Parses a data.txt written by jsPsych (a JSON array of trial objects)
into a tibble with one row per trial. A file that holds several JSON
values back to back, which is what repeated jatos.appendResultData()
calls leave behind (arrays after arrays, or one object per call), is
split into its top-level values first and read as one array. An empty
file gives a tibble with no rows and no columns.
Arguments
- file
Path of the file.
- flatten
If
TRUE, nested objects become columns namedouter.inner(seejsonlite::fromJSON()); otherwise they stay nested data-frame columns when every trial of the file carries the same object, and list columns otherwise.jatos_read_results()stores both as list columns before it binds files.
Value
A tibble. A key that is the empty string (an unnamed form input)
becomes a column unnamed_<position>.
See also
jatos_read_results() to read every file of a metadata tibble.
Examples
file <- system.file("extdata", "data.txt", package = "jatosr")
trials <- jatos_read_json(file)
trials[, c("trial_index", "trial_type", "rt", "correct")]
#> # A tibble: 9 × 4
#> trial_index trial_type rt correct
#> <int> <chr> <int> <lgl>
#> 1 0 survey-html-form 14250 NA
#> 2 1 html-keyboard-response 737 TRUE
#> 3 2 html-keyboard-response 824 TRUE
#> 4 3 html-keyboard-response 761 FALSE
#> 5 4 html-keyboard-response 848 TRUE
#> 6 5 html-keyboard-response 935 TRUE
#> 7 6 html-keyboard-response 872 TRUE
#> 8 7 html-keyboard-response 959 TRUE
#> 9 8 html-keyboard-response 1046 FALSE
jatos_read_json(file, flatten = TRUE)[1, ]
#> # A tibble: 1 × 11
#> rt response trial_type trial_index plugin_version time_elapsed
#> <int> <list> <chr> <int> <chr> <int>
#> 1 14250 <named list [2]> survey-html-fo… 0 2.0.0 14268
#> # ℹ 5 more variables: participant_id <chr>, stimulus <chr>, set_size <int>,
#> # change <lgl>, correct <lgl>