

Save the populated information file as a CSV. It can have any name.

Please note: Saving files in this format will delete any additional formatting, including other pages or tabs within the same spreadsheet. Copy and paste the data into a new, separate document if the file formatting and other pages or tabs are critical.

Step 4

---

Proceed to upload the CSV file on the "Upload an Information File for Machine Learning Prediction" page.