BENV0139 Chap.2 Summaries, Samples and Visual Distributions
Summaries, Samples and Visual Distributions
Summaries, Samples and Visual Distributions connects numerical summaries to the shapes they describe. Mean and standard deviation are sensitive to every observation and work naturally for roughly symmetric distributions. Median and IQR resist skew, uncertain dates and isolated extremes.
Histograms reveal modality and tail shape but depend on bin width, while boxplots show quartiles, median and flagged extremes without displaying every peak. The chapter also separates accuracy from precision: tightly clustered readings can still be centred on a biased sensor value.
Sampling matters because a precise sample summary may not represent the collection or year if accessible objects, opening hours or stable monitoring periods are overrepresented. Strong descriptive answers therefore name the variable, unit, shape, centre, spread and sampling boundary before interpreting what the values mean for heritage practice.
What this chapter covers
- 01
Mean and median
- 02
Variance and standard deviation
- 03
Quartiles and IQR
- 04
Histograms and bin width
- 05
Boxplots and outlier flags
- 06
Accuracy and precision
- 07
Choosing a robust summary for skewed, uncertain or heterogeneous heritage measurements
Worked example · free
Summarise a skewed dust series
- 2Order the readings and identify the median as 11 mg/m². The marks are original practice weighting.
- 2Calculate Q1 as 9 and Q3 as 12.5 using the halves around the median.
- 2Report IQR as 3.5 mg/m² and compare the mean of 13.33 mg/m².
- 1Explain that the high reading pulls the mean and should be investigated rather than automatically deleted.
Key terms
- Interquartile range
- The upper quartile minus the lower quartile, describing the middle half of ordered values.
- Skew
- Asymmetry in a distribution created by a longer or heavier tail on one side.
- Precision
- The reproducibility or spread of repeated measurements, distinct from closeness to truth.
Summaries, Samples and Visual Distributions FAQ
When should I use median and IQR?
Use them when a distribution is skewed, contains uncertain extremes or has outliers that would pull the mean and standard deviation away from the typical observation. Explain the shape and measurement context rather than invoking robustness as an automatic rule.
Does a boxplot prove a point is erroneous?
No. The usual fence is a display rule based on multiples of the IQR. A flagged point may be a real rare event, an unusual object or a data problem. Check provenance, method and context before correcting or excluding it.
Why are histograms sensitive to bins?
Bin width controls how values are grouped. Wide bins can hide peaks and tails, while narrow bins can make noise look like structure. Compare sensible widths and use a raw-value, boxplot or time display when the conclusion depends on shape.
Exam move
Create a summary decision table with rows for symmetric, skewed, multimodal and bounded data. For each row, name the leading centre, spread and display, then justify the choice in one sentence. Recalculate small datasets by hand so mean, median, sample variance and IQR are not black-box outputs.
Practise rounding the displayed mean to a precision justified by the standard deviation while retaining full precision during calculation. Compare the same data in a histogram, boxplot and time plot; write one feature that each view reveals and one feature it hides. Build heritage examples around estimated production dates, pollutant readings, visitor ratings and condition grades.
In every answer, distinguish an unusual observation from an error and accuracy from precision. Add a sampling sentence that identifies when time, location or accessibility could bias the summary. Finish by translating spread into a domain consequence, such as material heterogeneity or unstable environmental control, rather than stopping with a numerical label. Use a nine-value dataset to practise an audit trail.
Sort the values, mark the median and quartile halves, show the sum for the mean and compute every deviation before using the sample-variance denominator. Repeat after replacing the largest value with an extreme one; record which statistics move and which remain stable. Draw the values with two reasonable histogram widths, a boxplot and a time sequence, then explain how bin choice and ordering alter the visible story.
Compare two stores only after aligning units, observation windows and sampling frequency. Write the sample size beside every group and identify whether winter and summer are represented equally. For an apparent outlier, prepare three explanations: genuine event, recording fault and changed method. State the evidence that would distinguish them before considering deletion.
Finish with a one-minute conclusion containing variable, unit, shape, chosen summary pair, sampling boundary and heritage implication. Reject any conclusion that calls readings accurate merely because they are tightly clustered.
Working through Summaries, Samples and Visual Distributions in BENV0139? Sia is AskSia’s AI Statistics tutor — ask any BENV0139 Summaries, Samples and Visual Distributions question and get a clear, step-by-step explanation grounded in how BENV0139 is taught and assessed. Read this chapter free, then take your hardest questions to Sia.