What does GROUP BY do in SQL, and when would you use HAVING?

What does GROUP BY do in SQL, and when would you use HAVING?

• A. GROUP BY aggregates rows by one or more columns; HAVING filters groups after aggregation.
• B. GROUP BY sorts rows; HAVING limits results to top N.
• C. GROUP BY creates new tables; HAVING merges tables.
• D. GROUP BY unions datasets; HAVING counts groups.

Answer: (A)

Grouping collects rows that share the same values in one or more columns into distinct groups so you can compute an aggregate for each group. After those groups are formed, HAVING filters which groups to keep based on the results of those aggregates. For example, you might group by department and sum the sales in each department, then keep only departments where that total exceeds a threshold. A concrete query would be: select department, SUM(sales) as total_sales from orders group by department having SUM(sales) > 1000. Here, first the data is divided into departments, then total_sales is calculated for each, and finally groups with total_sales <= 1000 are removed by the HAVING clause. If you used a WHERE clause, you’d filter individual rows before grouping; HAVING is used specifically to filter after the aggregation has occurred, based on aggregate values.

Grouping collects rows that share the same values in one or more columns into distinct groups so you can compute an aggregate for each group. After those groups are formed, HAVING filters which groups to keep based on the results of those aggregates.

For example, you might group by department and sum the sales in each department, then keep only departments where that total exceeds a threshold. A concrete query would be: select department, SUM(sales) as total_sales from orders group by department having SUM(sales) > 1000. Here, first the data is divided into departments, then total_sales is calculated for each, and finally groups with total_sales <= 1000 are removed by the HAVING clause.

If you used a WHERE clause, you’d filter individual rows before grouping; HAVING is used specifically to filter after the aggregation has occurred, based on aggregate values.