Normalize responses — ar_normalize • associatoR

ar_normalize performs normalization steps for responses (and cues) and wraps around ar_normalize_manual.

Usage

ar_normalize(
  associations,
  case = "most_frequent",
  punct = "end",
  punct_replacement = " ",
  whitespace = "squish",
  process_cues = TRUE
)

Arguments

associations: an associatoR object containing association data as generated by ar_import.
case: a character specifying the normalization of cases or NULL to not change casing. Must be one of c("lower", "upper", "sentence", "most_frequent") or NULL. Setting case = "lower" normalizes all responses to lower case, case = "lower" to upper case, case = "sentence" to sentence case, and case = "most_frequent" to the most frequent casing for each string.
punct: a character specifying the normalization of punctuationor NULL to not change punctuation. Must be one of c("end", "all") or NULL. punct = "end" replaces all punctuation at the end (including following whitespaces) with punct_replacement (default is a single white space). punct = "all" replaces all punctuation.
punct_replacement: a character used as replacement for punctuation.
whitespace: a character specifying the normalization of white spaces or NULL to not change whitespace. Must be one of c("squish", "trim") or NULL. Setting whitespace = "squish" removes additional white spaces at the start, the end, and in-between words, whereas setting whitespace = "trim" removes white space from the start and end.
process_cues: a logical specifying, if cues should be processed as well. Defaults to TRUE.

Value

Returns an associatoR object containing a list of tibbles:

participants: A tibble of participants including a participant id and potential participant attributes.
cues: A tibble of cues including a cue variable and potential cue attributes.
responses: A tibble of responses including a participant id, the cues, the responses, the response level, and additional response attributes.

References

Aeschbach, S., Mata, R., Wulff, D. U. (in progress)

Examples


ar_import(intelligence,
          participant = participant_id,
          cue = cue,
          response = response,
          participant_vars = c(gender, education),
          response_vars = c(response_position, response_level)) %>%
  ar_normalize(case = "lower", punct = "all")
#> 197 entries were dropped from cues due to normalizing.
#> 
#> ── An associatoR object ────────────────────────────────────────────────────────
#> 
#> participants
#> # A tibble: 1,000 × 3
#>      id gender education  
#>   <dbl> <chr>  <chr>      
#> 1     1 male   high school
#> 2     2 male   high school
#> 3     3 male   high school
#> 4     4 male   high school
#> 5     5 male   high school
#> # ℹ 995 more rows
#> 
#> cues
#> # A tibble: 607 × 2
#>   cue          cue_original
#>   <chr>        <chr>       
#> 1 intelligence intelligence
#> 2 einstein     Einstein    
#> 3 books        books       
#> 4 iq tests     IQ tests    
#> 5 college      college     
#> # ℹ 602 more rows
#> 
#> responses
#> # A tibble: 29,882 × 7
#>      id cue          response     response_position response_level cue_original
#>   <dbl> <chr>        <chr>                    <dbl>          <dbl> <chr>       
#> 1     1 intelligence einstein                     1              1 intelligence
#> 2     1 intelligence books                        2              1 intelligence
#> 3     1 intelligence iq tests                     3              1 intelligence
#> 4     1 intelligence college                      4              1 intelligence
#> 5     1 intelligence smart people                 5              1 intelligence
#> # ℹ 29,877 more rows
#> # ℹ 1 more variable: response_original <chr>
#>