Hugging Face's logo

Datasets:
YXu120
/
NC_Education

Languages:
English
Size Categories:
1K<n<10K
Tags:
License:
Dataset card Viewer Files Community
1
NC_Education / README.md
YXu120's picture
YXu120
Update README.md
710d2fd verified
preview code
|
raw history blame contribute delete
No virus
5.27 kB
metadata
language:
  - en
license: cc-by-sa-4.0
pretty_name: NC Education Data
size_categories:
  - 1K<n<10K

netid: yx264

Dataset Description

Dataset Summary

The education datasets can be found at North Carolina Office of State Budget and Management (NC OSBM) open data portal.

The first data includes public school enrollment and finances, SAT score, higher education enrollment, educational attainment data for North Carolina and counties, and the second one includes educational attainment for the Black population aged 25 and over for North Carolina and counties. The processed final data will combine the information of them, providing a profound insight into NC educational level and equity.

Supported Tasks

The tasks/methodologies are:

  • (1) Data checking and cleaning

Since the first dataset is pretty large and includes some information we don't care, like the SAT score. Before proprecessing, we first check the datasets, do some visualization and summarizing, to decide whether to drop certain rows.

  • (2) Re-structure the datasets to present the values according to county-level distinctions

The original dataset architecture lacks depth, as all variables reside on a single plane. Considering the significance of these variables, it’s nature to contemplate reorganizing the hierarchy to reflect county-level distinctions and convey the data accordingly.

  • (3) Keywords-based data matching

The data matching process relies on all columns except the 'value' column. While 'area_name', 'area_type', and 'year' offer straightforward matching points across datasets, it's important to acknowledge the potential disparity in the 'variable' column. For instance, one dataset may contain the variable 'Less than High School Diploma', while another might list 'Less than High School Black'. Despite conveying the same meaning and requiring alignment, a direct join isn't feasible due to this discrepancy. One potential approach involves extracting keywords from the variables, such as 'Less' and 'High School', and cross-referencing them to determine potential matches between the datasets.

Languages

English

Dataset Structure

Data Instances

An example of records (note that specific details are subject to change):

{
"area_name":"Alexander County",
"area_type":"County",
"year":"2010",
"variable":"Elementary School Education or Less Age 25 Up",
"value":2325
}

An example of records (note that specific details are subject to change):

{
"area_name":"Burke County",
"area_type":"County",
"year":"2020",
"variable":"Less than High School Diploma",
"value":568
}
  • (3) Anticipated structure we aim to achieve:
{
"area_name":"Alamance County",
"area_type":"County",[("year": "2010", [
{"variable": "High School Graduate (includes Equivalency)", "value": 6443},
{"variable":"Some College or Associate's Degree","value":5242},
...
]),
...]
...
}

Data Fields

  • Area Name (string): name of the area
  • Area Type (string): type of the area
  • Year (integer): the year that the data was collected
  • Variable (string): level of education or other situations
  • Value (integer): population count

Dataset Creation

Curation Rationale

The datasets were collected and published to present the educational level of NC population in different area. The educational attainment for the black population data can rise the concern for educational equity issue in North Carolina. The combined dataset aims to offer a holistic perspective on educational levels and equity, with a specific focus on the educational attainment of the Black population aged 25 and over.

Source Data

The original data were published by North Carolina Department of Public Instruction, North Carolina Community College System and US Census Bureau.

Considerations for Using the Data

The records are only for limited years, making it challenging to discern timely patterns or trends.

The second data is quite small compared to the first one. Combining may result in many missing values.

Additional Information

Citation Information

Bureau, US Census. Census.Gov, 21 Jan. 2024, www.census.gov.

NC Budget & Management Open Data Portal Home Page - NC OSBM, https://linc.osbm.nc.gov/pages/home.

Contributions

Thanks North Carolina Department of Public Instruction, North Carolina Community College System and US Census Bureau for these datasets.