C207 Data-Driven Decision Making STUDY GUIDE FOR OA 2025
Western Governors University
Data Driven Decision Making
• Davenport-Kim three stage model:
o Framing the problem (aka problem recognition and review):
▪ Problem recognition :
1. Identify Stakeholders
2. Focusing on decisions
3. Identifying the kind of story you want to tell with the data
and your audience
4. Determine the scope of the problem
5. Getting specific about what you’re trying to find out. After
you have reviewed the big picture in this step you will then
narrow your data to focus on a specific set to analyze and
present.
▪ Review of previous findings
o Solving the problem:
▪ Involves intense statistics and data work.
▪ Where mathematical heavy lifting takes place.
▪ Solving the problem consist of 3 steps:
1. The Modeling Step
2. The Data Collection Step
3. The Data Analysis Step
o Communicating results
• Levels of Measurements
o Continuous data – CAN HAVE DECIMAL POINTS data can lay on any point
in range of data (Age: can be 22.67 years old depending on the time of the year. )
▪ Interval Data-(-type of continuous data)- has an order to it and all parts
of the data are equal and the difference between the two values will
be meaningful. DOES NOT USE ZERO DATA POINT (time, data
and temperature-Fahrenheit )
• Daily temperature in Fahrenheit or Celsius)
▪ Ratio data- Can use the zero data point (someone can be twice as old
as you) Kelvin scale in temperature is a ratio data
• Time it takes to travel the earth `
o Discrete data (N-O- Nominal and Ordinal- and WHOLE)-
WHOLE NUMBERS ONLY (cannot have 3.4 cars can only have 3 or
4 cars)
▪ Nominal Data (aka Categorical Data- type of discrete data)- used to
label subjects in a study (male=1 and females=2)
• Color choice of crayons
▪ Ordinal Data (- type of discrete data- Ranking data)- Places subject
in ORDER. (black belt is higher than green belt in karate)
• Level of education
, o Reliable data is consistent and repeatable
Western Governors University
Data Driven Decision Making
• Davenport-Kim three stage model:
o Framing the problem (aka problem recognition and review):
▪ Problem recognition :
1. Identify Stakeholders
2. Focusing on decisions
3. Identifying the kind of story you want to tell with the data
and your audience
4. Determine the scope of the problem
5. Getting specific about what you’re trying to find out. After
you have reviewed the big picture in this step you will then
narrow your data to focus on a specific set to analyze and
present.
▪ Review of previous findings
o Solving the problem:
▪ Involves intense statistics and data work.
▪ Where mathematical heavy lifting takes place.
▪ Solving the problem consist of 3 steps:
1. The Modeling Step
2. The Data Collection Step
3. The Data Analysis Step
o Communicating results
• Levels of Measurements
o Continuous data – CAN HAVE DECIMAL POINTS data can lay on any point
in range of data (Age: can be 22.67 years old depending on the time of the year. )
▪ Interval Data-(-type of continuous data)- has an order to it and all parts
of the data are equal and the difference between the two values will
be meaningful. DOES NOT USE ZERO DATA POINT (time, data
and temperature-Fahrenheit )
• Daily temperature in Fahrenheit or Celsius)
▪ Ratio data- Can use the zero data point (someone can be twice as old
as you) Kelvin scale in temperature is a ratio data
• Time it takes to travel the earth `
o Discrete data (N-O- Nominal and Ordinal- and WHOLE)-
WHOLE NUMBERS ONLY (cannot have 3.4 cars can only have 3 or
4 cars)
▪ Nominal Data (aka Categorical Data- type of discrete data)- used to
label subjects in a study (male=1 and females=2)
• Color choice of crayons
▪ Ordinal Data (- type of discrete data- Ranking data)- Places subject
in ORDER. (black belt is higher than green belt in karate)
• Level of education
, o Reliable data is consistent and repeatable
