minor spelling and markdown formatting fixes; valid has also been removed from the tests

Author: karenetheridge

output-tests/README.md

@@ -2,21 +2,21 @@ These tests are intended to validate that implementations are correctly generati
 
 Output was initially specified with draft 2019-09.  It remained largely unchanged for draft 2020-12, but will receive an update with the next release.
 
-***NOTE** Although the formats didn't change between 2019-09 and 2020-12, the tests are replicated for 2020-12 because the `$schema` is different and implementations may (but shouldn't) produce different output.*
+_**NOTE** Although the formats didn't change much between 2019-09 and 2020-12, the tests are copied for 2020-12 because the `$schema` is different and implementations may (but shouldn't) produce different output._
 
 ## Organization
 
 The tests are organized by specification release and then into two categories: content and structure.
 
-Content tests verify that the keywords are producing the correct annotations and/or error messages.  Since there are no requirements on the content of error messages, there's not much that can be verified for them, but it is possible to identify when a error message _could_ be present.  Primarily, these tests need to extensively cover the annotation behaviors of each keyword.  The only output format needed for these tests is `basic` for 2019-09/2020/12 and `list` for later versions.
+Content tests verify that the keywords are producing the correct annotations and/or error messages.  Since there are no requirements on the content of error messages, there's not much that can be verified for them, but it is possible to identify when a error message _could_ be present.  Primarily, these tests need to extensively cover the annotation behaviors of each keyword.  The only output format needed for these tests is `basic` for 2019-09/2020-12 and `list` for later versions.
 
-Structure tests verify that the structures of the various formats (i.e. `flag`, `basic`, `detailed`, `verbose` for 2019/2020 and `flag`, `list`, `hierarchical` for later versions) are correct.  These tests don't need to cover each keyword; rather they need to sufficiently cover the various aspects of building the output structures by using whatever keywords are necessary to do so.
+Structure tests verify that the structures of the various formats (i.e. `flag`, `basic`, `detailed`, `verbose` for 2019-09/2020-12 and `flag`, `list`, `hierarchical` for later versions) are correct.  These tests don't need to cover each keyword; rather they need to sufficiently cover the various aspects of building the output structures by using whatever keywords are necessary to do so.
 
 In each release folder, you'll also find an _output-schema.json_ file that contains the schema from the specification repo that describes output for that release.  This schema will need to be loaded as the tests reference it.
 
 ## Test Files
 
-The content of a test file is the same as the validation tests in `tests/`, however an `output` property has been added to each test case.
+The content of a test file is similar to the validation tests in `tests/`: for each test case, the `valid` property has been removed, and `output` has been added.
 
 The `output` property itself has a property for each of the output formats where the value is a schema that will successfully validate for compliant output.  For the content tests, only `basic`/`list` needs to be present.