Describing Data Numerically

Numerical Summaries and the Analytical Framework

  • Curricular Progression and Context:

    • Graphical and tabular summaries provide initial visual structures for categorical and quantitative distributions.
    • Numerical descriptive statistics quantify essential distributional properties, specifically central tendency, variability, and relative position.
    • These univariate measures establish the analytical foundation required to evaluate bivariate quantitative relationships.
  • Primary Divisions of Numerical Description:

    • Measures of Center: Mean, median, and mode.
    • Measures of Variability: Range, variance, and standard deviation.
    • Grouped and Weighted Calculations: Weighted mean.
    • Measures of Relative Position and Outlier Detection: zz-scores, percentiles, quartiles, and the interquartile range (IQR\text{IQR}).
    • Five-Number Summary and Boxplots: Order statistics, exploratory data analysis, and fence calculations.

Measures of Central Tendency

  • The Mean:

    • The mean is the arithmetic average and the most common measure of center.
    • Mathematical Notation:
    • NN denotes the total number of observations in a population.
    • nn denotes the total number of observations in a sample.
    • xx represents an observed variable value.
    • ∑\sum (capital Greek sigma) denotes summation across all observations.
    • Population Mean (μ\mu):
    • Represented by the Greek letter mu (μ\mu):       μ=∑xN\mu = \frac{\sum x}{N}
    • Sample Mean (xˉ\bar{x}):
    • Represented by xx-bar (xˉ\bar{x}):       xˉ=∑xn\bar{x} = \frac{\sum x}{n}
    • Computational procedure:
      1. Calculate the sum of all observed data values (∑x\sum x).
      2. Divide the resulting sum by the total sample size (nn).
  • Physical Interpretation as the Balance Point and Outlier Sensitivity:

    • The mean acts as the physical fulcrum or balance point of a distribution.
    • An extreme outlier heavily shifts the mean toward the tail, whereas the median remains stable.

Dotplot showing how an extreme value pulls the sample mean while the median remains stable

  • Step-by-Step Applications of the Mean:

    • Example 3.1: Calculating the Sample Mean:
    • Data set (hours of nightly sleep reported by students): 5,6,8,10,4,6,95, 6, 8, 10, 4, 6, 9.
    • Sample size: n=7n = 7.
    • Sum of observations:       ∑x=5+6+8+10+4+6+9=48\sum x = 5 + 6 + 8 + 10 + 4 + 6 + 9 = 48
    • Sample mean:       xˉ=487≈6.86 hours\bar{x} = \frac{48}{7} \approx 6.86\,\text{hours}
    • Example 3.2: Determining an Unknown Observation from a Given Mean:
    • Total songs purchased: n=5n = 5.
    • Target mean price: xˉ=$1.23\bar{x} = \$1.23.
    • Price of four individual songs: $1.29\$1.29 each.
    • Total expenditure required:       ∑x=5×1.23=$6.15\sum x = 5 \times 1.23 = \$6.15
    • Cost of the four known songs:       4×1.29=$5.164 \times 1.29 = \$5.16
    • Price of the fifth song (x5x_5):       x5=6.15−5.16=$0.99x_5 = 6.15 - 5.16 = \$0.99
    • Target Mean Exam Problem:
    • An academic course contains 55 exams. The first four percentage scores are 75%75\%, 87%87\%, 45%45\%, and 90%90\%.
    • Required mean for the course: 78%78\%.
    • Total points required:       5×78=3905 \times 78 = 390
    • Cumulative points obtained on exams 1 through 4:       75+87+45+90=29775 + 87 + 45 + 90 = 297
    • Required score on the fifth exam (x5x_5):       x5=390−297=93%x_5 = 390 - 297 = 93\%
  • The Median:

    • The median is the physical midpoint of a data set arranged in ascending numerical order.
    • Exactly 50%50\% of observations lie at or below the median, and 50%50\% lie at or above it.
    • Resistance Property: The median is robust and resistant to extreme values and outliers because its value depends solely on rank order rather than numerical magnitude.
    • Operational Algorithm:
    • Arrange all nn observations in non-decreasing order.
    • If nn is odd, the median is the unique single middle value located at position:       n+12\frac{n + 1}{2}
    • If nn is even, the median is the arithmetic mean of the two middle values located at positions:       n2andn2+1\frac{n}{2}\quad\text{and}\quad\frac{n}{2} + 1
    • Example 3.5: Finding the Median for Odd and Even Sample Sizes:
    • Sample A (Student absences): 3,4,6,7,2,8,93, 4, 6, 7, 2, 8, 9
      • Arranged data (n=7n = 7, odd): 2,3,4,6,7,8,92, 3, 4, 6, 7, 8, 9
      • Position: 7+12=4\frac{7 + 1}{2} = 4
      • Median=6\text{Median} = 6
    • Sample B (Student absences): 5,7,8,1,4,9,8,95, 7, 8, 1, 4, 9, 8, 9
      • Arranged data (n=8n = 8, even): 1,4,5,7,8,8,9,91, 4, 5, 7, 8, 8, 9, 9
      • Middle positions: 44 and 55, corresponding to values 77 and 88.
      • Median calculation:         Median=7+82=7.5\text{Median} = \frac{7 + 8}{2} = 7.5
  • The Mode:

    • The mode is the observation that appears with the highest frequency in a data set.
    • A data set can have a single mode, multiple modes, or no mode.
    • Modal Classifications:
    • No Mode: Occurs when every value appears with identical frequency (e.g., all values occur exactly once).
    • Unimodal: Exactly one value displays the greatest frequency.
    • Bimodal: Exactly two distinct values share the identical maximum frequency.
    • Multimodal: More than two distinct values share the identical maximum frequency.
    • Stating the category "bimodal" does not identify the modes; the specific numerical values must be identified.
    • Example 3.6: Identifying Modes:
    • Distribution A (Phone calls per hour): 6,4,6,1,7,8,7,2,5,76, 4, 6, 1, 7, 8, 7, 2, 5, 7
      • Frequency of 7 is 33; frequency of 6 is 22; all other values appear once.
      • Mode=7\text{Mode} = 7 (Distribution is unimodal).
    • Distribution B (Phone calls per hour): 3,4,7,8,1,6,93, 4, 7, 8, 1, 6, 9
      • Every value occurs exactly once.
      • Result: No mode.
    • Distribution C (Phone calls per hour): 2,2,3,3,4,4,5,52, 2, 3, 3, 4, 4, 5, 5
      • Each value appears with equal frequency (22).
      • Result: No mode (neither unimodal nor bimodal).
  • Comparative Case Study: Music Videos:

    • Counts of music videos by artist:
    • Michael Jackson: 3131
    • Taylor Swift: 2626
    • Usher: 2626
    • Katy Perry: 1515
    • Sample mean (n=4n = 4):     xˉ=31+26+26+154=984=24.5\bar{x} = \frac{31 + 26 + 26 + 15}{4} = \frac{98}{4} = 24.5
    • Sample median:
    • Ordered values: 15,26,26,3115, 26, 26, 31Median=26+262=26\text{Median} = \frac{26 + 26}{2} = 26
    • Sample mode: Value 2626 appears twice.     Mode=26\text{Mode} = 26
  • Worked Example: Retirement Ages:

    • Sample data (n=8n = 8): 84,80,82,77,78,80,79,4284, 80, 82, 77, 78, 80, 79, 42
    • Ordered array: 42,77,78,79,80,80,82,8442, 77, 78, 79, 80, 80, 82, 84
    • Mean calculation:     xˉ=42+77+78+79+80+80+82+848=6028=75.25\bar{x} = \frac{42 + 77 + 78 + 79 + 80 + 80 + 82 + 84}{8} = \frac{602}{8} = 75.25
    • Median calculation:     Median=79+802=79.5\text{Median} = \frac{79 + 80}{2} = 79.5
    • Mode calculation: 8080 appears twice.     Mode=80\text{Mode} = 80
  • Comparative Case Study: Hank Aaron vs. Barry Bonds Home Runs:

    • Hank Aaron Career Data (23 seasons):
    • Data: 10,12,13,20,24,26,27,29,30,32,34,34,38,39,39,40,40,44,44,44,44,45,4710, 12, 13, 20, 24, 26, 27, 29, 30, 32, 34, 34, 38, 39, 39, 40, 40, 44, 44, 44, 44, 45, 47
    • Total home runs: ∑x=755\sum x = 755
    • Mean:       xˉ=75523≈32.83\bar{x} = \frac{755}{23} \approx 32.83
    • Median (12th12^{\text{th}} ordered position):       Median=34\text{Median} = 34
    • Mode: 4444 (appears 44 times).
    • Barry Bonds Career Data (22 seasons):
    • Data: 5,16,19,24,25,25,26,28,33,33,34,34,37,37,40,42,45,45,46,46,49,735, 16, 19, 24, 25, 25, 26, 28, 33, 33, 34, 34, 37, 37, 40, 42, 45, 45, 46, 46, 49, 73
    • Total home runs: ∑x=762\sum x = 762
    • Mean:       xˉ=76222≈34.64\bar{x} = \frac{762}{22} \approx 34.64
    • Median (average of 11th11^{\text{th}} and 12th12^{\text{th}} positions, both 3434):       Median=34\text{Median} = 34
    • Mode: Multimodal with modes 25,33,34,37,45,4625, 33, 34, 37, 45, 46 (each appears twice).
    • Analytical Comparison:
    • Both players exhibit an identical median performance of 3434 home runs.
    • Bonds has a higher mean (34.6434.64 vs. 32.8332.83), influenced by the extreme upper value of 7373.
    • Aaron shows consistency at high productivity, with a single mode of 4444.
    • One cannot definitively conclude superior overall performance without considering variability and outlier effects.

Distribution Shape and Measure Selection

  • Geometric Relationships Between Mean, Median, and Mode:
    • Mode: The horizontal coordinate corresponding to the absolute peak (maximum height) of the density curve.
    • Median: The line that divides the total area under the density curve into two equal halves (50%50\% each).
    • Mean: Pulled away from the peak toward the tail containing extreme values or skewness.

Distributions illustrating the relationships among mean, median, and mode for symmetric and skewed distributions

Left-skewed distribution where mean is less than median which is less than mode

  • Rules for Skewness and Central Tendency:
    • Symmetric Unimodal Distribution:Mean≈Median≈Mode\text{Mean} \approx \text{Median} \approx \text{Mode}
    • Right-Skewed (Positively Skewed) Distribution:Mode<Median<Mean\text{Mode} < \text{Median} < \text{Mean}
    • Left-Skewed (Negatively Skewed) Distribution:Mean<Median<Mode\text{Mean} < \text{Median} < \text{Mode}

Right-skewed density curve with vertical markers A, B, and C

  • Graphical Deduction Example:

    • In a right-skewed unimodal distribution with vertical marker lines labeled AA, BB, and CC:
    • Line AA represents the peak: Mode.
    • Line BB splits the area in half: Median.
    • Line CC is pulled furthest into the long right tail: Mean.
  • Application to Skewed Economic Data:

    • The distribution of individual annual incomes in the United States is strongly skewed to the right.
    • In 2004, the income values for the top 1%1\% were reported as $315,000\$315,000 and $1,259,000\$1,259,000.
    • Because right-skewness pulls the mean above the median:
    • Mean=$1,259,000\text{Mean} = \$1,259,000
    • Median=$315,000\text{Median} = \$315,000
  • Criteria for Selecting the Appropriate Measure of Center:

    • Qualitative (Categorical) Data: The mode is the only valid measure of center.
    • Quantitative Data without Outliers / Approximately Symmetric: The mean is preferred because it incorporates the magnitude of every observation.
    • Quantitative Data with Outliers or Heavy Skewness: The median is preferred because of its resistance to extreme values.
  • Example 3.8: Evaluating Specific Data Scenarios:

    • Women's T-shirt sizes (S, M, L, XL): Categorical/ordinal data; use the Mode.
    • Professional baseball team player salaries: Highly right-skewed quantitative distribution with extreme high outliers; use the Median.
    • Home sales prices in a subdivision of similar tract homes: Homogeneous quantitative data without extreme outliers; use the Mean.
    • Professor evaluation ratings (Best, Average, Worst): Categorical/ordinal ranking; use the Mode.
  • Summary Properties of Central Tendency Measures:

PropertyMeanMedianMode
Common DescriptionArithmetic AverageMiddle ValueMost Frequent Value
Must Belong to Data Set?NoNo (can be average of middle pair)Yes
Number of Possible ValuesExactly one unique valueExactly one unique valueOne, multiple, or none
Sensitivity to OutliersHighly sensitiveHighly resistant (robust)Highly resistant (robust)
Optimal Data DomainQuantitative, symmetric, no outliersQuantitative, skewed, or with outliersQualitative / Categorical

Measures of Variability

  • The Significance of Variation:
    • Measures of center fail to describe the spread of a distribution.
    • Two distinct distributions can have identical means, medians, and modes while possessing completely different dispersion profiles.
    • Case Study: Women's Volleyball Team Heights (Inches):
    • Western Massachusetts University (WMU): 60,70,70,70,7560, 70, 70, 70, 75
      • xˉ=69\bar{x} = 69, Median=70\text{Median} = 70, Mode=70\text{Mode} = 70
    • Northern Connecticut University (NCU): 66,67,70,70,7266, 67, 70, 70, 72
      • xˉ=69\bar{x} = 69, Median=70\text{Median} = 70, Mode=70\text{Mode} = 70

Dotplots comparing the height variability and range of WMU and NCU volleyball teams

  • The Range:

    • The range is the difference between the maximum and minimum observations in a data set:     Range=Largest Value−Smallest Value\text{Range} = \text{Largest Value} - \text{Smallest Value}
    • Calculation for volleyball teams:
    • WMU Range:       75−60=15 inches75 - 60 = 15\,\text{inches}
    • NCU Range:       72−66=6 inches72 - 66 = 6\,\text{inches}
    • Analytical Limitations: The range relies solely on the two most extreme values, ignores all intermediate data points, and is vulnerable to outliers.
  • The Concept of Deviation:

    • A deviation quantifies the directed distance between an individual observation xx and the mean of the data set.
    • Sample Deviation:x−xˉx - \bar{x}
    • Population Deviation:x−μx - \mu
    • Directional Properties:
    • If x>meanx > \text{mean}, the deviation is positive.
    • If x<meanx < \text{mean}, the deviation is negative.
    • If x=meanx = \text{mean}, the deviation equals zero.
    • Fundamental Algebraic Property: The sum of all raw deviations around the mean always equals zero:     ∑(x−xˉ)=0\sum (x - \bar{x}) = 0
    • Raw deviations cannot be averaged directly without squaring or applying absolute values.
  • Population Variance (σ2\sigma^2) and Standard Deviation (σ\sigma):

    • Population Variance (σ2\sigma^2): The arithmetic mean of the squared deviations from the population mean:     σ2=∑(x−μ)2N\sigma^2 = \frac{\sum (x - \mu)^2}{N}
    • Population Standard Deviation (σ\sigma): The positive square root of the population variance, restoring the original unit of measurement:     σ=∑(x−μ)2N\sigma = \sqrt{\frac{\sum (x - \mu)^2}{N}}
    • Represents the typical distance an observation lies from the population mean.
  • Sample Variance (s2s^2) and Sample Standard Deviation (ss):

    • Sample statistics serve as point estimators for population parameters.
    • Dividing sample squared deviations by nn underestimates the population variance.
    • Degrees of Freedom Correction: Dividing by n−1n - 1 corrects for this downward bias.
    • Sample Variance (s2s^2):s2=∑(x−xˉ)2n−1s^2 = \frac{\sum (x - \bar{x})^2}{n - 1}
    • Sample Standard Deviation (ss):s=∑(x−xˉ)2n−1s = \sqrt{\frac{\sum (x - \bar{x})^2}{n - 1}}
    • ss represents the typical difference between an observation and the sample mean.
  • Computational (Shortcut) Formulas:

    • Reduce rounding errors caused by intermediate decimal calculations.
    • Population Parameters:σ2=∑x2−(∑x)2NN\sigma^2 = \frac{\sum x^2 - \frac{(\sum x)^2}{N}}{N}σ=∑x2−(∑x)2NN\sigma = \sqrt{\frac{\sum x^2 - \frac{(\sum x)^2}{N}}{N}}
    • Sample Statistics:s2=∑x2−(∑x)2nn−1s^2 = \frac{\sum x^2 - \frac{(\sum x)^2}{n}}{n - 1}s=∑x2−(∑x)2nn−1s = \sqrt{\frac{\sum x^2 - \frac{(\sum x)^2}{n}}{n - 1}}
  • Worked Example: Dog Shoulder Heights:

    • Heights of five dogs: 600 mm,470 mm,170 mm,430 mm,300 mm600\,\text{mm}, 470\,\text{mm}, 170\,\text{mm}, 430\,\text{mm}, 300\,\text{mm}.
    • Sample size: n=5n = 5.
    • Mean Calculation:xˉ=600+470+170+430+3005=19705=394 mm\bar{x} = \frac{600 + 470 + 170 + 430 + 300}{5} = \frac{1970}{5} = 394\,\text{mm}

Heights of five dogs plotted with their mean height of 394 millimeters

  • Deviations and Sum of Squared Deviations Table:
Value (xx)Deviation (x−xˉx - \bar{x})Squared Deviation ((x−xˉ)2(x - \bar{x})^2)
600 mm600\,\text{mm}600−394=206600 - 394 = 206(206)2=42,436(206)^2 = 42,436
470 mm470\,\text{mm}470−394=76470 - 394 = 76(76)2=5,776(76)^2 = 5,776
170 mm170\,\text{mm}170−394=−224170 - 394 = -224(−224)2=50,176(-224)^2 = 50,176
430 mm430\,\text{mm}430−394=36430 - 394 = 36(36)2=1,296(36)^2 = 1,296
300 mm300\,\text{mm}300−394=−94300 - 394 = -94(−94)2=8,836(-94)^2 = 8,836
Sum (Total)∑(x−xˉ)=0\sum (x - \bar{x}) = 0∑(x−xˉ)2=108,520\sum (x - \bar{x})^2 = 108,520
  • Sample Variance (s2s^2):s2=108,5205−1=108,5204=27,130 mm2s^2 = \frac{108,520}{5 - 1} = \frac{108,520}{4} = 27,130\,\text{mm}^2

  • Sample Standard Deviation (ss):s=27,130≈164.71 mms = \sqrt{27,130} \approx 164.71\,\text{mm}

  • An individual dog's height differs from the mean of 394 mm394\,\text{mm} by an average of 164.71 mm164.71\,\text{mm}.

    • Worked Example: Daily Temperature Variations:
  • Daily recorded temperatures: 12,14,15,15,16,17,18,2112, 14, 15, 15, 16, 17, 18, 21 (n=8n = 8).

  • Sum of values: ∑x=128\sum x = 128

  • Mean:     xˉ=1288=16\bar{x} = \frac{128}{8} = 16

  • Deviations (x−16x - 16): −4,−2,−1,−1,0,1,2,5-4, -2, -1, -1, 0, 1, 2, 5

  • Squared deviations: 16,4,1,1,0,1,4,2516, 4, 1, 1, 0, 1, 4, 25

  • Sum of squared deviations:     ∑(x−xˉ)2=52\sum (x - \bar{x})^2 = 52

  • Sample variance:     s2=528−1=527≈7.43s^2 = \frac{52}{8 - 1} = \frac{52}{7} \approx 7.43

  • Sample standard deviation:     s=527≈2.73s = \sqrt{\frac{52}{7}} \approx 2.73

    • Fundamental Mathematical Properties of Standard Deviation:
  • Measures the spread of data about the mean; must only be used when the mean describes the center.

  • s=0s = 0 if and only if all observations are identical (zero variability). Under all other circumstances, s>0s > 0.

  • As the dispersion around the mean increases, ss increases.

  • Standard deviation shares the vulnerability of the mean to extreme outliers.

    • Practical Interpretations of Standard Deviation:
  • Example 3.13: Stock Price Volatility:

    • Profacto Corporation daily closing price standard deviation: s=$1.02s = \$1.02
    • Yardsmoth Company daily closing price standard deviation: s=$9.67s = \$9.67
    • Interpretation: Profacto stock exhibits far tighter clustering around its average price, indicating lower volatility and risk for long-term investors. Yardsmoth exhibits wider swings, appealing to investors seeking rapid profit through high volatility.
  • Course Exam Section Comparison (Wayne State University STA 1020):

    • Morning section exam score standard deviation: s=3.5 pointss = 3.5\,\text{points}
    • Evening section exam score standard deviation: s=1.4 pointss = 1.4\,\text{points}
    • Interpretation: Scores in the evening section are clustered much more tightly around their mean than morning scores. This statistic does not show which section achieved a higher average score, as variability does not measure performance level.

Grouped Data and the Weighted Mean

  • The Weighted Mean:

    • Used when individual observations differ in relative importance or contribution to the final measure.
    • Mathematical Formula:xˉ=∑(xi⋅wi)∑wi\bar{x} = \frac{\sum (x_i \cdot w_i)}{\sum w_i}
    • xix_i represents the value of observation ii.
    • wiw_i represents the numerical weight assigned to observation ii.
    • Computational Steps:
    1. Multiply each observed value xix_i by its corresponding weight factor wiw_i.
    2. Sum all products: ∑(xi⋅wi)\sum (x_i \cdot w_i).
    3. Divide this sum by the total weight: ∑wi\sum w_i.
  • Step-by-Step Weighted Mean Problems:

    • Walter's U.S. History Course Grade:
    • Grading breakdown: Tests (40%40\%), Homework (20%20\%), Quizzes (10%10\%), Final Exam (30%30\%).
    • Walter's scores: Tests = 8383, Homework = 9898, Quizzes = 9090, Final Exam = 8787.
    • Calculation:       xˉ=(83×0.40)+(98×0.20)+(90×0.10)+(87×0.30)0.40+0.20+0.10+0.30\bar{x} = \frac{(83 \times 0.40) + (98 \times 0.20) + (90 \times 0.10) + (87 \times 0.30)}{0.40 + 0.20 + 0.10 + 0.30}xˉ=33.2+19.6+9.0+26.11.00=87.91.00=87.9%\bar{x} = \frac{33.2 + 19.6 + 9.0 + 26.1}{1.00} = \frac{87.9}{1.00} = 87.9\%
    • Heather's Target Course Grade Problem:
    • Grading breakdown: Tests (35%35\%), Class Assignments (20%20\%), Semester Project (35%35\%), Final Exam (10%10\%).
    • Known scores: Tests = 7878, Class Assignments = 9292, Semester Project = 9393.
    • Target course grade: 90%90\%.
    • Setting up the equation with final exam score xx:       (78×0.35)+(92×0.20)+(93×0.35)+(x×0.10)=90(78 \times 0.35) + (92 \times 0.20) + (93 \times 0.35) + (x \times 0.10) = 9027.30+18.40+32.55+0.10x=9027.30 + 18.40 + 32.55 + 0.10x = 9078.25+0.10x=9078.25 + 0.10x = 900.10x=90−78.25=11.750.10x = 90 - 78.25 = 11.75x=11.750.10=117.5%x = \frac{11.75}{0.10} = 117.5\%
    • Heather must score 117.5%117.5\% on the final exam to attain a final grade of 90%90\%.
    • Current Completed Course Average Problem:
    • Grade components completed:
      • Homework (25%25\% weight): Current average = 78%78\%
      • Quizzes (20%20\% weight): Current average = 68%68\%
      • Participation (5%5\% weight): Current average = 90%90\%
      • Midterm Exams (30%30\% weight): Current average = 71.5%71.5\%
      • Final Exam (20%20\% weight): Uncompleted
    • Cumulative weight of completed course components:       ∑wi=0.25+0.20+0.05+0.30=0.80\sum w_i = 0.25 + 0.20 + 0.05 + 0.30 = 0.80
    • Weighted sum of completed points:       ∑(xi⋅wi)=(78×0.25)+(68×0.20)+(90×0.05)+(71.5×0.30)\sum (x_i \cdot w_i) = (78 \times 0.25) + (68 \times 0.20) + (90 \times 0.05) + (71.5 \times 0.30)∑(xi⋅wi)=19.50+13.60+4.50+21.45=59.05\sum (x_i \cdot w_i) = 19.50 + 13.60 + 4.50 + 21.45 = 59.05
    • Current course grade:       xˉ=59.050.80=73.8125%≈74%\bar{x} = \frac{59.05}{0.80} = 73.8125\% \approx 74\%

Measures of Relative Position and Outlier Detection

  • Standard Scores (zz-Scores):

    • A zz-score indicates the number of standard deviations an individual observation xx falls above or below the mean.
    • Standardizing values allows direct comparison between observations from different distributions with distinct means and standard deviations.
    • Formulas:
    • For a Sample:       z=x−xˉsz = \frac{x - \bar{x}}{s}
    • For a Population:       z=x−μσz = \frac{x - \mu}{\sigma}
    • Sign Conventions:
    • Positive zz-score: Observation lies above the mean (x>meanx > \text{mean}).
    • Negative zz-score: Observation lies below the mean (x<meanx < \text{mean}).
    • Zero zz-score: Observation equals the mean (x=meanx = \text{mean}).
    • Recovering the Raw Value (xx) from a Given zz-Score:
    • Sample:       x=xˉ+z⋅sx = \bar{x} + z \cdot s
    • Population:       x=μ+z⋅σx = \mu + z \cdot \sigma
    • If μ=670\mu = 670, σ=70\sigma = 70, and z=−1z = -1:       x=670+(−1)(70)=600x = 670 + (-1)(70) = 600
  • Step-by-Step Applications of zz-Scores:

    • Example 3.25: Daily Summer Temperature:
    • Average summer daily high: μ=80∘\mu = 80^\circ, with σ=6∘\sigma = 6^\circ.
    • Temperature of interest: x=85.22∘x = 85.22^\circ.
    • Calculation:       z=85.22−806=5.226=0.87z = \frac{85.22 - 80}{6} = \frac{5.22}{6} = 0.87
    • Standardized SAT Performance:
    • Math SAT parameters: μ=500\mu = 500, σ=100\sigma = 100.
    • Jasmine's test score: x=650x = 650.
    • zz-score:       z=650−500100=150100=1.5z = \frac{650 - 500}{100} = \frac{150}{100} = 1.5

Normal curve showing Jasmine's SAT score of 650 with a z-score of 1.5

  • Comparative Performance: Emma vs. Sophia:

    • Emma scored 8888 in a class with xˉ=75\bar{x} = 75 and s=8s = 8:       zEmma=88−758=138=1.625z_{\text{Emma}} = \frac{88 - 75}{8} = \frac{13}{8} = 1.625
    • Sophia scored 8484 in a class with xˉ=70\bar{x} = 70 and s=6s = 6:       zSophia=84−706=146≈2.333z_{\text{Sophia}} = \frac{84 - 70}{6} = \frac{14}{6} \approx 2.333
    • Comparison: Sophia performed better relative to her peer group (z=2.333>1.625z = 2.333 > 1.625).
  • Comparative Performance: Maria vs. Elena (5K Footrace):

    • Maria ran 24 minutes24\,\text{minutes} in a race where xˉ=28 minutes\bar{x} = 28\,\text{minutes} and s=4 minutess = 4\,\text{minutes}:       zMaria=24−284=−44=−1.0z_{\text{Maria}} = \frac{24 - 28}{4} = \frac{-4}{4} = -1.0

    • Elena ran 20 minutes20\,\text{minutes} in a race where xˉ=25 minutes\bar{x} = 25\,\text{minutes} and s=3 minutess = 3\,\text{minutes}:       zElena=20−253=−53≈−1.67z_{\text{Elena}} = \frac{20 - 25}{3} = \frac{-5}{3} \approx -1.67

    • Comparison: In competitive racing, lower completion times indicate superior performance. Elena finished 1.671.67 standard deviations below her race average, outperforming Maria relative to their respective fields.

    • Outlier Detection via the zz-Score Method:

  • An outlier is an extreme data point relative to the rest of the distribution.

  • Standard Outlier Classification Thresholds:

    • Not Unusual: Observations where −2<z<2-2 < z < 2.
    • Moderately Unusual: Observations where −3<z≤−2-3 < z \le -2 or 2≤z<32 \le z < 3.
    • Outlier: Observations where z≤−3z \le -3 or z≥3z \ge 3.
  • Customer Volume Problem:

    • Normal daily volume: μ=150\mu = 150, σ=20\sigma = 20. Observed day: x=195x = 195.
    • Standard score calculation:       z=195−15020=4520=2.25z = \frac{195 - 150}{20} = \frac{45}{20} = 2.25
    • Conclusion: Because 2≤2.25<32 \le 2.25 < 3, serving 195195 customers is classified as moderately unusual.
  • Delivery Route Time Problem:

    • Route duration parameters: μ=60 minutes\mu = 60\,\text{minutes}, σ=8 minutes\sigma = 8\,\text{minutes}. Driver completion time: x=35 minutesx = 35\,\text{minutes}.

    • Standard score calculation:       z=35−608=−258=−3.125≈−3.13z = \frac{35 - 60}{8} = \frac{-25}{8} = -3.125 \approx -3.13

    • Conclusion: Because z≤−3z \le -3, completing the route in 35 minutes35\,\text{minutes} is an outlier.

    • Percentiles and Percentile Ranks:

  • Percentile Definition: Let pp be an integer such that 0<p<1000 < p < 100. The pthp^{\text{th}} percentile (PP) is the value such that at least p%p\% of all data values fall at or below it.

  • Location Algorithm for the pthp^{\text{th}} Percentile:

    • Sort data in ascending order.
    • Compute the position index ii:       i=(p100)⋅ni = \left(\frac{p}{100}\right) \cdot n
    • Rule 1 (Fractional/Decimal Result): If ii contains a decimal, round ii up to the next integer. The observation at that integer rank is the pthp^{\text{th}} percentile.
    • Rule 2 (Integer Result): If ii is an exact integer, the pthp^{\text{th}} percentile is the arithmetic mean of the value at position ii and the value at position i+1i + 1:       x(i)+x(i+1)2\frac{x_{(i)} + x_{(i+1)}}{2}
  • Percentile Rank of an Observed Value:

    • The percentile rank PP corresponding to a specific value xx is calculated as:       P=(Number of values in the data set ≤xn)⋅100P = \left(\frac{\text{Number of values in the data set } \le x}{n}\right) \cdot 100

    • The resulting value PP is rounded to the nearest integer.

    • Quartiles and the Interquartile Range (IQR\text{IQR}):

  • Quartiles divide an ordered distribution into four equal parts, each containing 25%25\% of the data.

    • First Quartile (Q1Q_1): The 25th25^{\text{th}} percentile.
    • Second Quartile (Q2Q_2): The 50th50^{\text{th}} percentile, identical to the Median.
    • Third Quartile (Q3Q_3): The 75th75^{\text{th}} percentile.

Normal curve divided into four equal 25 percent segments by Q1, Q2, and Q3

  • Interquartile Range (IQR\text{IQR}):

    • The difference between the third and first quartiles:       IQR=Q3−Q1\text{IQR} = Q_3 - Q_1

    • Measures the dispersion of the central 50%50\% of the distribution.

    • Highly resistant to extreme values and outliers.

    • Quartile Calculations for Motorist Speeds:

  • Sample A (n=14n = 14): 60,62,63,65,65,67,70,71,71,75,78,79,80,8160, 62, 63, 65, 65, 67, 70, 71, 71, 75, 78, 79, 80, 81

    • Median (Q2Q_2): Average of 7th7^{\text{th}} (7070) and 8th8^{\text{th}} (7171) values:       Q2=70+712=70.5Q_2 = \frac{70 + 71}{2} = 70.5
    • Lower half (7 observations: 60,62,63,65,65,67,7060, 62, 63, 65, 65, 67, 70):       Q1=65Q_1 = 65
    • Upper half (7 observations: 71,71,75,78,79,80,8171, 71, 75, 78, 79, 80, 81):       Q3=78Q_3 = 78
  • Sample B (n=15n = 15): 59,66,67,67,72,74,75,75,75,76,78,79,80,81,8559, 66, 67, 67, 72, 74, 75, 75, 75, 76, 78, 79, 80, 81, 85

    • Median (Q2Q_2): Located at position 15+12=8\frac{15 + 1}{2} = 8, which is 7575.

    • Lower half (7 observations below position 8: 59,66,67,67,72,74,7559, 66, 67, 67, 72, 74, 75):       Q1=67Q_1 = 67

    • Upper half (7 observations above position 8: 75,76,78,79,80,81,8575, 76, 78, 79, 80, 81, 85):       Q3=79Q_3 = 79

    • Quartile and IQR\text{IQR} Analysis for State Exports Data (n=12n = 12):

  • Ordered export values: 1.6,2.7,3.3,3.5,3.5,4.6,4.7,4.8,5.0,5.8,7.5,7.71.6, 2.7, 3.3, 3.5, 3.5, 4.6, 4.7, 4.8, 5.0, 5.8, 7.5, 7.7.

Ordered array of state exports showing positions and values of Q1, Q2, and Q3

  • First quartile (Q1Q_1): Average of positions 3 and 4 (3.33.3 and 3.53.5):     Q1=3.3+3.52=3.4Q_1 = \frac{3.3 + 3.5}{2} = 3.4

  • Median (Q2Q_2): Average of positions 6 and 7 (4.64.6 and 4.74.7):     Q2=Median=4.6+4.72=4.65Q_2 = \text{Median} = \frac{4.6 + 4.7}{2} = 4.65

  • Third quartile (Q3Q_3): Average of positions 9 and 10 (5.05.0 and 5.85.8):     Q3=5.0+5.82=5.4Q_3 = \frac{5.0 + 5.8}{2} = 5.4

  • Interquartile Range (IQR\text{IQR}):     IQR=Q3−Q1=5.4−3.4=2.0\text{IQR} = Q_3 - Q_1 = 5.4 - 3.4 = 2.0

    • Comprehensive Reference Study: Lake Water Pollutant Concentrations:
  • Ordered concentrations (n=40 mg/Ln = 40\,\text{mg/L}):

    • 7.5,7.8,7.9,8.2,8.4,8.4,8.5,8.7,8.8,8.97.5, 7.8, 7.9, 8.2, 8.4, 8.4, 8.5, 8.7, 8.8, 8.9
    • 9.1,9.1,9.3,9.5,9.6,9.7,9.8,10.1,10.2,10.39.1, 9.1, 9.3, 9.5, 9.6, 9.7, 9.8, 10.1, 10.2, 10.3
    • 10.4,10.6,10.6,10.6,10.9,11.1,11.2,11.3,11.4,11.610.4, 10.6, 10.6, 10.6, 10.9, 11.1, 11.2, 11.3, 11.4, 11.6
    • 11.8,11.9,12.1,12.3,12.5,12.5,12.7,13.0,13.0,13.511.8, 11.9, 12.1, 12.3, 12.5, 12.5, 12.7, 13.0, 13.0, 13.5
  • 16th16^{\text{th}} Percentile:

    • i=16100×40=6.4i = \frac{16}{100} \times 40 = 6.4
    • Since ii is fractional, round up to position 77.
    • Observation at position 7:       P16=8.5 mg/LP_{16} = 8.5\,\text{mg/L}
  • 35th35^{\text{th}} Percentile:

    • i=35100×40=14.0i = \frac{35}{100} \times 40 = 14.0
    • Since ii is an integer, average positions 14 and 15 (9.59.5 and 9.69.6):       P35=9.5+9.62=9.55 mg/LP_{35} = \frac{9.5 + 9.6}{2} = 9.55\,\text{mg/L}
  • Quartiles (Q1,Q2,Q3Q_1, Q_2, Q_3):

    • Position for Q1Q_1 (25th25^{\text{th}} percentile): i=0.25×40=10i = 0.25 \times 40 = 10. Average positions 10 and 11 (8.98.9 and 9.19.1):       Q1=8.9+9.12=9.0 mg/LQ_1 = \frac{8.9 + 9.1}{2} = 9.0\,\text{mg/L}
    • Position for Q2Q_2 (50th50^{\text{th}} percentile): i=0.50×40=20i = 0.50 \times 40 = 20. Average positions 20 and 21 (10.310.3 and 10.410.4):       Q2=Median=10.3+10.42=10.35 mg/LQ_2 = \text{Median} = \frac{10.3 + 10.4}{2} = 10.35\,\text{mg/L}
    • Position for Q3Q_3 (75th75^{\text{th}} percentile): i=0.75×40=30i = 0.75 \times 40 = 30. Average positions 30 and 31 (11.611.6 and 11.811.8):       Q3=11.6+11.82=11.7 mg/LQ_3 = \frac{11.6 + 11.8}{2} = 11.7\,\text{mg/L}
  • Percentile Rank of 12.1 mg/L12.1\,\text{mg/L}:

    • Values ≤12.1\le 12.1 span positions 1 through 33 (count = 3333).
    • Rank calculation:       P=(3340)⋅100=82.5%≈83rd percentileP = \left(\frac{33}{40}\right) \cdot 100 = 82.5\% \approx 83^{\text{rd}}\,\text{percentile}
  • Percentile Rank of 10.6 mg/L10.6\,\text{mg/L}:

    • Values ≤10.6\le 10.6 span positions 1 through 24 (count = 2424).
    • Rank calculation:       P=(2440)⋅100=60th percentileP = \left(\frac{24}{40}\right) \cdot 100 = 60^{\text{th}}\,\text{percentile}
  • Percentile Rank of 13.3 mg/L13.3\,\text{mg/L}:

    • Values ≤13.3\le 13.3 span positions 1 through 39 (count = 3939).
    • Rank calculation:       P=(3940)⋅100=97.5%≈98th percentileP = \left(\frac{39}{40}\right) \cdot 100 = 97.5\% \approx 98^{\text{th}}\,\text{percentile}

The Five-Number Summary and Boxplots

  • The Five-Number Summary:

    • A resistant summary of a quantitative distribution consisting of five order statistics:
    1. Minimum (Min): The smallest numerical observation in the data set.
    2. First Quartile (Q1Q_1): The 25th25^{\text{th}} percentile.
    3. Median (Q2Q_2): The 50th50^{\text{th}} percentile.
    4. Third Quartile (Q3Q_3): The 75th75^{\text{th}} percentile.
    5. Maximum (Max): The largest numerical observation in the data set.
  • Detecting Outliers Using the 1.5×IQR1.5 \times \text{IQR} Rule:

    • The 1.5×IQR1.5 \times \text{IQR} rule sets operational fences beyond which values are classified as outliers.
    • Interquartile Range:IQR=Q3−Q1\text{IQR} = Q_3 - Q_1
    • Lower Fence (LF):LF=Q1−1.5⋅IQR\text{LF} = Q_1 - 1.5 \cdot \text{IQR}
    • Upper Fence (UF):UF=Q3+1.5⋅IQR\text{UF} = Q_3 + 1.5 \cdot \text{IQR}
    • Outlier Criterion: Any observation x<LFx < \text{LF} or x>UFx > \text{UF} is an outlier.
  • Step-by-Step Construction of a Boxplot (Box-and-Whisker Plot):

    • Step 1: Construct a horizontal numerical scale spanning the data range, including the calculated fences.
    • Step 2: Plot vertical line segments above the axis corresponding to Q1Q_1, the median (Q2Q_2), and Q3Q_3.
    • Step 3: Connect the segments at Q1Q_1 and Q3Q_3 to form a central rectangle (box) with the median line inside.
    • Step 4: Draw horizontal whiskers extending from the ends of the box:
    • Lower whisker extends from Q1Q_1 down to the smallest data value greater than or equal to the lower fence.
    • Upper whisker extends from Q3Q_3 up to the largest data value less than or equal to the upper fence.
    • Step 5: Plot any individual observations located outside the fences as isolated points (outliers).

Completed box-and-whisker plot displaying the five-number summary and quartile regions

  • Worked Example: Creating a Boxplot from a Five-Number Summary:

    • Five-number summary: Min=12.1\text{Min} = 12.1, Q1=19.8Q_1 = 19.8, Median=23.6\text{Median} = 23.6, Q3=25.3Q_3 = 25.3, Max=35.9\text{Max} = 35.9.
    • Interquartile Range (IQR\text{IQR}):IQR=25.3−19.8=5.5\text{IQR} = 25.3 - 19.8 = 5.5
    • Fence Calculations:
    • 1.5×IQR=1.5×5.5=8.251.5 \times \text{IQR} = 1.5 \times 5.5 = 8.25
    • Lower Fence:       LF=19.8−8.25=11.55\text{LF} = 19.8 - 8.25 = 11.55
    • Upper Fence:       UF=25.3+8.25=33.55\text{UF} = 25.3 + 8.25 = 33.55
    • Outlier Evaluation:
    • The minimum value (12.112.1) exceeds the lower fence (11.5511.55); no lower outlier exists.
    • The maximum value (35.935.9) exceeds the upper fence (33.5533.55); 35.935.9 is an outlier.
    • In a modified boxplot, the upper whisker stops at the largest non-outlier value, and 35.935.9 is marked with an isolated symbol.
  • Worked Example: Outlier Analysis for Lake Water Data (n=40n = 40):

    • Previously calculated values:
    • Q1=9.0 mg/LQ_1 = 9.0\,\text{mg/L}
    • Q3=11.7 mg/LQ_3 = 11.7\,\text{mg/L}
    • IQR=11.7−9.0=2.7 mg/L\text{IQR} = 11.7 - 9.0 = 2.7\,\text{mg/L}
    • Fence Calculations:
    • 1.5×IQR=1.5×2.7=4.051.5 \times \text{IQR} = 1.5 \times 2.7 = 4.05
    • Lower Fence:       LF=9.0−4.05=4.95 mg/L\text{LF} = 9.0 - 4.05 = 4.95\,\text{mg/L}
    • Upper Fence:       UF=11.7+4.05=15.75 mg/L\text{UF} = 11.7 + 4.05 = 15.75\,\text{mg/L}
    • Outlier Assessment:
    • Data range: Min=7.5 mg/L\text{Min} = 7.5\,\text{mg/L}, Max=13.5 mg/L\text{Max} = 13.5\,\text{mg/L}.
    • Since 7.5>4.957.5 > 4.95 and 13.5<15.7513.5 < 15.75, all observations fall within the fences.
    • Result: The lake water pollutant distribution contains no outliers under the 1.5×IQR1.5 \times \text{IQR} criterion.