Module 3 Part 2 and P3 Quiz

Why is data visualization important?

It provides a holistic understanding of data

Which statistic measures central tendency?

Mean

Median is useful when data is:

Skewed or contains outliers

Anscombe’s Quartet shows:

Same statistics but different distributions highlighting the importance of visualizing data

Visualization helps detect:

Patterns and anomalies

Histogram is used for:

Distribution of one variable

Density plot is:

Continuous histogram

(smoothed-out, continuous version of a histogram. Instead of using discrete, stepped bars to show frequencies, a density plot uses kernel density estimation (KDE))

Scatterplot is used for:

Two variables (correlation between two numerical variables)

LOESS (Locally Estimated Scatterplot Smoothing) is used to:

fit smooth curve (non para-metric regression method)

Correlation measures:

Relationship between variables

Outliers may indicate:

Dirty data

Log transformation helps when data is:

Highly skewed

(specifically used to manage right-skewed (positively skewed) data)

Bimodal distribution means:

Two peaks

Unimodal distribution means:

One peak

Missing values appear as:

Zero spikes or gaps

Scatterplot overplotting problem occurs when:

Too much data

Hexbin plots help:

Visualize dense data

(used to address the issue of overplotting in scatter plots, which occurs when dealing with large datasets where many points overlap)

Boxplot shows:

Distribution summary

(min, 1 quartile, median, 3 quartile, max)

Whiskers in boxplot represent:

Range of data

Jitter is used to:

Reduce overlap

(adds a small amount of random noise to the position of data points in a plot to help improve clarity)

Pairwise plots (grid of scatter plots) show:

Relationships among variables (each plot represents the relationship between a unique pair of numerical variables.)

Time series (data plotted in chronological order) plot is used for:

Data over time

Seasonality means:

Repeated patterns over time

Trend shows:

General direction

Data exploration focuses on:

Understanding data

Data presentation focuses on:

Communicating results

Regression line represents:

Linear relationship

LOESS vs regression:

LOESS handles non-linear

Cloud-like scatterplot means:

Weak relationship

Exponential relationship becomes linear after:

Log transformation

Dirty data includes:

Incorrect or missing values

Saturated data means:

Capped or limited values

Histogram bins represent:

Frequency counts

Correlation close to 1 means:

Strong positive relationship

Correlation close to 0 means:

Weak relationship

Multiple clusters indicate:

Different populations

Data range shows:

Spread of values

Visualization before modeling helps:

Better understanding

Boxplot median is shown by:

Line inside box

R function for histogram:

hist()