Algebra I Unit 4: Linear Models & Statistics
This Algebra I unit, Linear Models & Statistics, has 3 lessons that cover TEKS A.4A, A.4B, A.4C. Each lesson below gives the goal, key vocabulary, a worked example and practice questions with answers you can check.
Practice this unit with interactive questions
STAAR connection: Algebra I is a subject tested by the State of Texas Assessments of Academic Readiness (STAAR). Texas lawmakers have voted to replace STAAR with shorter tests beginning in the 2027–28 school year; the tests are still built on the TEKS, so practicing the TEKS stays useful.
TEKS covered in this unit
- A.4A calculate, using technology, the correlation coefficient between two quantitative variables and interpret this quantity as a measure of the strength of the linear association
- A.4B compare and contrast association and causation in real-world problems
- A.4C write, with and without technology, linear functions that provide a reasonable fit to data to estimate solutions and make predictions for real-world problems
Lesson 1: Correlation coefficient
TEKS A.4A
What you will learn
I can use technology to calculate the correlation coefficient r between two quantitative variables and interpret it as the strength and direction of a linear association.
I can describe a relationship as strong, moderate or weak and positive or negative using r.
Key vocabulary
- correlation coefficient (r)
- A number from −1 to 1 that measures how closely data points follow a straight line.
Examples: r = 0.92 (strong positive); r = −0.78 (strong negative) - positive and negative association
- Positive: as x increases, y tends to increase (r > 0). Negative: as x increases, y tends to decrease (r < 0).
Examples: Study time and test score; Hours of TV and exercise time - strength of association
- How tightly points cluster around a line: |r| near 1 is strong, near 0 is weak.
Examples: |r| = 0.95 strong; |r| = 0.15 weak
Worked example
A graphing calculator gives r = −0.88 for a data set of outdoor temperature and heating bill. Interpret r.
- The sign is negative: as temperature rises the heating bill tends to fall.
The sign gives the direction. - |r| = 0.88 is close to 1: a strong linear association.
Close to 1 means tight around a line. - Statement: there is a strong negative linear association between temperature and heating bill.
Say it in the context. - r² = 0.77, so about 77% of the variation in the bill is explained by the line.
Optional deeper meaning.
Strategy: Sign, Size, Sentence
- Sign: positive or negative?
- Size: how close is |r| to 1?
- Write a sentence naming both variables.
Check your understanding
A calculator shows r = 0.92 for hours studied and test score. What does it mean?
- Studying causes 92% of the score
- A strong negative linear association
- A weak positive linear association
- A strong positive linear association
Show the answer
Answer: D. A strong positive linear association
r is positive and close to 1, so strong and positive.
Which correlation coefficient shows the STRONGEST linear association?
- −0.50
- 0.20
- 0.70
- −0.95
Show the answer
Answer: D. −0.95
|−0.95| is the largest, so it is the strongest, with a negative direction.
Which value is NOT possible for a correlation coefficient?
- 0
- −1
- 1.3
- 0.99
Show the answer
Answer: C. 1.3
r always falls between −1 and 1, so 1.3 is impossible.
Lesson 2: Association versus causation
TEKS A.4B
What you will learn
I can compare and contrast association and causation in real-world problems.
I can explain association, causation and lurking variable using evidence.
Key vocabulary
- association
- A relationship in which two variables change together in a pattern.
Examples: Ice cream sales and sunburns; Height and shoe size - causation
- A change in one variable directly produces a change in the other.
Examples: Pressing the gas pedal speeds a car up; Heating water makes it boil - lurking variable
- A hidden third variable that influences both variables and creates a misleading association.
Examples: Hot weather: ice cream sales and swimming accidents; Age: shoe size and reading level
Worked example
A town finds a strong positive correlation between ice cream sales and the number of sunburns. Does ice cream cause sunburns?
- Association: when sales rise, sunburns rise, so r is positive and strong.
The data show a pattern. - Is there a direct mechanism? Eating ice cream does not damage skin.
Ask whether one could produce the other. - Look for a lurking variable: hot, sunny weather increases both ice cream sales and time outdoors.
A third variable explains both. - Conclusion: association, not causation. To show causation you would need a controlled experiment.
State the correct claim.
Strategy: Pattern, Mechanism, Lurker
- Describe the pattern in the data.
- Ask: is there a believable mechanism?
- Look for a lurking variable before claiming cause.
Check your understanding
A study finds students who own more books score higher on reading tests. Which conclusion is best supported?
- There is no relationship
- Higher scores cause students to buy books
- Buying books causes higher scores
- There is an association; more evidence is needed to claim buying books causes higher scores
Show the answer
Answer: D. There is an association; more evidence is needed to claim buying books causes higher scores
The data show association only. Family income or reading habits could be lurking variables.
Which is an example of causation rather than just association?
- People who carry umbrellas see more rain
- Ice cream sales and drownings rise together in summer
- Shoe size and reading ability both rise with age
- Heating water causes it to boil
Show the answer
Answer: D. Heating water causes it to boil
Heat directly raises the temperature of water until it boils. In the other examples a lurking variable explains the pattern.
What is the lurking variable in “fire damage is greater when more firefighters are sent”?
- The number of firefighters
- The size of the fire
- The time of day
- The cost of the damage
Show the answer
Answer: B. The size of the fire
Bigger fires cause both more firefighters and more damage.
Lesson 3: Linear models and predictions
TEKS A.4C
What you will learn
I can write a linear function that fits a data set, with and without technology, and use it to estimate and predict in real-world problems.
I can explain a line of best fit, interpolation, extrapolation and the meaning of its slope and intercept.
Key vocabulary
- line of best fit (trend line)
- A line that follows the pattern of a scatterplot, with about as many points above as below.
Examples: y = 4x + 2 for study hours; The least-squares line - interpolation and extrapolation
- Interpolation predicts within the range of the data; extrapolation predicts beyond it and is less reliable.
Examples: Predict day 5 from days 1–8; Predict year 2050 from data to 2020 - residual
- The vertical distance between an actual data point and the line: actual y − predicted y.
Examples: Actual 30, predicted 26: residual 4; Actual 12, predicted 15: residual −3
Worked example
A scatterplot of study hours x and test score y has points near (2, 10) and (8, 34). Write a linear model and predict the score for 10 hours.
- Slope: (34 − 10) ÷ (8 − 2) = 24 ÷ 6 = 4 points per hour.
Use two points the line passes through. - Intercept: 10 = 4(2) + b, so b = 2. Model: y = 4x + 2.
Substitute one point. - Predict x = 10: y = 4(10) + 2 = 42.
Substitute into the model. - 10 hours is beyond the data (2 to 8), so this is extrapolation and may not be reliable.
Say how far you can trust the prediction.
Strategy: Two Points, Model, Predict, Judge
- Pick two points near the trend and find m and b.
- Use the model to predict.
- Judge: interpolation or extrapolation?
Check your understanding
A trend line passes near (2, 10) and (8, 34). What is its equation?
- y = 2x + 4
- y = 6x − 2
- y = 4x + 10
- y = 4x + 2
Show the answer
Answer: D. y = 4x + 2
m = 24 ÷ 6 = 4 and 10 = 4(2) + b gives b = 2.
The model y = 3.5x + 12 predicts a quiz score from x hours of tutoring. What does 3.5 mean?
- The maximum score is 3.5
- A student with no tutoring scores 3.5
- Each extra hour is predicted to raise the score by 3.5 points
- Tutoring takes 3.5 hours
Show the answer
Answer: C. Each extra hour is predicted to raise the score by 3.5 points
The slope is the predicted change in the score per hour.
Data for x = 1 to 10 gave y = 2x + 5. Predicting y at x = 25 is…
- Interpolation, and very reliable
- Extrapolation, and less reliable
- Extrapolation, and perfectly reliable
- Interpolation, and less reliable
Show the answer
Answer: B. Extrapolation, and less reliable
25 is well outside the data, so it is extrapolation and may not hold.