THE VR SCHOOL
  • Schools & Districts
    • License the academy
    • Free program for your students
    • Request a demo
  • Catalog
    • AI & Data
    • Cybersecurity & Web
    • Programming
    • Health Science
    • All Guilds
    • All courses
  • Families
  • Accreditation
  • Log In
  • Request a demo
  • Home
  • Courses

RLP Statistics A — Exploring Data, Study Design and Probability

Curriculum

  • 4 Sections
  • 20 Lessons
  • 10 Weeks
Expand all sectionsCollapse all sections
  • Unit 1: Exploring One-Variable Data
    5
    • 1.1
      Distributions and the Statistical Process
      50 mins
    • 1.2
      Measuring Center and Spread
      50 mins
    • 1.3
      Comparing Distributions Across Missouri
      50 mins
    • 1.4
      Standardizing, Transforming and Reporting
      50 mins
    • 1.5
      Performance Task — Climate Station Profile
      150 mins
  • Unit 2: Two-Variable Data and Regression
    5
    • 2.1
      Scatterplots and Correlation
      50 mins
    • 2.2
      Least-Squares Regression
      50 mins
    • 2.3
      Residuals, Influence and Causation
      50 mins
    • 2.4
      Exponential Models and the Limits of Extrapolation
      50 mins
    • 2.5
      Performance Task — Bivariate Investigation Report
      150 mins
  • Unit 3: Designing Studies: Samples and Experiments
    5
    • 3.1
      Populations, Samples and Bias
      50 mins
    • 3.2
      Simulating Sampling Variability and the Margin of Error
      50 mins
    • 3.3
      Experiments, Observational Studies and Confounding
      50 mins
    • 3.4
      Comparing Treatments with a Randomization Test
      50 mins
    • 3.5
      Performance Task — Design, Run and Report a Study
      150 mins
  • Unit 4: Probability, Random Variables and the Binomial Model
    5
    • 4.1
      Probability Rules and Two-Way Tables
      50 mins
    • 4.2
      The Multiplication Rule, Trees and Counting
      50 mins
    • 4.3
      Random Variables, Expected Value and Variance
      50 mins
    • 4.4
      The Binomial Model
      50 mins
    • 4.5
      Performance Task — Expected-Value Decision Analysis
      150 mins

Scatterplots and Correlation

Unit 2  ·  Launch & Concept Development  ·  Lesson 6 of 20

Scatterplots and Correlation

A1.DS.A.5RLP-STATSA
Two rain gauges and a blank scatterplot show paired wet years forming a strong upward pattern.
By the end of this lesson I can…

describe the association in a scatterplot by direction, form, strength and unusual points, and compute and interpret the correlation coefficient and its properties.

Instruction

When two quantitative variables are measured on the same cases, a scatterplot shows their relationship. Put the explanatory variable (the one you think helps explain or predict) on the horizontal axis and the response variable on the vertical axis.

Describing a scatterplot. Use four features, in context: direction (positive: as one rises the other tends to rise; negative: the opposite), form (linear, curved, clusters), strength (how tightly points follow the form), and unusual points (outliers far from the pattern).

A real example. Pair each year’s precipitation at St. Louis with the same year’s precipitation at Columbia, 1973-2024 (NOAA Global Summary of the Year; the two airports are about 99 miles apart). Wet years at one station tend to be wet at the other, because the same large weather systems cross both. The association is positive, roughly linear and fairly strong.

The correlation coefficient. Correlation r measures the direction and strength of a linear association. It is the average product of z-scores:

r = (1 / (n − 1)) Σ zx zy

Work it for five years, 2011-2015. St. Louis: 47.20, 32.33, 42.69, 43.45, 61.28 (mean 45.390, SD 10.457). Columbia: 39.70, 30.93, 41.53, 44.16, 49.77 (mean 41.218, SD 6.893).

YearzSTLzCOUproduct
20110.173−0.220−0.038
2012−1.249−1.4931.865
2013−0.2580.045−0.012
2014−0.1860.427−0.079
20151.5201.2411.886

The products sum to 3.622, so r = 3.622 / 4 ≈ 0.91. Two years do almost all the work: 2012 (very dry at both stations) and 2015 (very wet at both) each contribute a product near 1.9, while the other three years contribute almost nothing. With all 52 years, technology gives r = 0.777, and for a different five years, 2019-2023, the same pair gives only r = 0.26. A correlation from five points is unstable, which is why the technology value from all 52 years is the one to report.

Why the formula works. Points in the upper-right and lower-left quadrants (around the point of means) give positive products and push r up. Points in the other two quadrants give negative products. Standardizing removes the units, so r has none.

Properties of r (PS.13).

  • −1 ≤ r ≤ 1. r = ±1 only when all points lie exactly on a line.
  • r has no units, and it does not change if you change units (inches to millimeters) or swap x and y.
  • r measures only linear association. A strong curve can have r near 0.
  • r is not resistant: one outlier can raise or lower it sharply.
  • r describes association, not causation.

Comparing strengths. Other pairs from the same NOAA data: Springfield and Columbia (126 miles apart), r = 0.634; St. Louis and Kansas City (237 miles apart), r = 0.468. Stations farther apart share fewer weather systems. Kansas City precipitation against year gives r = 0.05. There is no linear trend in Kansas City rainfall over these 52 years, even though individual years vary widely, from 22.31 to 55.31 inches.

📖 Read for this lesson

The reading opens over the lesson; close it to come back to exactly where you were. The checks below draw on it.

  • Six Pairs of Stations: Distance Apart and the Correlation of Their RainfallCorrelations of calendar-year precipitation 1973-2024 for every pair of four NOAA GSOY Missouri stations (St. Louis, Kansas City, Springfield, Columbia), with great-circle distances from NOAA’s station coordinates; retrieved November 10, 2026 · about 4 min

    r for all six station pairs with the distance between them, which is itself a six-point scatterplot, and a second five-year hand computation to check your method against.

  • Four Stations, Fifty-Two Years: The Rainfall Table Behind the ComparisonNOAA NCEI Global Summary of the Year, calendar-year precipitation 1973-2024 at St. Louis Lambert, Kansas City International, Springfield and Columbia Regional (USW00013994, USW00003947, USW00013995, USW00003945); retrieved November 10, 2026 · about 5 min

    The 52 paired annual totals for every station, 1973-2024, for the scatterplot and =CORREL the task asks for.

Vocabulary in context

  • explanatory variable — The variable used to explain or predict changes in another; plotted on the horizontal axis.
  • response variable — The variable whose changes we want to explain or predict; plotted on the vertical axis.
  • correlation coefficient (r) — A unitless number from −1 to 1 measuring the direction and strength of a linear association.
  • form — The overall pattern of a scatterplot: linear, curved, or clustered.

Formative check

Work through these before moving on. They are not graded — they tell you whether the standard below has landed yet.

Make a prediction

Annual precipitation at St. Louis paired with the same year’s precipitation at Kansas City (about 237 miles apart). Guess r.

r = 0.468: clearly positive (wet years often reach both cities), but weaker than St. Louis with Columbia (0.777), because distant stations share fewer weather systems.
+50 XP

Every precipitation value in the St. Louis–Columbia data is converted from inches to millimeters (multiply by 25.4). What happens to r = 0.777?

r is computed from z-scores, which do not change under a change of units.
Myth or Fact?

A correlation near 0 means there is no relationship of any kind between the two variables.

r measures only linear association. A strong curved relationship can have r near 0, so always look at the scatterplot.

Choose two other stations from the Lesson 1.3 reading. Make the scatterplot, describe it (direction, form, strength, unusual points), and compute r with a spreadsheet (=CORREL). Then compute r by hand for any five consecutive years using z-scores, and explain why your five-year r differs from the 52-year r. Check yourself against the six-pair table in this lesson’s reading.

0 words
Quick self-check

How confident are you that you can describe a scatterplot and compute and interpret a correlation, including what it cannot tell me?

Not yetVery confident

Practice

Work these on paper or in your notebook, then open Check your answer. Aim for all of Fluency and Application; try at least one Challenge.

Printable version: this unit’s practice workbook (PDF)

Fluency

Build speed and accuracy with the core skill.

  1. Use technology to find the least-squares line and the correlation coefficient for (3, 9), (4, 12), (9, 20), (10, 20), (12, 24).
    Check your answer
    Answer: ŷ = 4.954 + 1.585x; r = 0.993
    Interpret the slope in context: the predicted change in y for each one-unit increase in x.
  2. Use technology to find the least-squares line and the correlation coefficient for (3, 10), (5, 18), (6, 19), (12, 28), (13, 31).
    Check your answer
    Answer: ŷ = 6.728 + 1.855x; r = 0.979
    Interpret the slope in context: the predicted change in y for each one-unit increase in x.
  3. Use technology to find the least-squares line and the correlation coefficient for (4, 7), (5, 10), (6, 8), (8, 12), (9, 13), (14, 24).
    Check your answer
    Answer: ŷ = −0.301 + 1.648x; r = 0.968
    Interpret the slope in context: the predicted change in y for each one-unit increase in x.
  4. A scatterplot shows that as x increases, y tends to decrease, with points close to a straight line. Describe the direction, form and strength.
    Check your answer
    Answer: Negative, linear, strong.
  5. Which value of r shows the strongest linear association: 0.62, −0.91, 0.05 or 0.88?
    Check your answer
    Answer: −0.91.
    Strength is the size of |r|; the sign gives direction.
  6. Every y-value is converted from inches to centimeters. What happens to r?
    Check your answer
    Answer: Nothing: r is unchanged.
    r is built from z-scores, which ignore units.
  7. True or false: if r = 0, there is no relationship of any kind between x and y.
    Check your answer
    Answer: False.
    r measures only linear association; a strong curve can have r = 0.
  8. You swap the x and y variables. What happens to r?
    Check your answer
    Answer: Nothing: r is symmetric in x and y.

Application

Use the skill in context. Show your reasoning.

  1. Precipitation (inches) at Kansas City and Springfield, 2019-2023 (NOAA): Kansas City 53.22, 34.10, 41.26, 34.57, 37.40; Springfield 54.49, 49.71, 46.97, 43.28, 43.08. Compute r by hand from z-scores, then check with technology.
    Check your answer
    Answer: z-products sum to 3.009; r = 3.009/4 ≈ 0.752.
    Kansas City mean 40.110, SD 7.864; Springfield mean 47.506, SD 4.779.
  2. Over all 52 years (1973-2024), r = 0.452 for Kansas City against Springfield precipitation. Interpret it in context and explain why it is positive.
    Check your answer
    Answer: A moderate positive linear association: years that are wet at Kansas City tend to be wet at Springfield.
    The two stations, about 160 miles apart, are reached by many of the same large weather systems, but not all of them, so the association is only moderate.
  3. Kansas City precipitation against year, 1973-2024, has r = 0.050. A student says ‘Kansas City’s rain doesn’t vary.’ Correct the statement.
    Check your answer
    Answer: r near 0 means there is no linear trend over time, not that rainfall does not vary. Annual totals vary widely (from 22.31 to 55.31 inches), just without a steady upward or downward drift.

Challenge

Stretch problems. Expect to think before you write.

  1. Compute r for the points (−3, 9), (−2, 4), (−1, 1), (0, 0), (1, 1), (2, 4), (3, 9). What does the result show?
    Check your answer
    Answer: r = 0.000. The relationship is perfectly determined (y = x²) but not linear, so r misses it entirely.
    The positive and negative z-products cancel by symmetry.

Review

Keep earlier skills sharp.

  1. For the data 3, 15, 15, 20, 21, 31, 32: find the mean, median, range and interquartile range. (Find the quartiles as the medians of the lower and upper halves, leaving out the overall median when the count is odd.)
    Check your answer
    Answer: mean = 19.57; median = 20.0; range = 29; IQR = 16.0 (Q1 = 15.0, Q3 = 31.0)
  2. (Unit 1) St. Louis mean annual temperature, 1950-2024 (NOAA), has mean 56.64°F and SD 1.88°F. The 2012 value was 61.2°F: find its z-score.
    Check your answer
    Answer: z ≈ 2.43.

Missouri Learning Standards — Mathematics (data and probability) addressed: A1.DS.A.5, A1.DS.A.7, A1.DS.A.1, MP.2, MP.7

Instructional focus: Description and measurement of association between quantitative variables

Reading

Six Pairs of Stations: Distance Apart and the Correlation of Their Rainfall

Correlations of calendar-year precipitation 1973-2024 for every pair of four NOAA GSOY Missouri stations (St. Louis, Kansas City, Springfield, Columbia), with great-circle distances from NOAA’s station coordinates; retrieved November 10, 2026 · Public domain (works of the United States government)

Why this reading. The lesson computes r for one pair of stations by hand and quotes it for two more, then asks you to choose two other stations and do the same. All six pairs among the four stations are here, each with the distance between the stations, which turns the six correlations into a small two-variable data set of their own. A second five-year hand computation is worked for a different pair so you can check your method. The raw 52-year table is in the Lesson 1.3 reading.

Table 1. The four stations

Coordinates as NOAA lists them for each station; mean and SD of the 52 calendar-year precipitation totals, 1973-2024 (the years for which all four stations have a complete record).

StationGHCN idLatitudeLongitudeMean PRCP (in)SD (in)
St. Louis (St. Louis Lambert International Airport)USW0001399438.752°N90.373°W41.048.395
Kansas City (Kansas City International Airport)USW0000394739.297°N94.731°W38.728.332
Springfield (Springfield Weather Service Office Airport)USW0001399537.240°N93.390°W45.197.956
Columbia (Columbia Regional Airport)USW0000394538.817°N92.215°W41.328.271

Table 2. Six pairs: distance apart and correlation of annual precipitation

Distances are great-circle distances in miles computed from the coordinates above. r is the correlation of the two stations’ 52 paired annual totals.

PairDistance (mi)r
St. Louis – Columbia990.777
Springfield – Columbia1260.634
Kansas City – Columbia1390.536
Kansas City – Springfield1600.452
St. Louis – Springfield1950.579
St. Louis – Kansas City2370.468

Six points, and a pattern with exceptions: the farther apart two stations are, the weaker the correlation of their rainfall, but not perfectly. The correlation between distance and r across these six pairs is r = -0.72, and the least-squares line is r̂ = 0.851 − 0.00174 × distance, roughly 0.17 less correlation per 100 miles. The pair that sits farthest above the line is St. Louis and Columbia (residual +0.099): only 99 miles apart, they correlate more strongly than the line predicts. The pair farthest below it is Kansas City and Springfield (residual −0.122). Two of the three pairs that include Kansas City sit below the line. With only six points, treat that as a question to investigate, not a finding: one more pair could change the picture.

A second hand computation: Kansas City and Springfield, 2019-2023

Kansas City: mean 40.110, SD 7.864. Springfield: mean 47.506, SD 4.779 (sample SDs, n = 5).

YearKC (in)SGF (in)zKCzSGFproduct
201953.2254.491.6671.4612.435
202034.1049.71−0.7640.461−0.352
202141.2646.970.146−0.112−0.016
202234.5743.28−0.705−0.8840.623
202337.4043.08−0.345−0.9260.319

The products sum to 3.009, so r = 3.009 / 4 = 0.75. Look at the first row: 2019 was very wet at both stations, and its product (2.435) is more than the sum of the other four rows combined. The 52-year correlation for this pair is 0.452. Five years happened to include one extreme year at both stations, which pulls a five-point r upward; a different five years could just as easily give a lower value. The lesson’s pair shows the instability in both directions: for St. Louis and Columbia the five years 2011-2015 give r = 0.91, and the five years 2019-2023 give only r = 0.26, against 0.777 for all 52 years. Small samples give unstable correlations, which is the point of the lesson’s last question.

One station against year

Kansas City’s annual precipitation against year, 1973-2024, gives r = 0.05 and a least-squares slope of 0.274 inches per decade: essentially no linear trend, even though the totals range from 22.31 to 55.31 inches. Use this as the model for reading an r near 0 in your own pairs.

What to do

  1. Choose a pair not worked in the lesson (not St. Louis – Columbia). Make the scatterplot from the Lesson 1.3 table, describe direction, form, strength and unusual points, and compute r with =CORREL. Check it against Table 2.
  2. Compute r by hand for any five consecutive years of your pair using z-scores, as above, and explain why it differs from the 52-year value.
  3. Plot r against distance for the six pairs. Describe the association, and name the pair with the largest residual from the line above.

Sources: NOAA National Centers for Environmental Information, Global Summary of the Year (GSOY), stations USW00013994, USW00003947, USW00013995, USW00003945 (ncei.noaa.gov); NOAA NCEI Access Data Service query used for the four stations (PRCP, units=standard) (ncei.noaa.gov)

Reading

Four Stations, Fifty-Two Years: The Rainfall Table Behind the Comparison

NOAA NCEI Global Summary of the Year, calendar-year precipitation 1973-2024 at St. Louis Lambert, Kansas City International, Springfield and Columbia Regional (USW00013994, USW00003947, USW00013995, USW00003945); retrieved November 10, 2026 · Public domain (works of the United States government)

Why this reading. A discourse lab needs the data behind the claims, not just the summary table. Here are all 208 numbers: 52 calendar-year precipitation totals at each of the four stations the lesson compares, in one table you can paste into CODAP or a spreadsheet. Below them are the outlier checks at each station by the 1.5 × IQR rule, and three claims about the table that are worth arguing over. The summary statistics in the lesson all reproduce from this table.

The data

NOAA Global Summary of the Year, calendar-year precipitation in inches, 1973-2024. Stations: St. Louis Lambert International Airport (USW00013994), Kansas City International Airport (USW00003947), Springfield Weather Service Office Airport (USW00013995) and Columbia Regional Airport (USW00003945). The window starts in 1973 because that is the first year with a complete annual value at all four stations (Kansas City’s record in this dataset begins in 1972 with no annual value). No year in the window is missing at any station. Retrieved November 10, 2026.

YearSTLKCSGFCOU
197339.8555.3159.4350.52
197436.8636.1549.8843.97
197540.2034.1053.9544.14
197623.4823.6931.3023.91
197743.4449.7838.9536.55
197837.7434.0043.9537.93
197929.5031.7848.9732.11
198027.5031.8127.3723.68
198145.5642.1043.7248.54
198254.9947.2347.6948.11
198344.8332.9045.0746.30
198451.6838.8145.8150.72
198550.7552.7556.5055.41
198634.9039.4640.2141.68
198738.4334.0055.5133.58
198833.9524.2648.5030.54
198928.6137.6531.5234.67
199045.1140.6263.2053.63
199133.5028.7337.5934.94
199233.5050.6748.0734.39
199354.7951.5055.8062.51
199434.7228.3549.0438.28
199541.6634.6541.8748.27
199643.7040.4644.8940.02
199731.2433.1038.5037.20
199843.6449.5748.4944.70
199934.0740.1141.5429.81
200037.3935.0135.3939.30
200135.3353.5145.3240.54
200240.9824.8037.8742.33
200346.0927.9842.6339.45
200442.3037.6043.2545.98
200537.8744.1535.3341.24
200629.9430.9038.9030.15
200730.5733.0544.3032.96
200857.9644.6960.1556.81
200950.9444.9651.8850.47
201039.1241.9246.1846.00
201147.2036.9440.6239.70
201232.3322.3130.9730.93
201342.6934.5051.8941.53
201443.4540.1038.6344.16
201561.2846.6159.7649.77
201641.4648.6736.2640.09
201736.6946.0447.5637.65
201842.6240.7844.8934.80
201954.0253.2254.4945.01
202050.0234.1049.7144.74
202142.5141.2646.9753.26
202248.9634.5743.2834.09
202333.1837.4043.0834.64
202450.8434.6743.2147.05

Outlier checks, station by station

Quartiles by the median-of-halves method, as in the lesson’s table.

StationQ1Q3IQRLower fenceUpper fenceOutliersSD / mean
St. Louis (Lambert)34.4045.8311.4317.2562.97none0.205
Kansas City (International)33.5544.8311.2816.6461.74none0.215
Springfield39.5849.389.8024.8964.07none0.176
Columbia (Regional)34.7446.6811.9416.8364.59none0.200

Read the outlier column across: no station has a value outside its fences in these 52 years. The wettest and driest years are different for different stations:

StationWettest year (in)Driest year (in)
St. Louis (Lambert)2015: 61.281976: 23.48
Kansas City (International)1973: 55.312012: 22.31
Springfield1990: 63.201980: 27.37
Columbia (Regional)1993: 62.511980: 23.68

No single year was the wettest at two stations, and the driest year (1980) was shared only by Springfield and Columbia. St. Louis’s 61.28 inches in 2015 is flagged as an outlier in the 75-year record of Lesson 1.2, whose upper fence is 60.47, but in this 52-year window the St. Louis upper fence is 62.97, so the same value is not flagged. The comparison group matters. How closely the four stations move together year by year is a question for Unit 2.

Three claims to argue about

Claim A. “Kansas City’s rainfall is steadier than St. Louis’s. Its standard deviation is 8.33 inches against St. Louis’s 8.40.” Check the last column of the outlier table before you agree. The two SDs differ by 0.07 inch, and Kansas City also has less rain. Decide whether “steadier” should be measured in inches or relative to the usual amount, and whether that tiny gap in inches means anything.

Claim B. “A typical St. Louis year has about 41 inches of rain.” The mean is 41.04 and the median is 41.22. But count the STL column: only 24 of the 52 years fall within 5 inches of the median, and the middle half of years runs from 34.40 to 45.83 inches. Is “about 41 inches” a fair description of a typical year, or only of the long-run average? Those are different questions, and each statistic answers one of them.

Claim C. “Springfield is wetter than St. Louis. Its mean is 45.19 inches and St. Louis’s is 41.04.” Compare the medians (44.89 against 41.22), the IQRs (9.80 against 11.43) and the year-by-year counts. Springfield out-rained St. Louis in 35 of the 52 years and St. Louis out-rained Springfield in 17. The biggest Springfield lead was 19.58 inches in 1973; the biggest St. Louis lead was 7.63 inches in 2024. Decide whether “wetter” describes the typical year, every year, or the average, and whether the claim survives.

What to do

  1. Build stacked box plots for the four stations on one scale. Check each median and IQR against the lesson’s table.
  2. Pick one of the three claims, or write your own about two stations, and support or challenge it with at least two statistics and one display. Name the statistic, give the numbers with units, and say what they are being compared with.
  3. Write a reply to the lesson’s two sample claims. Qualifying a claim (“true for the median, but…”) counts as much as challenging it. You can read your replies aloud to a parent or guardian if you like.

Sources: NOAA NCEI Global Summary of the Year (GSOY), station USW00013994 (St. Louis Lambert International Airport), data service query used here (ncei.noaa.gov); NOAA NCEI Global Summary of the Year (GSOY), station USW00003947 (Kansas City International Airport), data service query used here (ncei.noaa.gov); NOAA NCEI Global Summary of the Year (GSOY), station USW00013995 (Springfield Weather Service Office Airport), data service query used here (ncei.noaa.gov); NOAA NCEI Global Summary of the Year (GSOY), station USW00003945 (Columbia Regional Airport), data service query used here (ncei.noaa.gov)

Up NextComplete & Continue →

Leave a Reply Cancel reply

Performance Task — Climate Station Profile
Prev
Least-Squares Regression
Next

© 2026 The VR School | Career content by Wiley | Powered by Sejal Learning Systems.

Join free · Accreditation · Privacy Policy · Terms of Service

Modal title

Main Content