Theoretical Questions

1. What are the key responsibilities of a Data Analyst?

• Answer:
• Collecting and interpreting data.
• Cleaning and organizing data.
• Performing analysis to identify patterns and trends.
• Creating visualizations and dashboards to communicate insights.
2. Explain the difference between structured and unstructured data.
• Answer:
• Structured Data: Data that is organized in rows and columns (e.g.,
databases).
• Unstructured Data: Data that is not organized in a predefined format
(e.g., text files, images).
3. What is the importance of data cleaning in analysis?
• Answer: Data cleaning ensures accuracy, consistency, and reliability of
insights by removing errors, duplicates, and irrelevant information.
4. What is the difference between correlation and causation?
• Answer:
• Correlation: A relationship between two variables without implying
causation.
• Causation: One variable directly affects another.
5. Describe the concept of normalization in databases.
• Answer: Normalization is the process of organizing data to reduce
redundancy and improve integrity by dividing a database into smaller tables and
defining relationships between them.
6. What is the significance of KPIs (Key Performance Indicators) in data
analysis?
• Answer: KPIs measure the performance of specific objectives, helping
organizations track progress and make data-driven decisions.
Technical Questions
SQL
1. Write an SQL query to find the second highest salary in an employee
table.
• Answer:SELECT MAX(salary) AS second_highest_salary
• FROM employees
• WHERE salary < (SELECT MAX(salary) FROM employees);
•
2. Scenario: You need to calculate the retention rate of customers month
over month. How would you write this query?
• Answer:SELECT month, COUNT(DISTINCT customer_id) AS retained_customers
• FROM transactions
• WHERE customer_id IN (
• SELECT customer_id
• FROM transactions
• WHERE month = 'previous_month'
• )
• GROUP BY month;
•
3. Write a query to identify duplicate records in a table.
• Answer:SELECT column1, column2, COUNT(*)
• FROM table_name
• GROUP BY column1, column2
• HAVING COUNT(*) > 1;
•
4. Scenario: Identify the top-selling product in each category.**
• Answer:SELECT category, product, MAX(sales) AS max_sales
• FROM sales_data
• GROUP BY category, product
• ORDER BY category, max_sales DESC;
 •
Python
1. How would you handle missing data in a Pandas DataFrame?
• Answer:import pandas as pd
• df['column_name'].fillna(df['column_name'].mean(), inplace=True)
• df.dropna(inplace=True)
•
2. Scenario: Write a Python script to calculate the correlation matrix for
a given dataset and visualize it as a heatmap.**
• Answer:import pandas as pd
• import seaborn as sns
• import matplotlib.pyplot as plt
•
• corr_matrix = df.corr()
• sns.heatmap(corr_matrix, annot=True, cmap='coolwarm')
• plt.show()
•
3. Write a Python script to find the most frequent word in a text column
of a DataFrame.
• Answer:from collections import Counter
•
• most_common_word = Counter('
'.join(df['text_column']).split()).most_common(1)
• print(most_common_word)
•
4. Scenario: Automate email sending for a monthly sales report.**
• Answer:import smtplib
• from email.mime.text import MIMEText
•
• def send_email(subject, body, recipient):
• msg = MIMEText(body)
• msg['Subject'] = subject
• msg['From'] = '[email protected]'
• msg['To'] = recipient
•
• with smtplib.SMTP('smtp.example.com', 587) as server:
• server.login('[email protected]', 'password')
• server.sendmail('[email protected]', recipient,
msg.as_string())
•
Pandas
1. How would you group data by a column and calculate the mean for each
group?
• Answer:grouped_data = df.groupby('column_name')
['numeric_column'].mean()
•
2. Scenario: Identify the top 5 products by sales in each region.**
• Answer:top_products = df.groupby('region')['sales'].apply(lambda x:
x.nlargest(5))
•
3. How do you merge two DataFrames on a common column?
• Answer:merged_df = pd.merge(df1, df2, on='common_column', how='inner')
•
4. Scenario: Create a rolling average for sales data over a 7-day
window.**
• Answer:df['7_day_avg'] = df['sales'].rolling(window=7).mean()
•
Power BI
1. How would you create a calculated column to classify sales as "High" or
"Low" based on a threshold?
• Answer:SalesCategory = IF(Sales[Amount] > 10000, "High", "Low")
•
2. Scenario: Create a dashboard to show sales trends, regional
performance, and key KPIs.**
• Answer:
• Use line charts for trends.
• Create bar/column charts for regional performance.
• Use card visualizations for key KPIs.
Tableau
1. Explain how to create a calculated field in Tableau.
• Answer:
• Go to the Data pane, click on the drop-down menu, and select "Create
Calculated Field."
• Write the formula, e.g., IF SUM(Sales) > 10000 THEN 'High' ELSE 'Low'
END.
2. Scenario: Create a churn analysis dashboard with KPIs and
visualizations.**
• Answer:
• Filter churned customers by last transaction date.
• Visualize with heatmaps for geographic data.
• Use line charts for churn trends.
Scenario-Based Questions
1. Guesstimate: How many Uber rides are taken in New York City on a
typical weekday?**
• Answer Approach:
• Estimate the population of NYC.
• Assume a percentage of people using Uber.
• Calculate average trips per user.
2. Case Study: Analyze sales performance using a given dataset. What steps
would you take?**
• Answer:
• Perform data cleaning.
• Conduct exploratory data analysis.
• Segment sales by various categories.
• Visualize insights using Power BI/Tableau.
3. Scenario: Identify main factors driving customer satisfaction.**
• Answer:
• Use correlation analysis on feedback data.
• Perform regression or decision tree analysis.
• Present findings with actionable recommendations.
4. Scenario: Forecast demand for a new product launch.**
• Answer:
• Analyze historical sales data of similar products.
• Use predictive modeling (e.g., ARIMA or Prophet).
• Validate results with business teams.