Definition 9.
Qualitative data (also called categorical data) represents characteristics that can be observed but not measured numerically. Qualitative data can be classified into categories.
Section Understanding Data Types
Subsection Qualitative vs. Quantitative Data
One of the most fundamental distinctions in data is between qualitative and quantitative data:
Definition 10.
Quantitative data represents information that can be measured numerically and can be used in calculations.
Example 11.
Consider the following variables from our Community Health dataset:
- Qualitative: Neighborhood name, zip code, predominant housing type
- Quantitative: Asthma rate (%), air quality index, median household income ($), number of parks
Activity 6. Identifying Data Types.
In this activity, you’ll practice identifying qualitative and quantitative data.
(a)
For the sample Community Health dataset, identify whether each variable is qualitative or quantitative:
- Neighborhood
- Population density
- Predominant land use
- Percent green space
- Average temperature (°F)
- Healthcare access rating
(b)
Now examine your chosen dataset and classify each variable as qualitative or quantitative.
Checkpoint 12. Identifying Data Types.
- Hair color
- Hair color is qualitative (categorical) data as it represents a characteristic that can be categorized but not measured numerically.
- Zip code
- Although zip codes contain numbers, they are actually qualitative data because they represent categories (geographic regions) rather than measurements.
- Height in centimeters
- Correct! Height in centimeters is quantitative data because it represents a numerical measurement that can be used in calculations.
- Blood type
- Blood type (A, B, AB, O) is qualitative data as it represents categories rather than numerical measurements.
Which of the following is an example of quantitative data?
Subsection Measurement Scales
Data can be further classified by measurement scale, which affects what operations and analyses make sense for that data.
Definition 13.
Nominal data represents categories with no inherent order. The only valid operation is determining equality (same or different).
Definition 14.
Ordinal data represents categories with a meaningful order or ranking, but the differences between values may not be consistent or meaningful.
Definition 15.
Interval data has consistent differences between values, but lacks a meaningful zero point.
Definition 16.
Ratio data has consistent differences between values and a meaningful zero point (zero represents the absence of the quantity).
Example 17.
Examples from our Community Health dataset:
- Nominal: Neighborhood name, predominant land use (residential, commercial, industrial, mixed)
- Ordinal: Healthcare access rating (poor, fair, good, excellent), air quality category (unhealthy, moderate, good)
- Interval: Temperature (°F or °C) - the difference between 70°F and 80°F is the same as between 80°F and 90°F, but 0°F doesn’t represent the absence of temperature
- Ratio: Asthma rate (%), income ($), number of parks, population - zero means none, and the difference between values is consistent
Insight 18.
Understanding measurement scales helps you determine appropriate:
- Summary statistics (mean, median, mode)
- Visualization methods (bar charts, histograms, scatter plots)
- Analysis techniques (correlation, regression, categorical tests)
Activity 7. Classifying by Measurement Scale.
In this activity, you’ll practice identifying measurement scales.
(a)
For each variable below, identify its measurement scale (nominal, ordinal, interval, or ratio):
- ZIP code
- Education level (no high school, high school, bachelor’s, graduate)
- Annual rainfall (inches)
- Birth month
- Satisfaction rating (1-5 scale)
- Distance from city center (miles)
(b)
Now examine your chosen dataset and classify each variable by measurement scale.
Checkpoint 19. Measurement Scales.
Subsection Other Data Classifications
Beyond the qualitative/quantitative and measurement scale distinctions, there are other useful ways to classify data:
Definition 20.
Discrete data can only take specific values (usually whole numbers), while continuous data can take any value within a range.
Example 21.
From our Community Health dataset:
- Discrete: Number of parks, number of healthcare facilities, population count
- Continuous: Air quality index, median income, percent green space, asthma rate
Definition 22.
Structured data is organized in a predefined format (like a spreadsheet or database), while unstructured data doesn’t conform to a predefined data model (text, images, audio).
In this course, we’ll primarily work with structured data, but it’s important to know that unstructured data constitutes the majority of data generated today.
Checkpoint 23. Data Type Review.
- All numerical data is quantitative and all text data is qualitative.
- Not all numerical data is quantitative. For example, zip codes are numerical but are considered qualitative (nominal) data.
- Nominal and ordinal scales are types of quantitative data.
- Nominal and ordinal scales are actually types of qualitative (categorical) data.
- Interval data has a true zero point, while ratio data does not.
- This is reversed. Ratio data has a true zero point (representing the absence of the quantity), while interval data does not.
- Discrete data can only take specific values, while continuous data can take any value within a range.
- Correct! Discrete data is limited to specific values (usually whole numbers), while continuous data can take any value within a range, including fractional values.
Which of the following statements about data types and classifications is correct?
Definition 24.
Tidy data is a specific way of organizing data where:
- Each variable forms a column
- Each observation forms a row
- Each type of observational unit forms a table
Tidy data makes analysis and visualization more straightforward, as most analysis tools (including CODAP) are designed to work with data in this format.
Activity 8. Identifying Tidy Data.
In this activity, you’ll practice identifying tidy and untidy data.
(a)
Examine the following datasets in CODAP:
- Open a new CODAP document
- Click on "Example Documents" in the main menu
- Open both "Mammals" and "Speed Trap"
(b)
For each dataset, determine if it follows the principles of tidy data. If not, explain what would need to change to make it tidy.
(c)
Examine your chosen project dataset. Is it in tidy format? If not, what would need to change?
