From One Center to Shared Movement
How do mean, variance, and covariance describe increasingly richer geometry in data?
How this lesson unfolds.
Begin with How does mathematics see a cloud of data? The sequence then connects Begin on one number line and builds toward Read center, spread, and direction together.
How does mathematics see a cloud of data?
Mean locates a center, variance measures one-variable spread around that center, and covariance describes the net direction of paired movement. Each idea adds one layer without replacing the earlier one.
- Learning goalExplain concept map using center, distance, and paired geometry.
- Helpful foundationsarithmetic mean · signed deviations · squared distance · paired observations
- Common trapVariance describes one-variable spread; covariance describes net linear co-movement, and zero covariance does not generally prove independence.
- Next connectionBegin on one number line
More context
The portrait diagram reads from an exact relationship to a visual model, three interpretation cues, and a concise conclusion.
What can one variable tell us by itself?
A single variable places every observation on one axis. From those positions we can ask where the observations balance and how far they spread, but not how two quantities move together.
- Learning goalExplain one variable number line using center, distance, and paired geometry.
- Helpful foundationsarithmetic mean · signed deviations · squared distance · paired observations
- Common trapVariance describes one-variable spread; covariance describes net linear co-movement, and zero covariance does not generally prove independence.
- Next connectionThe mean is the balance point
More context
The portrait diagram reads from an exact relationship to a visual model, three interpretation cues, and a concise conclusion.
Where is the center of repeated zero-one observations?
For A=(1,0,1,1,0), the population mean is 3/5=0.6. It is the balancing location of all five observations, even though 0.6 is not itself an observed value.
For A=(1,0,1,1,0), the population mean is 3/5=0.6.
- Learning goalExplain mean balance point using center, distance, and paired geometry.
- Helpful foundationsarithmetic mean · signed deviations · squared distance · paired observations
- Common trapVariance describes one-variable spread; covariance describes net linear co-movement, and zero covariance does not generally prove independence.
- Next connectionVariance keeps distance, not direction
More context
The portrait diagram reads from an exact relationship to a visual model, three interpretation cues, and a concise conclusion.
How far do the observations sit from the mean?
Subtract the mean, square each deviation, then average. For A=(1,0,1,1,0), the squared deviations sum to 1.20, so the population variance is 1.20/5=0.24.
For A=(1,0,1,1,0), the squared deviations sum to 1.20, so the population variance is 1.20/5=0.24.
- Learning goalExplain variance squared distance using center, distance, and paired geometry.
- Helpful foundationsarithmetic mean · signed deviations · squared distance · paired observations
- Common trapVariance describes one-variable spread; covariance describes net linear co-movement, and zero covariance does not generally prove independence.
- Next connectionA second variable changes the question
More context
The portrait diagram reads from an exact relationship to a visual model, three interpretation cues, and a concise conclusion.
What becomes possible when Bob joins Alice?
With one value from Alice and one from Bob per observation, every record becomes an ordered pair (A,B). We can still measure each variable's center and spread, and now ask how their deviations align.
- Learning goalExplain second variable transition using center, distance, and paired geometry.
- Helpful foundationsarithmetic mean · signed deviations · squared distance · paired observations
- Common trapVariance describes one-variable spread; covariance describes net linear co-movement, and zero covariance does not generally prove independence.
- Next connectionPairs become points in two dimensions
More context
The portrait diagram reads from an exact relationship to a visual model, three interpretation cues, and a concise conclusion.
Where does a paired observation live?
The pair (A,B) is one point on a plane. The vertical and horizontal mean lines meet at the joint center (E[A],E[B]); each point's position relative to that center supplies two signed deviations.
- Learning goalExplain paired coordinate plane using center, distance, and paired geometry.
- Helpful foundationsarithmetic mean · signed deviations · squared distance · paired observations
- Common trapVariance describes one-variable spread; covariance describes net linear co-movement, and zero covariance does not generally prove independence.
- Next connectionVariance and covariance ask different questions
More context
The portrait diagram reads from an exact relationship to a visual model, three interpretation cues, and a concise conclusion.
What is the graphical difference between spread and shared movement?
Variance measures how far one variable lies from its own mean. Covariance multiplies the two deviations for each pair, so its sign records orientation while its magnitude depends on both distances and units.
- Learning goalExplain variance versus covariance using center, distance, and paired geometry.
- Helpful foundationsarithmetic mean · signed deviations · squared distance · paired observations
- Common trapVariance describes one-variable spread; covariance describes net linear co-movement, and zero covariance does not generally prove independence.
- Next connectionSame-side deviations create positive covariance
More context
The portrait diagram reads from an exact relationship to a visual model, three interpretation cues, and a concise conclusion.
What pattern produces an upward lean?
When pairs mostly fall in the lower-left and upper-right quadrants around the joint center, the deviations share a sign. Their products are positive, so covariance is positive.
- Learning goalExplain positive covariance geometry using center, distance, and paired geometry.
- Helpful foundationsarithmetic mean · signed deviations · squared distance · paired observations
- Common trapVariance describes one-variable spread; covariance describes net linear co-movement, and zero covariance does not generally prove independence.
- Next connectionOpposite-side deviations create negative covariance
More context
The portrait diagram reads from an exact relationship to a visual model, three interpretation cues, and a concise conclusion.
What pattern produces a downward lean?
When pairs mostly fall in the upper-left and lower-right quadrants, one deviation is positive and the other negative. Their products are negative, so covariance is negative.
- Learning goalExplain negative covariance geometry using center, distance, and paired geometry.
- Helpful foundationsarithmetic mean · signed deviations · squared distance · paired observations
- Common trapVariance describes one-variable spread; covariance describes net linear co-movement, and zero covariance does not generally prove independence.
- Next connectionZero covariance means zero net linear lean
More context
The portrait diagram reads from an exact relationship to a visual model, three interpretation cues, and a concise conclusion.
What does a balanced scatter cloud tell us?
Positive and negative deviation products can cancel. Covariance zero says the average signed product is zero; it does not by itself prove independence or rule out a nonlinear pattern.
- Learning goalExplain zero covariance geometry using center, distance, and paired geometry.
- Helpful foundationsarithmetic mean · signed deviations · squared distance · paired observations
- Common trapVariance describes one-variable spread; covariance describes net linear co-movement, and zero covariance does not generally prove independence.
- Next connectionAdding variables creates a new total
More context
The portrait diagram reads from an exact relationship to a visual model, three interpretation cues, and a concise conclusion.
What changes when Alice and Bob are combined?
Define the total T=A+B for every paired observation. Its possible values and variability depend not only on the separate behavior of A and B, but also on how their deviations occur together.
Define the total T=A+B for every paired observation.
- Learning goalExplain sum random variable using center, distance, and paired geometry.
- Helpful foundationsarithmetic mean · signed deviations · squared distance · paired observations
- Common trapVariance describes one-variable spread; covariance describes net linear co-movement, and zero covariance does not generally prove independence.
- Next connectionThe total has its own center
More context
The portrait diagram reads from an exact relationship to a visual model, three interpretation cues, and a concise conclusion.
How does the total move around its mean?
Linearity gives E[T]=E[A]+E[B]. Therefore T-E[T]=(A-E[A])+(B-E[B]): the total deviation is exactly the sum of the two individual deviations.
Linearity gives E[T]=E[A]+E[B].Therefore T-E[T]=(A-E[A])+(B-E[B]): the total deviation is exactly the sum of the two individual deviations.
- Learning goalExplain center of sum using center, distance, and paired geometry.
- Helpful foundationsarithmetic mean · signed deviations · squared distance · paired observations
- Common trapVariance describes one-variable spread; covariance describes net linear co-movement, and zero covariance does not generally prove independence.
- Next connectionSquaring the total creates a cross term
More context
The portrait diagram reads from an exact relationship to a visual model, three interpretation cues, and a concise conclusion.
Why does shared movement enter total variance?
Squaring the sum of deviations produces two individual squares and twice their product. That cross-product is the covariance contribution after taking expectations.
- Learning goalExplain squared sum expansion using center, distance, and paired geometry.
- Helpful foundationsarithmetic mean · signed deviations · squared distance · paired observations
- Common trapVariance describes one-variable spread; covariance describes net linear co-movement, and zero covariance does not generally prove independence.
- Next connectionTotal variance includes interaction
More context
The portrait diagram reads from an exact relationship to a visual model, three interpretation cues, and a concise conclusion.
What determines the spread of a sum?
The exact identity is Var(A+B)=Var(A)+Var(B)+2Cov(A,B). Positive covariance enlarges total variance, negative covariance reduces it, and zero covariance leaves only the two individual variance terms.
The exact identity is Var(A+B)=Var(A)+Var(B)+2Cov(A,B).
- Learning goalExplain variance of sum identity using center, distance, and paired geometry.
- Helpful foundationsarithmetic mean · signed deviations · squared distance · paired observations
- Common trapVariance describes one-variable spread; covariance describes net linear co-movement, and zero covariance does not generally prove independence.
- Next connectionRead center, spread, and direction together
More context
The portrait diagram reads from an exact relationship to a visual model, three interpretation cues, and a concise conclusion.
What should you look for first in a data picture?
Locate each mean, inspect each marginal spread, then examine the paired cloud around the joint center. Mean tells where, variance tells how far, and covariance tells the net linear direction of shared movement.
- Learning goalExplain graphical reading checklist using center, distance, and paired geometry.
- Helpful foundationsarithmetic mean · signed deviations · squared distance · paired observations
- Common trapVariance describes one-variable spread; covariance describes net linear co-movement, and zero covariance does not generally prove independence.
- Next connectionConnect covariance to correlation, then regression slope
More context
The portrait diagram reads from an exact relationship to a visual model, three interpretation cues, and a concise conclusion.