Cronbach's Alpha (α)
The core engine metric. A statistical coefficient ranging from 0 to 1 that mathematically measures the internal consistency and reliability of a survey scale.
The Analogy:
Think of a choir singing. If all singers (items) hit the right notes together, the harmony is tight (high alpha). If they all sing different tunes, the harmony breaks (low alpha).
Internal Consistency
The degree to which different test items that are intended to probe the exact same underlying construct produce similar or correlated results.
The Analogy:
A rowing team. If everyone paddles in the exact same rhythm and direction, the boat moves powerfully. If they pull out of sync, consistency is lost.
Variance (s²_i)
A statistical measurement of the spread between numbers in a data set. It measures how far each number is from the mean (average) and from every other number.
The Analogy:
Arrows on a target. High variance means arrows are scattered everywhere. Low variance means they are tightly clustered in one spot.
Pearson Correlation (r)
A measure of the linear relationship between two specific items, ranging from -1 to 1. Alpha uses this to check if questions "move" together.
The Analogy:
Two dancers in a duet. If one moves left and the other moves left simultaneously, they have a positive correlation. If they move randomly, correlation drops.
Correlation Matrix (Heatmap)
A visual grid showing the Pearson correlation coefficients between all variables in a survey simultaneously to spot structural flaws.
The Analogy:
A corporate relationship chart. It shows exactly which team members work well together (dark blue) and who severely clashes (red).
Redundancy (r > 0.95)
When two items are virtually identical statistically. This artificially inflates the Alpha score and causes severe respondent survey fatigue.
The Analogy:
Installing two steering wheels in the same car. It doesn't make the car more reliable; it just adds unnecessary weight and confusion.
Inverse Correlation (Red Trap)
When an item measures the exact opposite of the rest of the scale, usually due to a negatively worded question that was not reverse-coded by the analyst.
The Analogy:
One rower paddling backwards while the rest of the team paddles forward, bringing the boat to a grinding halt.
Zero Variance
When every single respondent gives the exact same answer to a question. The calculation involves division by variance, so a zero causes a mathematical crash.
The Analogy:
A rigged thermostat that always reads 72°F regardless of the room. It provides absolutely no useful information about changes in the environment.
CAPA Protocol
Corrective and Preventive Action. A structured methodology used in Lean Six Sigma to investigate and fix systemic issues in the measurement system.
The Analogy:
Taking a car to the mechanic to fix a misfiring cylinder (corrective) and changing the oil to ensure it doesn't happen again (preventive).