The flowchart of spatial analysis for private residence for seniors:

Our base raster map is CT96.RST, and 65nonfam is obtained from database table of Census 96.  For detail on how the data is being handled please refer to Data Collection and Manipulation.  65nonfam represents the number of non-family persons 65 and over.  Total65+ represents the total number of persons 65 and over.  By dividing the two, we could get the ratio of where there is a highest number of seniors not living with a family, thereby concluding our result of the proposed private residence suite for senior.  Division by zero is not impossible within a census data, there may be times where a Census tract does not have a single senior.  Therefore, division by zero yields zero is chosen.  For such special case, the area would naturally be taken out of the statistical calculation.  And finally, SNRSUITE is the resultant output file.
 
 

The flowchart of spatial analysis for adult day care facility:

Here, we start off using the same raster file - CT96.  From there, I produced four raster images: 5ORLESS, 5TO9, 10NMORE, TOTALCARE.  The metadata are also similarly gathered from Census 96 Access database file.  Now we'll talk about each file and what they mean in an analysis point of view.  5ORLESS is data of population 15 years and over, giving less than 5 (but not zero) hours of unpaid care to seniors.  It should be emphasized that paid caregiver, nurses (registered and non), nanny, housemaid, etc are not counted in this category.  This group might consist of people of the following cases:

1) a senior married couple, where the wife/husband is taking care of his/her husband/wife;

2) children taking care of their elderly parents (as this is true with non-Western family)

3) any kind unpaid care-giving to seniors not mentioned above (e.g.: by relatives, close friends, neighbour?)

5TO9 is simply the same with the file 5ORLESS, but consist of the population where the hours of care is between 5 to 9 hours.  10NMORE represents the population giving 10 or more hours of unpaid care to seniors.

First, I started by adding the first two categories, 5ORLESS and 5TO9 together using OVERLAY (First + Second) and created an intermediate OVERLAY1.  Then I added the OVERLAY1 intermediate file with the last group of unpaid care givers, yielding the total number of unpaid caregiver population in the GIVECARE raster file.  The GIVECARE file itself does show some usefulness in analyzing the problem, for detail please visit Spatial Analysis II - The Search for Adult Day Care Facility.

Anyhow, because we need to further refine the data into more sensible and usable form, I divided the total number of unpaid care givers by the total sample population, in GIVECARE and TOTALCARE files respectively.  Again, in this analysis there is a possibility that a Census tract could contain no unpaid caregivers.  Therefore, the "division by zero yields zero" option is picked.  The resultant DAYCARE file gives the most suitable location for establishing an adult day care centre.