Help & Manual
Search the concepts you need, with short examples instead of a long handbook.
Getting started
A first merge in five steps.
Choose the files that belong to one study, select the base table, choose the ID column for every source, review anything uncertain, then merge and export. The app does not require the optional post-merge tools before you export.
Example / 예시 / Ejemplo
Base table → choose IDs → review → merge → export
What is the base table?
The table that defines the final participant rows.
The base table is the source that answers “who belongs in the final dataset?” Research Data Merger keeps those rows as the final row set. IDs found only in other source files are reported rather than silently added.
Example / 예시 / Ejemplo
Roster: P001, P002, P003 Source: P001, P002, P999 Final rows: P001, P002, P003 Reported source-only ID: P999
ID cleanup & possible mismatches
Review formatting differences without letting the app guess.
The review step can surface spaces, padding differences, conservative prefix cleanup, duplicates, and possible near-ID matches. Safe-looking cleanup still remains visible. If a cleanup would collide with an existing ID, the app blocks it.
Example / 예시 / Ejemplo
P0004 → P004 formatting review P006 → P006 whitespace cleanup P008X → P008 possible near match; review required
Same-name columns
Decide whether a repeated column is a duplicate, a conflict, or another measurement.
A same-name column is not automatically an error. If all compared values are identical, keeping the base avoids a duplicate copy. If values differ, review what the column means before deciding whether to keep the base, use the source, fill blanks, or preserve both.
Example / 예시 / Ejemplo
Site = Site in both files → usually keep base only Age differs in 1 row → review as a conflict
Repeated measurements
Keep the same variable from another timepoint instead of overwriting it.
Use repeated-measurement handling when the same variable was collected again at another visit or wave. If many columns repeat from one source, apply a source label such as Week4 once, then review only the individual exceptions you care about.
Example / 예시 / Ejemplo
Baseline.csv: Score, Q1, Q2 Week4.csv: Score, Q1, Q2 Source label: Week4 Result: Score | Score_Week4 | Q1 | Q1_Week4 | Q2 | Q2_Week4
Completion filter
Create a separate copy using the completion threshold you choose.
Select the survey columns that should count toward completion, choose a minimum percentage, and preview how many rows would be kept or excluded. The merged dataset itself is not deleted or changed.
Example / 예시 / Ejemplo
Selected: Week8 Q1–Q10 Minimum completion: 70% Preview: 117 kept · 7 excluded
De-identified copy
Remove or replace identifying fields in a separate export.
Choose Keep, Remove, or Replace with study code for each field. A crosswalk is requested only when you actually replace a field with study codes. You can also drag columns into the export order you want. The original merged result remains unchanged.
Example / 예시 / Ejemplo
Name → Remove Email → Replace with study code Age → Keep
Wide ↔ long reshape
Create another layout after the merge.
Use reshape only when your downstream analysis or software needs a different table structure. Repeated measurements often begin in wide form and can be reshaped to long form afterward.
Example / 예시 / Ejemplo
Wide: ID | Score_Baseline | Score_Week4 Long: ID | Time | Score
Export, provenance & merge record
Keep the result and the decisions together.
Exported workbooks can include the merged data plus documentation of source files, reviewed ID decisions, completion filtering, de-identification, and provenance where applicable. Review the workbook before beginning analysis.
Example / 예시 / Ejemplo
Merged Data Merge Record Provenance Completion Filter / Excluded Rows (when used)
Supported files & PDFs
Know what the app can read.
Research Data Merger supports Excel (.xlsx), CSV, TSV, SPSS (.sav), Stata (.dta), and text-based PDF tables. Image-only or scanned PDFs are not automatically OCR’d. If a PDF cannot be read as text, convert or export the table from the source system first.
Example / 예시 / Ejemplo
Excel · CSV · TSV · SPSS · Stata · text-based PDF
Security & privacy
Research files stay on the computer where you run the app.
Research Data Merger processes study files locally. The app does not send research datasets to IMPLEMON, a cloud processor, or an AI service. See the Security & Privacy page for deployment details and institutional-use notes.
Example / 예시 / Ejemplo
Local processing · No AI · No cloud processing · No research data uploads
Troubleshooting
Start with the smallest explanation.
If a result looks wrong, first confirm the base table and ID columns, then review duplicates and source-only IDs. For repeated measurements, confirm that the source label describes the correct visit or wave. Keep a copy of the original source files and compare the exported merge record with your intended decisions.
Example / 예시 / Ejemplo
1. Check base table 2. Check ID columns 3. Review duplicates / source-only IDs 4. Check repeated-measurement labels 5. Review export record
No matching help topic. Try a shorter search.