Interesting article on something that is relevant for many people here

http://www.theglobeandmail.com/news/national/poor-data-could-hinder-key-feature-of-perimeter-security-pact-border-agency/article4903320/

Another interesting article close to this topic (albeit from 2005 and focused on privacy aspects primarily) - "What happens when you book an airline ticket" by Colin J. Bennett

http://www.colinbennett.ca/Recent%20publications/What_happens_when_you_book_an-airline_ticket.pdf

Observations:

That which lies in-between the two linked sources (recent article and the 2005 article), the flaws and gaps and inaccuracies in data-gathering on one hand, and the extensive, intrusive capturing of personal information on many levels on the other hand, illuminates the difference between accessing data sources to identify false, misleading, or otherwise inaccurate information, versus any attempt to create a comprehensive, exhaustive record for particular individuals.

There are, as has been oft discussed, other reasons why no database is relied upon to represent a complete and accurate accounting of an individual's international travel, the biggest reason being rooted in the capacity of those with mal-intent to manipulate or deceive. But, between data-gathering flaws on one hand, and the massive amount of personal information actually collected from individuals on the other, loom the practical and technological difficulties of building an airtight, foolproof system to maintain personal border-crossing histories that are reliably accurate, absolutely complete, and yet sufficiently protected from serious breaches of privacy rights.

A system which captures almost every cross-border event, but not every-single event, falls short and is unreliable. And, even that system can be, most likely would be, vulnerable to serious breaches of personal privacy rights unless access is strictly limited, very strictly limited.

So, let's get the big one out of the way: we have no idea to what extent CSIS and the American FBI and CIA (and other security/intelligence agencies in the U.S.) have access to widespread, highly detailed, information about travel to and from North America, or between the U.S. and Canada. We can surmise that they have massive resources and extensive capacity to do data-mining on many, many levels. For the vast majority of us, this is not an issue. We are not on their radar. We are not perceived to be any sort of security concern. No one is tracking our movements in particular.

But they could be. They might be. For some of us, there may be perceived reasons, or even substantial reasons, why they would be. For those who fall into this (probably quite small) group, there should be no surprise: an application for citizenship is going to get tangled up some along the way. And this is likely to be true whether or not there are residency concerns.   

The other side of this coin has to do with statements like:

the notion that Canada doesn't collect information on people exiting by air or its govt doesn't have access to it is b.s.

I doubt anyone is suggesting that "Canada" is not collecting vast amounts of personal information including in particular such information as exits and entries into Canada.

But no modern government in the West, or otherwise industrialized country, is a monolithic entity. There is no central brain (or database) storing all that is known by the "government."

So, such an obversation is largely, if not entirely, irrelevant to how residency is or could be assessed in the processing of an application for Canadian citizenship.

The practical questions have to do with what information is being captured by who, and how that information is stored, updated, and accessed, including related questions such as who has access and what are the conditions of access, including information-sharing protocols (which akella alludes to, including such considerations as information sharing agreements and consent, as well as related privacy protections).

An aspect of the practical questions regarding information capturing, is that many of the particulars, including both the means and the content of information captured by this or that government agency, are deliberately NOT transparent but protected, largely secret procedures. Law-enforcement, in particular, including national security agencies, does not want the public to know the details about the what, when, where, how, or so on, of their intelligence/information gathering.

Suffice it to say: here or there, in this or that government agency, and in all of them combined, sure, anyone who spends any time at all inside Canada probably has had all sorts of detailed information about them gethered and stored. So, for any given PR there is, almost certainly, a huge amount of information about them somewhere in government databases.

CIC does not have easy access to a lot of that. For good reason. This has a lot to do with the protection of individual privacy. But, generally, it is rooted in limited access related to a need-to-know hierarchy which protects not just personal information but government processes and government information generally.

This leads to CIC access and use of information captured and stored by other govt agencies.

CIC access & use of information captured & stored by other govt agencies

The CBSA travel history is a prime example. One might say that there is a particular database maintained by CBSA in which traveler entries into Canada are recorded. But, it is more likely that the database is significantly more robust, more extensive than that. Rather, what is likely is that among designed queries for accessing a much broader, more extensive CBSA database, one query is specifically designed to generate a report indicating a particular individual's entries into Canada. It may or not be complete, because the capturing of the underlying information (all entries into Canada) may or may not be complete. The trend is toward making the capture of this information more complete, but there are still occasions in which this information is not captured and this is sometimes true even if the individual is not deliberately attempting to evade the capture of a particular entry. Since it is probably impossible to maintain open-borders and be certain no one is evading the capture of every time they cross the border, such data will always fall short of being absolutely complete . . . and thus there is little prospect that CIC will ever totally rely upon this information to be fully complete.

I suspect that CSIS, RCMP, and the U.S. FBI, have the capacity to either make a more expansive query into the CBSA database, or that in specified conditions, a referral or inquiry can be formally made which will generate a far more expansive record regarding a specifically targeted individual. I am quite certain, as well, that CIC too can make a formal request for more expansive, detailed information, but I think they are limited in this, and probably have to justify the request, as in specify grounds which will meet prescribed restrictions.

What they can obtain may be expanded some by consent. So a request for consent (allowing direct access) is now part of the citizenship application and (it appears) the most recent version of the RQ form.

I would bet that similarly CIC can request investigations or inquiries from other government agencies in specific cases, based on prescribed criteria which will justify access for that individual in that case.

Again, for most of us, the vast majority of us, this will be irrelevant. We are not under investigation. (Of course I do not really know, but . . . well, I think this is fairly obvious if one works their way through the relevant considerations, balances the equation, and checks the math, so to say).

Bottom-line: while to a large extent the CBSA report is used to corroborate an applicant's residency declarations, it does not absolutely confirm an individual's declaration of absences, so its primary purpose is to expose inconsistencies, inaccuracies, or omissions. It is not used to determine how many days an applicant was absent or present. It is used to compare with what the applicant declared and identify if the CBSA record indicates any absence which the applicant failed to disclose.