slinger | shell extension for throwing windows

Data in fire-base database is as below:

my intention is to interact with a single dictionary at a time, such as delete it from fire-base or update it. To do this i have to filter through the database with multiple filters such as by home_team and away_team

edit

I have been successful in filtering through the data to the specific dictionary i would like to remove. ps. add rules to firebase as followed here: adding rules to firebase

current code is as below:

I want to find an easy way to combine 3 ratings together for a mongoDB query. Specifically, is there some way to combine 3 aggregate calls in one. So for example i have the below data. Which lists ratings for movies, either in imbd.rating, tomato.rating or metacritic, I'd like to find the top 2 movies for each of these ratings. The bottom query is how i'm fetching the top 2 results based on the tomato.rating

data

I have the following object in a mongodb. I'm wanting to find the genre that has won the highest number of awards, and out of the whole collection find the top 3 most found genres. I'm not really sure how to go about targeting specific fields within a collection like this, is it better to treat it as a large array? or is that a ridiculous comment.

Tried query which fails because the genre field is not an accumulator

I have a dataset called data_words that looks as follows (these are just a few lines):

I did a word learning experiment with 54 participants. The design was such that each participant would be learning 12 cognate words and 12 non-cognate words. However, I had to remove some observations from the dataset, ending up with 1591 cognate observations and 1816 non-cognate observations.

Now, I want to calculate the correlation between scores for cognates and non-cognates. This means that I will have to reduce the 1816 non-cognate observations to 1591, because correlations are always performed with equal samples.

I could just remove rows 1592-1816, but that would not be ideal, because I would loose all non-cognates for the later participants (the data are sorted by participant number).

What I'd rather do, is loop through all the participants, and for each participant remove as many 'surplus' non-cognates so that the number of cognates and non-cognates becomes equal for that participant.

This is an example of the data for 2 participants (note that words are tested at multiple points in time, ideally this would also be taken into account when removing data):