Skip to contents

Exporting grades to Canvas

Source: vignettes/canvas-export.Rmd
canvas-export.Rmd

Scope

This guide shows how to prepare a Canvas Gradebook-style CSV grade file from learnrTrackR. The package does not call the Canvas API. The workflow is a manual CSV import into the Canvas Gradebook.

Instructure’s Canvas Instructor Guide states that Gradebook CSV uploads can update existing assignment grades or create new assignments, but they cannot update other Gradebook areas such as assignment status, comments, or grade posting policies.

Create a small gradebook 

library(learnrTrackR)

db_path <- tempfile(fileext = ".sqlite")
con <- init_tracking_db(db_path, overwrite = TRUE)

register_students(
  con,
  data.frame(
    student_id = c("student_001", "student_002"),
    student_label = c("Student 1", "Student 2"),
    email = c("student.001@example.test", "student.002@example.test"),
    group_id = c("A", "A")
  )
)
#> # A tibble: 2 × 5
#>   student_id  student_label email                    group_id created_at        
#>   <chr>       <chr>         <chr>                    <chr>    <chr>             
#> 1 student_001 Student 1     student.001@example.test A        2026-06-24T12:25:…
#> 2 student_002 Student 2     student.002@example.test A        2026-06-24T12:25:…

register_questions(
  con,
  tutorial_id = "module_01",
  questions = data.frame(
    question_id = c("q1", "q2"),
    max_score = c(1, 1)
  )
)
#> # A tibble: 2 × 6
#>   question_id tutorial_id question_label question_type max_score created_at     
#>   <chr>       <chr>       <chr>          <chr>             <dbl> <chr>          
#> 1 q1          module_01   q1             NA                    1 2026-06-24T12:…
#> 2 q2          module_01   q2             NA                    1 2026-06-24T12:…

track_attempt(
  con,
  student_id = "student_001",
  tutorial_id = "module_01",
  question_id = "q1",
  submitted_answer = "mean(x)",
  score = 1,
  max_score = 1,
  require_registered_student = TRUE
)

track_attempt(
  con,
  student_id = "student_002",
  tutorial_id = "module_01",
  question_id = "q1",
  submitted_answer = "sd(x)",
  score = 0,
  max_score = 1,
  require_registered_student = TRUE
)

Build the Canvas table

canvas_grades() returns a wide table with the standard leading columns used in Canvas Gradebook CSV files and one assignment grade column.

canvas_grades(
  con,
  tutorial_id = "module_01",
  assignment = "Module 01 quiz"
)
#> # A tibble: 2 × 6
#>   Student   ID    `SIS User ID` `SIS Login ID` Section `Module 01 quiz`
#>   <chr>     <chr> <chr>         <chr>          <chr>              <dbl>
#> 1 Student 1 ""    student_001   ""             A                      1
#> 2 Student 2 ""    student_002   ""             A                      0

By default, learnrTrackR writes student_id values to the SIS User ID column. If your Canvas course matches users with another field, change the identifier mapping.

canvas_grades(
  con,
  tutorial_id = "module_01",
  assignment = "Module 01 quiz",
  student_id_column = "SIS Login ID",
  student_id_source = "email"
)
#> # A tibble: 2 × 6
#>   Student   ID    `SIS User ID` `SIS Login ID`          Section `Module 01 quiz`
#>   <chr>     <chr> <chr>         <chr>                   <chr>              <dbl>
#> 1 Student 1 ""    ""            student.001@example.te… A                      1
#> 2 Student 2 ""    ""            student.002@example.te… A                      0

The default grade value is the raw score, because Canvas assignment imports are interpreted as assignment points. If the Canvas assignment is intended to receive a 0 to 100 value, use grade_value = "percent".

canvas_grades(
  con,
  tutorial_id = "module_01",
  assignment = "Module 01 percent",
  grade_value = "percent"
)
#> # A tibble: 2 × 6
#>   Student   ID    `SIS User ID` `SIS Login ID` Section `Module 01 percent`
#>   <chr>     <chr> <chr>         <chr>          <chr>                 <dbl>
#> 1 Student 1 ""    student_001   ""             A                        50
#> 2 Student 2 ""    student_002   ""             A                         0

Export the CSV file 

csv_path <- tempfile(fileext = ".csv")

export_canvas_grades(
  con,
  path = csv_path,
  tutorial_id = "module_01",
  assignment = "Module 01 quiz",
  grade_value = "score",
  digits = 2
)

readr::read_csv(csv_path, show_col_types = FALSE)
#> # A tibble: 2 × 6
#>   Student   ID    `SIS User ID` `SIS Login ID` Section `Module 01 quiz`
#>   <chr>     <lgl> <chr>         <lgl>          <chr>              <dbl>
#> 1 Student 1 NA    student_001   NA             A                      1
#> 2 Student 2 NA    student_002   NA             A                      0

If the database contains several registered groups, use group_id to export only one cohort:

canvas_grades(
  con,
  tutorial_id = "module_01",
  assignment = "Module 01 quiz",
  group_id = "A"
)
#> # A tibble: 2 × 6
#>   Student   ID    `SIS User ID` `SIS Login ID` Section `Module 01 quiz`
#>   <chr>     <chr> <chr>         <chr>          <chr>              <dbl>
#> 1 Student 1 ""    student_001   ""             A                      1
#> 2 Student 2 ""    student_002   ""             A                      0

Suggested Canvas import workflow

Before importing into a real course:

  1. Export the current Canvas Gradebook first.
  2. Compare the Canvas identifier columns with the learnrTrackR export.
  3. Confirm whether your course uses ID, SIS User ID, or SIS Login ID.
  4. Create the Canvas assignment first when possible, especially when grading periods are enabled.
  5. Generate a learnrTrackR CSV with the matching identifier column.
  6. Import the CSV in Canvas’s Gradebook import interface.
  7. Review the Canvas import preview before saving changes.
  8. Validate the workflow on a small test group before importing a full cohort.

Privacy check before import

The Canvas grade CSV should contain only the identifiers needed for matching, the section when operationally useful, and the assignment grade column. Do not include attempts, feedback, free-text answers, or pseudonymisation keys in a Canvas import file.

References