Data Analytics Flashcards

Which SQL aggregate function counts only non-NULL values in a column?

Answer: COUNT(column_name)

COUNT(column_name) counts only the non-NULL values in that column, while COUNT(*) counts all rows regardless of NULL values.

What is the difference between structured and unstructured data?

Answer: Structured data fits a predefined schema (e.g., tables); unstructured data lacks a fixed format (e.g., text, images)

Structured data is organized in rows and columns with a consistent schema, while unstructured data (emails, images, video) has no predefined format.

In data analytics, what is 'feature engineering'?

Answer: Creating or transforming input variables to improve model performance

Feature engineering involves creating new variables or transforming existing ones (e.g., extracting month from a date) to provide more useful information to a model.

What does the HAVING clause do in SQL that WHERE cannot?

Answer: Filters groups based on aggregate function results

HAVING filters groups after aggregation (e.g., HAVING COUNT(*) > 5), while WHERE filters individual rows before aggregation occurs.

A scatter plot shows data points that form a U-shaped curve. What type of relationship does this suggest?

Answer: Non-linear (quadratic) relationship

A U-shaped scatter plot indicates a non-linear (quadratic) relationship where linear correlation would be near zero but a clear pattern exists.

Which pandas method is used to read a CSV file into a DataFrame?

Answer: pd.read_csv()

pd.read_csv() is the standard pandas function for loading comma-separated value files into a DataFrame object.

In the context of data quality, what does 'data completeness' refer to?

Answer: All required data is present with no missing values

Data completeness measures whether all expected records and fields are present; missing values are a completeness issue.