Investigate repeated submissions

The question

An attendance upload may contain repeated submissions. Prepare the evidence without deleting any records.

df Duplicate evidence

Your inputs

`df`

submission_id identifies one submission. Identical IDs are repeated uploads. Do not remove or change source records.

Field meanings and units

submission_id: Attendance-submission identifier.
student: Learner label.
group: Comparison group.
hours: Study duration in hours.
score: Assessment score in points.
attended: Whether the learner attended.
date: Calendar date supplied as YYYY-MM-DD text unless the task states it is invalid.

`df` · first 5 of 16 rows
`submission_id`	`student`	`group`	`hours`	`score`	`attended`	`date`
S001	Learner 1	A	2	55	false	2026-04-01
S002	Learner 2	B	5	78	true	2026-04-01
S003	Learner 3	A	3	68	true	2026-04-02
S004	Learner 4	B	7	90	true	2026-04-02
S005	Learner 5	A	4	72	false	2026-04-03

View all 16 rows

`df` · complete input
`submission_id`	`student`	`group`	`hours`	`score`	`attended`	`date`
S001	Learner 1	A	2	55	false	2026-04-01
S002	Learner 2	B	5	78	true	2026-04-01
S003	Learner 3	A	3	68	true	2026-04-02
S004	Learner 4	B	7	90	true	2026-04-02
S005	Learner 5	A	4	72	false	2026-04-03
S006	Learner 6	B	6	85	true	2026-04-03
S007	Learner 7	A	8	94	true	2026-04-04
S008	Learner 8	B	1	49	true	2026-04-04
S009	Learner 9	A	2	56	false	2026-04-05
S010	Learner 10	B	5	79	true	2026-04-05
S011	Learner 11	A	3	69	true	2026-04-06
S012	Learner 12	B	7	91	true	2026-04-06
S004	Learner 4	B	7	90	true	2026-04-02
S014	Learner 14	B	6	86	true	2026-04-07
S009	Learner 9	A	2	56	false	2026-04-05
S016	Learner 16	B	1	50	true	2026-04-08

The editable setup in Your Python creates these inputs when you run it.

Your task · Investigate repeated submissions

An attendance upload may contain repeated submissions. Prepare the evidence without deleting any records.

repeatedDataFramerepeated_idsnumber

Deliverables

Make the outputs listed below. Use each name exactly as shown. Then choose Check answer.

Python name key

Teal outlines name table columns. Violet left rules name Python objects. Dotted rules mark syntax, functions and parameters. Use each name exactly as shown.

Every repeated memberOutputrepeatedDataFrame
1. Create repeated with every row whose submission_id appears more than once, including the first occurrence.
2. Keep all columns and source order.
Affected submissionsOutputrepeated_idsnumber
1. Store the number of distinct submission_id values in repeated_ids.

Data policy Keep all supplied source tables unchanged. Work on separate outputs.

Help Hints & explained solution

Choose the amount of help you need. Opening one hint does not reveal the others.

Hint 1 — Think

Would flagging only later copies show the whole investigation?

Hint 2 — Tools

duplicated(), nunique() and boolean selection

Revisit: Find duplicate rows · What values exist?

Hint 3 — Approach

Find every member of repeated submission IDs, retain their full records, then count distinct affected IDs.

Explained solution

Add these lines below # Your work in the editor. The supplied input setup is already there.

repeated = df[df.duplicated(subset=['submission_id'], keep=False)]
repeated_ids = repeated.submission_id.nunique()
repeated

View supplied setup

This code is already at the top of the editor.

import pandas as pd

# Supplied input: df (16 rows)
df = pd.DataFrame(
    [
        ('S001', 'Learner 1', 'A', 2, 55, False, '2026-04-01'),
        ('S002', 'Learner 2', 'B', 5, 78, True, '2026-04-01'),
        ('S003', 'Learner 3', 'A', 3, 68, True, '2026-04-02'),
        ('S004', 'Learner 4', 'B', 7, 90, True, '2026-04-02'),
        ('S005', 'Learner 5', 'A', 4, 72, False, '2026-04-03'),
        ('S006', 'Learner 6', 'B', 6, 85, True, '2026-04-03'),
        ('S007', 'Learner 7', 'A', 8, 94, True, '2026-04-04'),
        ('S008', 'Learner 8', 'B', 1, 49, True, '2026-04-04'),
        ('S009', 'Learner 9', 'A', 2, 56, False, '2026-04-05'),
        ('S010', 'Learner 10', 'B', 5, 79, True, '2026-04-05'),
        ('S011', 'Learner 11', 'A', 3, 69, True, '2026-04-06'),
        ('S012', 'Learner 12', 'B', 7, 91, True, '2026-04-06'),
        ('S004', 'Learner 4', 'B', 7, 90, True, '2026-04-02'),
        ('S014', 'Learner 14', 'B', 6, 86, True, '2026-04-07'),
        ('S009', 'Learner 9', 'A', 2, 56, False, '2026-04-05'),
        ('S016', 'Learner 16', 'B', 1, 50, True, '2026-04-08'),
    ],
    columns=['submission_id', 'student', 'group', 'hours', 'score', 'attended', 'date'],
)

Why these steps exist

keep=False marks the first upload as well as later copies, so the investigation can compare every member.
Distinct repeated IDs count affected submissions; counting rows would count their copies too.

Another valid approach

Use duplicate membership with all copies included; inspecting only later copies misses part of the evidence.