Skip to contents

Score a SWAN Questionnaire

Source: vignettes/articles/score_a_swan_questionnaire.Rmd
score_a_swan_questionnaire.Rmd

The Strengths and Weaknesses of ADHD Symptoms and Normal Behavior Rating Scale (SWAN) is a validated instrument for measuring attention-deficit/hyperactivity disorder (ADHD) traits (Burton et al., 2018). The sfsScorer package provides an easy way to automatically score your SWAN tests.

Quick Start

The code below shows how to score a SWAN questionnaire using the score_swan() function.

library(sfsScorer)
#> sfsScorer Copyright (C) 2026 Jonathan Clutton
#> This program comes with ABSOLUTELY NO WARRANTY
#> This is free software, and you are welcome to redistribute it under certain
#> conditions
#> See conditions: <https://jclutton.github.io/sfsScorer/LICENSE.html>

#' Here's how we expect the data to be formatted
#' age (5-18)
#' gender 
#'  for non-gendered t-scores (Any gender values are acceptable)
#'  for gendered t-scores (1 = boy, 2 = girl)
#' p_respondent (1 = parent / caregiver responded, 0 = child / youth respondent)
#' All swan questions 1-18 as swan1...swan18
head(random_data, 1)
#>   age gender p_respondent swan1 swan2 swan3 swan4 swan5 swan6 swan7 swan8 swan9
#> 1  13      1            1    -1     2    -2     0     3    -1     0     1     2
#>   swan10 swan11 swan12 swan13 swan14 swan15 swan16 swan17 swan18
#> 1      2      0     -2      2     -1     -1      2      0     -2

# Score from a data frame
scores <- score_swan(df = random_data)
#> ✔ The model scored 5 observations.

# OR
# Score from a a csv or xlsx file
csv <- system.file("extdata", "sample_swan.csv", package = "sfsScorer")
scores_csv <- score_swan(file = csv)
#> ✔ The model scored 5 observations.

# Data are validated by default
 df_mod <- random_data |> 
   dplyr::mutate(swan1 = 6)
 try(scores_csv <- score_swan(df = df_mod))
#> There are 5 impossible values in the file.
#> The only valid values are -3, -2, -1, 0, 1, 2, 3, and NA.
#> 
#> Error in clean_file(df, test = "swan", ignore_check = ignore_check) : 
#>   Please correct or remove these rows - "Row 1: swan1 - 6", "Row 2: swan1
#> - 6", "Row 3: swan1 - 6", "Row 4: swan1 - 6", and "Row 5: swan1 - 6"
 
 # To ignore validation and introduce NAs use `ignore_check = TRUE`
 scores_csv <- score_swan(df = df_mod, ignore_check = TRUE)
#> ! 5 impossible values were changed to NA. This could impact scores. 
#> The only valid values are -3, -2, -1, 0, 1, 2, 3, and NA. To correct, review the following rows before running - "Row 1: swan1 - 6", "Row 2: swan1 - 6", "Row 3: swan1 - 6", "Row 4: swan1 - 6", and "Row 5: swan1 - 6"
#> ✔ The model scored 5 observations.

Notes about the data

  • The score_swan() function will ask you to upload a spreadsheet with your SWAN data. If you have any additional columns beyond the necessary columns described below, i.e. an identifier, those columns will remain untouched in the output. For example, if your input file has two additional columns, an identifier and parent’s education, those columns will pass through to the output file.

  • Scores are automatically reversed before calculating the t-score so that a higher score is associated with higher ADHD trait.

  • The test is split into two subdomains. Questions 1-9 measure inattentiveness. Questions 10-18 measure hyperactivity and impulsivity. If more than one question is missing from a subdomain the test will not be scored.

  • T-scores will be generated based on gendered and non-gendered norms. Please feel free to include children who are trans or non-binary in your dataset and leave the codes for their gender as appropriate for the individuals in your study. Non-gendered t-scores will be generated for all individuals. To generate gendered t-scores for binary gendered participants be sure to code gender as 1 = boy and 2 = girl. Any gender not coded as 1 or 2 will not receive a gendered t-score. Any gender not coded as 1 or 2 will not receive a gendered t-score.

Instructions

Formatting Your Data

Our first step is to prepare your raw SWAN data.

  1. Prepare a spreadsheet, preferably a .csv file, with a row for each of the tests you’d like to score.

  2. Be sure the following columns are present in the spreadsheet and rename the columns to match the reference guide below. All of the following columns are necessary for the model. If age or p_respondent are missing from a row, the model will not return a t-score for that row.

    Column Name Additional Information Example
    age The child’s age 5 - 18

    Note: Age in a numeric form, i.e. with decimals, will provide a more accurate t-score compared to an integer-based age
    gender The child’s gender.

    - A full range of gender options can be included with the questionnaire.
    - If you want to generate non-gendered norms, any gender identities can be included and coded in any way (e.g., boy, girl, transgender, non-binary, etc).
    - If you want to generate gendered norms, gender must be binary and coded as 1 = boy and 2 = girl    		 Non-gendered t-scores: any value

    Gendered t-scores: 1 = Boy, 2 = Girl

    Note: Any gender not coded as 1 or 2 will not receive a gendered t-score.
    p_respondent Whether the survey was filled out by the parent or the child 1 = Parent / Caregiver Respondent, 0 = Child / Youth Self-Respondent
    swan1 1. Give close attention to detail and avoid careless mistakes -3 (Far Below) to 3 (Far Above)
    swan2 2. Sustain attention on tasks and play activities -3 (Far Below) to 3 (Far Above)
    swan3 3. Listen when spoken to directly -3 (Far Below) to 3 (Far Above)
    swan4 4. Follow through on instructions and finish schoolwork / chores -3 (Far Below) to 3 (Far Above)
    swan5 5. Organize tasks and activities -3 (Far Below) to 3 (Far Above)
    swan6 6. Engage in tasks that require sustained mental effort -3 (Far Below) to 3 (Far Above)
    swan7 7. Keep track of things necessary for activities -3 (Far Below) to 3 (Far Above)
    swan8 8. Ignore extraneous stimuli (able to ignore background noise/distractions) -3 (Far Below) to 3 (Far Above)
    swan9 9. Remember daily activities -3 (Far Below) to 3 (Far Above)
    swan10 10. Sit still (control movements of hands / feet or control squirming) -3 (Far Below) to 3 (Far Above)
    swan11 11. Stay seated (when required by class rules / social conventions) -3 (Far Below) to 3 (Far Above)
    swan12 12. Modulate motor activity (inhibit inappropriate running / climbing) -3 (Far Below) to 3 (Far Above)
    swan13 13. Play quietly (keep noise level reasonable) -3 (Far Below) to 3 (Far Above)
    swan14 14. Settle down and rest (control constant activity) -3 (Far Below) to 3 (Far Above)
    swan15 15. Modulate verbal activity (control excess talking) -3 (Far Below) to 3 (Far Above)
    swan16 16. Reflect on questions (control blurting out answers) -3 (Far Below) to 3 (Far Above)
    swan17 17. Await turn (stand in line and take turns) -3 (Far Below) to 3 (Far Above)
    swan18 18. Enter into conversations and games (control interrupting / intruding) -3 (Far Below) to 3 (Far Above)

Generate Scores

Use the code below to generate your t-scores. First, you will be prompted to select your file. Second, we will check that your data are formatted properly. Third, t-scores for the full test as well as the two subdomains (inattentive and hyperactive) will be generated.

If you receive an error, please correct the issue in your file, save your file, then run the score_swan() function again.

library(sfsScorer)

# The score_swan checks that your data are formatted correctly and generates the t-scores
swan_tscores <- score_swan()

Additional options 

# Score from a data.frame in case the data do not exist in a csv file, i.e. a REDCap API
swan_tscores <- score_tocs2(df = sample_data)

# Example of how to specify the input file
swan_tscores <- score_swan(file = here("test_scores.csv"))

# Example of how to specify an output folder
swan_tscores <- score_swan(output_folder = file.path("C:","Users",..."yourpath"))

# Data are automatically validated. 
# To ignore validation and introduce NA, ignore_check = TRUE
swan_tscores <- score_swan(df = sample_data, ignore_check = TRUE)

Understanding the Output

Reversed SWAN Scores

  • Columns swan1 to swan18 are reverse-scored and returned as swan1_reversed to swan18 reversed respectively.

Summary Values

Column Description
swan_tot A summed score of the answered questions across the entire test
swan_miss A count of missing values across the entire test
swan_pro A prorated score by dividing swan_tot by the number of answered questions across the entire test
swan_ia_tot A summed score of the answered questions across the inattentive subdomain
swan_ia_miss A count of missing values across the inattentive subdomain
swan_ia_pro A prorated score by dividing swan_tot by the number of answered questions across the inattentive subdomain
swan_hi_tot A summed score of the answered questions across the hyperactive subdomain
swan_hi_miss A count of missing values across the hyperactive subdomain
swan_hi_pro A prorated score by dividing swan_tot by the number of answered questions across the hyperactive subdomain

T-scores for generic model

Column Description
swan_tot_gender_tscores A t-score across the entire SWAN test that adjusts for age, respondent, and gender
swan_tot_tscores A t-score across the entire SWAN test that adjusts for age and respondent
swan_ia_gender_tscores A t-score of the inattentive subdomain (questions 1-9) that adjusts for age, respondent, and gender
swan_ia_tscores A t-score of the inattentive subdomain (questions 1-9) that adjusts for age and respondent
swan_hi_gender_tscores A t-score of the hyperactive subdomain (questions 10-18) that adjusts for age, respondent, and gender
swan_hi_tscores A t-score of the hyperactive subdomain (questions 10-18) that adjusts for age and respondent