Friday, June 2, 2006

Several questions re: NPHS and CCHS

Questions

NPHS

1. The IMDB entry explains that "The NPHS Cycle 3 (1998-1999) cross-sectional sample contains all members of the NPHS longitudinal sample and 2 top-up samples." So does this mean that the cross-sectional data from each cycle is gathered from the same "longitudinal" respondents each time (plus the top-up, and minus the longitudinal non-respondents, during the 2nd and 3rd cycles)?

2. Am I correct in understanding that the NPHS cross-sectional data is available for three cycles, 94/95 to 98/99, only? And that this is what I'm able to access through the DLI ftp server?

3. Is the longitudinal data available only through the RDCs?

4. Is the NPHS still an active survey? From what I've read it seems the cross-sectional part was conducted for 3 cycles, but I see reference to Cycle 4 and 5 in the longitudinal description in the IMDB. Cycle 5 was 2002/2003, and I don't see any reference to the survey being administered since then. I also read that "These same persons will be interviewed every two years over a period of 18 years i.e. 10 cycles." So should I be expecting to see Cycles 6-10 in the future? Or has something changed for this survey?

CCHS
1. What is the main difference between these two surveys? I once thought that the CCHS was the successor of the NPHS but I'm not sure where I got that idea, as I haven't seen it anywhere in the documentation. The objectives of the surveys seems very similar though.

2. The IMDB says that "only cross-sectional data are available on such files [the PUMFs], because longitudinal information can lead to the identification of respondents." I don't see any other references to longitudinal data for the CCHS - is there a longitudinal component? The description of sampling methods for 2.1 appears to be the same type of sampling as for 1.1, and doesn't say that it's using the same respondents as 1.1. Am I missing something? And if there is longitudinal data, is that only available through the RDCs?

Answer

Answer to NPHS Question 1:
The best way to understand the samples is to look at the table found in IMDB:
http://www.statcan.ca/cgi-bin/imdb/p2SV.pl?Function=getSurvey&SDDS=3236&lang=en&db=IMDB&dbg=f&adm=8&dis=2
Go under the section: SAMPLING.

Answer to NPHS Question 2&3:
Yes.

Answer to NPHS Question 4:
Yes it is still active, will publish results of cycle 6 longitudinal in September. Still cycles 7, 8, 9 , 10 to do.

Answer to CCHS Question 1:
Yes, the CCHS took the cross-sectional component of the NPHS. So the CCHS is cross-sectional only, and has small geography and a big sample, NPHS is strictly longitudinal, has a small sample of a panel of respondents, and is provincial. They complement each other.

Answer to CCHS Question 2:
CCHS is strictly cross-sectional. Having the same sampling methods from one cycle to the other does not mean it is longitudinal. The big difference is the people: are they the same or not? In NPHS it is the SAME respondents every cycle, for CCHS they are always DIFFERENT.

2 comments:

  1. Everyone loves what you guys are up too. This sort of clever work
    and reporting! Keep up the very good works guys I've added you guys to our blogroll.

    ReplyDelete
  2. magnificent issues altogether, you just gained a new reader.
    What would you suggest in regards to your put up that you just made some days ago?
    Any certain?

    ReplyDelete

Note: Only a member of this blog may post a comment.