Student Growth Model (SGM) End-of-Course Results

Student growth percentiles on Georgia Milestones end-of-course tests by school, district, and subject, 2015-2023.

Source: Georgia Insights (GaDOE) — Georgia Student Growth Model on End-of-Course assessments
Year coverage: 2015–2023
Updated: annual
Version: 1.1.0
Schema hash: b61cbc13276d

Detail levels available

StateDistrictSchool · default

Overview

Georgia Student Growth Model (GSGM / Student Growth Percentile) results on Georgia Milestones End-of-Course assessments, at the state, district/system, and school level per subject. Core metrics are the median Student Growth Percentile (1-99), the count of students receiving an SGP, and growth-share metrics. 2015-2019 publish the count and share of tested students with an SGP, two achievement-level cumulative shares (Proficient / Developing Learner or above), and the share with Typical-or-High growth (SGP >= 35). The Full-Year 2022-2023 report (year 2023) redesigns the metric set: a three-band SGP split (Low < 35, Typical 35-65, High > 65) replaces the single typical-or-high share, and the tested counts, received-SGP share, and achievement shares are no longer reported. No GSGM was published for 2020-2022 (COVID). Subject coverage narrows from 8-10 subjects (2015-2016) to six (2017-2019) to two algebra courses (2023).

Row grain: One row per year, district_code, school_code, subject.

Key metric

sgp_median — SGP Median (percentile, 1–99)

The single headline value most answers want.

JSON schema →Data contract (YAML) →

Variables

Every column in the fact table. The code is the exact field name the API returns.

Column	Type	Unit	Range	Null?	Description
`year`Year	integer	—	—	no	Ending calendar year of the school year (e.g., 2017 = school year 2016-2017). Published years: 2015-2019 and 2023 (the Full-Year 2022-2023 report). No 2020-2022 data exists. Full description▾ Ending calendar year of the school year (e.g., 2017 = school year 2016-2017). Published years: 2015-2019 and 2023 (the Full-Year 2022-2023 report). No 2020-2022 data exists.
`district_code`District Code	string	—	—	yes	3-digit GOSA district/system code (FK to districts dimension). School-level rows for State/Commission Charter campuses use the 7-digit campus code (system 782/783 + school code) — the 2015-2017 School files published those rows under the bare system code inside the compound Key and they are promoted here. NULL for state-level rows. Full description▾ 3-digit GOSA district/system code (FK to districts dimension). School-level rows for State/Commission Charter campuses use the 7-digit campus code (system 782/783 + school code) — the 2015-2017 School files published those rows under the bare system code inside the compound Key and they are promoted here. NULL for state-level rows. Joins districts; adds its label columns.
`school_code`School Code	string	—	—	yes	4-digit GOSA school code (FK to schools dimension, composite with district_code). NULL for district- and state-level rows. Full description▾ 4-digit GOSA school code (FK to schools dimension, composite with district_code). NULL for district- and state-level rows. Joins schools; adds its label columns.
`subject`Subject	string	—	—	no	The end-of-course subject (e.g. algebra_i, biology, american_literature_and_composition). Full description▾ Snake-case EOC subject. 2015 reports eight subjects (9th_grade_literature_and_composition, american_literature_and_composition, coordinate_algebra, analytic_geometry, biology, physical_science, us_history, economics_business_free_enterprise); 2016 adds algebra_i and geometry (ten); 2017-2019 report the six literature/math subjects; 2023 reports only coordinate_algebra and algebra_i. 10 allowed values▾ `9th_grade_literature_and_composition` `algebra_i` `american_literature_and_composition` `analytic_geometry` `biology` `coordinate_algebra` `economics_business_free_enterprise` `geometry` `physical_science` `us_history`
`num_tested`Number Tested	integer	count	—	yes	Number of students who took the EOC assessment. Never suppressed where published. NULL in 2023 (the Full-Year 2022-2023 report does not publish tested counts). Full description▾ Number of students who took the EOC assessment. Never suppressed where published. NULL in 2023 (the Full-Year 2022-2023 report does not publish tested counts).
`num_sgp_received`Number SGP Received	integer	count	—	yes	Number of students with enough prior-year scores to receive a Student Growth Percentile (a subset of num_tested). Suppressed (NULL) for small cohorts in the 2019 files only. Full description▾ Number of students with enough prior-year scores to receive a Student Growth Percentile (a subset of num_tested). Suppressed (NULL) for small cohorts in the 2019 files only.
`sgp_received_rate`SGP Received Rate	number	proportion	—	yes	Share of tested students who received an SGP (num_sgp_received / num_tested), 0-1 scale (bronze 0-100 divided by 100). NULL in 2023 (not reported). The 9999 sentinel note on sgp_median also applies (11 NULLed values in 2016 Physical Science system rows). Full description▾ Share of tested students who received an SGP (num_sgp_received / num_tested), 0-1 scale (bronze 0-100 divided by 100). NULL in 2023 (not reported). The 9999 sentinel note on sgp_median also applies (11 NULLed values in 2016 Physical Science system rows).
`sgp_median`SGP Median	number	percentile	1–99	yes	Median student growth percentile (1-99); 50 is average growth, higher means faster growth than similar peers. Full description▾ Median Student Growth Percentile on the 1-99 percentile-rank scale (preserved verbatim — not a share; there is no SGP of 0 or 100, so the range guard is tightened from the percentile default [0, 100] to [1, 99]). Float64 because bronze publishes half-point medians (e.g., 43.5). The 2016 System Physical Science sheet stored an undocumented 9999 suppression sentinel in 11 rows — impossible on this scale, NULLed per §4b.
`pct_proficient_learner_or_above`Percent Proficient Learner or Above	number	proportion	—	yes	Share of tested students at the Proficient Learner achievement level or higher on the EOC assessment itself (0-1 scale). Published 2015-2019 only; NULL in 2023. The 9999 sentinel note on sgp_median also applies (11 NULLed values in 2016 Physical Science system rows). Full description▾ Share of tested students at the Proficient Learner achievement level or higher on the EOC assessment itself (0-1 scale). Published 2015-2019 only; NULL in 2023. The 9999 sentinel note on sgp_median also applies (11 NULLed values in 2016 Physical Science system rows).
`pct_developing_learner_or_above`Percent Developing Learner or Above	number	proportion	—	yes	Share of tested students at the Developing Learner achievement level or higher on the EOC assessment (0-1 scale). Published 2015-2019 only; NULL in 2023. The 9999 sentinel note on sgp_median also applies (11 NULLed values in 2016 Physical Science system rows). Full description▾ Share of tested students at the Developing Learner achievement level or higher on the EOC assessment (0-1 scale). Published 2015-2019 only; NULL in 2023. The 9999 sentinel note on sgp_median also applies (11 NULLed values in 2016 Physical Science system rows).
`pct_sgp_typical_or_high_growth`Percent SGP Typical or High Growth	number	proportion	—	yes	Share of SGP-scored students with SGP >= 35 (Typical or High growth), 0-1 scale. Published 2015-2019 only — 2023 replaces it with the three-band split. NOT equal to pct_sgp_typical_growth, which excludes High. The 9999 sentinel note on sgp_median also applies (11 NULLed values in 2016 Physical Science system rows). Full description▾ Share of SGP-scored students with SGP >= 35 (Typical or High growth), 0-1 scale. Published 2015-2019 only — 2023 replaces it with the three-band split. NOT equal to pct_sgp_typical_growth, which excludes High. The 9999 sentinel note on sgp_median also applies (11 NULLed values in 2016 Physical Science system rows).
`pct_sgp_low_growth`Percent SGP Low Growth	number	proportion	—	yes	Share of SGP-scored students in the Low Growth band (SGP < 35), 0-1 scale. 2023 only; NULL for 2015-2019. Full description▾ Share of SGP-scored students in the Low Growth band (SGP < 35), 0-1 scale. 2023 only; NULL for 2015-2019.
`pct_sgp_typical_growth`Percent SGP Typical Growth	number	proportion	—	yes	Share of SGP-scored students in the Typical Growth band (35 <= SGP <= 65), 0-1 scale. 2023 only; NULL for 2015-2019. Excludes High growth — not comparable to pct_sgp_typical_or_high_growth. Full description▾ Share of SGP-scored students in the Typical Growth band (35 <= SGP <= 65), 0-1 scale. 2023 only; NULL for 2015-2019. Excludes High growth — not comparable to pct_sgp_typical_or_high_growth.
`pct_sgp_high_growth`Percent SGP High Growth	number	proportion	—	yes	Share of SGP-scored students in the High Growth band (SGP > 65), 0-1 scale. 2023 only; NULL for 2015-2019. Full description▾ Share of SGP-scored students in the High Growth band (SGP > 65), 0-1 scale. 2023 only; NULL for 2015-2019.

Filters

Query parameters the fact endpoint accepts. Comma-separated values are OR within a parameter; multiple parameters AND together.

Parameter	Kind	Allowed values	Notes
`year`	year_exact	any value	—
`year_min`	year_range	any value	—
`year_max`	year_range	any value	—
`detail`	detail	`districtsschoolsstates`	enum-enforced, default: schools
`district_code`	foreign_key	any value	multi-value
`school_code`	foreign_key	any value	multi-value
`subject`	categorical	10 allowed values▾ `9th_grade_literature_and_composition` `algebra_i` `american_literature_and_composition` `analytic_geometry` `biology` `coordinate_algebra` `economics_business_free_enterprise` `geometry` `physical_science` `us_history`	multi-value, enum-enforced, pick one
`district_type`	dimension_attribute	7 allowed values▾ `commission_charter` `resa` `standard` `state_agency` `state_charter` `state_school` `state_special`	multi-value, enum-enforced
`school_level`	dimension_attribute	`elementarymiddlehighk12other`	multi-value, enum-enforced
`locale`	dimension_attribute	`citysuburbtownrural`	multi-value, enum-enforced

Example requests

Run these against the base URL — they return live data.

All rows (first page), schools level

curl "https://georgiacivicdata.org/api/v1/education/georgia_student_growth_model_end_of_course"

schools data for 2023

curl "https://georgiacivicdata.org/api/v1/education/georgia_student_growth_model_end_of_course?year=2023"

Filtered to subject = 9th_grade_literature_and_composition for 2023

curl "https://georgiacivicdata.org/api/v1/education/georgia_student_growth_model_end_of_course?year=2023&subject=9th_grade_literature_and_composition"

One district (code 601), 2023

curl "https://georgiacivicdata.org/api/v1/education/georgia_student_growth_model_end_of_course?district_code=601&year=2023&detail=districts"

Download 2023 as CSV

curl "https://georgiacivicdata.org/api/v1/education/georgia_student_growth_model_end_of_course?year=2023&format=csv"

Notes & limitations

NULL handling

Suppressed cells (too few students to report) are NULL, never zero.
Zero is a real, reported value.

Limitations

Suppressed cells are NULL (not zero). State rows have NULL district_code and school_code. District rows have NULL school_code. No GSGM data exists for 2020-2022 (not reported during the COVID years); gold emits partitions only for 2015-2019 and 2023. 2023 (Full-Year 2022-2023) covers only coordinate_algebra and algebra_i and redesigns the metric set: the three SGP band shares replace pct_sgp_typical_or_high_growth, and num_tested / sgp_received_rate / the achievement cumulatives are not reported. All percent columns are 0-1 decimal scale (bronze ships 0-100); sgp_median keeps its natural 1-99 percentile-rank scale and can carry half-point values. Suppression markers vary by era ('----' in 2015-2016 and residually in 2017, 'TFS' in 2017-2019, '--' in 2023, 33 stray 'NULL' literals in 2017 School, and an undocumented 9999 numeric sentinel in the 2016 System Physical Science sheet that is NULLed in all five rate/median columns on 11 rows) - all land as NULL in gold. School-level rows for State/Commission Charter campuses (systems 782/783) are keyed on the 7-digit campus district code; 2015-2017 bronze published them under the bare system code and they are promoted during transform (ledgered as manifest reclassified events).