Welcome to Reasoning with Data: Data Collection, Population Sampling, and Representation
Have you ever wondered how scientists predict weather patterns, how streaming platforms recommend songs, or how researchers estimate wildlife populations? It all begins with one foundational skill: gathering and organizing data.
In this chapter, we explore how data is defined, how to select a fair sample from a large population, and how to represent raw information using tables, frequency charts, and diagrams. Don't worry if mathematical investigations sometimes seem daunting—we will take this step-by-step with clear real-world examples.
Note: Later chapters in this section cover summary statistics (like mean, median, range, and standard deviation), cumulative frequency, box plots, and probability. Here, our focus is strictly on data collection, sampling methods, and fundamental representations.
---1. Understanding Data: The Building Blocks
Before we can collect or display information, we need to know what kind of data we are handling. Data can be classified based on whether it is descriptive or numerical, and whether it can take any value or only specific values.
Qualitative vs. Quantitative Data
Qualitative (Categorical) Data: Describes qualities, characteristics, or categories. It is non-numerical.
Examples: Favorite subject, eye color, type of transport used to travel to school.
Quantitative (Numerical) Data: Involves quantities that can be counted or measured using numbers.
Examples: Number of books in a backpack, running time for a \(100\text{ m}\) sprint, outdoor temperature.
Discrete vs. Continuous Data
Quantitative data is further split into two distinct types:
Discrete Data: Can only take specific, distinct values (often whole numbers obtained by counting). You cannot have values in between.
Examples: The number of siblings a student has (\(0, 1, 2, 3\)), or goals scored in a match. You cannot score \(2.4\) goals!
Continuous Data: Can take any numerical value within a given interval or range, typically obtained by measuring. The precision depends on the measuring instrument.
Examples: Height (\(162.5\text{ cm}\)), mass (\(54.28\text{ kg}\)), or time taken to complete a test (\(42.15\text{ minutes}\)).
Primary vs. Secondary Data
Primary Data: Data collected directly by you (first-hand) for your own specific investigation.
Examples: Conducting your own survey in the school cafeteria, carrying out a science experiment, or counting cars passing through an intersection.
Secondary Data: Data that has already been gathered, published, or recorded by someone else.
Examples: Using census records from a government website, reading climate figures from a published scientific paper, or consulting online sports databases.
Quick Review & Key Takeaway:
Ask yourself two simple questions:
1. "Can I measure it with infinite precision?" If yes, it is continuous. If you count it in steps, it is discrete.
2. "Did I collect it myself?" If yes, it is primary. If someone else compiled it, it is secondary.
2. Populations and Sampling
Imagine you are a chef making a giant pot of vegetable soup. Do you have to drink the entire pot to know if it needs salt? Of course not! You stir the pot thoroughly and taste a single spoonful. That single spoonful is your sample, and the giant pot of soup is your population.
Key Definitions
Population: The entire group of individuals, objects, or events that you want to draw conclusions about. A population does not have to mean an entire country; it could be "all students in Year 9 at our school."
Census: A study or survey that collects data from every single member of the entire population.
Advantages: Completely accurate; no sampling error.
Disadvantages: Time-consuming, costly, difficult to manage for large populations, and sometimes practically impossible.
Sample: A subset or smaller group selected from the population to represent the whole.
Advantages: Fast, inexpensive, practical, and easier to organize.
Disadvantages: May not capture the entire variety of the population if not chosen carefully.
Representative Samples and Bias
A sample must be representative: it should reflect the key characteristics, diversity, and proportions of the wider population.
If a sample overrepresents or underrepresents certain groups, it contains bias. A biased sample leads to invalid conclusions.
Example of Bias: If you survey only students waiting in line at the music practice rooms about their favorite school activity, your findings will likely be biased toward music and will not represent the whole student body.
---3. Sampling Methods
How we choose our sample determines whether our findings are reliable and valid. Let us look at the primary sampling techniques used in mathematical reasoning.
Simple Random Sampling
Every member of the population has an equal chance of being chosen, and every combination of members has an equal chance of selection.
How it works: Assign every member of the population a unique number from \(1\) to \(N\). Use a random number generator or draw numbers from a hat to select \(n\) individuals.
Strengths: Free from researcher bias.
Limitations: Can be impractical or difficult if the population is widely scattered or if a complete list of members does not exist.
Systematic Sampling
Selecting elements from an ordered population list at regular intervals.
How it works: If a population has size \(N\) and you need a sample of size \(n\), calculate the interval \(k = \frac{N}{n}\). Choose a random starting point between \(1\) and \(k\), and then select every \(k\)-th person on the list.
Example: To pick \(20\) students from a list of \(200\), \(k = \frac{200}{20} = 10\). Pick a random number between \(1\) and \(10\) (e.g., \(4\)). Then choose students \(4, 14, 24, 34, \dots, 194\).
Strengths: Simple and spreads the sample evenly across the list.
Limitations: If the list has a hidden periodic pattern that matches \(k\), bias can occur.
Stratified Sampling
The population is divided into distinct subgroups (called strata) based on shared characteristics (such as age group, year level, or gender). A random sample is then drawn from each stratum in exact proportion to the stratum's size in the whole population.
The formula for the number of individuals to select from a specific stratum is:
\(\text{Stratum sample size} = \frac{\text{Size of stratum}}{\text{Total population size}} \times \text{Total desired sample size}\)
Step-by-Step Stratified Sampling Example:
A secondary school has \(600\) students in total across three year groups:
MYP 3: \(180\) students
MYP 4: \(240\) students
MYP 5: \(180\) students
A researcher wishes to take a stratified sample of \(50\) students. How many students should be chosen from each year group?
Step 1: Calculate the fraction for MYP 3:
\(\text{Sample for MYP 3} = \frac{180}{600} \times 50 = \frac{3}{10} \times 50 = 15\text{ students}\)
Step 2: Calculate the fraction for MYP 4:
\(\text{Sample for MYP 4} = \frac{240}{600} \times 50 = \frac{4}{10} \times 50 = 20\text{ students}\)
Step 3: Calculate the fraction for MYP 5:
\(\text{Sample for MYP 5} = \frac{180}{600} \times 50 = \frac{3}{10} \times 50 = 15\text{ students}\)
Step 4: Check your total:
\(15 + 20 + 15 = 50\). The proportions match perfectly!
Convenience Sampling
Selecting participants who are easiest to reach or readily available.
Example: Standing outside the school entrance and interviewing the first \(20\) people who walk past.
Strengths: Fast, inexpensive, and easy to carry out.
Limitations: Prone to substantial bias; rarely representative of the entire population.
Quota Sampling
Similar to stratified sampling, the researcher divides the population into categories (e.g., \(25\) boys and \(25\) girls) but chooses participants within each category using non-random methods (like convenience) until the target quota is filled.
Key Takeaway: Stratified sampling guarantees that subgroups are represented in fair proportion to their actual population numbers, while simple random sampling gives every single member an identical chance of being picked.
---4. Organizing and Representing Data
Once data is gathered, it often exists as an unorganized list of numbers or words, known as raw data. To make sense of it, we organize it into structured tables and visual representations.
Tally Charts and Frequency Tables
A frequency table summarizes how often each value or category appears.
For example, suppose the number of pets owned by \(20\) families is recorded as follows:
\(0, 1, 2, 0, 1, 1, 3, 2, 0, 1, 4, 1, 2, 0, 1, 2, 1, 0, 3, 1\)
We organize this discrete data into a frequency table:
Number of pets \(0\): Frequency = \(5\)
Number of pets \(1\): Frequency = \(8\)
Number of pets \(2\): Frequency = \(4\)
Number of pets \(3\): Frequency = \(2\)
Number of pets \(4\): Frequency = \(1\)
Total Frequency (\(\sum f\)): \(5 + 8 + 4 + 2 + 1 = 20\)
Grouped Frequency Tables
When dealing with a large range of continuous data (such as heights, times, or masses), individual values are grouped into continuous numerical intervals called class intervals.
Consider the time, \(t\) in seconds, taken by \(30\) athletes to finish a sprint:
\(10.0 \le t < 11.0\): Frequency = \(4\)
\(11.0 \le t < 12.0\): Frequency = \(11\)
\(12.0 \le t < 13.0\): Frequency = \(10\)
\(13.0 \le t < 14.0\): Frequency = \(5\)
Total: \(30\)
Notice the inequality symbols: \(10.0 \le t < 11.0\) means that any time from \(10.0\) up to (but not including) \(11.0\) seconds falls into this group. A value of exactly \(11.0\) belongs to the next interval: \(11.0 \le t < 12.0\). This prevents ambiguity!
Bar Charts vs. Histograms
Students often mix these two charts up, but they have distinct differences:
Bar Charts:
- Used for categorical (qualitative) or discrete numerical data.
- There are visible spaces or gaps between the bars to show that categories are distinct.
- The height of each bar represents the frequency.
Histograms (Equal Class Widths):
- Used for continuous quantitative data.
- There are no gaps between bars because continuous intervals flow seamlessly into one another (e.g., from \(10\) to \(11\), then \(11\) to \(12\)).
- The horizontal axis shows a continuous number line with class intervals.
Pie Charts
A pie chart is a circular diagram divided into sectors. It displays how a total quantity is split into parts or percentages.
Because a full turn in a circle is \(360^\circ\), the angle for each sector is proportional to its frequency:
\(\text{Sector Angle} = \frac{\text{Frequency of Category}}{\text{Total Frequency}} \times 360^\circ\)
Worked Example: In a class of \(30\) students, \(12\) walk to school. What angle should represent walkers on a pie chart?
\(\text{Angle} = \frac{12}{30} \times 360^\circ = \frac{2}{5} \times 360^\circ = 144^\circ\)
Stem-and-Leaf Diagrams
A stem-and-leaf diagram organizes numerical data in order of magnitude while retaining every individual data value.
Each value is split into two parts:
- A stem: the leading digit(s).
- A leaf: the final digit.
Example: Represent the following test scores: \(52, 58, 61, 64, 67, 70, 73, 73, 79, 85\)
Stem | Leaf
\(5\) | \(2\), \(8\)
\(6\) | \(1\), \(4\), \(7\)
\(7\) | \(0\), \(3\), \(3\), \(9\)
\(8\) | \(5\)
Key: \(6 \mid 1\) represents \(61\text{ marks}\)
Crucial Rule: A stem-and-leaf diagram must always have a key! Without a key, a reader cannot tell if \(6 \mid 1\) means \(61\), \(6.1\), or \(610\).
Two-Way Tables
Two-way tables display relationships between two sets of categorical variables simultaneously.
Example: Surveying \(100\) students about club participation:
Boys: \(28\) In a club, \(22\) Not in a club | Total = \(50\)
Girls: \(34\) In a club, \(16\) Not in a club | Total = \(50\)
Totals: \(62\) In a club, \(38\) Not in a club | Grand Total = \(100\)
Two-way tables make it easy to see marginal totals and conditional subsets at a glance.
---5. Evaluating Validity and Accuracy in Real-World Contexts (Criterion D)
In MYP Mathematics, gathering data is not just about crunching numbers; it is about thinking critically. When reviewing data collection in real-world scenarios, always reflect on these three questions:
1. Is the sample size adequate?
A sample of \(5\) people from a school of \(1000\) is too small to provide a reliable model, even if chosen randomly.
2. Could the method of data collection introduce bias?
- Leading questions: "Don't you agree that school lunches are delicious?" encourages a positive response.
- Location bias: Asking people at a train station how they commute to work will drastically over-represent public transport users.
3. Is the degree of accuracy appropriate?
Measuring the length of a desk to the nearest millimeter is sensible. Measuring walking distance from home to school to the nearest millimeter is not sensible—the nearest ten or one hundred meters is much more appropriate.
Common Mistakes to Avoid
- Forgetting the Key in a Stem-and-Leaf Plot: Always write a clear key indicating the units (e.g., \(4 \mid 2 = 42\text{ cm}\)).
- Leaving Gaps in Continuous Data Histograms: Never leave gaps between the bars of continuous data intervals unless a specific interval has a frequency of zero.
- Equal Slices in Pie Charts: Do not draw pie chart sectors by guessing! Always calculate the exact angle using \(\frac{\text{frequency}}{\text{total}} \times 360^\circ\) and measure accurately.
- Confusing Sample and Population: Remember that the population is the entire group of interest; the sample is the fraction you actually observe.
Chapter Review & Summary Checklist
Use this checklist to confirm your understanding:
[ ] I can distinguish between qualitative and quantitative data.
[ ] I can identify whether numerical data is discrete (counted) or continuous (measured).
[ ] I understand the difference between a population (the whole group) and a sample (a representative part).
[ ] I can select and describe sampling techniques: simple random, systematic, stratified, and convenience.
[ ] I can calculate proportional numbers for a stratified sample.
[ ] I can construct and interpret frequency tables, two-way tables, pie charts, and stem-and-leaf diagrams.
[ ] I can identify potential sources of bias and evaluate the validity of a data collection plan in real-world contexts.