Question:
I have a question from a grad student in Political Studies and, given my lack of expertise dealing with the PCCF, I thought I’d post it here for help. The problem stems largely from the fact that some FSAs are classified as both urban and rural.:
"I am using survey data to determine whether or not feelings of western alienation are higher in rural places relative to urban places in Western Canada. My survey asked respondents for the first three digits of their postal code (FSA), which I plan to code as either rural or urban. To do so, I plan on using the Postal Code Conversion File (PCCF) to code the FSAs of Western Canada as rural or urban, according to the "PopCntr_RA_size_class" variable. Unfortunately, each FSA has multiple values for the "PopCntr_RA_size_class" variable, making it difficult to code each FSA as either rural or urban. How can I effectively sort each FSA into the geographic categories?
Also, once I classify each FSA as rural or urban, how can I integrate that information into the survey data set? Is there a way to create an SPSS syntax file that sorts the respondents into geographic categories based on their reported FSA?"
Answer:
I’ve received the following from subject matter:
“There is no easy way to classify the FSAs as either rural/urban. FSAs are designed as part of Canada Post Corporation’s mail delivery system, which conceptually is operational, not geographical. That makes it difficult to match it with conceptually geographic systems, like StatCan’s census geographies. The PCCF does the best it can, and at the lowest levels of geography, it’s not too bad. However, moving to higher levels of geography, like Population Centres, or higher in the delivery system, like FSAs, causes more gray areas. To facilitate the sorting of a lot of mail, they cover delivery to many different types of geography: villages, indian reserves, cities, towns, rural routes, and so on. It is not surprising that they cannot be classified as either rural or uban.
The best that can be done is some analysis of how the FSAs link up with PopCentres in the PCCF. Then, the researcher will need to make some decisions as to how to define each FSA, based on that analysis.
So, given all that, we have done the following:
— selected all non-retired FSAs from the western provinces (Manitoba to British Columbia) à 455
— found those that overlapped more than 1 PopCentreSizeClass or more than 1 PopCentreRAType à 304
and then we provided further information on those with the postal code counts. See the attached table. With that, the client can decided which way to classify each FSA.
This is only one example of the types of analysis that can be done. There are also the SAC and SACType attributes that may be useful, as they indicate whether the FSA is influenced by a CMA / CA or not. Hmm, I just took a quick look at that, and that may prove more useful. I’ve included that table in the workbook as well.
Anyway, there is no easy way to classify the FSAs are rural/urban with the PCCF. The client will have to do some sort of analysis of the attributes (PopCentre or SAC or perhaps something else that makes sense to them), then make some decisions as to how they will classify each FSA.”
