pipeline.cip_crosswalk

The code in pipeline.cip_crosswalk handles transformations related to CIP (Classification of Instructional Program) codes:

• The CIPCodeCrosswalk handles transforming old CIP codes (CIP90, CIP2k, and CIP2010) to the current standard (CIP2020)
• The DHSClassifier classifies CIP codes according to whether or not the appear in the Department of Homeland Security's STEM Designated Degree Program List
• The NCSESClassifier classifies CIP codes according to the NCSES Alternate Classification and the broad fields included in the NSF's Diversity and STEM reports

CIPCodeCrosswalk(crosswalk_dir: Path = pipeline.settings.RAW_DATA_DIR / pipeline.settings.CROSSWALKS_DIRNAME)

Handles cross-walking of CIP codes from the past to 2020 CIP codes

walk(year_range: Tuple[int, int], codes: pd.Series, titles: Optional[pd.Series] = None) -> Tuple[pd.Series, pd.Series]

Map from old set of codes (in old year range) to newer set of codes

convert_to_cip2020(year: int, codes: Union[str, List[str], pd.Series], titles: Optional[Union[str, List[str], pd.Series]] = None) -> pd.DataFrame

Convert from old CIP codes to CIP 2020 CIP Codes and Titles

NCSESClassifier(filepath: Path = PIPELINE_ASSETS / 'ncses_stem_classification_table.csv')

Class for converting CIP codes to NCSES hierarchical classification

Read the NCSES file into an internal df to use for classification

get_titles(codes: Union[str, Iterable[str]], fill_na: bool = True) -> pd.Series

Return NCSES title strings corresponding to 2020 CIP Codes

Parameters:

Name	Type	Description	Default
codes	Union[str, List[str], Series]	CIP 2020 codes	required
fill_na	bool	Whether to fill NA values with "Unknown". Default: True	True

Returns:

Type	Description
Series	pd.Series: NCSES title strings

classify(original_codes: Union[str, List[str], pd.Series], codes_2020: str | List[str] | pd.Series | None = None) -> pd.DataFrame

Classify CIP code(s) in the NCSES classification.

In all cases, prefer the classification of the CIP2020, but use the original version if the 2020 version is unclassified.

Parameters:

Name	Type	Description	Default
original_codes	Union[str, List[str], Series]	CIP code(s)	required
codes_2020	Union[str, List[str], Series]	CIP 2020 code(s) to classify, optional. Default: None	None

Returns:

Type	Description
DataFrame	pd.DataFrame: Data frame indexed by CIP code with each level of NCSES classifcation as columns

DHSClassifier(filepath: Path = PIPELINE_ASSETS / 'dhs_stem_classification_table.csv')

Read the NCSES file into an internal df to use for classification

classify(codes: Union[str, List[str], pd.Series, pd.Index]) -> pd.DataFrame

Classify a set of CIP codes as belonging (True) or not belonging (False) to the DHS set of STEM CIP codes

Parameters:

Name	Type	Description	Default
codes	Union[str, List[str], Series]	CIP code(s)	required

Returns:

Type	Description
DataFrame	pd.DataFrame: DataFrame indexed by input codes with one bool column indicating DHS STEM