•   about 5 years ago

Adjusted Cohort Graduation Rate (ACGR)

The Grad Nation report uses the Adjusted Cohort Graduation Rate. However, I could not find this variable in the data dictionary when searched for it.

The footnote references it here:
[1] The four-year adjusted cohort graduation rate is the number of students who graduate in four years with a regular high school diploma divided by the number of students who form the adjusted cohort for the graduating class. For any given cohort, students who are entering grade 9 for the first time form a cohort that is subsequently “adjusted” by adding any students who transfer into the cohort later during the next three years and subtracting any students who transfer out, emigrates to another country, or dies during that same period.

Can we get this variable in the data sets, or some guidance on what variables are used to calculate the adjusted cohort graduation rate?


  • Manager   •   about 5 years ago

    Hi Rob,

    Thanks for your question!

    We are fairly certain that the ACGR is synonymous with the ALL_RATE_1011 variable located in the data dictionary. This is the variable we are trying to explain. All of the rate variables coming from the ACGR dataset are assumed to be adjusted per the footnote you've referenced.

    Please let me know if this does not answer your question!

  •   •   about 5 years ago

    Hi Serena,

    Thank you for the clarification. That's a big help.

    - Rob

  •   •   almost 5 years ago

    Hi Serena,

    The schema says that the variables end with the digits 1011. But in the actual dataset, the variables end in 1112. I'm assuming these indicate school years: 2010-11 vs 2011-12. Do you know why these are different?

    Do you also know which year they used in the benchmark analysis? We tried replicating the results and got different answers.


  • Manager   •   almost 5 years ago

    Hi Azi,

    Thanks for your question. I'll circle back with the AT&T team and respond to you shortly.


  • Manager   •   almost 5 years ago

    Hi Azi,

    Thanks again for your question. I've heard back from the team and it seems that the discrepancy between schema name and column names in the dataset was simply a writeup error. The schema SHOULD say 1112 as oppose to 1011.

    The data used for the benchmark is the same data that is in the dataset provided. Please note, however, that the Benchmark results could vary depending on computer calculation precision and/or feature engineering and cleaning of the dataset.

    I hope this helps!

Comments are closed.