README.md

---
language:
- en
license: mit
dataset_info:
  features:
  - name: id
    dtype: string
  - name: passage
    dtype: string
  - name: question
    dtype: string
  - name: choice_A
    dtype: string
  - name: choice_B
    dtype: string
  - name: choice_C
    dtype: string
  - name: choice_D
    dtype: string
  - name: correct_answer
    dtype: string
  - name: rationale
    dtype: string
  - name: difficulty
    dtype: string
  - name: domain
    dtype: string
  splits:
  - name: train
    num_bytes: 706258.0
    num_examples: 397
  download_size: 346011
  dataset_size: 706258.0
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---
# Dataset Card for SAT Reading and Writing Dataset

This dataset card aims to be a base template for the SAT Reading and Writing Dataset, optimized for use with Hugging Face's datasets library.

## Dataset Details

### Dataset Description

This dataset contains SAT Reading and Writing assessment questions sourced from the College Board's SAT Suite Question Bank, intended for use in training and evaluating Language Models like LLMs.

- **Curated by:** College Board
- **License:** Creative Commons Attribution-ShareAlike 4.0 International License

### Dataset Sources

- **Repository:** [College Board SAT Suite Question Bank](https://satsuitequestionbank.collegeboard.org/)

## Uses

### Direct Use

The dataset can be used for SAT exam preparation and educational purposes. It is suitable for training and evaluating Language Models for SAT-style reading comprehension and writing tasks.

## Dataset Structure

The dataset contains questions with passages, choices, correct answers, and rationales, making it ideal for training and evaluating Language Models on SAT-style reading comprehension and writing tasks.

## Dataset Creation

### Curation Rationale

The dataset was created to provide a resource for students preparing for the SAT exam and for researchers and developers working on Natural Language Processing tasks related to standardized testing.

### Source Data

#### Data Collection and Processing

The questions were extracted from the College Board SAT Suite Question Bank using automated scraping and filtering processes.

#### Who are the source data producers?

The College Board is the producer of the source data.

## Bias, Risks, and Limitations

Users should be aware of the limitations of using this dataset for predicting SAT scores and should use it in conjunction with other resources for a comprehensive SAT preparation.

### Recommendations

Users should use this dataset as a supplementary resource for SAT exam preparation and as a benchmark for evaluating the performance of Language Models on SAT-style reading comprehension and writing tasks.

## Citation

**APA:**
College Board. SAT Reading and Writing Dataset. Retrieved from [College Board SAT Suite Question Bank](https://satsuitequestionbank.collegeboard.org/)