vicharanashala-org/participation-points-remote-internship

Participation Points in a Remote Internship

Points, attendance, polls and badges of 4,476 learners, May–Sep 2026

Eight tables from a large online internship programme run at a National Institute in India, 15 May to 14 September 2026. Learners joined on a rolling basis and earned participation points for attending live sessions, answering live polls, peer learning, answering peer queries and completing a reviewed project. Points unlocked levels and badges, and learners could share their badges. Every learner appears only as an anonymous token (S00001 …), which is the same in every file.

Files

learners.csv — 4,476 rows, one per learner. The list of learners; every other file joins to it.

column	meaning	purpose
learner_id	anonymous learner token	join key for all files
start_date	date the learner joined the programme	align learners by time since joining
planned_end_date	planned end date of the internship (filled for 40%)	plan, not the outcome
started	true = at least one point earned after the initial grant	keep only learners who actually took part

points_ledger.csv — 108,314 rows, one per points event. The full points history of every learner.

column	meaning	purpose
learner_id	anonymous learner token	join key
event_date	date of the event (UTC)	order events; rebuild points over time
category	initial_grant, attendance, poll, peer_learning, query, project, goal_setting, faq_reuse, manual_adjustment	which activity earned the points
session_label	live session the row belongs to; empty if not tied to a session	join to sessions, attendance and poll_participation
points_requested	points the rule computed	shows where a category cap cut the points
points_applied	points actually added after caps (can be negative for penalties)	sum this column to rebuild a learner's total
balance_after	learner's running total after this row	check totals
reason	system-written sentence explaining the row	read the exact rule applied

sessions.csv — 121 rows, one per live session.

column	meaning	purpose
session_label	session name; primary key	join key
session_date	date of the session	place sessions in time
slot	morning, afternoon, evening, followup; empty when a day had one session	tell apart two sessions on one day
total_minutes	official length of the session	denominator for attendance rates
n_attendance_rows	rows for this session in attendance.csv	check completeness
n_poll_rows	rows for this session in poll_participation.csv	check completeness

attendance.csv — 106,844 rows, one per learner per live session.

column	meaning	purpose
learner_id	anonymous learner token	join key
session_label	session name	join to sessions.csv
session_date	date of the session	place in time
attended_minutes	minutes counted present	attendance measure
total_minutes	official length of the session	denominator
attendance_pct	attended_minutes / total_minutes × 100	attendance rate
qualified	true = met that day's attendance threshold	the attendance outcome that earned points

poll_participation.csv — 96,352 rows, one per learner per live session.

column	meaning	purpose
learner_id	anonymous learner token	join key
session_label	session name	join to sessions.csv
session_date	date of the session	place in time
total_questions	poll questions asked in the session	denominator
attempted	questions the learner answered	participation measure
missed	total_questions − attempted	participation measure

poll_responses.csv — 257,135 rows, one per learner per poll question, from 16 July.

column	meaning	purpose
learner_id	anonymous learner token	join key
poll_date	date of the poll	join to sessions on date
day_number	programme day number	order polls
question_no	question number within that day's poll	question-level analysis (question text not included)
result	correct, incorrect or blank	correctness measure
points_earned	the poll tool's own speed-weighted score for the whole poll (not participation points)	performance within the poll
rank_in_poll	learner's rank in that poll (1 = top)	relative performance
n_learners_in_poll	learners who took part in that poll	denominator for rank

badges.csv — 1,546 rows, one per badge.

column	meaning	purpose
learner_id	anonymous learner token	join key
badge_id	badge identifier, e.g. ms:level:5 (milestone) or rank:total:week:2026-08-10:2 (weekly top 3)	join to badge_shares
kind	milestone or rank	badge type
board	leaderboard for rank badges: total, attendance, poll, query; empty for milestones	which leaderboard
place	1–3 for rank badges, 0 for milestones	podium position
period	period label, e.g. All-time, Week of Jul 6–12	readable period
period_key	all, or the start date of the week or period	machine key for the period
title	badge title as shown to the learner	badge name
detail	detail line shown on the badge	what the badge was for
first_seen_at	when the badge was first recorded (UTC)	badge timing (see Notes)

badge_shares.csv — 1,095 rows, one per badge share, from 12 August.

column	meaning	purpose
learner_id	anonymous learner token	join key
badge_id	badge shared	join to badges.csv
kind, board, place, period, period_key	as in badges.csv	describe the shared badge
shared_at	when the share happened (UTC)	sharing timing

Notes

• Join on learner_id. Join session-level files on session_label, not the date, because some days had two sessions. Join poll_responses.csv on learner_id and poll_date.
• Every learner starts with one initial_grant row of +100 points. Summing points_applied in file order reproduces balance_after.
• About 741 learners were re-registered between 30 June and 5 July. Their start_date is the later date, and their earlier attendance and poll points are not in the ledger.
• Project points (+500) were added some time after the review and dated back.
• From the end of July, attendance minutes were derived from poll answers rather than measured session minutes.
• first_seen_at is when a badge was first recorded, often when the learner first opened the badges page, not the moment the threshold was crossed.
• Dates are calendar days in India time (UTC+05:30); timestamps are UTC.
• Licence: MIT.

Citation

Also published on IEEE DataPort: https://dx.doi.org/10.21227/6aqq-vm56

@data{6aqq-vm56-26,
  doi       = {10.21227/6aqq-vm56},
  url       = {https://dx.doi.org/10.21227/6aqq-vm56},
  author    = {Sharma, Sakshi and Iyengar, S. R. S.},
  publisher = {IEEE Dataport},
  title     = {Participation Points in a Remote Internship},
  year      = {2026}
}