pbPassingBI
/
power bi/quizzes/power-bi-quiz-transform
12 questions

Data import & transformation — quiz

12 questions covering this module. Connecting to sources, Power Query, cleaning, merging and appending.

  1. Where should you remove unneeded columns?

    1. In DAX with a measure
    2. In Power Query before load
    3. In the report visual
    4. In the Service after publishing

    Answer: In Power Query before load — Columns removed in Power Query never enter the model, so they cost nothing in size or refresh.

  2. When does DAX evaluate?

    1. At refresh time
    2. At query time, when a visual is rendered
    3. When the file is saved
    4. Only on publish

    Answer: At query time, when a visual is rendered — DAX measures are computed on demand as users interact with visuals.

  3. In a composite model, dimension tables joined to DirectQuery facts should usually be:

    1. Import
    2. DirectQuery
    3. Dual
    4. Live connection

    Answer: Dual — Dual lets the engine serve from memory when possible and query the source when a join requires it.

  4. Which is a genuine limitation of DirectQuery?

    1. No relationships allowed
    2. Many DAX functions are unavailable or slow
    3. No visuals can be used
    4. Refresh is impossible

    Answer: Many DAX functions are unavailable or slow — DirectQuery restricts the DAX surface and pushes work to the source on every interaction.

  5. Which action commonly breaks query folding?

    1. Filtering rows
    2. Removing columns
    3. Adding an index column
    4. Renaming a column

    Answer: Adding an index column — An index column has no SQL equivalent, so folding stops there and later steps run locally.

  6. Wide spreadsheet data with one column per month should be:

    1. Pivoted
    2. Unpivoted
    3. Merged
    4. Grouped

    Answer: Unpivoted — Unpivot converts the month columns into attribute/value rows, which the model can actually use.

  7. Which is valid M?

    1. table.selectrows(Sales, each [Amt] > 0)
    2. Table.SelectRows(Sales, each [Amt] > 0)
    3. SELECT * FROM Sales WHERE Amt > 0
    4. Table.selectRows(Sales, [Amt] > 0)

    Answer: Table.SelectRows(Sales, each [Amt] > 0) — M is case sensitive, and each supplies the per-row context.

  8. How do you stop one unparseable value failing a refresh?

    1. Remove the column
    2. Wrap the conversion in try ... otherwise
    3. Use DirectQuery
    4. Disable load

    Answer: Wrap the conversion in try ... otherwise — try ... otherwise substitutes a fallback instead of raising a step error.

  9. What does query folding mean?

    1. Collapsing steps
    2. Power Query translates steps into a query the source runs
    3. Compressing the file
    4. Merging queries

    Answer: Power Query translates steps into a query the source runs — It keeps the work in the database.

  10. View Native Query is greyed out on a step. This means:

    1. The step is invalid
    2. Folding has stopped at or before it
    3. The source is a file
    4. The query is slow

    Answer: Folding has stopped at or before it — Everything after is processed locally.

  11. Which join kind returns only first-table rows with no match?

    1. Inner
    2. Left anti
    3. Left outer
    4. Full outer

    Answer: Left anti — It answers "which of these are missing" directly.

  12. A merge matches nothing on identical-looking values. Fix:

    1. Change join kind
    2. Trim and lowercase both key columns
    3. Use Append
    4. Add an index

    Answer: Trim and lowercase both key columns — Whitespace and case are invisible in the preview.