2.2 What is test.results?
The autograder infrastructure has provided a test.results data frame to store and display feedback from our Checks.
When writing autograder code for any question, we always first initialize the question’s default test.results.
For example, to initialize Question 1’s test.results, we write the following line:
#Testing Student's Question 1 Against Question 1 Solution
test.results[1, ] <- c("Part 1 Question 1 (Public)", 0, 20, "Try again.")where each element of test.results[#, ] is as follows:
| Element | Description |
|---|---|
| 1 | the question’s number |
| “Part 1 Question 1 (Public)” | the question’s displayed part, number and type (Private/Public) |
| 0 | the question’s default score |
| 20 | the question’s maximum score |
| “Try again.” | the default feedback message |
It is important to note that a question’s displayed number may differ from its question number, especially in the case of multi-part questions.
Recommendation: If students would greatly benefit from a very specific hint (e.g., using a certain function), you can include it in the default feedback message to ensure they see it in the case that none of the Checks trigger.
After initializing a question’s test.results[#, ], we can update it through indexing.
- For example, if the student’s Question 1 is correct, we can update
test.results[1, ]to award them full score with the following line:
Similarly, we can modify the feedback message through the following line:
#Modifying the feedback message for Question 1
test.results[1, 4] <- "Insert feedback message here!"Note:
For most of the sample code provided for Public Questions, we will be updating test.results[2, ] (i.e., Question 2).