Skip to contents

synr: Quick start

Source: vignettes/quick-start.Rmd
quick-start.Rmd

This is the ‘quick start’ guide to synr. For an in-depth tutorial, please see the main tutorial.

Rolling up the data

If you have long format data (otherwise see Creating ParticipantGroup objects):

pg <- create_participantgroup(
  raw_df=synr_exampledf_long_small,
  n_trials_per_grapheme=2,
  id_col_name="participant_id",
  symbol_col_name="trial_symbol",
  color_col_name="response_color",
  time_col_name="response_time", # optional, not necessary for core functionality
  color_space_spec="Luv"
)

The resulting object (pg) is a nested structure implemented with reference classes. With this, you can call various methods and access various attributes. Examples of common use cases are included below.

Calculate participant consistency scores

Consistency scores for all participants, looking only at trials involving letters.

cons_scores_letters <- pg$get_mean_consistency_scores(symbol_filter=LETTERS)
print(cons_scores_letters)
#> [1] 119.2646 129.1678 145.5045

Produce a plot of a single participant’s responses

Plot of single participant’s per-grapheme consistency scores, looking only at trials involving the graphemes ‘A’ and ‘7’.

pg$participants[['1']]$get_plot(symbol_filter=c('A', '7'))

Export relevant participant data to a data frame 

# get mean consistency scores for all participants, filtering first by letters, then digits
mean_cscores_letters <- pg$get_mean_consistency_scores(symbol_filter=LETTERS)
mean_cscores_digits <- pg$get_mean_consistency_scores(symbol_filter=0:9)

# get number of graphemes where all response colors were non-missing, 
# filtering first by letters, then digits
# (in the example data frame, all participants have all-valid responses)
num_valid_letters <- pg$get_numbers_all_colored_graphemes(symbol_filter=LETTERS)
num_valid_digits <- pg$get_numbers_all_colored_graphemes(symbol_filter=0:9)

p_ids <- pg$get_ids()

mean_scores_df <- data.frame(
  participant_id=p_ids, 
  cscore_letters=mean_cscores_letters,
  cscore_digits=mean_cscores_digits,
  num_valid_letters=num_valid_letters,
  num_valid_digits=num_valid_digits
)
print(mean_scores_df)
#>   participant_id cscore_letters cscore_digits num_valid_letters
#> 1              1       119.2646       155.089                 2
#> 2              2       129.1678        45.786                 2
#> 3              3       145.5045       164.870                 2
#>   num_valid_digits
#> 1                1
#> 2                1
#> 3                1

Finding more information

Apart from the aforementioned vignettes, you can find help documentation for fields/attributes and methods of involved objects by running help(ParticipantGroup), help(Participant) or help(Grapheme).