Data that is measured on a continuous, infinite scale. It represents "how much" of something. The Histogram groups these infinite decimals into chunks called "Bins".
EXAMPLE: The exact fill volume of a soda bottle (330.12ml, 329.84ml). We bucket them into ranges like [329.00 - 330.00] to see the distribution shape.
Data that is counted in distinct, separate, un-mergeable categories. It answers "how many" items fall under a specific label.
EXAMPLE: The number of defective fans produced by Factory A vs Factory B. A fan is either Factory A or B; there are no continuous "in-between" decimals.
In a healthy, normal process, 95% of data falls within ±2 Sigma, and 99.7% falls within ±3 Sigma.
• Warning Zone (Yellow): Items strictly between 2σ and 3σ. The process is drifting but hasn't failed.
• Critical Zone (Red): Items beyond 3σ. This is a statistical anomaly requiring immediate investigation.
HOW TO SPOT IT ON THE CURVE (ANALOGY): Imagine a highway. The Mean is the center dotted line. The 2-Sigma Limits are the outer painted lanes. The 3-Sigma Limits are the concrete guardrails.
If you look at the Bell Curve overlay on the Histogram, driving on the outer painted lines (Yellow Warning) means you are drifting out of the main "hump" of the curve. Hitting the concrete guardrail (Red Alert) means a data bar has completely detached and is sitting far outside the visible bell shape. A crash has occurred!
The mathematical center or balancing point of your dataset. Calculated by adding all values and dividing by the count.
ANALOGY: Think of it as the balancing point (fulcrum) on a seesaw. If you have heavy weights on one side, you need to shift the center to keep it balanced.
A measure of how "spread out" or volatile your data is around the Mean. A lower number indicates high consistency.
ANALOGY: Throwing darts. If all darts hit the bullseye, standard deviation is near zero. If they are scattered randomly across the board, it is very high.
Measures the asymmetry of your Bell Curve.
• 0 is perfectly balanced.
• > 0 (Positive) means the tail pulls to the right.
• < 0 (Negative) means the tail pulls to the left.
ANALOGY: A slide at a playground. If it's steep on the left and has a long, slow slope to the right, that is "Positive Skew". It usually means a physical barrier is preventing data from dropping below a certain limit (like 0 seconds).
Measures the "peakedness" and the heaviness of the tails (outliers).
• High Kurtosis = Sharp peak, fat tails (more extreme outliers).
• Low Kurtosis = Flat, plateau-like curve.
ANALOGY: Comparing buildings. High kurtosis is a tall, incredibly sharp skyscraper (90% of your data is perfect, but the remaining 10% are extreme, dangerous outliers). Low kurtosis is a flat, sprawling warehouse (very predictable, but mediocre across the board).
The absolute Upper and Lower boundaries of acceptable process variation. In standard Six Sigma, these are automatically set at exactly ±3 Standard Deviations from the Mean, but you can override them with custom boundaries.