NCA-ADS - Descriptive Analysis and Visualization (13% of the exam) - Section 4.2
Select appropriate visualisations.
Select appropriate chart types - histograms, scatter plots, box plots, heatmaps, and line charts - based on the variable type and the analytical question. Recognise when a visualisation choice obscures rather than reveals the underlying pattern.
Practice question for this objective
lock_openFree sampleDescriptive Analysis and Visualizationeasy
An analyst has a cuDF DataFrame holding the daily order count for a single warehouse over three years and wants one chart that shows how that count changes from day to day and reveals any seasonal pattern across the period. The horizontal axis represents time and the vertical axis represents the count. Which chart type best fits this goal?
AA pie chart, with one slice for each day so the relative share of every single day's order count is visible at a glance.
BA histogram of the order counts, grouping the daily values into bins to show how often each order count level occurs.
CA line chart, plotting the order count against the date so successive daily values are connected in temporal order.check_circle Correct
DA scatter plot of order count against warehouse identifier, placing one point per day to compare across the sites.
A line chart is the appropriate visualisation for showing how a single numeric value changes over a continuous time axis. Trend and seasonality are properties of an ordered sequence of values across time. A line chart maps time to a continuous axis and joins consecutive points, making rises, falls, and repeating cycles directly visible in a way that distribution or proportion charts cannot.
Why A is wrong: A pie chart shows parts of a whole at one moment and is tempting because it encodes proportion, but with hundreds of days it cannot convey order or trend over time and becomes unreadable.
Why B is wrong: A histogram summarises the distribution of the counts and is tempting because the data is numeric, but it discards the date axis entirely and so cannot show change over time or seasonality.
Why C is correct: A line chart connects ordered observations along a continuous time axis, which is exactly how trends and seasonal cycles in a single time series are read; it is the standard choice for one numeric value tracked over time.
Why D is wrong: A scatter plot relates two variables and is tempting for numeric data, but there is only one warehouse here and no second continuous variable, so it cannot display the day-to-day temporal trend.
Exam traps in Descriptive Analysis and Visualization
Answers that look right on this material and are not. Each one is a distractor from a different question in the NCA-ADS bank for this domain.
A histogram of the payment methods, binning the 600,000 values so the frequency of each binned range of methods is shown.
errorWhy it is wrong: A histogram bins a continuous numeric variable and is tempting because it also shows frequency, but payment method is nominal with no numeric range to bin, so a histogram does not apply.
Bar chart
errorWhy it is wrong: A bar chart is designed for comparing discrete categories, not for showing how a continuous numerical variable is spread across a range of values.
A bar chart with one bar per campaign, where the bar height encodes the sign-ups, ordered along the axis by marketing spend.
errorWhy it is wrong: A bar chart compares a value across categories and is tempting because spend can order the bars, but 5,000 bars are unreadable and the chart shows only one variable per bar rather than the joint relationship between two.