In regards to the data files, it is important
to highlight the following:
- Datasets are available in three formats (SAS, SPSS, and STATA),
and are divided into sections, corresponding to sections in the
questionnaire. The following table is an example of how the datasets
will be divided into sections. NIS datasets may be downloaded
from the NIS Data File Download Area.
Registration (free) is required for access.
| NIS 2003 |
SAS Files |
Comp'd Size |
File(s) Size |
Section |
Section Name |
File Type |
Type of Interview (*) |
2082816 |
104188928 |
Roster |
Roster |
Zip Archive |
MA, CP, SP |
152200 |
1705984 |
Section CIS |
Preload |
Zip Archive |
MA, CP |
5195254 |
279589888 |
Section A |
Demographics |
Zip Archive |
MA, CP |
1645226 |
78302208 |
Section B |
Pre-Immigration Experiences |
Zip Archive |
MA, CP, SP |
303693 |
3296256 |
Section B-ppp |
Pre-Immigration Experiences |
Zip Archive |
MA, CP, SP |
| etc... |
|
|
|
|
|
| (*) MA=adult; CP=Child Proxy; SP=Spouse; CH=Child. |
- The variables that correspond to the questions are named following
the convention of section and question numbers (refer to NIS
2003-1 Questionnaires and NIS
2003-1 Codebook).
- Each section dataset has the variable PU_ID, which is a six-digit
unique individual identification number, the first four-digits
are the unique household identification number, and the last
two-digits refer to the individual within the household (10=Adult
Immigrant or Proxy of Child Immigrant, 20=Spouse, 21=Sampled
Child, 22+=additional children). This variable is necessary to
merge the various files from NIS 2003-1 from the questionnaire
and to merge files from the Main Adult with Child Proxy respondents.
- The first four-digits from the variable PU_ID are necessary
to merge files from the Spouse and Child datasets to the Adult
Immigrant or Proxy of Child Immigrant datasets.
- The Preload database has several variables collected from the
INS records, some of which were collapsed, or recoded to protect
the identity of respondents. Among them are CISADJUST (if status
of respondent was adjusted or new arrival), CISCOBINSmo (country
of birth), and CISSTATEmo (state where green card was sent).
Additionally, sampling weights have been included in this file
(NISWGTSAMP1) as well as the various files from the questionnaire.
For more information about the construction of the design weights,
refer to the Sampling
Weights document.
- The variable INTTYPE identifies the type of interview that
was conducted (1=Adult Immigrant; 2=Proxy of Child Immigrant;
3=Overseas Immigrant; or 4=Spouse); the variable WHOKNOW states
who is the most knowledgeable to answer about the household finances
(1=Respondent; 2=Spouse).
- Sections B, C, G, H, and I have each a separate PPP file containing
information of amounts in different currencies converted to US
dollars using purchasing power parity over consumption.
Not all original amounts were converted into US dollars due to
lack of information. These files are put in the rows following
each section as shown in
the table
above
(Section
B
and Section
B-ppp).
|
