Salt Shaker Press: Graph Analysis

Identify the chart's main title and any subtitles.
Define the primary question this graph is trying to answer.
Summarize the graph's main takeaway in a single sentence.
Identify the source of the data and its collection/publication date.
Determine the intended audience for this visualization (e.g., expert, general public).
Read the caption or any accompanying text for context.
Identify the type of graph being used (e.g., bar, line, scatter).
State the "5 Ws" of the graph: Who, What, Where, When, Why.
Identify the key variables being presented.
Describe the overall purpose of the visualization (e.g., comparison, distribution, relationship, composition).

Describe the primary trend (e.g., increasing, decreasing, cyclical, volatile, stable).
Identify the highest (peak) and lowest (trough) data points on the chart.
Identify any significant outliers or anomalies that deviate from the main pattern.
Describe the distribution of the data (e.g., normal, skewed left/right, bimodal).
Locate the data point for [specific category/time] and state its value.
Identify any periods of rapid growth or sharp decline.
Find the mean, median, and mode (if apparent or calculable).
Identify any gaps, plateaus, or missing data in the series.
Describe the volatility or consistency of the data over time.
Identify any seasonality or cyclical patterns in the data.
Pinpoint the location of the [75th/25th] percentile (if possible).
Describe the density of data points in different regions of the chart.
Analyze the "shape" of the data.
Identify any points of inflection where the trend changes direction.
State the value at the beginning and end of the time series.
Calculate the absolute change between the first and last data point.
Calculate the percentage change from the beginning to the end.
Formulate a hypothesis to explain the most significant outlier.
Identify any "step changes" (sudden, permanent jumps) in the data.
Find the point where the data first crosses a [specific threshold].

Compare the value of [Series A] to [Series B] at a [specific point in time].
Describe the relationship between the two variables on the X and Y axes (e.g., positive correlation, negative correlation, no correlation).
Identify any intersection points where two data series cross.
Analyze the spread or gap between [Series A] and [Series B] over time.
Calculate the percentage difference between [Category A] and [Category B].
Identify which category accounts for the largest/smallest share of the total.
Group the data points into [N] visual clusters.
Describe the "part-to-whole" relationship (composition) shown in the graph.
Find the strongest and weakest correlations shown on the chart.
Compare the current value to a historical average or benchmark.

Identify the top 3 and bottom 3 categories by value.
Check the sorting of the bars (e.g., alphabetical, by value) and assess its effectiveness.
Calculate the approximate sum of all bars.
In a stacked bar chart, analyze the composition of the [tallest/shortest] bar.
In a histogram, identify the modal class (the tallest bin).
Describe the skewness of the distribution based on the histogram's shape.

Describe the strength and direction of the correlation (e.g., strong positive, weak negative).
Visually draw a line of best fit for the data points.
Identify any clusters or groupings of data points.
In a bubble chart, identify the bubble with the largest [size variable].
Analyze the four quadrants (e.g., high-X, high-Y) of the scatter plot.

In a heatmap, identify the "hot spots" (cells with the highest intensity).
In a heatmap, identify the "cold spots" (cells with the lowest intensity).
Analyze the patterns in the rows and columns of a heatmap.
In a treemap, identify the largest and smallest rectangles.
Analyze the hierarchy of a treemap: What is the parent category of [Sub-category A]?

Critique the choice of graph type: Is it the most effective way to show this data?
Identify any misleading elements (e.g., 3D effects, truncated axes, inappropriate scales).
Assess the "data-ink ratio": Is the chart cluttered with non-essential design elements ("chart junk")?
Assess the overall clarity and readability (e.g., font size, label quality).
Propose an alternative graph type that might reveal a different insight.
Suggest an improvement to the title or subtitle to make it more descriptive.
Suggest an improvement to the labels, legend, or color scheme.
Formulate three actionable insights or business decisions based on the graph.
Generate two new questions that are raised by this analysis.
Describe any potential sources of bias in how the data is presented. 9Next. Predict the next data point or future trend based on the pattern.
Write a brief narrative (a "story") that explains the graph from start to finish.
Determine if the graph supports or contradicts a [given hypothesis].
Explain the "so what?" of the graph.
Redesign the graph to emphasize a [different, specific] insight.
Combine the findings from this graph with [another piece of information] to form a new conclusion.

Salt Shaker Press