Evidence of meeting #5 for Indigenous and Northern Affairs in the 40th Parliament, 2nd session. (The original version is on Parliament’s site, as are the minutes.) The winning word was information.
A recording is available from Parliament.
Director, Social and Aboriginal Statistics Division, Statistics Canada
We would be pleased to, yes.
Conservative
The Chair Conservative Bruce Stanton
Similarly, the department also has a deck from the same session pertaining to that June 16 presentation, and we will get a copy of that to the committee members, so you will have that.
The question I have arises from that June 16 presentation. This was from the department, and I wonder if you could comment on it.
This is in the area of the issue of individual coverage when you're measuring first nations statistics. It indicated the data was suppressed in 166 inhabited reserves because of the low coverage, the uptake. Presumably, unless you get a certain threshold of participation, the data wouldn't be statistically relevant.
I am wondering what that threshold would be. Clearly there were communities where the participation was high enough that it wouldn't have to be suppressed, and I am wondering what that threshold would be in terms of participation.
Director, Social and Aboriginal Statistics Division, Statistics Canada
There are different aspects to this, one of which I have spoken to, about those first nations communities that we did not get on to at all. Those are what we call incompletely enumerated. So we have no data for those, and that was 22. That's one aspect.
Then we may get on the reserve and we may not get complete information. We may just get a population count. Because we have a long questionnaire we may not get that complete information of the quality we put out. As you know, Statistics Canada has high levels of quality.
So I would have to check in terms of our threshold of what we call...they haven't answered so many questions. I've forgotten what that threshold would be. It's a percentage of the questions they have not answered or where the data quality is not of our quality.
On a number of reserves the population is so small that if we released the information we would be releasing individual information. Our concerns about confidentiality of information are there as well.
So I think within those numbers, all of that is a part of it.
Conservative
The Chair Conservative Bruce Stanton
On that last one you're saying that if the volume of data is too low, if there's not a large enough sample, it could be crossing a line in terms of confidentiality.
Director, Social and Aboriginal Statistics Division, Statistics Canada
Yes, it may be 15 people, for example, and it is not a sample. We go house to house, to every house on the reserve. We have to protect confidentiality. That's in our statistics act.
But then there may be data quality issues as well for other areas. So that is where I have to check on the threshold to see where that point would be.
Conservative
The Chair Conservative Bruce Stanton
Great. That is all I had.
We are going to take a brief recess and allow our witnesses to finish up.
I thank you all for your appearance here today and the information you have provided the committee. I am sure it's going to be very helpful as we continue with our study of these important issues.
So we'll take a brief recess and then reconvene in three or four minutes. Thank you.
[Proceedings continue in camera]