Real-life Examples of Data Cleaning in Data Science Projects for High School Students

🧠 Quick Study Guide: Data Cleaning Essentials

• 💡 What is Data Cleaning? It's the process of detecting and correcting (or removing) corrupt, inaccurate, irrelevant, or incomplete records from a dataset. Think of it as tidying up your data so it's ready for analysis.
• 📊 Why is it Important? "Garbage in, garbage out!" Clean data leads to accurate insights, reliable models, and better decision-making. Messy data can lead to skewed results and wrong conclusions.
• 🚫 Common Data Issues:
  • ❓ Missing Values: Gaps in your data (e.g., a student's grade is missing).
  • 👯 Duplicate Records: The same entry appears multiple times (e.g., two identical rows for the same product).
  • 🔀 Inconsistent Formats: Data entered differently (e.g., "USA", "U.S.A.", "United States" for the same country).
  • 📈 Outliers: Data points significantly different from others, possibly due to errors or rare events (e.g., a student scoring 500% on a test).
  • ✏️ Typographical Errors: Simple spelling mistakes (e.g., "Appel" instead of "Apple").
• 🛠️ Basic Cleaning Techniques:
  • 🗑️ Handling Missing Values:
    • ✂️ If few: Remove rows/columns.
    • ➕ If many: Impute (fill in) with mean, median, mode, or a predicted value.
  • ♻️ Removing Duplicates: Identify and delete identical records.
  • ✅ Standardizing Data: Convert inconsistent formats to a uniform one (e.g., all dates to YYYY-MM-DD).
  • 🧐 Validating Data: Check if data falls within expected ranges or follows certain rules (e.g., age cannot be negative).
  • 🔍 Dealing with Outliers: Investigate if they are errors, remove them if they are, or transform the data if they represent genuine extremes.

📝 Practice Quiz: Test Your Data Cleaning Skills!

1. A high school teacher is analyzing student test scores and notices that one student's score is completely blank for a major exam. To avoid losing the student's other valuable data, what is a common data cleaning approach for this missing value?
   A) Remove the entire student's record from the dataset.
   B) Randomly assign a score between 0 and 100.
   C) Impute the missing score with the average (mean or median) score of the class.
   D) Assign a score of 0, assuming the student didn't take the test.
2. Imagine you're analyzing a dataset of daily temperatures for your city. You find an entry that reads "Temperature: -500°C". Given that the lowest recorded temperature on Earth is around -89°C, what kind of data issue does this likely represent?
   A) A missing value.
   B) A duplicate record.
   C) An outlier or a typographical error.
   D) An inconsistent date format.
3. You're building a simple recommendation system for books based on user ratings. Some users rate books on a scale of 1-5, while others use "Excellent", "Good", "Fair", "Poor". What data cleaning step is crucial before combining these ratings?
   A) Remove all ratings that are not numbers.
   B) Convert all text ratings into their corresponding numerical scale (e.g., "Excellent" to 5).
   C) Delete any user who used text ratings.
   D) Treat text ratings as missing values.
4. In a dataset tracking high school football game statistics, you notice that a specific player's tackle count for a game appears twice, with identical values and timestamps. What is the most appropriate data cleaning action?
   A) Average the two duplicate tackle counts.
   B) Remove one of the duplicate records.
   C) Flag both records as outliers.
   D) Impute a new tackle count for that player.
5. A student is collecting data on the favorite colors of their classmates. One student accidentally types "blle" instead of "blue". This is an example of what type of data quality issue?
   A) Missing value.
   B) Inconsistent format.
   C) Typographical error.
   D) Outlier.
6. You are analyzing data from a school's energy consumption sensors. For a particular hour, the sensor recorded "N/A" for electricity usage. This situation is a clear example of what common data cleaning challenge?
   A) Data inconsistency.
   B) Duplicate data.
   C) Missing values.
   D) Data outlier.
7. A data scientist is preparing a dataset of customer feedback. Some feedback is written in all caps, some in title case, and some in lowercase. To ensure consistency for text analysis, what technique should be applied?
   A) Remove all feedback written in all caps.
   B) Standardize the text by converting all feedback to a uniform case (e.g., lowercase).
   C) Identify and remove any feedback longer than 100 characters.
   D) Impute missing words in the feedback.