Data is information describing some phenomenon.
A factor situation that is observed to exist or happen, especially one whose cause or explanation is in question.
A factor situation that is observed to exist or happen, especially one whose cause or explanation is in question.
A factor situation that is observed to exist or happen, especially one whose cause or explanation is in question.
A factor situation that is observed to exist or happen, especially one whose cause or explanation is in question.
Anything and everything are phenomena!
Discrete Objects
Distinct boundaries
Chat can be exactly measured
Finite
They are countable and cannot be infinitely subdivided
Continuous Fields
No distinct boundaries
Everywhere has a value
Infinitely divisible
They are not countable and can be infinitely subdivided
When is a phenomenon discrete or continuous?
A strike is a discrete object, what about a lighting bolt?
A strike is a discrete object, what about a lighting bolt?
Continuous field at large scale
Discrete object at small scale
Unless you change the time scale
Most things don’t fall perfectly into one category or the other.
Discrete objects: (select all that apply)
Buildings are a great example.
Political Boundaries are also a great example.
Elevation is a great example.
Density of tweets is also a great example.
Everywhere has this too
Derived from something countable
But not countable itself
Not a physical property
Frequently we’ll end up working with both discrete objects and continuous fields.
We’ll talk more about spatial data models later. For now, lets think about data more broadly.
Digital information is represented as bits (0’s and 1’s)
There are numerous ways to translate human readable data to binary, such as ASCII.
Modern computers use 64-bit “architecture”. The central processing unit (CPU) can handle 64 bits (8 bytes) of information at a time.
i.e., an individual piece of information
18 quintillion unique combinations of 1’s and 0’s
Within the context of a GIS, every piece of information describing a phenomenon is referred to as an Attribute.
There are multiple ways to classify/think about attributes. One important distinction we must make
All data (attributes), spatial and non-spatial, can be either qualitative or quantitative.
Qualitative data is Categorical. It is strictly descriptive and lacks any meaningful numeric value.
Names or categories with no ranking or direction. Categories are not more/less, better/worse, they just different. Some examples include:
Names or categories with no ranking or direction. Categories are not more/less, better/worse, they just different. Some examples include:
Names or categories with no ranking or direction. Categories are not more/less, better/worse, they just different. Some examples include:
With nominal data we can:
Names or categories with a ranking. The differences are relative. Categories are more/less, better/worse, etc.
Names or categories with a ranking. The differences are relative. Categories are more/less, better/worse, etc.
Names or categories with a ranking. The differences are relative. Categories are more/less, better/worse, etc.
All the same operations as nominal data + more. With ordinal data we can:
Sometimes we can calculate the median.
Exceptions that blur the lines. Where to draw the line between forest/alpine?
In practice, lots of qualitative data we work with, especially for natural phenomena, are actually graded membership.
Which of the following would be examples of Nominal Data? (select all that apply)
Quantitative data is Numeric. It describe the quantities associated with a phenomenon. Key properties include:
Discrete
Continuous
Discrete
Continuous
Both Interval and Ratio data can consist of discrete or continuous numbers. These types of quantitative data are closely related, but have one important distinction.
°C = K-273.15.
°C = K-273.15.
Interval data has an arbitrary zero point.
Interval data has an arbitrary zero point.
Ratio data has a fixed, absolute zero point.
Ratio data has a fixed, absolute zero point.
Match the value to the type measurement scale and type of number:
Length a hiking trail | Interval (Discrete) |
Temperature in Fahrenheit | Ratio (Discrete) |
Global Orca Population | Ratio (Continuous) |
Change in Global Orca Population from 2000 to 2022 | Interval (Continuous) |
Sometimes called normalizing or standardizing, we calculate derived ratios to account for the influence of a confounding variable over a variable of interest. e.x. Housing affordability (Ha):
In Lab, you are going to work with two derived ratios:
In Lab, you are going to work with two derived ratios:
Speed is another example of a derived ratio. If a line of thunderstorm takes 5 hours to travel from Brandon, MB to Winnipeg, MB (200 km), what is the storm’s speed in km/hr?
Operation | Nominal | Ordinal | Interval | Ratio |
Equality | x | x | x | x |
Counts/Mode | x | x | x | x |
Rank/Order | x | x | x | |
Median | ~ | x | x | |
Add/Subtract | x | x | ||
Mean | x | x | ||
Multiply/Divide | x |