Math 1153 - 29 September 2026 - Sections 5.1, 5.2

Overview

Mike Jacobsen finishes Section 5.1 by using z-scores to interpret an unusually short-lived light bulb, recover an unknown lifetime, and compare performance across tests with different scales. He introduces Section 5.2’s transformation rules: adding a constant shifts measures of position but leaves spread unchanged, while multiplying by a constant rescales both; examples include converting Fahrenheit statistics to Celsius and setting up an ACT-to-SAT conversion.

Key takeaways

A z-score standardizes an observation relative to its own distribution, allowing meaningful comparisons across tests with different score scales; in the example, z = 1.875 on Test 2 beats z = 1.5 on Test 1.
To recover a raw value from a z-score, rearrange z = (y − ȳ) / s; with z = 2.18, ȳ = 885.33, and s = 69.68, the light bulb’s estimated runtime is 1,037.2 hours.
Adding or subtracting a constant from every observation shifts the mean, median, quartiles, minimum, and maximum by that constant but leaves standard deviation, IQR, and range unchanged.
Multiplying all observations by a factor rescales both position and spread statistics; multiplying the pine-needle data by 10 changes the mean from 9.59 to 95.9 and the standard deviation from 1.59 to 15.9.
For Fahrenheit-to-Celsius conversion, C = (5/9)F − 160/9 applies in full to position measures, but spread measures such as standard deviation are multiplied only by 5/9.
When calculating a z-score in one calculator entry, parentheses around the full numerator are essential: (45 − 30) / 8 = 1.875, not 45 − 30 / 8 = 41.25.

Chapters

0:00 Z-Scores Measure Distance from the Mean

Mike Jacobsen reviews the formula z = (y − ȳ) / s: subtract the mean from a data value and divide by the sample standard deviation.
A negative z-score means the observation is below the mean; a positive score means it is above.
The light bulb with a runtime of 693 hours has z = −2.7, an unusual observation that merits attention.

6:30 Recovering a Light Bulb’s Lifetime from z = 2.18

A z-score of 2.18 directly means the light bulb is 2.18 standard deviations above the mean.
To recover its runtime, Jacobsen substitutes ȳ = 885.33 and s = 69.68 into 2.18 = (y − 885.33) / 69.68.
Solving gives an estimated lifetime of 1,037.2 hours; showing the substituted equation can earn partial credit on an exam.

11:30 Comparing Scores on Two Different Aptitude Tests

The example gives Test 1 a mean of 475 and standard deviation of 100, and Test 2 a mean of 30 and standard deviation of 8.
A student scores 625 on Test 1 and 45 on Test 2, but the raw scores cannot be compared directly because the tests use different scales.
Jacobsen uses z-scores to compare each result relative to its test’s mean and standard deviation.

14:10 Organizing Test Data Before Applying the z-Score Formula

Jacobsen recommends separating the information into Test 1 and Test 2 columns before calculating.
The word “respectively” pairs each test’s first listed statistic with the mean and its second with the standard deviation.
The organized inputs are (y, ȳ, s) = (625, 475, 100) for Test 1 and (45, 30, 8) for Test 2.

20:00 Calculator Parentheses Prevent a False Test Comparison

Test 1’s score has z = (625 − 475) / 100 = 1.5; Test 2’s has z = (45 − 30) / 8 = 1.875.
Entering 45 − 30 / 8 without parentheses gives 41.25, an implausible result that signals a calculator order-of-operations error.
Because 1.875 is larger than 1.5, the student performed better relative to other test takers on Test 2.

24:30 Section 5.2: Distinguishing Position from Spread

Jacobsen introduces the rule that adding or subtracting a constant changes measures of position but not measures of spread.
Measures of position include the sample mean, median, first and third quartiles, minimum, and maximum.
Measures of spread include sample standard deviation, interquartile range, and range; standard deviation is used for roughly symmetric data, while IQR is useful for skewed data.

32:00 How Shifting and Rescaling Affect Statistical Summaries

Shifting means adding or subtracting the same constant from every data value: position measures shift by that constant, while spread measures remain unchanged.
Rescaling means multiplying or dividing every data value by the same factor; the position and spread measures scale with the data.
The key distinction is that a shift moves a distribution without changing its spread, whereas rescaling changes its spread.

35:30 Aleppo Pine Needles Illustrate Data Transformations

The example uses a random sample of 15 Aleppo pine needles from trees in Southern California, measured in centimeters.
The histogram is described as slightly right-skewed, with sample mean 9.59 cm and sample standard deviation 1.59 cm.
Jacobsen uses the sample to visualize how shifting and rescaling affect summary statistics.

37:30 Subtracting 5 cm Moves the Mean but Preserves Spread

Subtracting 5 from every pine-needle measurement shifts the histogram left; for example, an observation of 9 cm becomes 4 cm.
The mean changes from 9.59 cm to 4.59 cm, while the sample standard deviation stays 1.59 cm.
The interquartile range and range also remain unchanged because shifting does not alter distances between observations.

43:00 Multiplying Pine-Needle Lengths by 10 Rescales Statistics

Multiplying each observation by 10 changes a 7.2 cm measurement to 72 cm and expands the histogram’s horizontal scale.
The mean increases from 9.59 to 95.9, and the standard deviation from 1.59 to 15.9.
Unlike shifting, rescaling affects measures of both position and spread.

45:10 Converting Body-Temperature Statistics from Fahrenheit

A study of 130 healthy men and women reports a mean temperature of 98.25°F and asks for summary statistics in Celsius.
The conversion formula is C = (5/9)F − 160/9, combining a rescaling by 5/9 with a shift of −160/9.
Jacobsen first classifies the mean, minimum, and maximum as position measures and standard deviation as a spread measure.

50:00 Apply the Full Fahrenheit Formula to Position Measures

For the mean, both parts of C = (5/9)F − 160/9 apply, giving 36.8°C from 98.25°F.
The factor 5/9 rescales statistics, while −160/9 shifts position measures.
For standard deviation, only the rescaling factor applies: multiply the Fahrenheit spread by 5/9, yielding about 0.4°C; omit the shift.

58:40 Convert Minimum and Maximum Without Shifting Spread

Jacobsen explains that applying the shift to standard deviation would produce an impossible negative spread, confirming that the shift must be omitted.
The minimum of 96.3°F converts using the full formula to 35.7°C.
The worked maximum calculation uses 100.8°F and produces 38.2°C; position measures use both rescaling and shifting.

1:04:40 Set Up an ACT-to-SAT Transformation for Next Class

The example notes that SAT scores reach 1,600 while ACT composite scores reach 36, making raw-score comparisons difficult.
The proposed conversion is SAT-equivalent score = 40 × ACT score + 150: multiplication rescales, and addition shifts.
Jacobsen begins setting up how to transform an admissions sample’s minimum, mean, standard deviation, quartiles, median, and IQR; the calculations continue in the next class.

Summary, takeaways, and chapters were generated by AI from the video's transcript and may contain errors. The video belongs to its creator, Mike Jacobsen.