ballet / predict-life-outcomes

Collaborating to solve the Fragile Families Challenge using the Ballet framework
5 stars 11 forks source link

[PARTITION] Parents' Education #48

Open ChengFR opened 3 years ago

ChengFR commented 3 years ago

Name

patients' education

Specification

Feature engineered from the questions asked to the Father or Mother of the child on their completed or in-progress education in all 5 waves.

from ballet import b
import fragile_families.analysis.metadata as metadata
X_df, _ = b.api.load_data()
cols = metadata.search([{'name': 'subtopics', 'op': 'like', 'val': '%education%'}, 
                        {'or': [{'name': 'respondent', 'op': 'eq', 'val': 'Father'}, 
                                {'name': 'respondent', 'op': 'eq', 'val': 'Mother'}]}])

There are 386 input columns in this partition.

Background

The parents' education information, including education level, skill training, etc., has been collected from Wave-1 (birth of the child) to Wave-5 (year 9). This information should directly influence the primary caregiver participation in job training at year 15. It may also indicate the family's finicial conditions and the child's performance in school.


I'm proposing this new feature development partition, who will take up the mantle to develop features from this partition? Whoever replies to this issue with the response me will "claim" it -- it might even be myself! It's okay for multiple people to claim this partition, but in that case, make sure you stay in touch directly or via the project chat, or follow each other's accepted (and rejected) feature contributions.

ChengFR commented 3 years ago

me