EDC data entry, queries & lock
Enter data, resolve queries and lock forms correctly.
20 project-authored proposals · taxonomy v1.1 · 0 new benchmark runs. Verifiers and partial-credit rules below are designs, not executed results.
- Typical applications
- REDCap · Medidata Rave EDC
- Class GUI-need label
G3· GUI-only
Task proposals · 20
Expand to inspect the complete original wording and proposed checks. Original proposal text remains English in both interfaces.
R09.05.01.02-T01 Enter one subject's screening vitals from a source CRF into REDCap
Original instruction
REDCap project HTN-ALPHA (sandbox) is open as a site data-entry user. Transcribe input/crf_1003_screening.pdf (synthetic subject 1003) into the Screening event's Vital Signs instrument: height, weight, 3 seated BP readings, pulse, temperature unit. Answer any edit-check pop-up, then set the form status to Complete.
- Platforms
web· Any OS (web)- Applications
- REDCap
- Interfaces
brow· Browser- GUI need · task setting
G2· GUI-essential- GUI need · ordinary access
G2· GUI-essential- Interaction primitives
type· Text & formsclick· Point & click- Work product
WP02· Records/table- Conditions
- None declared
- Proposed difficulty
- 2 / 5
Proposed verifier
redcap_data rows for record 1003/Screening compared field-by-field with gold (numeric exact, units coded); form_complete=2. API token is not issued and Data Import Tool is disabled for the role; every change must appear in redcap_log_event with page DataEntry/index.php.
Proposed partial credit
60% field values exact; 20% edit-check pop-up answered without overriding valid data; 20% status Complete.
R09.05.01.02-T02 Enter Week 4 and Week 12 visits for three subjects in a longitudinal REDCap arm
Original instruction
In REDCap project HTN-ALPHA (Arm 1, events Screening/Baseline/Week 4/Week 12), enter Week 4 and Week 12 Vital Signs and Labs for subjects 1004, 1006 and 1009 from input/visit_worksheets_wk4_wk12.pdf. Subject 1006 missed Week 12: mark that event's forms per the CRF completion guidelines in input/ccg_v2.pdf instead of entering data.
- Platforms
web· Any OS (web)- Applications
- REDCap
- Interfaces
brow· Browser- GUI need · task setting
G2· GUI-essential- GUI need · ordinary access
G2· GUI-essential- Interaction primitives
type· Text & formsclick· Point & clicknav· Pan/zoom/scroll- Work product
WP02· Records/table- Conditions
- None declared
- Proposed difficulty
- 3 / 5
Proposed verifier
Per-field accuracy of redcap_data for 3 records x 2 events vs gold; 1006/Week 12 has the CCG missed-visit coding and no lab values. Data entered only via the web UI (redcap_log_event page DataEntry, no API or import events). Pass if accuracy = 100%.
Proposed partial credit
35% 1004+1009 values; 25% 1006 Week 4 values; 25% missed-visit coding; 15% all forms status set per CCG.
R09.05.01.02-T03 Resolve five open queries on a Rave subject against source documents
Original instruction
In a Medidata Rave EDC training study (site 104), subject 104-007 has 5 open queries. Using input/source_104-007.pdf, correct the 3 values that the source contradicts (with reason for change 'Transcription error') and answer the other 2 with 'Value confirmed per source document' without changing data. Do not close queries; leave them Answered for the monitor.
- Platforms
web· Any OS (web)- Applications
- Medidata Rave EDC
- Interfaces
brow· Browser- GUI need · task setting
G3· GUI-only- GUI need · ordinary access
G2· GUI-essential- Interaction primitives
click· Point & clicktype· Text & formsnav· Pan/zoom/scroll- Work product
WP08· App/system state- Conditions
licensed· Licensed pro app- Proposed difficulty
- 3 / 5
Proposed verifier
Rave audit trail export for 104-007: 3 data changes with reason code, 2 query responses with exact text, 5 queries in status Answered, 0 closed, no other datapoints changed. The training-site role is denied Rave Web Services (RWS) access and batch upload, so every change must come from the eCRF UI (audit source = EDC).
Proposed partial credit
45% three corrected values; 30% two confirmed responses with data untouched; 25% statuses Answered and no stray edits.
R09.05.01.02-T04 Log four adverse events and trigger the SAE form from a clinical narrative
Original instruction
Rave training study, subject 104-012. From input/ae_narrative_104-012.pdf enter the Adverse Events log: 4 events with verbatim term, onset/stop dates, ongoing flag, CTCAE grade, seriousness, relationship and action taken. One event meets SAE criteria; complete the dynamically added SAE form (hospitalization dates, outcome). Resolve every edit check the forms fire.
- Platforms
web· Any OS (web)- Applications
- Medidata Rave EDC
- Interfaces
brow· Browser- GUI need · task setting
G3· GUI-only- GUI need · ordinary access
G2· GUI-essential- Interaction primitives
type· Text & formsclick· Point & clicknav· Pan/zoom/scroll- Work product
WP02· Records/table- Conditions
licensed· Licensed pro app- Proposed difficulty
- 4 / 5
Proposed verifier
Clinical-view export of AE and SAE forms vs gold: 4 rows, dates exact, grade/seriousness/relationship coded; SAE form present only for the qualifying event; zero open system queries remain. RWS access and batch upload are disabled for the site role, and the audit trail must show eCRF entry by the site user.
Proposed partial credit
40% four AE rows correct; 25% correct event flagged serious; 20% SAE form complete; 15% no open edit checks.
R09.05.01.02-T05 Enter a seven-row concomitant medication log in OpenClinica
Original instruction
OpenClinica 4 sandbox study CARDIO-2, participant SS-0211, event Baseline. Enter the Concomitant Medications repeating group from input/conmed_list_ss0211.pdf: 7 medications with dose, unit, route, frequency, start date and ongoing flag. Two entries have partial start dates (month/year only); use the UNK convention from input/ccg_cardio2.pdf.
- Platforms
web· Any OS (web)- Applications
- OpenClinica
- Interfaces
brow· Browser- GUI need · task setting
G2· GUI-essential- GUI need · ordinary access
G2· GUI-essential- Interaction primitives
type· Text & formsclick· Point & click- Work product
WP02· Records/table- Conditions
- None declared
- Proposed difficulty
- 3 / 5
Proposed verifier
Clinical data export (ODM XML) for SS-0211: 7 item-group repeats matching gold in order-insensitive comparison; partial dates use UNK-DD format; form marked complete. REST/ODM import endpoints are disabled for the account and the audit log must show form-based entry.
Proposed partial credit
60% seven rows correct (pro rata per row); 25% partial dates per CCG; 15% form complete.
R09.05.01.02-T06 Clean a full Rave casebook to lock-ready state across nine forms
Original instruction
Rave training study, subject 104-019 (completed study). Source binder input/source_104-019/ (12 PDFs) and the monitor's findings in input/sdv_findings.xlsx. Correct every confirmed discrepancy with reasons, answer all 23 open queries, enter the 2 missing lab panels, then mark all forms SDV-complete and apply Freeze. Do not apply Lock; that needs the data manager.
- Platforms
web· Any OS (web)- Applications
- Medidata Rave EDC
- Interfaces
brow· Browser- GUI need · task setting
G3· GUI-only- GUI need · ordinary access
G2· GUI-essential- Interaction primitives
click· Point & clicktype· Text & formsnav· Pan/zoom/scrollmulti· Cross-app- Work product
WP08· App/system state- Conditions
licensed· Licensed pro applong· Long-horizonmultiapp· Multi-app- Proposed difficulty
- 5 / 5
Proposed verifier
Audit-trail and status export for 104-019: datapoint values vs gold (100%), 23 queries Answered/Closed as appropriate, lab panels present, every form SDV=true and Frozen=true, Locked=false. RWS access is denied to the training role and the audit trail is checked for non-EDC sources; any RWS call or Lock action fails the task.
Proposed partial credit
35% discrepancies corrected; 25% queries resolved; 15% missing labs entered; 25% SDV+Freeze set with no Lock.
R09.05.01.02-T07 Lock the Baseline instruments for one REDCap record
Original instruction
REDCap project HTN-ALPHA has Record Locking enabled. For record 2001, Baseline event, confirm the 4 instruments are marked Complete, then lock each instrument using the lock checkbox on the form. Do not lock other events and do not e-sign the record.
- Platforms
web· Any OS (web)- Applications
- REDCap
- Interfaces
brow· Browser- GUI need · task setting
G3· GUI-only- GUI need · ordinary access
G2· GUI-essential- Interaction primitives
click· Point & clicknav· Pan/zoom/scroll- Work product
WP08· App/system state- Conditions
risk· Irreversible/risky- Proposed difficulty
- 2 / 5
Proposed verifier
redcap_locking_data contains exactly 4 rows for record 2001, event Baseline, one per instrument; redcap_esignatures has no row for 2001; no other record/event locked. Locking has no API route in core REDCap, so the UI log entries are the only valid source.
Proposed partial credit
70% four instruments locked; 30% nothing else locked or e-signed.
R09.05.01.02-T08 Work six Data Resolution Workflow queries in REDCap
Original instruction
REDCap project HTN-ALPHA uses the Data Resolution Workflow. Open the Resolve Issues page, filter to your site, and handle the 6 open queries using input/source_excerpts.pdf: correct the 4 that are true discrepancies (then respond 'Corrected per source'), and reply to 2 with justification that the value is clinically verified. Do not close queries; leave them for the monitor.
- Platforms
web· Any OS (web)- Applications
- REDCap
- Interfaces
brow· Browser- GUI need · task setting
G3· GUI-only- GUI need · ordinary access
G2· GUI-essential- Interaction primitives
click· Point & clicktype· Text & formsnav· Pan/zoom/scroll- Work product
WP08· App/system state- Conditions
- None declared
- Proposed difficulty
- 3 / 5
Proposed verifier
redcap_data_quality_status/resolutions tables: 6 threads with a site response, status 'Responded', none Closed; 4 field values changed to gold, 2 unchanged. DRW has no API, and value edits must log via DataEntry pages.
Proposed partial credit
45% four corrections exact; 25% two justified replies without data change; 30% statuses Responded, none closed.
R09.05.01.02-T09 E-sign and lock five REDCap records but hold back the one with an incomplete form
Original instruction
REDCap project DIAB-REG: the PI delegated record locking (not signing) for a test signer account. For records 3001-3005, verify every instrument in every event is Complete, then lock the whole record. One record has an instrument still Incomplete on a later event; do not lock it, and open a DRW query on that field explaining what is missing.
- Platforms
web· Any OS (web)- Applications
- REDCap
- Interfaces
brow· Browser- GUI need · task setting
G3· GUI-only- GUI need · ordinary access
G2· GUI-essential- Interaction primitives
click· Point & clicknav· Pan/zoom/scrolltype· Text & forms- Work product
WP08· App/system state- Conditions
risk· Irreversible/riskyhidden· Hidden/stale state- Proposed difficulty
- 4 / 5
Proposed verifier
redcap_locking_records shows 4 locked records (the complete ones); the incomplete record unlocked; one DRW query opened on the correct field with non-empty text. No e-signature rows created. No API token is issued to the signer account, so locking and the DRW query must appear in redcap_log_event from the web session.
Proposed partial credit
40% four complete records locked; 30% incomplete record found and left unlocked; 30% query opened on the right field.
R09.05.01.02-T10 Enter mixed-unit lab results with unit selectors in Castor EDC
Original instruction
Castor EDC sandbox study RENAL-01, participant 110-004, visit Month 3. Enter the chemistry panel from input/lab_report_110-004.pdf exactly as reported, selecting the reported unit in each unit dropdown (creatinine in umol/L, glucose in mmol/L, urea in mg/dL). Respond to the out-of-range warning for potassium by confirming the value and entering the comment 'Confirmed with lab'.
- Platforms
web· Any OS (web)- Applications
- Castor EDC
- Interfaces
brow· Browser- GUI need · task setting
G2· GUI-essential- GUI need · ordinary access
G2· GUI-essential- Interaction primitives
type· Text & formsclick· Point & click- Work product
WP02· Records/table- Conditions
- None declared
- Proposed difficulty
- 3 / 5
Proposed verifier
Castor export for 110-004/Month 3: 9 analyte values and unit codes exact; potassium validation message acknowledged with the exact comment. Castor API client credentials are withheld for this study, and the record audit trail must show web-form entry by the site user.
Proposed partial credit
50% values exact; 30% units selected correctly; 20% warning confirmed with comment.
R09.05.01.02-T11 Record an out-of-window visit and its protocol deviation in Rave
Original instruction
Rave training study, subject 104-015: the Week 8 visit occurred on day 66 (window 56 +/- 5 days) per input/visit_note_104-015.pdf. Enter the visit date and vitals, answer the auto-fired out-of-window query with the reason from the note, and add a Protocol Deviation form entry with category 'Visit schedule', deviation date and description.
- Platforms
web· Any OS (web)- Applications
- Medidata Rave EDC
- Interfaces
brow· Browser- GUI need · task setting
G3· GUI-only- GUI need · ordinary access
G2· GUI-essential- Interaction primitives
type· Text & formsclick· Point & clicknav· Pan/zoom/scroll- Work product
WP08· App/system state- Conditions
licensed· Licensed pro app- Proposed difficulty
- 4 / 5
Proposed verifier
Export of Visit Date, Vitals and Protocol Deviation forms for 104-015 vs gold; the out-of-window query has a site response and is Answered; PD row category/date exact, description contains the reason keyword. The site role is denied RWS access and batch upload; audit source must be the eCRF UI.
Proposed partial credit
30% visit date and vitals; 30% query answered with reason; 40% protocol deviation entry correct.
R09.05.01.02-T12 Transcribe handwritten CRFs for three subjects and query illegible fields
Original instruction
Rave training study, site 207. input/scans/ holds handwritten paper CRFs (synthetic) for subjects 207-001..003, 6 forms each. Enter all legible data. Where a value is illegible or out of logical range, do not guess: leave it blank, mark it per the CCG 'not done/unknown' convention and open a site-to-site query naming the page and field.
- Platforms
web· Any OS (web)- Applications
- Medidata Rave EDC
- Interfaces
brow· Browser- GUI need · task setting
G3· GUI-only- GUI need · ordinary access
G2· GUI-essential- Interaction primitives
type· Text & formsvis· Visual judgmentnav· Pan/zoom/scroll- Work product
WP02· Records/table- Conditions
licensed· Licensed pro applong· Long-horizon- Proposed difficulty
- 5 / 5
Proposed verifier
Field-level accuracy vs gold over 18 forms (pass >= 99% of legible fields); the 7 seeded illegible fields are blank with UNK coding and each has a query; any guessed value on a seeded field fails it. RWS access and batch upload are disabled for the site role; eCRF audit source only.
Proposed partial credit
50% legible-field accuracy; 30% illegible fields handled without guessing; 20% queries cite page and field.
R09.05.01.02-T13 Collect two participant surveys offline in the REDCap Mobile App and sync
Original instruction
On the Android emulator, the REDCap Mobile App is set up with project WELLBEING-PILOT and the device is in airplane mode. Enter the PHQ-9 and GAD-7 for new records P-031 and P-032 from input/paper_responses.pdf, then turn networking on and upload both records, resolving the sync review screen without overwriting server data for existing records.
- Platforms
and· Android- Applications
- REDCap Mobile App
- Interfaces
mob· Mobile touch- GUI need · task setting
G3· GUI-only- GUI need · ordinary access
G3· GUI-only- Interaction primitives
touch· Touch gesturestype· Text & formsclick· Point & click- Work product
WP02· Records/table- Conditions
xdevice· Cross-device- Proposed difficulty
- 3 / 5
Proposed verifier
Server-side redcap_data has P-031 and P-032 with all 16 item values exact and computed scores matching; redcap_log_event shows mobile-app sync as source; no existing record modified. Pass requires both records synced. The user has no API token and web data-entry rights are revoked, so records can arrive only through the Mobile App sync.
Proposed partial credit
40% offline entries exact; 40% both records synced; 20% no overwrite of other records.
R09.05.01.02-T14 Correct a transcription error with a required reason for change
Original instruction
REDCap project HTN-ALPHA has 'Require a reason when making changes to existing records' on. Record 1007's Baseline weight was entered as 812 kg instead of 81.2 kg (see input/source_1007_baseline.pdf). Fix the value, enter the reason 'Transcription error - decimal omitted', and leave every other field unchanged.
- Platforms
web· Any OS (web)- Applications
- REDCap
- Interfaces
brow· Browser- GUI need · task setting
G2· GUI-essential- GUI need · ordinary access
G2· GUI-essential- Interaction primitives
type· Text & formsclick· Point & click- Work product
WP08· App/system state- Conditions
- None declared
- Proposed difficulty
- 1 / 5
Proposed verifier
redcap_data weight for 1007/Baseline = 81.2; redcap_log_event row has the exact change reason; diff of all other fields for 1007 is empty. Change must come from DataEntry page (API disabled).
Proposed partial credit
50% corrected value; 30% reason text exact; 20% no collateral changes.
R09.05.01.02-T15 Perform risk-based SDV on critical fields and raise discrepancy queries
Original instruction
As a CRA in a Rave training study, perform targeted SDV for subjects 104-021, 104-022 and 104-023 per input/monitoring_plan.pdf: verify only eligibility, informed consent, AE and primary-endpoint fields against input/source/ PDFs, tick SDV on those fields only, and open a query on every discrepancy you find (the number is not given).
- Platforms
web· Any OS (web)- Applications
- Medidata Rave EDC
- Interfaces
brow· Browser- GUI need · task setting
G3· GUI-only- GUI need · ordinary access
G2· GUI-essential- Interaction primitives
click· Point & clickvis· Visual judgmentnav· Pan/zoom/scrollmulti· Cross-app- Work product
WP08· App/system state- Conditions
licensed· Licensed pro appmultiapp· Multi-app- Proposed difficulty
- 4 / 5
Proposed verifier
Rave SDV status export: critical fields verified (recall 100%), non-critical fields untouched (precision >= 95%); queries opened vs 5 seeded discrepancies scored by F1 (pass >= 0.8) on field identity; no data values changed by the CRA role. The CRA account is denied RWS access, so SDV flags and queries must carry EDC-UI audit source.
Proposed partial credit
35% SDV flags on critical fields only; 45% discrepancy query F1; 20% no data edits.
R09.05.01.02-T16 Import the same Week 4/12 visit data through the REDCap API (control)
Original instruction
Control pair of T02. In a fresh copy of project HTN-ALPHA with an API token provided in env REDCAP_TOKEN, load the Week 4 and Week 12 data for subjects 1004, 1006 and 1009 from input/visit_worksheets_wk4_wk12.csv via the REDCap API (import records), applying the same missed-visit coding for 1006 Week 12.
- Platforms
lin· Linux- Applications
- REDCap API · Python
- Interfaces
api· API/SDKterm· Terminal- GUI need · task setting
AP· API-sufficient- GUI need · ordinary access
AP· API-sufficient- Interaction primitives
code· Code & commands- Work product
WP02· Records/table- Conditions
- None declared
- Proposed difficulty
- 3 / 5
Proposed verifier
Same gold comparison as T02 on redcap_data; log events must be API imports. Used to measure the GUI-vs-API gap for identical content.
Proposed partial credit
60% values exact; 25% missed-visit coding; 15% form status fields set.
R09.05.01.02-T17 Close out a Vault EDC site: answer queries, request PI signature, lock signed casebooks
Original instruction
Veeva Vault EDC training study, site 301 (4 completed subjects). Answer the 9 remaining open queries using input/site301_source/, mark remaining forms SDV-complete, and send the PI an e-signature request for each casebook. You must not sign as the PI. Lock only the casebooks that already carry a PI signature, then list in output/closeout.txt which subjects still await signature.
- Platforms
web· Any OS (web)- Applications
- Veeva Vault EDC
- Interfaces
brow· Browser- GUI need · task setting
G3· GUI-only- GUI need · ordinary access
G2· GUI-essential- Interaction primitives
click· Point & clicktype· Text & formsnav· Pan/zoom/scrollmulti· Cross-app- Work product
WP08· App/system state- Conditions
licensed· Licensed pro apprisk· Irreversible/riskylong· Long-horizon- Proposed difficulty
- 5 / 5
Proposed verifier
Vault export: 9 queries answered with source-consistent values; SDV complete; 4 signature requests sent; locked set = signed set exactly; no signature by the agent account; closeout.txt lists the unsigned subject IDs exactly. Vault API access is not granted to the agent's role, so all actions must be made in the Vault EDC web session.
Proposed partial credit
35% queries answered; 15% SDV complete; 20% signature requests; 30% lock set exact and closeout list correct.
R09.05.01.02-T18 Record an early termination and close remaining scheduled events in OpenClinica
Original instruction
OpenClinica 4 sandbox study CARDIO-2: participant SS-0230 withdrew consent on 2026-05-14 (input/withdrawal_letter.pdf). Complete the End of Study form (reason, date, last dose), set every future scheduled event to 'not scheduled'/skipped per input/ccg_cardio2.pdf, and leave already-entered data intact.
- Platforms
web· Any OS (web)- Applications
- OpenClinica
- Interfaces
brow· Browser- GUI need · task setting
G2· GUI-essential- GUI need · ordinary access
G2· GUI-essential- Interaction primitives
click· Point & clicktype· Text & formsnav· Pan/zoom/scroll- Work product
WP08· App/system state- Conditions
risk· Irreversible/risky- Proposed difficulty
- 4 / 5
Proposed verifier
ODM export for SS-0230: End of Study items exact; event statuses for 5 future events = skipped/not scheduled; participant status per CCG; prior event data checksum unchanged; no participant removal. Audit log must show UI actions.
Proposed partial credit
40% End of Study form; 40% future events statuses; 20% prior data untouched.
R09.05.01.02-T19 Enter eligibility and randomize a subject with the REDCap randomization module
Original instruction
REDCap project HTN-ALPHA (sandbox) has a stratified randomization module (strata: site, diabetes). For new record 1012, enter the Eligibility instrument from input/eligibility_1012.pdf; if all criteria are met, randomize the record once using the Randomize button. If any criterion fails, do not randomize; mark the record screen-failed per the CCG.
- Platforms
web· Any OS (web)- Applications
- REDCap
- Interfaces
brow· Browser- GUI need · task setting
G3· GUI-only- GUI need · ordinary access
G2· GUI-essential- Interaction primitives
type· Text & formsclick· Point & click- Work product
WP08· App/system state- Conditions
risk· Irreversible/risky- Proposed difficulty
- 3 / 5
Proposed verifier
Eligibility values vs gold; redcap_randomization_allocation shows record 1012 allocated exactly once with correct strata (or not allocated, with screen-fail coding, if the seeded variant has an exclusion). No API token is issued (blocking the API randomize method); redcap_log_event must show the Randomize action from the web session.
Proposed partial credit
40% eligibility fields exact; 20% correct eligibility decision; 40% randomization or screen-fail handled correctly.
R09.05.01.02-T20 Register a participant and fix a consent-date edit check in Castor EDC
Original instruction
Castor EDC sandbox study RENAL-01, site 110. Create participant 110-009 and enter the Informed Consent form (consent date, ICF version, signatory) and the Screening form from input/screening_110-009.pdf. The source lists screening labs dated one day before consent; enter dates as recorded, answer the resulting edit check with the explanation from the source note, and do not alter the dates.
- Platforms
web· Any OS (web)- Applications
- Castor EDC
- Interfaces
brow· Browser- GUI need · task setting
G2· GUI-essential- GUI need · ordinary access
G2· GUI-essential- Interaction primitives
type· Text & formsclick· Point & click- Work product
WP02· Records/table- Conditions
- None declared
- Proposed difficulty
- 3 / 5
Proposed verifier
Castor export: participant 110-009 exists at site 110; consent and screening fields exact including the out-of-order dates; the validation query has a response containing the explanation keyword; no backdating. Castor API credentials are withheld for the study; the audit trail must show web-form entry only.
Proposed partial credit
25% participant created at site; 45% fields exact; 30% edit check answered without changing dates.
Mapped research references
No benchmark reference is assigned to this class in the current mapping.