[CTIS] New child schooling and child age indicators for Wave 12 #1541
Conversation
nmdefries
changed the title
There was a problem hiding this comment.
Looks good, but I want to push back on a couple of the API additions because I don't think they're really necessary for the API, just for contingency tables.
Also, we're going to need to add these all to the documentation pages for the API and contingency tables, which should be... fun.
facebook/delphiFacebook/R/binary.R
Outdated
| # misc children | ||
| "smoothed_whas_child_under_18", "weight", "ch_has_child_under_18", 6, compute_binary_response, jeffreys_binary, | ||
|
|
||
| "smoothed_woldest_child_under_5", "weight", "ch_oldest_child_under_5", 6, compute_binary_response, jeffreys_multinomial_factory(4), | ||
| "smoothed_woldest_child_5_to_11", "weight", "ch_oldest_child_5_to_11", 6, compute_binary_response, jeffreys_multinomial_factory(4), | ||
| "smoothed_woldest_child_12_to_15", "weight", "ch_oldest_child_12_to_15", 6, compute_binary_response, jeffreys_multinomial_factory(4), | ||
| "smoothed_woldest_child_16_to_17", "weight", "ch_oldest_child_16_to_17", 6, compute_binary_response, jeffreys_multinomial_factory(4) |
There was a problem hiding this comment.
Not sure why we need daily updates on these in the API, unless we want to track the COVID baby boom... seems like something for demographic contingency tables. Was this part of a list from Facebook?
There was a problem hiding this comment.
Yup, they're all from FB. I'll loop you in on Asana.
There was a problem hiding this comment.
Removed from API and weekly/monthly contingency tables with plan to include in demographic contingency tables.
facebook/delphiFacebook/R/binary.R
Outdated
| "smoothed_wchild_school_public", "weight", "s_child_school_public", 6, compute_binary_response, jeffreys_multinomial_factory(5), | ||
| "smoothed_wchild_school_private", "weight", "s_child_school_private", 6, compute_binary_response, jeffreys_multinomial_factory(5), | ||
| "smoothed_wchild_school_homeschool", "weight", "s_child_school_homeschool", 6, compute_binary_response, jeffreys_multinomial_factory(5), | ||
| "smoothed_wchild_school_not", "weight", "s_child_school_not", 6, compute_binary_response, jeffreys_multinomial_factory(5), | ||
| "smoothed_wchild_school_other", "weight", "s_child_school_other", 6, compute_binary_response, jeffreys_multinomial_factory(5), |
There was a problem hiding this comment.
Not sure these need to be in the API, since we don't expect them to change rapidly over time
There was a problem hiding this comment.
Removed from API. Added to monthly contingency tables only.
| @@ -193,6 +193,7 @@ load_response_one <- function(input_filename, params, contingency_run) { | |||
| input_data <- code_activities(input_data, wave) | |||
| input_data <- code_vaccines(input_data, wave) | |||
| input_data <- code_schooling(input_data, wave) | |||
| input_data <- code_children(input_data, wave) | |||
There was a problem hiding this comment.
this function name sounds ominous
|
@krivard This is ready to merge and deploy |
Description
Add new indicators to both API and contingency tables at FB's request.
Add new variable creation function for non-schooling, non-vaccine children-related items (e.g. age breakdown).
Changelog
binary.Rcontingency_indicators.Rvariables.Rresponses.R