Making Sense of Data: Descriptive Analytics in SQL
- Kavita
- Sep 3
- 5 min read
Updated: Sep 9
In my view, data is powerful, but it alone rarely tells us a meaningful story. Here is when analytics comes into the picture. When we source the raw data and perform the necessary Data Cleaning operations, the next step in Data Analysis is Descriptive analysis. It is to know and understand the data in depth which lets us uncover patterns, trends, and summaries. Then comes the, Prescriptive analysis which goes a step further on bringing new insights and stating actions to improve the Outcomes. Descriptive Analysis can be performed using SQL, pandas(Python), Excel/Google sheets or any BI Tool like Tableau/ Power BI.
In this blog, we’ll explore Descriptive analysis using SQL.
I have used the Hearing Survey dataset from Kaggle. This dataset is a collection of survey responses designed to capture people’s experiences, perceptions, and behaviors around hearing wellness. It provides valuable insights into aspects such as daily headphone use, perceived barriers to hearing tests, instances of missed important sounds, and interest in hearing care apps.
*Dataset- The reference link is provided at the end of this blog.
In my previous blog, I demonstrated the Data Cleaning operations in PostgreSQL—removing inconsistencies, standardizing values, and addressing missing information. With these preparations complete, the dataset is now well-structured and reliable for further analysis.
*Data Cleaning using SQL- The reference link is provided at the end of this blog.
This cleaned dataset serves as an excellent foundation to explore, Descriptive analysis. For the demonstration, I am using PostgreSQL and pgAdmin to showcase Data Cleaning in SQL.
Descriptive Analysis
Descriptive analysis helps us understand what the data is telling us by summarizing patterns, trends, and behaviors. Using SQL, we can explore key insights from the hearing survey dataset, uncovering how respondents interact with hearing care and technology."
Lets explore the columns in hearing_survey dataset one by one.
Age Distribution
To understand the demographics, we first examine the age spread of respondents.
SELECT age_group, COUNT(*) AS respondents
FROM hearing_survey
GROUP BY age_group
ORDER BY respondents DESC;
Takeaway:
Majority of responses (287) are from 18–24 year olds, which is ~77% of the dataset.
The next biggest group is 35–44 (30 responses), but that’s still just ~8% of the 18–24 group.
Very few respondents are over 45 or under 18.
2. Belief in Early Hearing Care
Determine how strong is the belief in Early Hearing care in people
SELECT belief_early_hearing_care, COUNT(*) AS respondents
FROM hearing_survey
GROUP BY belief_early_hearing_care
ORDER BY belief_early_hearing_care;
Takeaway:
Most respondents (186) selected 5, showing strong belief in early hearing care.
Very few respondents selected 1 or 2, suggesting low disagreement.
3. Daily Headphone Use:
Since headphone use is a potential risk factor, we analyze reported daily usage.
SELECT daily_headphone_use, COUNT(*) AS respondents
FROM hearing_survey
GROUP BY daily_headphone_use
ORDER BY respondents DESC;
Takeaway:
Most respondents (~148 people) use headphones less than 1 hour daily, indicating low exposure.
A sizable group (192 people) use headphones 1–4 hours daily, showing moderate use, while 44 respondents exceed 4 hours (potential risk group).
Missed Important Sounds
Hearing difficulties often show up in missed communication moments.
SELECT missed_important_sounds, COUNT(*) AS frequency
FROM hearing_survey
GROUP BY missed_important_sounds
ORDER BY frequency DESC;
Takeaway:
Majority (142 respondents) reported “No, I usually hear things well”, showing most participants do not perceive major hearing difficulties.
A significant portion (185 combined) admitted missing sounds in public spaces, work/school meetings, or family conversations, highlighting real-world listening challenges.
Feeling Left Out Due to Hearing
Social exclusion offers another lens into perceived hearing challenges.
SELECT left_out_due_to_hearing, COUNT(*) AS frequency
FROM hearing_survey
GROUP BY left_out_due_to_hearing
ORDER BY frequency DESC;
Takeaway:
131 respondents said they feel left out “only in noisy places”, showing hearing challenges are mostly environment-dependent.
A large segment (107 people) answered “Never”, meaning many participants don’t perceive exclusion due to hearing.
Smaller groups (95 “Sometimes”, 29 “Yes, often”, 25 “I try to avoid such situations”) reflect varying degrees of social impact, with a notable minority actively avoiding situations.
Barriers to Hearing Tests
A crucial barrier analysis highlights why people avoid hearing tests.
SELECT category, COUNT(*) AS barrier_count
FROM hearing_test_barrier_categorized
GROUP BY category
ORDER BY barrier_count DESC;
Takeaway:
Awareness gaps dominate with 260 respondents, showing lack of knowledge is the biggest hurdle in hearing test adoption.
Psychological barriers (78) like stigma or fear are the second-largest challenge, pointing to mindset-related obstacles.
Financial and logistical issues (34 financial, 1 time, 1 infrastructure) are relatively minor, suggesting access is less of a problem than awareness and perception.
Methods of Last Hearing Test
Exploring how respondents last checked their hearing gives clues about healthcare access.
SELECT last_hearing_test_method, COUNT(*) AS respondents
FROM hearing_survey
GROUP BY last_hearing_test_method
ORDER BY respondents DESC;
Takeaway:
Majority (307 respondents) have never taken a hearing test, showing a huge gap in proactive hearing care.
Traditional hospital or clinic tests (41) remain the most common method among those who got tested.
Digital solutions (mobile apps & online tests, 36 total) are starting to emerge but still represent a small fraction of adoption.
Interest in Hearing App
Since digital tools are emerging, we check openness to hearing apps.
SELECT interest_in_hearing_app, COUNT(*) AS respondents
FROM hearing_survey
GROUP BY interest_in_hearing_app
ORDER BY respondents DESC;
Takeaway:
"Maybe" (178 respondents) is the largest group, showing uncertainty but openness toward hearing apps.
Strong interest (135 respondents saying “Yes”) indicates a significant potential adoption base.
Skeptics (74 respondents saying “No”) remain, highlighting the need for awareness, trust-building, and clear value demonstration.
Desired App Features
Features people want in an app highlight expectations from digital health.
SELECT feature, COUNT(*) AS frequency
FROM desired_features_clean
GROUP BY feature
ORDER BY frequency DESC;
Takeaway:
Quick tests (245 respondents) stand out as the top priority, showing demand for fast and easy screening.
Doctor consultation (181) and game-based interaction (164) highlight the balance between professional trust and engagement-driven approaches.
Features like earphone calibration (152), soft guidance (150), and regular testing reminders (148) emphasize the need for personalized, supportive, and continuous care.
Advanced features such as detailed reports (140) and personalized volume adjustments (131) show interest in deeper insights and customization.
Ear Discomfort After Use
Lastly, we check if technology use itself creates physical discomfort.
SELECT ear_discomfort_after_use, COUNT(*) AS respondents
FROM hearing_survey
GROUP BY ear_discomfort_after_use
ORDER BY respondents DESC;
Takeaway:
142 respondents reported “No” discomfort, indicating most users tolerate headphone use well.
137 respondents experience discomfort occasionally, showing a significant portion with mild issues.
74 respondents experience discomfort regularly (“Yes”), representing a group at higher risk.
