Skip to content

Data Analysis

Jump to bottom
Adam VanIwaarden edited this page Aug 24, 2016 · 8 revisions

The objective of the student growth percentile (SGP) analysis is to describe how (a)typical a student's growth is by examining his/her current achievement relative to students with a similar achievement history; i.e his/her academic peers (see this presentation). The estimation of this norm-referenced growth quantity is conducted using quantile regression to model curvilinear functional relationships between student's prior and current scores. One hundred such regression models are calculated for each separate analysis (defined as a unique year by content area by grade by prior order combination). The end product of these 100 separate regression models is a single coefficient matrix, which serves as a look-up table to relate prior student achievement to current achievement for each percentile. This process ultimately leads to tens of thousands of model calculations (and many more when SIMEX measurement error corrections are performed) during each of PARCC's annual batch of analyses.

The 2015 PARCC SGP analyses follow a work flow that includes the following 4 steps:

  1. Update the PARCC assessment meta-data required for SGP calculations using the SGP package.
  2. Create annual SGP configurations for analyses.
  3. Conduct all EOG and EOCT SGP analyses at both the PARCC Consortium and individual State levels.
  4. Combine results into the master longitudinal data set, summarize/visualize results.
  5. Format the PARCC Consortium and individual State results according to the requested specifications and submit to Pearson.

Data analysis is conducted using the R Software Environment in conjunction with the SGP Package. Broadly, the analysis takes place in 6 steps.

  1. prepareSGP
  2. analyzeSGP
  3. combineSGP
  4. summarizeSGP
  5. visualizeSGP
  6. outputSGP

Because these steps are almost always conducted simultaneously, the SGP Package has wrapper functions abcSGP and updateSGP that "wrap" the above 6 steps into a single function call and simplify the source code associated with the analysis.

For a project as large and complicated as running growth analyses for the PARCC consortium, SGP analyses are conducted at both the consortium and state level. R source code associated with the PARCC consortium or state's SGP analysis are contained in the relevant folder (PARCC Consortium, Colorado, Illinois, Maryland, Massachusetts, New Jersey, New Mexico, Rhode Island and Washington DC.)

For annual state testing systems like PARCC and its member states, SGP analyses are conducted each year using a similar script. Links to the annual data analysis scripts and accompanying documentation are provided as follows:

SGP - Student Growth Percentiles SGP Blog | SGP GitHub Repo | SGP on CRAN

Clone this wiki locally