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 B — Probability Models and Statistical Inference

Curriculum

  • 4 Sections
  • 20 Lessons
  • 10 Weeks
Expand all sectionsCollapse all sections
  • Unit 1: Continuous Distributions and Sampling Distributions
    5
    • 1.1
      Continuous Random Variables and Density Curves
      50 mins
    • 1.2
      The Normal Distribution
      50 mins
    • 1.3
      When Is a Normal Model Appropriate?
      50 mins
    • 1.4
      Sampling Distributions and the Central Limit Theorem
      50 mins
    • 1.5
      Performance Task — Sampling Distribution Simulation Study
      150 mins
  • Unit 2: Estimating with Confidence
    5
    • 2.1
      Estimating a Proportion with Confidence
      50 mins
    • 2.2
      Confidence Intervals for a Mean
      50 mins
    • 2.3
      What Confidence Means, and What It Does Not
      50 mins
    • 2.4
      Comparing Two Groups with Confidence Intervals
      50 mins
    • 2.5
      Performance Task — Estimation Study
      150 mins
  • Unit 3: Testing Claims: Significance Tests
    5
    • 3.1
      The Logic of a Significance Test
      50 mins
    • 3.2
      t-Tests for Means and Paired Data
      50 mins
    • 3.3
      Errors, Power and Significance in the Real World
      50 mins
    • 3.4
      Two-Sample Tests from Experiments and Observational Data
      50 mins
    • 3.5
      Performance Task — Significance Test Investigation
      150 mins
  • Unit 4: Chi-Square, Regression Inference and the Capstone Study
    5
    • 4.1
      Chi-Square Goodness of Fit
      50 mins
    • 4.2
      Chi-Square Tests for Two-Way Tables
      50 mins
    • 4.3
      Inference for the Slope of a Regression Line
      50 mins
    • 4.4
      Choosing the Right Procedure and Writing It Up
      50 mins
    • 4.5
      Capstone — Full Statistical Inference Study
      200 mins

Comparing Two Groups with Confidence Intervals

Unit 2  ·  Modeling & Reporting Workshop  ·  Lesson 9 of 20

Comparing Two Groups with Confidence Intervals

PS.17RLP-STATSB
Student compares paired weather observations and uncertainty intervals for rainy and dry days.
By the end of this lesson I can…

construct and interpret confidence intervals for a difference in proportions, a difference in means and a mean difference from paired data, and state what the study design allows you to conclude.

Instruction

Most real questions compare groups. The standard error of a difference between two independent estimates is √(SE12 + SE22), because variances of independent quantities add even when you subtract.

1. Difference in proportions. From the St. Louis winter table (NOAA daily data, December-February 2000-2024, each day paired with the day before, 2,256 pairs): after a rainy day, 294 of 688 next days were rainy (p̂1 = 0.427); after a dry day, 396 of 1,568 were (p̂2 = 0.253).

SE = √(0.427 × 0.573/688 + 0.253 × 0.747/1,568) ≈ 0.0218

95% interval: 0.175 ± 1.96(0.0218) = 0.132 to 0.218. After a rainy winter day, the chance of rain the next day is between about 13 and 22 percentage points higher than after a dry day. Condition caveat: consecutive days are not independent random draws, so treat the interval as describing this 25-winter record, and be modest about its precision.

2. Difference in means (two-sample t). St. Louis annual mean temperature (NOAA Global Summary of the Year): 1950-1979, x̄1 = 55.32°F, s1 = 1.36; 1995-2024, x̄2 = 57.89°F, s2 = 1.68; n = 30 each.

SE = √(1.362/30 + 1.682/30) ≈ 0.394

Technology uses a df formula (the Welch approximation) that gives df ≈ 55.5 and t* ≈ 2.004. The interval for μ2 − μ1 is 2.57 ± 2.004(0.394) = 1.78 to 3.36°F. By hand, the conservative choice df = smaller n − 1 = 29 (t* = 2.045) gives 1.77 to 3.38°F. Interpretation and scope: the years are not a random sample from a population. The interval treats each era’s years as draws from the climate conditions of that era. It says the difference between eras is far larger than year-to-year variation would produce. It does not say why: regional warming and the growth of the St. Louis area around the airport are both candidate explanations.

3. Paired data. When each value in one group is matched to one in the other, analyze the differences. St. Louis minus Springfield annual mean temperature, same year, 1995-2024 (n = 30 pairs): mean difference d̄ = 0.810°F, sd = 0.569. SE = 0.569/√30 ≈ 0.104, t* = 2.045 (df = 29), interval 0.598 to 1.022°F. St. Louis averages between about 0.6 and 1.0°F warmer than Springfield in the same year. Pairing by year removes the shared year-to-year swings, such as unusually warm or cool years across the whole state, which is why this interval is so much narrower than a two-sample interval on the same data would be (-0.02 to 1.64°F).

Choosing the procedure. Are the data proportions or means? Are the two groups independent, or paired (same year, same person before and after, twins)? Using a two-sample procedure on paired data wastes the pairing. Using a paired procedure on independent groups is simply wrong.

Experiments (A2.DS.A.2 / A2.DS.A.3). The same intervals compare treatments in a randomized experiment, where random assignment supports a causal reading. For the music experiment in RLP Statistics A (illustrative data), silence minus music was 2.2 words, and a two-sample t-interval gives roughly 0.4 to 4.0 words. Because treatments were randomly assigned, it is an interval for the causal effect on those volunteers.

📖 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.

  • Two Eras and Two Cities: The Annual Temperatures Behind the Comparison IntervalsNOAA NCEI Global Summary of the Year, element TAVG, stations USW00013994 (St. Louis), USW00013995 (Springfield) and USW00013987 (Joplin) for 1950-1979 and 1995-2024; retrieved 2026-11-10 · about 5 min

    The 180 annual values behind the lesson’s three intervals, the Welch degrees of freedom evaluated by hand, what pairing bought against a two-sample interval on the same years, and two more pairings for the workshop.

Vocabulary in context

  • two-sample t-interval — (x̄₁ − x̄₂) ± t*·√(s₁²/n₁ + s₂²/n₂), for the difference between two independent means.
  • paired data — Two measurements linked one-to-one (same unit, same year); analyzed through their differences.
  • Welch approximation — The degrees-of-freedom formula technology uses for two-sample t-procedures.
  • scope statement — A sentence saying to whom a conclusion applies and whether it supports causation.

Formative check

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

+50 XP

St. Louis and Springfield temperatures are measured in the same 30 years. Which procedure fits the question ‘how much warmer is St. Louis?’

Each St. Louis year is matched to the same Springfield year, so analyze the differences.
Fill in the blank

For the St. Louis eras, SE = √(1.36²/30 + 1.68²/30) ≈ , and the 95% interval is about to 3.36°F.

Myth or Fact?

Because the St. Louis interval (1.78 to 3.36°F) is entirely above 0, it shows that regional climate change caused the difference.

The data are observational. The interval shows a difference far larger than year-to-year variation, but not its cause; urban growth around the station is one alternative explanation.

Using NOAA Global Summary of the Year data (the Lesson 2.4 reading has three Missouri stations), construct one paired interval (two stations, same years) and one two-sample interval (one station, two eras) for a variable you choose. For each, state the parameter, check conditions, compute by hand and with technology, interpret, and write a one-sentence scope statement.

0 words
Quick self-check

How confident are you that you can choose and build the right interval to compare two groups, and state its scope?

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. Paired or independent? Annual temperature at two stations in the same 30 years.
    Check your answer
    Answer: Paired (by year).
  2. Paired or independent? Test scores of 30 students taught with method A and 30 different students taught with method B.
    Check your answer
    Answer: Independent groups.
  3. Paired or independent? Each volunteer’s reaction time with and without a notification sound, in random order.
    Check your answer
    Answer: Paired (each volunteer is measured twice).
  4. Two independent sample proportions: p̂₁ = 0.40 (n = 200), p̂₂ = 0.30 (n = 250). Find the SE of the difference and the 95% interval.
    Check your answer
    Answer: SE ≈ 0.0452; interval ≈ (0.011, 0.189).
  5. Two independent samples: x̄₁ = 52, s₁ = 8, n₁ = 40; x̄₂ = 47, s₂ = 10, n₂ = 50. Find the SE of the difference.
    Check your answer
    Answer: ≈ 1.897.
    √(s₁²/n₁ + s₂²/n₂).
  6. Why is a paired interval usually narrower than a two-sample interval on the same data?
    Check your answer
    Answer: Pairing removes variation shared by both members of a pair (such as a wet year at both stations), leaving only the variation in the differences.

Application

Use the skill in context. Show your reasoning.

  1. Springfield winter days (Dec-Feb) 2000-2024 (NOAA): after a rainy day, rain followed on 234 of 584 days; after a dry day, on 353 of 1,672. Construct a 95% interval for the difference and interpret it.
    Check your answer
    Answer: p̂₁ ≈ 0.401, p̂₂ ≈ 0.211; SE ≈ 0.0226; interval ≈ (0.145, 0.234). Rain is between about 15 and 23 percentage points more likely after a rainy winter day. Consecutive days are not independent draws, so treat the precision with caution.
  2. Joplin annual mean temperature (NOAA): 1950-1979 mean 57.39°F, SD 1.39; 1995-2024 mean 58.77°F, SD 1.36 (30 years each). Construct a 95% interval for the difference (recent − early) using technology’s df, and write a scope statement.
    Check your answer
    Answer: Difference 1.38°F; SE ≈ 0.356; df ≈ 58.0; interval ≈ (0.67, 2.09)°F. Scope: an observational station record; the interval shows a difference larger than year-to-year variation, not its cause (regional change and the town around the airport are both candidates).
  3. Springfield minus Kansas City precipitation, paired by year, 1995-2024: mean difference 5.339 in, SD 7.204 in. Construct the 95% paired interval. Then compare its SE with the two-sample SE on the same data (1.874 in) and explain the difference.
    Check your answer
    Answer: Paired interval ≈ (2.65, 8.03) in (SE ≈ 1.315). The two-sample SE (1.874) is about 1.4 times larger, because it treats the shared wet and dry years as noise; pairing removes them.

Challenge

Stretch problems. Expect to think before you write.

  1. In a randomized experiment, the 95% interval for (treatment − control) is (0.4, 4.0) words recalled. In an observational comparison of two eras, the 95% interval is (2.37, 3.68)°F. Write the scope statement for each.
    Check your answer
    Answer: Experiment: random assignment supports a causal effect of the treatment, for the participants (volunteers, so generalizing further is a judgment). Observational: the interval describes a difference between eras larger than year-to-year variation, but gives no causal conclusion.

Review

Keep earlier skills sharp.

  1. A random sample of 15 has mean 42.4 and standard deviation 11.9. Construct a 90% confidence interval for the population mean.
    Check your answer
    Answer: t* ≈ 1.761 (df = 14); margin ≈ 5.41; interval ≈ (36.99, 47.81). We are 90% confident that the population mean is between 36.99 and 47.81.
    Use t, not z, because the population standard deviation is unknown. Conditions: a random sample, less than 10% of the population, and a roughly symmetric sample without strong outliers (or n ≥ 30).
  2. (Lesson 2.3) An ACS 90% margin is ±1.5 points. What is the 95% margin?
    Check your answer
    Answer: About ±1.79 points.

Missouri Learning Standards — Mathematics (data and probability) addressed: A2.DS.A.4, A2.DS.A.2, A2.DS.A.3, A2.DS.A.5, A1.DS.A.8, MP.4

Instructional focus: Two-sample and paired confidence intervals with attention to study design

Reading

Two Eras and Two Cities: The Annual Temperatures Behind the Comparison Intervals

NOAA NCEI Global Summary of the Year, element TAVG, stations USW00013994 (St. Louis), USW00013995 (Springfield) and USW00013987 (Joplin) for 1950-1979 and 1995-2024; retrieved 2026-11-10 · Public domain (works of the United States government), with SLS original worked computation

Why this reading. The lesson’s three worked intervals rest on 120 annual values it never prints: St. Louis in two eras, and St. Louis against Springfield year by year. The workshop then asks you to build one paired and one two-sample interval of your own from the same NOAA product. Both jobs need the numbers. They are here, arranged so the lesson’s figures can be recomputed, and so that other pairings (Joplin’s two eras, St. Louis minus Joplin) can be built from the same table.

Table 1. Annual mean temperature (°F), NOAA Global Summary of the Year

Stations USW00013994 (St. Louis Lambert International Airport), USW00013995 (Springfield-Branson National Airport) and USW00013987 (Joplin Regional Airport). Left block: the earlier era, 1950–1979. Right block: the later era, 1995–2024, with the paired St. Louis minus Springfield difference. Retrieved 2026-11-10; no year is missing.

YearSt. LouisSpringfieldJoplinYearSt. LouisSpringfieldJoplinSt. Louis − Springfield
195054.454.356.0199557.056.258.1+0.8
195154.454.256.0199654.954.357.4+0.6
195257.256.758.8199755.154.457.3+0.7
195358.558.260.3199858.757.959.7+0.8
195458.558.761.2199957.957.659.4+0.3
195557.256.158.3200056.155.958.1+0.2
195657.056.959.0200157.756.659.2+1.1
195756.356.058.0200257.856.258.6+1.6
195854.154.256.4200356.555.858.3+0.7
195956.355.557.7200457.556.758.6+0.8
196054.654.756.3200558.057.859.4+0.2
196153.854.756.8200658.558.860.6−0.3
196254.655.558.1200758.358.159.1+0.2
196354.357.258.1200855.455.157.4+0.3
196456.056.757.6200956.655.256.7+1.4
196556.056.958.9201058.056.558.3+1.5
196654.154.556.2201158.757.059.0+1.7
196754.655.356.8201261.259.561.5+1.7
196854.554.255.8201356.455.456.3+1.0
196954.655.156.9201455.855.356.1+0.5
197055.054.657.1201558.857.858.7+1.0
197156.256.758.4201660.459.460.3+1.0
197254.655.857.8201760.259.160.4+1.1
197355.956.657.7201857.457.658.1−0.2
197454.855.756.5201957.257.457.7−0.2
197555.355.756.8202058.257.258.9+1.0
197654.055.054.9202159.057.759.2+1.3
197755.456.957.4202257.857.158.9+0.7
197853.253.956.4202360.459.160.4+1.3
197954.153.455.5202461.259.761.4+1.5

Summary statistics, for checking

SeriesnMeanSample SD
St. Louis 1950–19793055.3171.358
St. Louis 1995–20243057.8901.681
Springfield 1950–19793055.6631.298
Springfield 1995–20243057.0801.516
Joplin 1950–19793057.3901.394
Joplin 1995–20243058.7701.365

The lesson’s two-sample interval, with the Welch degrees of freedom shown

St. Louis, later minus earlier: 57.89 − 55.32 = 2.573°F. SE = √(1.3582/30 + 1.6812/30) = √(0.06145 + 0.09415) = 0.3945. Technology’s df comes from the Welch formula, which you can evaluate yourself: with v1 = s12/n1 and v2 = s22/n2,

df = (v1 + v2)2 / (v12/(n1 − 1) + v22/(n2 − 1)) = (0.06145 + 0.09415)2 / (0.061452/29 + 0.094152/29) = 55.5.

From the critical-values reading, t* at df = 56 is about 2.003 (technology, at 55.55, gives 2.0036). Interval: 2.573 ± 2.004(0.3945) = 1.78 to 3.36°F. The conservative df = 29 gives t* = 2.045 and 1.77 to 3.38. Welch df always lands between the smaller n − 1 and n1 + n2 − 2; when the two SDs are close it is near the top of that range.

The lesson’s paired interval

The 30 differences in the last column sum to 24.3, so d̄ = 0.8100; sd = 0.5689; SE = 0.5689/√30 = 0.1039; t* = 2.045 (df = 29); interval 0.598 to 1.022°F. 27 of the 30 differences are positive: St. Louis ran warmer than Springfield in 27 of the 30 years (the airport sits in a large urban area, and Springfield’s station is on a higher plateau).

What pairing bought. Treat the same 60 values as two independent samples instead. SE = 0.4133, Welch df 57.4, t* 2.002, interval -0.02 to 1.64°F: the same center and a width of 1.65 against 0.42. Nearly four times wider, and for nothing. Warm years and cool years hit both airports together, and the paired analysis subtracts that shared swing out before it measures the gap.

For the workshop: two more pairings this table supports

Joplin, two eras (two-sample). 58.770 − 57.390 = 1.380; SE = 0.3562; Welch df 58.0. Finish it: the interval, and a scope sentence. Compare its width with St. Louis’s and say why the two stations’ differences are so different in size.

St. Louis minus Joplin, paired, 1995–2024. Compute the 30 differences from the right block. Before you compute an interval, plot them against year: is there a trend? A paired t-interval assumes the differences are draws from one stable process, and if they drift, the interval describes an average of something that is changing. (For reference, the mean of the 30 differences is -0.880°F.)

Whatever you build, the scope statement is the same shape as the lesson’s: these are not random samples of anything; the interval treats the years as draws from the climate of an era at a fixed instrument, and it says how big a difference is relative to year-to-year variation, not what caused it.

Sources: NOAA NCEI, Global Summary of the Year data search (ncei.noaa.gov); NOAA NCEI Access Data Service (the query used here) (ncei.noaa.gov)

Up NextComplete & Continue →

Leave a Reply Cancel reply

What Confidence Means, and What It Does Not
Prev
Performance Task — Estimation Study
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