使用Pandas计算分数变化并添加新列：个体心理健康分数跨时间点差值计算方案问询

To calculate the mental health score change between Timepoint 3 and Timepoint 1 for each user (and display this value only in the Timepoint 3 row), here's a straightforward approach using pandas:

Step-by-Step Breakdown #

1. Isolate Key Timepoints: Extract the mental health scores for each user at Timepoint 1 and Timepoint 3.
2. Compute Score Differences: Calculate the difference between each user's Timepoint 3 score and Timepoint 1 score.
3. Integrate Changes into Original Data: Merge the computed differences back into your DataFrame, then clear the values for non-Timepoint 3 rows to match your desired output format.

Code Implementation #

import pandas as pd

# Initialize your sample DataFrame
data = {
    'User': ['Bill', 'Bill', 'Bill', 'Wiz', 'Wiz', 'Wiz', 'Sam', 'Sam', 'Sam'],
    'Timepoint': [1, 2, 3, 1, 2, 3, 1, 2, 3],
    'Mental Health Score': [5, 10, 15, 10, 10, 15, 5, 5, 5]
}
df = pd.DataFrame(data)

# Calculate the score change (TP3 - TP1) for each user
tp1_scores = df[df['Timepoint'] == 1].set_index('User')['Mental Health Score']
tp3_scores = df[df['Timepoint'] == 3].set_index('User')['Mental Health Score']
change_values = (tp3_scores - tp1_scores).reset_index()
change_values.columns = ['User', 'Change in Mental Health (TP1 and 3)']

# Merge the change data into the original DataFrame
df = df.merge(change_values, on='User', how='left')

# Empty the change column for rows that aren't Timepoint 3
df.loc[df['Timepoint'] != 3, 'Change in Mental Health (TP1 and 3)'] = ''

# View the final result
print(df)

Final Output #

User  Timepoint  Mental Health Score Change in Mental Health (TP1 and 3)
0  Bill          1                    5                                     
1  Bill          2                   10                                     
2  Bill          3                   15                                     10
3   Wiz          1                   10                                     
4   Wiz          2                   10                                     
5   Wiz          3                   15                                      5
6   Sam          1                    5                                     
7   Sam          2                    5                                     
8   Sam          3                    5                                      0

This method scales well if you add more users to your dataset and ensures the change value only appears in the relevant row, just as you requested.

内容的提问来源于stack exchange，提问作者leahyota