# Extraction Report: Koncewicz 2019

## Source Crosswalk

- Thesis PDF: `pdf/paper.pdf`
- Text extraction: `out/text.txt`, `out/fulltext.md`
- Raw scenario datasets: `data/*.csv`
- Shared demographic/end files: `data/demo/625672.csv`, `data/end/255249.csv`
- Manual numeric analysis: `data/raport_natka.rmd`
- Reproducible computation: `analysis/effect_sizes.qmd`

The thesis is treated as the source of truth for scenario wording, procedure, scale wording,
sample description, and moderator coding. The raw CSV exports and report Rmd are treated
as the source of truth for group descriptives and effect sizes.

## Extraction Decisions

- Included single-scenario effects: five between-subject high-vs-low contrasts for
  `przechodzien`, `znaki`, `policjant`, `dziecko`, and `pijany`.
- Included paired effects: five within-subject high-vs-low contrasts for the same evidence
  sources in the two-scenario presentation.
- Effect sign convention: `d = mean(low) - mean(high)`.
- Between-subject effects: SMD via `esc::esc_mean_sd`.
- Within-subject effects: SMCRP via `metafor::escalc(measure = "SMCRP", correct = FALSE)`,
  using the within-person correlation computed from the raw paired responses.

## Issues For Review

### HUMAN_CHECK_KONCEWICZ_EFFECT_UNIT

The extraction includes both the single-scenario between-subject contrasts and the
two-scenario paired contrasts. These are different presentation conditions and may or may not
both belong in the same meta-analytic model.

Required human answer: confirm whether to include both sets, only single-scenario effects,
or only paired/two-scenario effects.

### HUMAN_CHECK_KONCEWICZ_SAMPLE_N

The thesis reports 753 participants and no exclusions. The copied raw scenario files produce
effect-level Ns that are used directly in `analysis/effect_sizes.qmd`; these scenario-level Ns
do not sum to 753 because the files represent scenario/presentation records from the combined
survey structure.

Required human answer: confirm that thesis-level `N = 753` should remain at study level.

### HUMAN_CHECK_KONCEWICZ_PIJANY_PAIRED_COLUMNS

The original Rmd notes that the paired drunk-passerby dataset uses `PrzechodzenOba` column
names for the `pijany_oba` file and states that this is intentional despite a naming mistake.
The QMD follows that original Rmd behavior.

Required human answer: confirm that `823783.csv` with `PrzechodzenOba.SA001/SA002`
is indeed the paired drunk-passerby source.

## Validation Notes

The YAML uses thesis pages for stimuli and question-wording provenance. Dataset-derived
groups use `table_ref` values pointing to the report Rmd and reproducible QMD.