Hi @Pabblo_96,
The error occurs due to the 'All' record not having the same structure as the Dataverse choice column. We can add an additional option by looping through the Dataverse option set and matching that record schema. This approach will, however, also require some changes to your Filter statement.
(1) Add 'All' to your dropdown
//Option 1 -> create a collection e.g. in the App.Onstart and use that collection as your dropdown items property
ClearCollect(
colSubject,
{Value: "All"},
ForAll(
Choices(Students.Subject),
{Value: Text(ThisRecord.Value)}
)
);
//Option 2 - use the code below in the Items property of your dropdown (use option 1 if you need this choice list across multiple dropdowns / screens)
//Ungroup nested table structure (merge tables)
Ungroup(
[
[{Value: "All"}],
ForAll(
Choices(Students.Subject),
{Value: Text(ThisRecord.Value)}
)
],
"Value"
)
(2) Change Filter condition that references the choice column
Due to the record schema adjustment, your filter conditions referencing this dropdown will result in an 'incompatible type' error. We can resolve this by creating a mapping to the correct choice via the With & Switch functions:
With(
{
wChoice: Switch(
//Change 'DropdownName'
DropdownName.Selected.Value,
"English",
//Should the choice name not be recognized - see comment below this code block
'Subject (Students)'.'English',
"Maths",
'Subject (Students)'.'Maths',
"Geography",
'Subject (Students)'.'Geography'
)
},
Filter(
Students,
//Change 'DropdownName'
//Filter by choice if 'All' is not selected
DropdownName.Selected.Value = "All" || Subject = wChoice
)
)
The name of your choice column will vary depending on whether you sync your choice set (see column settings). Should they be synced then you will have to replace 'Subject (Students)' with the name provided for that choice set.

The code block above expects 'No' to be selected. In this case the name will have the following structure:
'ColumnName (TableName)'
If this solves your question, would you be so kind as to accept it as a solution.
Thanks!