On this R-data statistics page, you will find information about the USclassifiedDocuments data set which pertains to Official Secrecy of the United States Government . The USclassifiedDocuments data set is found in the Ecdat R package. You can load the USclassifiedDocuments data set in R by issuing the following command at the console data("USclassifiedDocuments"). This will load the data into a variable called USclassifiedDocuments. If R says the USclassifiedDocuments data set is not found, you can try installing the package by issuing this command install.packages("Ecdat") and then attempt to reload the data. If you need to download R, you can go to the R project website. You can download a CSV (comma separated values) version of the USclassifiedDocuments R data set. The size of this file is about 823 bytes.
Official Secrecy of the United States Government
Data on classification activity of the United States government.
Fitzpatrick (2013) notes that the dramatic jump in derivative classification activity (
DerivClassActivity) that occurred in 2009 coincided with "New guidance issued to include electronic environment". Apart from the jump in 2009, the
DerivClassActivity tended to increase by roughly 12 percent per year (with a standard deviation of the increase in the natural logarithm of
DerivClassActivity of 0.18).
A dataframe containing :
the calendar year
Number of people in the government designated as Original Classification Authorities for the indicated
Original classification activity for the indicated year: These are the number of documents created with an original classification, i.e., so designated by an official Original Classification Authority.
OCActivity covered by the 10 year declassification rules.
Derivative classification activity for the indicated year: These are the number of documents created that claim another document as the authority for classification.
The lag 1 autocorrrelation of the first difference of the logarithms of
DerivClassActivity through 2008 is
-0.52. However, because there are only 13 numbers (12 differences), this negative correlation is not statistically significant.
Fitzpatrick, John P. (2013) Annual Report to the President for 2012, United States Information Security Oversight Office, National Archives and Record Administration, June 20, 2013 (https://www.archives.gov/isoo/reports)
## 1. plot DerivClassActivity
# Exponential growth? plot(DerivClassActivity~year, USclassifiedDocuments,
# A jump in 2009 as discussed by Fitzpatrick (2013).
# Otherwise plausibly a straight line. ##
## 2. First difference?
# Jump in 2009 but otherwise on distribution ##
## 3. autocorrelation?
sel <- with(USclassifiedDocuments,
(1995 < year) & (year < 2009) )
# lag 1 autocorrelation = (-0.52).
# However, with only 12 numbers,
# this is not statistically significant.
Dataset imported from https://www.r-project.org.