Basic joins: incomplete movie_comments field

Hello Everyone! I am with recreating the movie_comments example from the video. I have got the movie_comments field created, but it only adds one comment regardless of the number of comments a movie has. Find my code below and an attached screenshot. The first sample document should have 2 comment objects, but only has 1. Any help would be much appreciated.

{ from: “comments”,
let: {‘id’: ‘$_id’},
pipeline: [
{’$expr’:{’$eq’:[’$movie_id’, ‘$$id’]}
as: ‘movie_comments’

Update: I’ve found most of the movie_id 's from the comments collection do not actually correspond to any _id 's in the movies collection.

Could I have missed a step uploading the sample dataset into the cluster?