3.4 Quick Table Calculations & Analytics Tools

3.4 Quick Table Calculations & Analytics Tools

Tableau Desktop offers powerful built-in analytical tools that allow users to perform secondary calculations on query results and add statistical reference elements without writing complex database code. Understanding Table Calculations and the Analytics Pane is essential for answering execution and configuration questions on the Tableau Desktop Specialist exam.

---

Quick Table Calculations

A Table Calculation is a secondary calculation that Tableau computes on the aggregated result set stored in Tableau's local data cache after initial query results return from the underlying data source.

Primary Quick Table Calculations

Tableau provides pre-built Quick Table Calculations accessed by right-clicking any measure pill in the view and selecting Quick Table Calculation:

1. Running Total: Accumulates measure values sequentially across marks (e.g., cumulative sales over months).
2. Difference: Calculates the absolute difference between the current mark and a prior/subsequent mark.
3. Percent Difference: Calculates the percentage change relative to a prior/subsequent mark.
4. Percent of Total: Computes each mark's value as a percentage of the total aggregated partition sum.
5. Rank: Assigns an ordinal rank (1, 2, 3...) to marks based on measure magnitude.
6. Percentile: Calculates the percentile rank of each mark (0% to 100%).
7. Moving Average: Smooths data fluctuations by calculating an average over a defined rolling window of preceding/following marks.
8. Year over Year Growth: Compares a mark's value to the corresponding period in the prior year (requires a date field in the view).

Addressing vs. Partitioning Dimensions

To control how a table calculation computes across your visualization, right-click the pill and select Compute Using:

• Addressing Dimensions (Direction): Dimensions that define the direction and order of computation. Tableau iterates along addressing dimensions.
• Partitioning Dimensions (Scope/Group): Dimensions that group and isolate data. The table calculation resets its computation at every partition boundary.

Compute Using Options

• Table (across): Computes horizontally across the entire width of the table, resetting at each row.
• Table (down): Computes vertically down the height of the table, resetting at each column.
• Table (across then down): Computes across the row, then moves to the next row down continuously.
• Cell: Computes within each individual cell (useful for cell-level percent of total).
• Specific Dimensions: Allows you to explicitly check which dimensions act as Addressing fields vs. Partitioning fields.

Delta Symbol: When a measure pill has an active Table Calculation applied, a small triangle delta icon (Δ) appears on the right edge of the pill capsule (e.g., SUM(Sales) Δ).

---

The Analytics Pane

The Analytics Pane (accessible via the tab next to the Data Pane on the left sidebar) provides drag-and-drop analytical capabilities.

1. Constant & Reference Lines

• Constant Line: Draws a straight reference line across an axis at a fixed numeric value.
• Reference Line: Adds a line at a calculated statistic (Mean, Median, Minimum, Maximum, Sum) for an entire table, pane, or cell.
• Reference Band: Shades the region on an axis between two statistical values (e.g., shading between Min and Max sales).
• Distribution Band: Shades regions based on statistical distributions such as Percentiles, Quantiles, Standard Deviations, or Confidence Intervals.
• Bullet Graphs: Adds target reference lines and distribution background bands to compare measures against benchmarks.

2. Totals and Subtotals

• Grand Totals: Drag Totals from the Analytics pane to add Column Grand Totals or Row Grand Totals.
• Subtotals: Adds subtotals for discrete category breaks across panes.
• Total Aggregation Method: By default, Grand Totals calculate using the measure's underlying aggregation (e.g., calculating SUM across all rows). You can change how totals calculate by navigating to Analysis -> Totals -> Total Using and selecting Sum, Average, Minimum, Maximum, Automatic.

3. Trend Lines

Trend lines add statistical regression models to continuous scatter plots or line charts.

• Requirements: Both X and Y axes must be continuous fields (numeric or continuous date).
• Model Options: Linear, Logarithmic, Exponential, Polynomial, Power.
• Statistical Review: Hovering over a trend line displays a tooltip with the regression equation, R-Squared value (how well the model fits the data, 0 to 1), and p-value (statistical significance, where p < 0.05 indicates significance).

4. Forecasting

Forecasting predicts future values using exponential smoothing models.

• Requirements: Requires at least 1 continuous date dimension, 1 measure, and at least 5 periods of historical data. Aggregate Measures must be toggled on.
• Menu Path: Analysis -> Forecast -> Show Forecast or drag Forecast from the Analytics Pane.

5. The Summary Card

• Path: View -> Cards -> Summary (or Analysis -> Summary).
• Function: Displays statistical metrics for the selected marks or entire worksheet, including Count, Sum, Average, Minimum, Maximum, Median, Standard Deviation, and First/Third Quartiles.

---

Analytics Tools Quick Reference