Year 2 · Notes: Organising, Representing and Interpreting Data – Research, Sampling and Data Analysis
Topic: S.S1 Y2 ORGANISING, REPRESENTING AND INTERPRETING DATA
Census versus sample, sampling methods and bias, planning a small research project, cumulative frequency curves, quartiles and percentiles, box plots, outliers and comparing data sets.
Year 2 · Notes: Organising, Representing and Interpreting Data – Research, Sampling and Data Analysis
Ghana counts every person in a national census only once every ten years, because it is costly and slow. Most of the time, researchers study a smaller group, a sample, and use it to say something about everyone. In Year 2 you plan your own small research project: deciding who to ask, collecting the data fairly, analysing it and explaining what you found.
IN THIS NOTE
Census and sample; the population being studied.
Sampling methods: simple random, systematic, stratified, cluster and convenience.
Bias, and how to avoid it.
The steps of a statistical research project.
Cumulative frequency curves, quartiles and percentiles.
Box plots, outliers and comparing two data sets.
1. Census or sample?
The population is the whole group you want to know about, for example all SHS students in your district. A census collects data from every member of the population: it is complete but expensive and slow. A sample uses part of the population: it is quicker and cheaper, and if chosen well it represents the population fairly.
2. Sampling methods
Method | How it works | Example |
|---|---|---|
Simple random | Every member has an equal chance; use a lottery or random numbers | Draw 30 names from a box of all students |
Systematic | Pick every kth member from a list after a random start | From 1000 names, a random start then every 20th |
Stratified | Split into groups (strata) and sample each in proportion | Sample each form in proportion to its size |
Cluster | Pick whole groups at random and study everyone in them | Choose 3 classes at random and survey all their students |
Convenience | Ask whoever is easy to reach | Ask the first 20 people at the gate (often biased) |
WORKED EXAMPLE · A stratified sample
A school has 1200 students: 500 in SHS 1, 400 in SHS 2 and 300 in SHS 3. A stratified sample of 60 takes \(\tfrac{60}{1200} = \tfrac{1}{20}\) of each form: 25, 20 and 15 students.
WATCH OUT
A sample is biased if some members of the population are more likely to be chosen than others. Asking only students in the library about study habits, or only people with smartphones in an online poll, gives misleading results.
3. A research project, step by step
Question: state clearly what you want to find out, e.g. "Do students who live further from school arrive later?"
Population and sample: decide who to study and how to choose the sample.
Collect: design a short questionnaire or observation sheet, then test it on a few people first.
Organise: put the data in frequency tables (by hand or in a spreadsheet).
Represent and analyse: draw suitable graphs and calculate averages and spread.
Conclude: answer your question in words, and say what limits your conclusion (sample size, possible bias).
4. Cumulative frequency, quartiles and percentiles
The cumulative frequency is a running total of the frequencies. Plot it against the upper class boundaries and join the points with a smooth S-shaped curve, the ogive. From the ogive of \(n\) values you can read the median (at \(\tfrac{n}{2}\)), the lower quartile \(Q_1\) (at \(\tfrac{n}{4}\)), the upper quartile \(Q_3\) (at \(\tfrac{3n}{4}\)) and any percentile (the 90th percentile is at \(\tfrac{90}{100}n\)).
WORKED EXAMPLE · Times taken to reach school
Times in minutes for 50 students: 0–10 (5), 10–20 (12), 20–30 (18), 30–40 (10), 40–50 (5). Cumulative frequencies: 5, 17, 35, 45, 50.
Using the same formula as for the median: median (25th value) \(= 20 + \tfrac{25 - 17}{18}\times 10 \approx 24.4\) min; \(Q_1\) (12.5th) \(= 10 + \tfrac{12.5 - 5}{12}\times 10 = 16.25\) min; \(Q_3\) (37.5th) \(= 30 + \tfrac{37.5 - 35}{10}\times 10 = 32.5\) min. The interquartile range is about \(16.3\) min.
Interpretation: half the students take less than about 24 minutes, and the middle half take between about 16 and 33 minutes.
5. Box plots and outliers
A box plot shows five numbers: the smallest value, \(Q_1\), the median, \(Q_3\) and the largest value. The box covers the middle half of the data; the "whiskers" reach out to the extremes.
A common rule marks a value as an outlier if it is more than \(1.5 \times \text{IQR}\) below \(Q_1\) or above \(Q_3\). If \(Q_1 = 20\) and \(Q_3 = 30\), the IQR is 10, so values below \(20 - 15 = 5\) or above \(30 + 15 = 45\) are outliers. Check outliers: they may be mistakes in recording, or genuinely unusual cases worth explaining.
KEY IDEA
To compare two data sets, compare an average (which is typically higher?) and a spread (which is more consistent?), and always write the comparison in the context of the question.
WORKED EXAMPLE · Comparing two schools
School A's journey times have median 24 min and IQR 16 min; School B's have median 18 min and IQR 6 min. Students at School B usually take less time, and their times are more consistent (smaller IQR).
Check yourself
Additional Mathematics
Senior High
More on ORGANISING, REPRESENTING AND INTERPRETING DATA
Other resources you can find on TudHub.