2.3: Structured Data types: Categorical vs. Numerical

Question 1

What are the two broad categories of structured data?

Answer 1

The two broad categories of structured data are categorical and numerical.

Question 2

What is nominal data?

Answer 2

Nominal data refers to categorical data that cannot be ranked or ordered, such as country of origin and transaction type.

It is summarized using counting, grouping, and proportion methods.

Question 3

How can you analyze categorical nominal data like the “Online or In-Person” column?

Answer 3

Categorical nominal data can be analyzed by counting and grouping transactions by their categories (e.g., online or in-person) and calculating proportions.

Proportions are found by dividing the number of observations in a category by the total number of observations in the sample.

Question 4

What is proportion?

Answer 4

proportion
The number of observations in one category divided by the grand total of all available observations.

Question 5

What is ordinal data, and what distinguishes it from nominal data?

Answer 5

Ordinal data allow counting, grouping, calculating proportions, and ranking based on a natural order or ranking associated with the variable.

This distinguishes ordinal data from nominal data, which lack a natural order.

Question 6

What are the three primary methods of summarizing ordinal data?

Answer 6

The three primary methods for summarizing ordinal data are counting and grouping, calculating proportions, and ranking (or sorting).

Question 7

Can you provide examples of ordinal data?

Answer 7

Examples of ordinal data include letter grades (A, B, C, D, F) and Olympic medals (gold, silver, bronze). They have a natural order that allows ranking.

Question 8

Why is sorting by rank valuable in analyzing ordinal data?

Answer 8

Sorting by rank adds value to the analysis of ordinal data because it reflects the natural order associated with the data, providing insights beyond numerical or alphabetical sorting.

Question 9

What are numerical data, and what types of analysis can be performed on them?

Answer 9

Numerical data consist of meaningful numbers, and analysis can include operations such as summing, multiplying, or dividing.

Question 10

What are the four primary methods for summarizing numerical data?

Answer 10

The four primary methods for summarizing numerical data are counting and grouping, proportion, summing, and averaging.

Question 11

What is interval data, and why does it lack a meaningful zero?

Answer 11

Interval data has an equal interval between data points but lacks a meaningful zero because zero in interval data does not represent the absence of something, and it is simply another number.

Examples include temperature and SAT scores.

Question 12

What distinguishes ratio data from interval data, and why is it considered the most sophisticated form of data?

Answer 12

Ratio data has an equal interval between data points, but it also has a meaningful zero, allowing for the calculation of ratios.

It is considered the most sophisticated form of data due to its ability to measure relationships and financial positions.

Question 13

Provide examples of data types that fall under ratio data.

Answer 13

Examples of ratio data include transaction amounts, expenses, revenues, assets, salary, and taxes.

Question 14

What is a “flag”?

Answer 14

flag
A term used to describe a data set in which there are only two options.

Question 15

What is the string data type, and what types of characters can it include?

Answer 15

The string data type consists of a collection of characters, which can be letters, numbers, or a combination.

However, the numbers in a string are not meaningful values and cannot be used in calculations.

Question 16

What is the date data type, and how is it typically formatted?

Answer 16

The date data type is a string of characters formatted in a traditional date format, such as mm/dd/yyyy or mm/dd/yy.

Question 17

What is the number data type reserved for, and what types of characters can be stored as numbers?

Answer 17

The number data type is reserved for numeric data, typically ratio data. Any characters stored as numbers can be used in calculations.

Question 18

What is geographic data, and how is it linked to maps?

Answer 18

Geographic data are attributes linked to a map, such as state, city, country, or zip/postal code.

These attributes are linked with latitude and longitude numbers to represent them on maps.

Question 19

What is a dimension, and where might you encounter this term, such as in Tableau?

Answer 19

A dimension is any attribute characterized as categorical.

The term dimension is often used in applications like Tableau to refer to categorical attributes.

Question 20

What is a measure, and where is it commonly used, such as in Tableau?

Answer 20

A measure is an attribute characterized as numerical. The term measure is often used in applications like Tableau to refer to numerical attributes.

Question 21

What do ETL and ELT stand for in the context of data preparation?

Answer 21

ETL stands for “Extract, Transform, Load,” while ELT stands for “Extract, Load, Transform.” Both are processes used for preparing data for analysis.

Question 22

What are the four steps involved in preparing data for analysis, often described as part of the ETL or ELT process?

Answer 22

The four steps in preparing data for analysis are

(1) ensuring data quality,

(2) validating data for completeness and integrity,

(3) cleansing the data, and

(4) performing preliminary exploratory analysis.