Data Cleaning and Preprocessing

Active learning works well for data cleaning because students need to experience the frustration of messy data to truly understand why cleaning matters. Hands-on activities make abstract concepts like outliers and missing values concrete and memorable, preparing students for real-world data work.

• 10th Grade
• Computer Science
• 4 activities
• 15 min–35 min

Learning Objectives

1. Identify common data inconsistencies such as missing values, duplicate entries, and formatting errors in a given dataset.
2. Analyze the impact of specific data quality issues, like outliers or incorrect data types, on statistical calculations and visualizations.
3. Formulate a step-by-step plan to clean a provided messy dataset, justifying each cleaning decision.
4. Evaluate the effectiveness of different data cleaning strategies for a specific analytical goal.
5. Demonstrate the application of data cleaning techniques using a programming tool or spreadsheet software.

Gallery Walk

⏱ 35 min·Small Groups

Gallery Walk: The Messy Dataset Museum

Print five different messy datasets and post them around the room, each with a different type of data quality problem (duplicates, missing values, format mismatches, outliers, impossible values). Groups rotate through stations with sticky notes to identify the problem type and propose a cleaning strategy before moving on.

Prepare & details

Explain the common types of data inconsistencies and errors.

Facilitation Tip: During the Gallery Walk, position students as curators who must explain their cleaning decisions to peers using the provided rubric.

Setup: Wall space or tables arranged around room perimeter

Materials: Large paper/poster boards, Markers, Sticky notes for feedback

• Understand
• Apply
• Analyze
• Create
• Relationship Skills
• Social Awareness

Think-Pair-Share

⏱ 20 min·Pairs

Think-Pair-Share: Should We Delete It?

Give students a dataset with 15% missing age values and ask them individually to decide whether to delete rows, fill with the mean, or flag the records. Pairs compare decisions and discuss trade-offs, then share cases where they disagreed and why.

Prepare & details

Analyze the impact of dirty data on analytical results.

Facilitation Tip: For Think-Pair-Share, insist that pairs produce a single list of deletion criteria and a justification before sharing with the class.

Setup: Standard classroom seating; students turn to a neighbor

Materials: Discussion prompt (projected or printed), Optional: recording sheet for pairs

• Understand
• Apply
• Analyze
• Self-Awareness
• Relationship Skills

Inquiry Circle

⏱ 30 min·Small Groups

Inquiry Circle: Before-and-After Analysis

Small groups receive the same raw sales dataset and a pre-cleaned version. They must reverse-engineer which cleaning steps were applied by comparing the two versions, then write a short cleaning log documenting each transformation in order.

Prepare & details

Construct a plan for cleaning a given messy dataset.

Facilitation Tip: In the Collaborative Investigation, assign each group a different cleaning technique so the class can compare outcomes and discuss trade-offs.

Setup: Groups at tables with access to source materials

Materials: Source material collection, Inquiry cycle worksheet, Question generation protocol, Findings presentation template

• Analyze
• Evaluate
• Create
• Self-Management
• Self-Awareness

Collaborative Problem-Solving

⏱ 15 min·Whole Class

Structured Discussion: The Cost of Dirty Data

Share a real case study (e.g., a hospital billing error or a census miscoding) where uncleaned data led to a costly mistake. The class discusses what preprocessing step could have caught the error, then identifies which step from their cleaning toolkit would apply.

Prepare & details

Explain the common types of data inconsistencies and errors.

Facilitation Tip: During the Structured Discussion, provide a list of real-world consequences of dirty data to guide the conversation.

Setup: Groups at tables with problem materials

Materials: Problem packet, Role cards (facilitator, recorder, timekeeper, reporter), Problem-solving protocol sheet, Solution evaluation rubric

• Apply
• Analyze
• Evaluate
• Create
• Relationship Skills
• Decision-Making
• Self-Management

Teaching This Topic

Teachers should model mistakes in datasets and demonstrate their own thought process when cleaning, making the invisible work visible. Avoid presenting cleaning as a checklist; instead, emphasize context and consequences. Research shows that students learn best when they see data cleaning as a detective story with multiple possible solutions rather than a single correct answer.

What to Expect

Students will confidently identify data errors, justify their cleaning choices, and explain how clean data supports reliable analysis. They will move beyond simple deletions to use multiple strategies and recognize cleaning as an ongoing process.

Watch Out for These Misconceptions

Common Misconception

During the Gallery Walk, watch for students who assume all problematic rows should be deleted without considering context or consequences.

What to Teach Instead

Use the Gallery Walk debrief to push students to explain why they chose deletion over other strategies like imputation or transformation, using the examples they observed.

Common Misconception

During the Think-Pair-Share activity, listen for students who say data errors are always easy to spot through visual inspection alone.

What to Teach Instead

In the pair phase, require students to use statistical summaries (min, max, unique counts) to find subtle errors before deciding on a cleaning method.

Common Misconception

During the Collaborative Investigation, some students may treat preprocessing and analysis as separate phases that don’t overlap.

What to Teach Instead

Use the before-and-after analysis to highlight how new issues often appear during analysis, requiring students to revisit their cleaning steps iteratively.

Assessment Ideas

Exit Ticket

After the Gallery Walk, provide students with a messy dataset (CSV snippet) and ask them to identify two specific data quality issues and suggest one cleaning step for each before leaving class.

Quick Check

During the Think-Pair-Share activity, present students with a scenario about a dataset of student test scores with missing values and text entries. Ask them to list three potential problems this data could cause and propose one method to address each problem.

Discussion Prompt

After the Structured Discussion, pose the question about product prices with extreme values. Facilitate a class discussion on critical thinking in data cleaning, using student responses to assess their understanding of context and thresholds.

Extensions & Scaffolding

• Challenge: Ask students to design a new dataset with intentional errors and write a cleaning guide for another student to follow.
• Scaffolding: Provide a partially cleaned dataset so students focus on identifying remaining issues rather than starting from scratch.
• Deeper exploration: Have students research how data cleaning is handled in a specific industry (e.g., healthcare, finance) and present their findings to the class.

Key Vocabulary

Missing Values	Data points that are absent or not recorded for a particular observation. These can be represented as blank cells, NA, or null.
Duplicate Records	Identical or near-identical entries for the same entity within a dataset. These can inflate counts and skew analysis.
Data Type Mismatch	Occurs when a column contains values that do not conform to the expected data type, such as text in a numerical field.
Outlier	A data point that significantly differs from other observations in the dataset. Outliers can be genuine extreme values or errors.
Data Imputation	The process of replacing missing data points with substituted values, such as the mean, median, or a predicted value.