Mixed Deck — All DAC Topics Flashcards

Which metric summarizes model performance across all classification thresholds?

Answer: AUC-ROC

The AUC-ROC measures a classifier's ability to rank positives above negatives across every threshold.

A histogram is most useful for showing what about a single numeric variable?

Answer: Its distribution

Histograms group values into bins to reveal the shape of a distribution.

In Power BI or Tableau, what is the main benefit of drill-down functionality?

Answer: Letting users navigate from summary to detailed levels

Drill-down lets users move from aggregated summaries into finer detail.

What does standard deviation measure in a dataset?

Answer: The average distance of data points from the mean

Standard deviation quantifies how spread out the data values are around the mean — a larger value indicates more variability.

A report double-counts revenue because the same order appears twice. What cleaning step fixes this?

Answer: Deduplication

Deduplication removes repeated records so each entity is counted once.

What is 'data-ink ratio' as described by Edward Tufte?

Answer: The proportion of ink used to display actual data versus decoration

Data-ink ratio measures how much of a chart's ink represents real data versus non-essential decoration.

What is a dashboard in data analytics?

Answer: An interactive visual display of key metrics and KPIs, providing at-a-glance monitoring of business performance

Analytics dashboards consolidate key metrics, KPIs, and visualizations into a single interactive interface, enabling stakeholders to monitor performance, identify trends, and make data-driven decisions quickly.

When choosing between showing a data table and a chart, you should generally use a chart when:

Answer: The goal is to reveal patterns, trends, or comparisons

Charts excel at revealing patterns and trends, while tables are better for exact values.

What is a data steward's primary responsibility?

Answer: Overseeing data quality and enforcing data policies within a domain

A data steward is responsible for maintaining data quality, enforcing governance policies, and acting as a subject-matter expert for a specific data domain.

A correlation coefficient of -0.9 indicates:

Answer: A strong negative relationship

A value near -1 indicates a strong inverse relationship between variables.

In data governance, what is a 'data dictionary'?

Answer: A document defining the meaning, format, and rules for each data element in a system

A data dictionary is a centralized repository that defines all data elements including their names, definitions, data types, and business rules.

Grouping continuous ages into ranges like 0-18, 19-35, 36-60 is an example of what?

Answer: Binning

Binning buckets continuous values into discrete interval categories.

In HDFS, what is the default behavior for ensuring data durability across the cluster?

Answer: Replicating each block across multiple nodes

HDFS replicates each data block (default 3 copies) across DataNodes for fault tolerance.

A chart shows sales rising every December for three years. This repeating pattern is called what?

Answer: Seasonality

Seasonality is a pattern that repeats at regular calendar intervals.

Which of the following best describes a data lake?

Answer: A repository that stores vast amounts of raw data in its native format until needed

A data lake stores large volumes of raw, unstructured, semi-structured, and structured data in its native format, offering flexible schema-on-read capabilities.

Which chart type is best for displaying the distribution of a single continuous variable?

Answer: Histogram

A histogram bins a continuous variable to show its frequency distribution.

Why is consistent formatting (fonts, colors, number formats) important across a report suite?

Answer: It builds trust and reduces misinterpretation

Consistent formatting builds credibility and helps readers interpret data correctly across reports.

In a balanced scorecard approach to analytics strategy, which of the following is NOT one of the four perspectives?

Answer: Technology Infrastructure

The Balanced Scorecard's four perspectives are Financial, Customer, Internal Business Processes, and Learning & Growth — technology infrastructure is not one of the four.

A dual-axis chart is most useful when you need to:

Answer: Show two measures with different scales together

Dual-axis charts let you plot two measures with different scales on the same chart for comparison.

A data analyst is publishing a report using a dataset that includes personally identifiable information (PII). Which technique best protects individual privacy while preserving data utility?

Answer: Anonymization or de-identification of the PII fields

Anonymization or de-identification removes or masks PII so individuals cannot be identified, allowing the data to be used analytically without exposing personal information.