Statikali

Research data preparation toolkit

A blank variable dictionary and a short checklist for documenting your own research file. The template contains headers only: no participant records, example answers, formulas or automatic checks. Consultation and ordering on Statikali are currently closed; these files can be used independently and do not upload anything.

Use the files

  1. Save an unchanged original of your dataset. Open codebook-template.csv in your spreadsheet application or import it as UTF-8 with comma separation. Use one row per variable, including derived variables. This is a dictionary, not the research data table.
  2. Keep the English header names unchanged if sharing the dictionary with others. You may write descriptions in your preferred language. Both language versions have the same headers; the Hebrew version is paired with Hebrew instructions.
  3. Import dictionary columns as text to preserve codes and leading zeros. Describe calculation rules in words. CSV does not preserve cell formatting; check import settings when reopening it.
  4. Keep one header row and save a working copy. If your spreadsheet application uses a different separator, select comma in the import/export settings. CSV preserves text values, not formatting, validation, worksheets or checkboxes.
  5. Mark genuinely unresolved rules as unknown or not yet decided; use not applicable when appropriate. A blank cell is unfinished documentation, not a decision about missing-data handling. Do not alter research values to fill the dictionary.
  6. Complete checklist.html, on screen or after printing. Attach the questionnaire/manual and a separate change log to your own project materials. Do not place participant names, email addresses, identification numbers or a re-identification key in this dictionary.

Columns

Header What to record
variable_name Exact column name in the relevant data file.
variable_label Meaning of the variable or the full question wording.
data_type Storage type, such as numeric, text or date; include the date format if applicable.
measurement_level Nominal, ordinal or quantitative, with clarification when uncertain. Storage as a number does not establish measurement level.
units Measurement unit, or not applicable.
allowed_values_and_labels Permitted range or each category code and meaning. Describe restrictions in text.
missing_codes_and_meanings Every recorded missing-value code, including what a blank means if known. Distinguish missingness from a valid zero; do not invent the reason.
instrument_and_item Questionnaire/instrument version and source item number or reference.
scale_membership Scale or subscale membership; list the contributing item names for a derived score.
reverse_key_status forward, reverse, unknown or not applicable, based on the instrument. Recording this does not reverse values. State in notes whether data have already been reversed.
score_or_derivation_rule Document the intended calculation, valid item range, direction, minimum answered items and handling of missing items, with a manual/source reference. Describe the rule; do not execute it in this CSV.
analysis_role Intended role for a specific research question, such as outcome, predictor, covariate, group or wave. A role can differ across questions.
source_file_and_sheet Non-identifying file and worksheet names; distinguish original and derived sources.
notes Unresolved questions, discrepancies, relevant restrictions and where a decision is documented.

Use the dictionary to document research decisions and retain open questions for clarification. Only transfer information that your study permissions allow; replacing names with codes does not by itself make research data anonymous.