DSE - Data Science Exploratory Data Analysis Techniques Questions and Answers — Questions and Answers
Question 1: A data scientist is beginning to work with a new dataset of customer transactions. Which of the following is the PRIMARY goal of Exploratory Data Analysis (EDA) in this initial phase?
- To build a predictive machine learning model with high accuracy.
- To summarize the main characteristics of the data and uncover initial patterns. (Correct answer)
- To formally test predefined hypotheses about customer behavior.
- To deploy the data into a production database for real-time analysis.
Correct answer: To summarize the main characteristics of the data and uncover initial patterns.
The primary goal of EDA is to understand the dataset before formal modeling or hypothesis testing. This involves using statistical summaries and visualizations to discover patterns, spot anomalies, check assumptions, and understand the data's structure, which is crucial for subsequent analysis.
Question 2: During the EDA process for a housing price dataset, you create a scatter plot of 'square footage' versus 'price' and observe a strong positive linear relationship. This is an example of which type of EDA technique?
- Univariate non-graphical
- Univariate graphical
- Multivariate graphical (Correct answer)
- Multivariate non-graphical
Correct answer: Multivariate graphical
A scatter plot that examines the relationship between two variables ('square footage' and 'price') is a form of multivariate graphical analysis. Univariate analysis would focus on a single variable, and non-graphical methods involve summary statistics rather than plots.
Question 3: A data analyst is using a box plot to examine the distribution of salaries for a specific job role. The plot reveals several data points located far beyond the whiskers of the box. What is the most likely interpretation of these points?
- The median salary
- The interquartile range (IQR)
- Potential outliers (Correct answer)
- Missing data values
Correct answer: Potential outliers
Box plots are a standard graphical EDA technique used to display the distribution of numerical data. The 'whiskers' typically extend to 1.5 times the interquartile range (IQR) from the first and third quartiles. Data points that fall outside of these whiskers are considered potential outliers that may require further investigation.
Question 4: Which of the following techniques is most suitable for understanding the relationships between multiple continuous variables simultaneously during EDA?
- Histogram
- Bar chart
- Pie chart
- Correlation matrix with a heatmap (Correct answer)
Correct answer: Correlation matrix with a heatmap
A correlation matrix calculates the correlation coefficient between all pairs of continuous variables in a dataset. Visualizing this matrix as a heatmap provides a quick, intuitive way to spot strong positive or negative correlations between multiple variables at once, making it a powerful multivariate EDA tool.
Question 5: A data science team is analyzing a dataset with many features (high dimensionality). To simplify the data and aid in visualization, they apply a technique that reduces the number of variables while retaining most of the original information. Which EDA technique are they most likely using?
- Outlier detection
- Principal Component Analysis (PCA) (Correct answer)
- Hypothesis testing
- Data profiling
Correct answer: Principal Component Analysis (PCA)
Principal Component Analysis (PCA) is a classic dimensionality reduction technique used in EDA. It transforms a large set of correlated variables into a smaller set of uncorrelated variables called principal components, making it easier to visualize and analyze high-dimensional data without significant information loss.
Question 6: What is a key difference between Exploratory Data Analysis (EDA) and Confirmatory Data Analysis (CDA), such as formal hypothesis testing?
- EDA uses graphical techniques, while CDA exclusively uses statistical calculations.
- EDA is an open-ended process of generating hypotheses, while CDA aims to rigorously test pre-specified hypotheses. (Correct answer)
- EDA is only performed on small datasets, while CDA is used for big data.
- EDA focuses on data cleaning, while CDA focuses on model building.
Correct answer: EDA is an open-ended process of generating hypotheses, while CDA aims to rigorously test pre-specified hypotheses.
EDA is an exploratory, open-ended approach where the goal is to discover patterns, generate questions, and form hypotheses from the data without preconceived notions. In contrast, Confirmatory Data Analysis (which includes hypothesis testing) is a more rigid process focused on evaluating the evidence for or against specific hypotheses that were formulated beforehand.
DSE - Data Science Exploratory Data Analysis Techniques Questions and Answers
A data scientist is beginning to work with a new dataset of customer transactions.
Which of the following is the PRIMARY goal of Exploratory Data Analysis (EDA) in this initial phase?