SQL Data Hygiene: Hearing Wellness Survey

Overview

This project demonstrates data cleaning and preprocessing of a hearing wellness survey dataset using PostgreSQL.
Clean, standardized data ensures reliable analysis, visualization, and insights.

Dataset: Hearing Wellness Survey on Kaggle

---

Steps

1. Setup

   • Created database hearing_wellness_db and table hearing_survey.
   • Imported survey CSV into PostgreSQL.

2. Data Inspection

   • Checked structure, types, and missing values.
   • Previewed first 10 rows.

3. Text Standardization

   • Replaced special characters (curly quotes, dashes, ellipsis).
   • Removed non-ASCII characters.

4. Duplicates & Primary Key

   • Verified no duplicate rows.
   • Added id SERIAL PRIMARY KEY.

5. Column Cleaning & Normalization

   • daily_headphone_use: grouped time intervals.
   • interest_in_hearing_app: standardized Yes/No/Maybe.
   • missed_important_sounds: corrected typos.
   • hearing_test_barrier: split multi-responses and categorized.
   • desired_app_features: normalized multi-select features.
   • Created mapping/normalized tables for structured analysis.

---

Outcome

• Dataset is clean, consistent, and free of duplicates.
• Textual and categorical data standardized for meaningful analysis.
• Normalized tables allow aggregation and visualization of survey responses.

---

Usage

1. Clone the repository.
2. Run hearing_survey_data_cleaning.sql in PostgreSQL.
3. Query normalized tables for analysis and visualization.