First, please create an account

Already have a Sophia account?

Cross Sectional and Time Series Data

Author: Sophia

what's covered

In this lesson, you will learn about cross-sectional and time series data. The differences between the two types of data will be discussed and specific examples will be used to illustrate how businesses use these two types of data. Specifically, this lesson will cover:

1. Overview of Data
2. Cross-Sectional Data
3. Time Series Data
4. Differences Between Cross-Sectional and Time Series Data

1. Overview of Data

Data are values that are collected about some phenomena. In business data analytics, this phenomenon is a business process that occurs. A variable is something of interest being measured or observed. An observation is the value that the variable takes on. A data set is the collection of variables with corresponding observations. Variables are the columns in a data set and observations are the rows.

The figure below shows a snapshot of a typical data set. This data is product data related to baby and child health products that are sold at Walgreens, a national retail pharmacy. The table shown has five observations (rows) and eight variables (columns). Each row is an observation and represents a baby and child healthcare product that was sold at a Walgreens store in a certain region of the country. The format of the data shown in the figure below is known as structured data. Structured data is data that is stored in a column and row format like that shown in the figure below.

terms to know

Data

Facts and figures collected from business processes.

Variable

An attribute or characteristic that can be measured.

Observation

The value of a variable.

Data Set

The collection of variables and their observations.

Structured Data

Organized data set that follows a predefined structure like rows and columns.

2. Cross-Sectional Data

Cross-sectional data is data collected at one point in time. For example, if you were told that the baby and children's product data from Walgreens were products sold at Walgreens stores on June 24th, then this data would be cross-sectional data.

Businesses like Walgreens could use this cross-sectional data in a variety of ways.

Pricing: Walgreens could perform a web scrape for prices of their competitor’s websites for the same products for a particular day (say June 24th) and use this information to optimize their pricing. Pricing decisions can be made by analyzing how different products or services are priced in the market.
Inventory Management: The table shown below could be generated from the cross-sectional data provided previously. The table displays the count of the number of items for each brand of cough medicine and whether it is in stock or not across 15 stores in the southwestern region of the United States.

	Number of Stores in Southwestern Region In/Out-of-Stock for Children’s Medicines
Brand	In-Stock	Out-of-Stock
Children’s Tylenol	5	10
Little Remedies	11	4
Children’s Delsym	8	7

The table constructed can be used to assist stores in stocking the correct number of brands of cough syrup. A data analyst could produce this table daily or weekly to provide visibility of the stock for the product. This visibility enables quicker restocking decisions and efficient merchandise replenishment.

term to know

Cross-Sectional Data

Data collected at a single point in time or over a short period, capturing a snapshot of multiple subjects, such as individuals, companies, or countries.

3. Time Series Data

Time series data are collected at regular intervals and over time. Time series data can be used to discover what has happened in the past, identify trends and patterns over time, and help projections for the future.

The structure of time series data looks different from cross-sectional data. Continuing with the Walgreens example, the table at right is an example of time series data for children’s cough syrup. Unlike the cross-sectional data table shown in the previous section, there are multiple rows for the same product. Each row contains sales information related to a certain children's cough product for a different time (each month in this case).

The way you use time series data is different from cross-sectional data. Time series data can track products over time so Walgreens could use the data in the table to forecast and understand how the sales of the products change over time.

Using the data in the table above, a time series graph like the one shown below could be constructed. The time series graph shows the sales of the three cough syrup products for each month in 2023. This graph could be used for trend analysis. For example, if Walgreens sees that most sales occur during December and January, store inventory can be adjusted accordingly. Walgreens can clearly see which product has sold the most in the past (Children’s Tylenol) and ensure there is plenty of that product in stock. Children’s Tylenol is clearly the market leader for children’s cough medicine. If Walgreens wanted to measure the effectiveness of an advertising campaign in August, they would want to be sure they were considering the regular seasonal trends.

The seasonality of cough medicine sales can inform effective marketing strategies for ancillary products like cough drops. The marketing team at Walgreens can align their campaigns with these time periods to maximize visibility and sales. For example, care packages could be created that contain cough syrup and cough drops. These care packages could be set up using an eye-catching display that is strategically located in the pharmacy aisles. Walgreens could create flu prevention tips on social media or in-store signage. Cough drops could be positioned as part of a proactive health strategy.

terms to know

Time Series Data

A collection of observations recorded sequentially over time at regular intervals. This type of data captures how a variable changes over time.

Time Series Graph

A visualization that displays data over time where the x-axis is a time component, and the y-axis is a quantitative value.

4. Differences Between Cross-Sectional and Time Series Data

Cross-sectional data captures a view of the data at a specific point in time. Whereas, time series data tracks a variable’s values over a continuous sequence of time intervals (e.g., days, months, years).

	Cross-Sectional	Time Series
Collection	One point in time	Regular intervals of time
Purpose	Characterize the data at one moment in time	Identifying trends, patterns, and cycles

try it

You decide you want to start tracking the number of steps you walk each day using a fitness tracker app. Every day, the app records the number of steps you take. Over several weeks, you accumulate a series of data points, each corresponding to a specific day.

Is this cross-sectional or time series data?

This data is a time series because it tracks steps over time, and may be helpful for identifying trends, patterns, and cycles in your daily movement.

You are a Human Resources (HR) data analyst in a large company. The company has decided it is going to start offering snacks in the break rooms on each floor. You hold a session for employees to provide their feedback on their favorite snacks. You ask each employee to select their top three favorite snacks (for example, fruit, granola bars, and Fig Newtons) from a list.

Is this cross-sectional or time series data?

This data is cross-sectional because it captures preferences at a single moment in time.

summary

In this lesson, you were introduced to a formal definition of data and the different elements like variables and observations that make up a data set. You learned about the difference between cross-sectional and time series data. A pharmacy retail scenario was used to illustrate how businesses use and apply analytics using cross-sectional and time series data. In this lesson, you focused on the idea that in business data analytics, data are values collected about some phenomena, typically a business process. This allows the examination of cross-sectional data, which captures a snapshot at a single point in time, such as survey results from a group of people on their current health status. And in contrast, time series data, which tracks changes over a period, like Walgreens' monthly sales of different children’s cough syrup brands over the year. Cross-sectional data is useful for comparing different subjects at one time, while time series data is essential for analyzing trends and making predictions. Understanding these differences is crucial for effective data analysis and decision-making.

Source: THIS TUTORIAL WAS AUTHORED BY SOPHIA LEARNING. PLEASE SEE OUR TERMS OF USE.

REFERENCES

Product data from Walgreens - dataset by Data-Hut. (2024, July 8). data.world. www.data.world/data-hut/product-data-from-walgreens

Terms to Know

Cross-Sectional Data: Data collected at a single point in time or over a short period, capturing a snapshot of multiple subjects, such as individuals, companies, or countries.
Data: Facts and figures collected from business processes.
Data Set: The collection of variables and their observations.
Observation: The value of a variable.
Structured Data: Organized data set that follows a predefined structure like rows and columns.
Time Series Data: A collection of observations recorded sequentially over time at regular intervals. This type of data captures how a variable changes over time.
Time Series Graph: A visualization that displays data over time where the x-axis is a time component, and the y-axis is a quantitative value.
Variable: An attribute or characteristic that can be measured.