Closed skadauke closed 2 years ago
rsh52
commented
2 years ago Sounds good, everyone loves superheroes and it'll be easier to follow than "repeat/nonrepeat/nonrepeat2." It might be easier to just rename variables and content in the current databases than to make whole new ones.
The current vignette also seems like it was auto-generated from the README and should probably be reworked anyway.
rsh52
commented
2 years ago I made a test database and formatted/uploaded the heroes information in a non-repeating instrument. I opted to make all of the fields text fields. It would have required a lot of "find and replace"s to change the dataset labels to newly assigned numeric values (i.e. "Marvel Comics" = 1, "Dark Horse" = 2) since REDCap only allows for raw data uploads, not label uploads.
What were your thoughts on how to import the powers dataset? For context:
Did you have in mind that there would be one question about "power" and it would be set as a repeating instrument? Will probably require a bit of some rework to get this sheet into an importable format.
Other option is to make a way more trimmed down version of this and forgo wrapping our arms around the entire dataset 😅
skadauke
commented
2 years ago Yes I was thinking for rework you can pivot longer then filter for power == TRUE? Would be good to have this as a repeat instrument to demo how redcaptidier handles nonrepeat and repeat tables differently!
Get Outlook for iOShttps://aka.ms/o0ukef
From: Rich Hanna @.> Sent: Wednesday, August 3, 2022 4:58:33 PM To: CHOP-CGTDataOps/REDCapTidieR @.> Cc: Stephan Kadauke @.>; Author @.> Subject: Re: [CHOP-CGTDataOps/REDCapTidieR] Create a demo database to make the vignette easier to understand (Issue #4)
I made a test database and formatted/uploaded the heroes information in a non-repeating instrument. I opted to make all of the fields text fields. It would have required a lot of "find and replace"s to change the dataset labels to newly assigned numeric values (i.e. "Marvel Comics" = 1, "Dark Horse" = 2) since REDCap only allows for raw data uploads, not label uploads.
What were your thoughts on how to import the powers dataset? For context: [image]https://user-images.githubusercontent.com/38384694/182706738-f65bd141-8079-4e67-8fb4-7cf49fa90733.png
Did you have in mind that there would be one question about "power" and it would be set as a repeating instrument? Will probably require a bit of some rework to get this sheet into an importable format.
Other option is to make a way more trimmed down version of this and forgo wrapping our arms around the entire dataset 😅
— Reply to this email directly, view it on GitHubhttps://github.com/CHOP-CGTDataOps/REDCapTidieR/issues/4#issuecomment-1204476446, or unsubscribehttps://github.com/notifications/unsubscribe-auth/ACTGHWSOQU2HYQSLYRSMZPDVXLMPTANCNFSM547AE7IQ. You are receiving this because you authored the thread.Message ID: @.***>
rsh52
commented
2 years ago Just flexed some tidyverse data manipulation magic and imported the super heroes datasets into the REDCap. Just added you as an admin so you should get an e-mail.
Let me know what you think, I tested the code below and got the expected result:
read_redcap_tidy(redcap_uri = Sys.getenv("REDCAP_URI"),
token = Sys.getenv("SUPERHEROES_REDCAP_API")) %>%
bind_tables()bind_tables output:
> heroes_information
# A tibble: 734 × 12
recor…¹ name gender eye_c…² race hair_…³ height weight publi…⁴ skin_…⁵ align…⁶ form_…⁷
<dbl> <chr> <chr> <chr> <chr> <chr> <dbl> <dbl> <chr> <chr> <chr> <dbl>
1 0 A-Bo… Male yellow Human No Hair 203 441 Marvel… - good 0
2 1 Abe … Male blue Icth… No Hair 191 65 Dark H… blue good 0
3 2 Abin… Male blue Unga… No Hair 185 90 DC Com… red good 0
4 3 Abom… Male green Huma… No Hair 203 441 Marvel… - bad 0
5 4 Abra… Male blue Cosm… Black -99 -99 Marvel… - bad 0
6 5 Abso… Male blue Human No Hair 193 122 Marvel… - bad 0
7 6 Adam… Male blue - Blond -99 -99 NBC - … - good 0
8 7 Adam… Male blue Human Blond 185 88 DC Com… - good 0
9 8 Agen… Female blue - Blond 173 61 Marvel… - good 0
10 9 Agen… Male brown Human Brown 178 81 Marvel… - good 0
# … with 724 more rows, and abbreviated variable names ¹record_id, ²eye_color,
# ³hair_color, ⁴publisher, ⁵skin_color, ⁶alignment, ⁷form_status_complete
# ℹ Use `print(n = ...)` to see more rows
> super_hero_powers
# A tibble: 5,966 × 4
record_id redcap_repeat_instance power form_status_complete
<dbl> <dbl> <chr> <dbl>
1 0 1 Accelerated Healing 0
2 0 2 Durability 0
3 0 3 Longevity 0
4 0 4 Super Strength 0
5 0 5 Stamina 0
6 0 6 Camouflage 0
7 0 7 Self-Sustenance 0
8 1 1 Agility 0
9 1 2 Accelerated Healing 0
10 1 3 Cold Resistance 0
# … with 5,956 more rows
# ℹ Use `print(n = ...)` to see more rowsI would probably keep the current database as a test database and create a new one for superheroes. Would make it a classic with one nonrepeated instrument (superhero info) and one repreated instrument (superpowers)
Get Outlook for iOShttps://aka.ms/o0ukef
From: Rich Hanna @.> Sent: Friday, July 29, 2022 9:43:38 AM To: CHOP-CGTDataOps/REDCapTidieR @.> Cc: Stephan Kadauke @.>; Author @.> Subject: Re: [CHOP-CGTDataOps/REDCapTidieR] Create a demo database to make the vignette easier to understand (Issue #4)
Sounds good, everyone loves superheroes and it'll be easier to follow than "repeat/nonrepeat/nonrepeat2." It might be easier to just rename variables and content in the current databases than to make whole new ones.
The current vignette also seems like it was auto-generated from the README and should probably be reworked anyway.
— Reply to this email directly, view it on GitHubhttps://github.com/CHOP-CGTDataOps/REDCapTidieR/issues/4#issuecomment-1199302971, or unsubscribehttps://github.com/notifications/unsubscribe-auth/ACTGHWWIOZMZCWRTQJ3AX4LVWPNYVANCNFSM547AE7IQ. You are receiving this because you authored the thread.Message ID: @.***>
Feature Request Description
The current test database is being used to showcase the functionality of REDCapTidieR in the README.Rmd. Let's create a more fun demo database, I'm thinking of this one here → https://www.kaggle.com/claudiodavi/superhero-set/home
Proposed Solution
Let's build a classic database "superheroes" with one record per hero and two instruments: heroes_information (nonrepeating) and superpowers (repeating) and use those to showcase the idea of having different granularities in different instruments.
Checklist