Skip to content
Data and analysis supporting the BuzzFeed News article, "The Agriculture Department Hired More Than 50 Political Appointees. They All Say They're White.," published November 15, 2017
Branch: master
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
data
notebooks
.gitignore
Makefile
README.md
requirements.txt

README.md

Analysis of Political Appointee Demographics

This repository contains data and analysis supporting the BuzzFeed News article, "The Agriculture Department Hired More Than 50 Political Appointees. They All Say They're White.," published November 15, 2017. See below for more details.

Data

The data in this repository comes from the Office of Personnel Management's FedScope tool. The data on racial/ethnic diversity come from FedScope's "Diversity Cubes", and the data on gender come from the "Employment Cubes".

Raw data

The files in the data/raw directory were obtained by using each of those "cubes" to cross-tabulate the employment counts for each "Type of Appointment" by demographic.

A few relevant links:

Processed data

The files in the data/processed calculate the proportions of employees who identified as minorities/female, for each fiscal quarter, for two aggregated groups:

  • "Permanent" employees (see link above for definition)
  • Mid-level political appointees (see below for definition)

The Python code used to make these calculations can be found here.

Defining "mid-level political appointees"

According to correspondence with the Office of Personnel Management, there are two "type of appointment" categories that are composed entirely of political appointees:

When the analysis refers to mid-level political appointees, it refers to the combination of these two categories.

Note re. other political appointees

In addition, according to OPM, there are four other "type of appointment" categories that may, at times, include political appointees:

  • Permanent "Excepted Service - Executive"
  • Non-permanent "Excepted Service - Executive"
  • "Senior Executive Service - Limited Term
  • "Senior Executive Service - Limited Emergency"

Unfortunately, among these categories, it is not possible to distinguish between political appointees and non-appointees in the FedScope data.

Reproducibility

To reproduce the calculations, you'll need to do the following:

  • Ensure that you have installed Python and the Python libraries listed in requirements.txt.
  • Clear the data/processed directory. (Shortcut: run make clear.)
  • Re-run the data-processing notebook in the notebooks/. (Shortcut: run make reproduce; requires Python 3.)

Feedback / Questions?

Contact Jeremy Singer-Vine at jeremy.singer-vine@buzzfeed.com.

Looking for more from BuzzFeed News? Click here for a list of our open-sourced projects, data, and code.

You can’t perform that action at this time.