NIS Restricted Data
Restricted data comprises those fields and variables that potentially
enable identification of survey respondents either directly or
through deductive identification. Determination of which variables
are included in the restricted and public use data sets is based
on the disclosure risk and the consequences of disclosure of
each variable. The public use data include only variables with
low disclosure risk and low levels of sensitivity.
The NIS has
created two versions of Restricted Data Files, each of which
involves access requirements and data security:
Mid- to high-level of Restricted Data. This file includes individual characteristics
reported in the interviews. In an effort to maximize the data available to data
users, most of these variables were included in the public use file after some
sort of transformation, such as top coding, categorization of continuous variables,
or collapsing of categorical responses. Version 1 of the Restricted Data File
will contain unmodified variables for country of origin, destination states
of previous trips,
education, health conditions, employment, religion, occupation
and industry, among others.
Highest level of Restricted Data. This file adds geographic
identifiers describing the characteristics of the place at time
of interview (state, city, and zip code), as well as some selected
variables from the INS records about the sampled respondents (i.e.
class of admission, visa category, country of birth, country of
chargeability, country of last residence, etc.). Release of this
version of the restricted data requires that it be stored within
a secure data enclave.
These versions are discussed in detail in the NIS Restricted
Data Access Policy.