4. Employee Performance and Improvement Rate (easy)

Problem Statement

Table: Performance
Each row in this table represents a performance evaluation of an employee, detailing the employee’s name, the evaluation result, the evaluation round, and the performance score.


+----------------+---------+
| Column Name    | Type    |          
+----------------+---------+
| employee_name  | varchar |           
| evaluation     | varchar |       
| round          | int     |          
| score          | int     |    
+----------------+---------+
This table may have duplicate rows.
This table contains information collected from performance evaluations of employees.
The round column has a value from 1 to 10.
The score column has a value from 1 to 100. An evaluation with a score less than 50 is considered a poor evaluation.

Develop a solution to find each employee_name, their performance, and improvement_rate.

We define employee performance as:

The average of the ratio between employee evaluation score and its round.

We also define improvement rate as:

The percentage of all evaluations with a score improvement over the previous rounds, and are greater than 50.

Both performance and improvement_rate should be rounded to 2 decimal places.

Return the result table in any order.

Example

Input:


Performance table:
+---------------+-------------+-------+-------+
| employee_name | evaluation  | round | score |
+---------------+-------------+-------+-------+
| Alice         | Efficiency  | 1     | 80    |
| Alice         | Productivity| 2     | 90    |
| Alice         | Creativity  | 3     | 60    |
| Bob           | Efficiency  | 1     | 40    |
| Bob           | Productivity| 2     | 60    |
| Bob           | Creativity  | 3     | 70    |
+---------------+-------------+-------+-------+

Output:


+---------------+-----------+-------------------+
| employee_name | performance | improvement_rate |
+---------------+-----------+-------------------+
| Alice         | 48.33      | 33.33             |
| Bob           | 31.11       | 66.67             |
+---------------+-----------+-------------------+

Output Explanation

Step 1: Calculate Employee Performance

For each employee:

Divide the score by the round for all evaluations.
Calculate the average of these ratios.

Alice:

$\text{Performance} = \frac{80}{1} + \frac{90}{2} + \frac{60}{3} = 80 + 45 + 20 = 145$

$\text{Average} = \frac{145}{3} = 48.33$

Bob:

$\text{Performance} = \frac{40}{1} + \frac{60}{2} + \frac{70}{3} = 40 + 30 + 23.33 = 93.33$ $\text{Average} = \frac{93.33}{3} = 31.11$

Step 2: Calculate Improvement Rate

For each employee:

Check if the score in each round:
- Is greater than 50.
- Is greater than the score in the previous round.
Count the number of evaluations that meet the criteria.

Alice:
- Round 1: 80 (No previous round, does not count).
- Round 2: 90 (Improved and >50).
- Round 3: 60 (Not improved but >50).
- Total: 1 improvement out of 3 evaluations.

\text{Improvement Rate} = \frac{1}{3} \times 100 = 33.33

Bob:
- Round 1: 40 (No previous round, does not count).
- Round 2: 60 (Improved and >50).
- Round 3: 70 (Improved and >50).
- Total: 2 improvements out of 3 evaluations.

\text{Improvement Rate} = \frac{2}{3} \times 100 = 66.67

Try It Yourself

Database Exercise

Database Schema:


-- Database schema would be rendered here

Exercise Script:


-- Exercise script would be rendered here

Available actions: Execute

Solution

To solve this problem, we need to calculate each employee’s performance and improvement rate based on their evaluation scores across different rounds. The approach involves systematically breaking down the problem into manageable steps, utilizing SQL window functions and aggregation techniques to derive the required metrics.

Calculate Performance: Determine the average ratio of each employee’s score to the corresponding evaluation round.
Determine Improvement Rate: Identify the percentage of evaluations where an employee’s score improved compared to the previous round and is greater than 50.
Round the Results: Ensure that both performance and improvement rate are rounded to two decimal places for precision.

SQL Query


WITH LaggedPerformance AS (
    SELECT 
        employee_name,
        round,
        score,
        LAG(score) OVER (PARTITION BY employee_name ORDER BY round) AS previous_score
    FROM Performance
)
SELECT 
    employee_name,
    ROUND(AVG(score * 1.0 / round), 2) AS performance,
    ROUND(
        (SUM(CASE 
                 WHEN score > 50 AND previous_score IS NOT NULL AND score > previous_score THEN 1 
                 ELSE 0 
             END) * 100.0 / COUNT(*)), 2
    ) AS improvement_rate
FROM LaggedPerformance
GROUP BY employee_name;

Step-by-Step Approach

Step 1: Calculate Performance

Compute the average of the ratio between each employee’s score and the evaluation round to assess overall performance.

SQL Query:


SELECT 
    employee_name, 
    AVG(score * 1.0 / round) AS performance
FROM 
    Performance
GROUP BY 
    employee_name;

Explanation:

AVG(score * 1.0 / round):
- Calculates the average of the score divided by the round number for each employee.
- Multiplying by 1.0 ensures that the division is performed using floating-point arithmetic, preserving decimal values.
GROUP BY employee_name:
- Aggregates the calculations for each employee individually.

Output After Step 1:


+---------------+-------------+
| employee_name | performance |
+---------------+-------------+
| Alice         | 48.3333333  |
| Bob           | 31.111111   |
+---------------+-------------+

Step 2: Determine Improvement Rate

Calculate the percentage of evaluations where an employee’s score improved compared to the previous round and is greater than 50.

SQL Query:


WITH LaggedPerformance AS (
    SELECT 
        employee_name,
        round,
        score,
        LAG(score) OVER (PARTITION BY employee_name ORDER BY round) AS previous_score
    FROM 
        Performance
)
SELECT 
    employee_name,
    ROUND(
        (SUM(CASE 
                 WHEN score > 50 
                      AND previous_score IS NOT NULL 
                      AND score > previous_score THEN 1 
                 ELSE 0 
             END) * 100.0 / COUNT(*)), 2
    ) AS improvement_rate
FROM 
    LaggedPerformance
GROUP BY 
    employee_name;

Explanation:

WITH LaggedPerformance AS (...):
- Creates a Common Table Expression (CTE) that adds a column previous_score representing the score from the previous round for each employee.
LAG(score) OVER (PARTITION BY employee_name ORDER BY round):
- Retrieves the score from the preceding round for each employee, enabling comparison between consecutive rounds.
SUM(CASE WHEN ... THEN 1 ELSE 0 END):
- Counts the number of evaluations where:
  - The current score is greater than 50.
  - There is a valid previous_score (i.e., not the first round).
  - The current score is higher than the previous_score.
(SUM(...) * 100.0 / COUNT(*)):
- Calculates the improvement rate as a percentage by dividing the count of improved evaluations by the total number of evaluations for each employee.
ROUND(..., 2):
- Rounds the improvement rate to two decimal places.

Output After Step 2:


+---------------+-------------------+
| employee_name | improvement_rate  |
+---------------+-------------------+
| Alice         | 33.33             |
| Bob           | 66.67             |
+---------------+-------------------+

Step 3: Combine Performance and Improvement Rate

Objective:
Merge the performance and improvement rate calculations to present a consolidated view for each employee.

SQL Query:


WITH LaggedPerformance AS (
    SELECT 
        employee_name,
        round,
        score,
        LAG(score) OVER (PARTITION BY employee_name ORDER BY round) AS previous_score
    FROM 
        Performance
)
SELECT 
    employee_name,
    ROUND(AVG(score * 1.0 / round), 2) AS performance,
    ROUND(
        (SUM(CASE 
                 WHEN score > 50 
                      AND previous_score IS NOT NULL 
                      AND score > previous_score THEN 1 
                 ELSE 0 
             END) * 100.0 / COUNT(*)), 2
    ) AS improvement_rate
FROM 
    LaggedPerformance
GROUP BY 
    employee_name;

Explanation:

Combining CTE and Aggregations:
- The CTE LaggedPerformance is used to access both the current and previous scores within the same query.
Calculating Performance and Improvement Rate Simultaneously:
- AVG(score * 1.0 / round) computes the performance metric.
- The SUM(CASE ...) construct calculates the number of improved evaluations, which is then converted into a percentage to represent the improvement rate.
Rounding the Results:
- Both performance and improvement_rate are rounded to two decimal places for clarity and consistency.

Final Output:


+---------------+-------------+-------------------+
| employee_name | performance | improvement_rate  |
+---------------+-------------+-------------------+
| Alice         | 48.33       | 33.33             |
| Bob           | 31.11       | 66.67             |
+---------------+-------------+-------------------+