Lesson 6 of 6 · 16 min
Reporting a comparison in context
Answer a statistical question about groups with a systematic, concise report that compares centre, spread, shape and outliers in context and ends with a justified conclusion.
What you will be able to do
- Describe the stages of a statistical investigation that compares groups.
- Write a systematic comparison: centre, spread, shape and outliers, then a conclusion.
- Make claims that the statistics support, and state their limits.
Comparisons in this topic answer a statistical question: a question about a group that is answered with data that varies, such as "Do Year 12 students spend less time on their phones than Year 10 students?" The statistics are the evidence; the report is the answer. QCAA asks for that report to be systematic and concise: the same features in the same order, each in a sentence or two, with numbers and context.
The statistical investigation process
Pose a question
Name the variable and the groups: "Is daily phone screen time different for Year 10 and Year 12 students at our school?"
Collect the data
Decide how many people, how they are chosen and how the variable is measured (for example, the screen-time report on each phone for one school day).
Analyse
Draw parallel box plots (or dot plots), test for outliers, and calculate summary statistics for each group.
Interpret and report
Compare the groups, answer the question in context, and note any limitations, such as a small sample.
| Feature | What to write | Example | |
|---|---|---|---|
| Centre | Which group is typically higher, by how much, with the medians (or means) | "The median screen time for Year 12 (3.75 h) was 1.5 h lower than for Year 10 (5.25 h)." | |
| Spread | Which group is more variable, with the IQRs (or standard deviations) and ranges | "Year 10 times were slightly more variable (IQR 2 h compared with 1.5 h)." | |
| Shape | Symmetric or skewed | for each group | "Year 12 times were positively skewed." |
| Outliers | Name each one in context | "One Year 12 student's 9 h was a possible outlier." | |
| Conclusion | Answer the question; note limits | "Year 12 students in the sample typically spent less time on their phones, though the samples were small." |
Daily phone screen time of ten Year 10 and ten Year 12 students. Check each sentence in the table above against the box plots and statistics.
Put the parts of a systematic comparison in the order this lesson recommends.
A weak comparison
"Year 10 has a median of 5.25 and Year 12 has a median of 3.75. Year 10 has an IQR of 2 and Year 12 has an IQR of 1.5. Year 12 has an outlier. So Year 10 is bigger."
- States values but never compares them.
- No units or context ("bigger" what?).
- Doesn't answer the question.
A strong comparison
"The median daily screen time for Year 12 students (3.75 h) was 1.5 h lower than for Year 10 students (5.25 h). Year 10 times were slightly more variable (IQR 2 h compared with 1.5 h). Year 12 times were positively skewed, with one possible outlier of 9 h. So Year 12 students in this sample typically spent less time on their phones, although with only ten students in each group the result may not hold for the whole school."
A different sample of students gave these summary statistics for daily phone screen time, in hours.
| Minimum | Median | Maximum | Possible outliers | |||
|---|---|---|---|---|---|---|
| Year 10 | 2.0 | 3.5 | 4.5 | 5.5 | 7.5 | none |
| Year 12 | 1.0 | 2.0 | 3.0 | 5.0 | 8.0 | none |
Compare the screen times of the two year levels in a short, systematic report.
Worked example
Does travel time depend on how students get to school?
A Year 11 class investigated the question "Does the time it takes to get to school depend on how students travel?" They surveyed 90 students at their Townsville school and drew parallel box plots of the travel times for students who came by car, by bus, and by walking or cycling.
Write a report that answers the question.
Travel time to school by mode of travel.
Step 1: Read the summaries
Min Median Max IQR Outliers Car 5 10 14 18 24 8 none Bus 15 25 32 38 52 13 70 Walk or cycle 4 8 12 20 35 12 none Step 2: Compare centres
The median bus time (32 minutes) was 18 minutes longer than the median car time (14 minutes) and 20 minutes longer than the median walk or cycle time (12 minutes). Car and walk or cycle medians were similar.
Step 3: Compare spreads
Bus times were the most variable (IQR 13 minutes), then walk or cycle (12 minutes); car times were the most consistent (8 minutes).
Step 4: Shape and outliers
Walk or cycle times were positively skewed (median close to , long upper whisker). Car and bus times were roughly symmetric, apart from one bus trip of 70 minutes, a possible outlier (bus upper fence ).
Step 5: Use the overlap
The bus box (25 to 38 minutes) lies entirely above the car box (10 to 18) and the walk or cycle box (8 to 20), so the difference for bus students is clear-cut.
Step 6: Conclude and note limitations
Travel time does depend on how students travel: bus students typically took much longer. The data don't show why; bus students may simply live further away, and the sample is from one school.
"Justify" and "evaluate" marks go to conclusions that stay within what the data show.
Try it yourself first, then check the working one step at a time.
Answer
Bus students typically took much longer (median 32 minutes, compared with 14 for car and 12 for walking or cycling) and their times were the most variable (IQR 13 minutes), with one possible outlier of 70 minutes. Car and walk or cycle times were typically similar, though walk or cycle times were more variable and positively skewed. So travel time does depend on mode of travel, though distance from school may explain much of the difference.
Method
Read and tabulate each five-number summary.
Compare centres, then spreads, quoting differences.
Describe shapes and outliers; use overlap to judge how clear the difference is.
Answer the question and state limitations (cause, sample).
Watch out for
Claiming that catching the bus causes longer trips.
Describing each box plot separately without ever comparing them.
Common mistake
Claims the data can't support
- "All" or "every". A lower median means typically lower, not lower for every person.
- Cause. The data show that the groups differ, not why. "Year 12 students are busier" is a guess, not a finding.
- Beyond the sample. Ten students from one school may not represent all Queensland students. Say so when the sample is small.
- Ignoring overlap. If the boxes overlap a lot, the difference is less convincing.
Exam tip
Systematic and concise
Markers look for each feature compared, with the statistic quoted and interpreted in context. Keep to one or two sentences per feature, in the same order every time: centre, spread, shape and outliers, conclusion. Don't repeat the five-number summaries back; use them to say which group is higher or more variable, and by how much.
Summary
A statistical investigation poses a question, collects data, analyses it and reports the findings in context.
A systematic comparison covers centre, spread, shape and outliers, then a conclusion, in that order.
Each sentence compares the groups with a statistic, its units and what it means in context.
The conclusion answers the question asked, and notes limitations such as a small sample.
Claim only what the data support: "typically", not "always"; difference, not cause.
Finished the lesson?
Try the 2 quick checks above and the lesson completes itself, or mark it complete now.