Timedora
Free Practice Quiz Question List

Data Preparation, Validation, and Documentation Online Quiz Questions

Use this free practice quiz with 20 questions to review Data Preparation, Validation, and Documentation, test your knowledge, and prepare for your next test or exam.

20 questions
01
Choose one
1 point

Which approach best validates prepared data before it is used for analysis?

  1. A

    Inspect the output once and proceed if it looks reasonable.

  2. B

    Run checks only when the preparation process reports an error.

  3. C

    Use repeatable tests to compare the output with expectations before analysis.

02
Choose all
1 point

Which practices help profile data before changing it? Select all that apply.

  1. A

    Establish what one row represents and examine the fields and data types.

  2. B

    Delete unusual values before checking whether they are valid.

  3. C

    Check completeness, validity, uniqueness, consistency, or distribution.

  4. D

    Check whether a visual profile covers the full dataset or only a sample.

03
True or false
1 point

When cleaning data, unusual values should always be deleted because they are errors.

  1. A

    True

  2. B

    False

04
Written response
1 point

Before grouping sales by month, what data type should text dates be converted to?

05
Fill in the blank
1 point

Before joining two datasets, check that their matching keys have .

06
Open ended
1 point

What information should be documented about prepared data so others can interpret it and reproduce the results?

07
Choose one
1 point

Which documentation approach best helps others interpret prepared data and reproduce its results?

  1. A

    Record only the final row count and omit how the data was prepared.

  2. B

    Record the source, row grain, preparation decisions, limitations, and preparation version or date.

  3. C

    Use generic names for fields and omit the definitions so they remain concise.

08
Choose all
1 point

Which checks can be used to validate prepared data against expectations before analysis? Select all that apply.

  1. A

    Check row counts and whether transformations or joins changed the intended grain.

  2. B

    Check key uniqueness where expected and inspect allowed values or missingness.

  3. C

    Treat a join as validated if it completes without an error.

  4. D

    Compare totals against a trusted source.

09
Written response
1 point

If each product_id is expected to identify one row in a product table, what should you check for product_id before joining it to sales transactions?

10
Fill in the blank
1 point

Convert text dates to a before grouping sales by month. A transformation should preserve a field’s underlying .

11
Choose one
1 point

A sales dataset contains a value far above the expected range. What is the most appropriate first step when cleaning the data?

  1. A

    Delete it immediately because unusual values are errors.

  2. B

    Investigate it before deciding how to handle it.

  3. C

    Replace it with the median without checking its source.

  4. D

    Remove every record containing an unusual value.

12
True or false
1 point

Every missing value should be imputed, regardless of the analysis or the likely effect on results.

  1. A

    True

  2. B

    False

13
Written response
1 point

What data-preparation term describes what one row in a dataset represents and should be recorded in its documentation?

14
Choose one
1 point

You need to retain every sales transaction from the left dataset and add product details wherever a matching product is found. Which join type fits this requirement?

  1. A

    An inner join, which keeps only rows with matches in both datasets.

  2. B

    A left join, which keeps every left-side row and adds matching right-side values.

  3. C

    A full outer join, which keeps only rows with matches in both datasets.

  4. D

    A left join, which keeps every right-side row and discards unmatched left-side rows.

15
True or false
1 point

A visual data profile can be assumed to cover the entire dataset because it displays a summary.

  1. A

    True

  2. B

    False

16
Written response
1 point

Before using prepared data for analysis, what should validation checks be treated as rather than one-time inspections?

17
Choose one
1 point

A sales dataset stores dates as text. What should you do before grouping sales by month?

  1. A

    Group the text values into months without converting them.

  2. B

    Convert the dates into currency values before grouping.

  3. C

    Convert the text dates to a date type before grouping.

  4. D

    Change the business definition of the date field before grouping.

18
Choose one
1 point

A location field uses “NY,” “N.Y.,” and “New York” for the same place. How should you handle these inconsistent values during cleaning?

  1. A

    Use a documented mapping to standardize the equivalent forms.

  2. B

    Keep each form as a separate category because the labels differ.

  3. C

    Replace all forms with a guessed value without recording the rule.

  4. D

    Delete every record containing one of the different forms.

19
Choose one
1 point

A team is preparing a dataset for analysis. What should it do first to assess whether the data is fit for its intended purpose?

  1. A

    Clean the data until every field is complete and every dataset looks perfect.

  2. B

    Establish what the data contains and assess its quality and fitness for the intended analysis before changing it.

  3. C

    Remove unusual values before checking how the data was collected.

  4. D

    Change the data to match common formats, then decide what analysis to perform.

20
Choose one
1 point

Which documentation set would best help another analyst interpret prepared data and reproduce the results?

  1. A

    Only the final row count and the name of the person who prepared the data.

  2. B

    A list of every value in the dataset, without details about how it was prepared.

  3. C

    The dataset’s source and date range, row grain, field definitions, checks, exclusions, transformations, join details, limitations, and preparation version or date.

  4. D

    The analysis conclusions, leaving out the preparation steps so the results remain concise.