Populations, Samples, and Bias - Sixth Grade Mathematics

Valid statistical analysis depends upon sound data collection methods that accurately represent a broader group without systematic distortions known as bias.

Populations and Samples

In statistics, researchers often wish to draw conclusions about large collections of individuals or items, but examining every member is usually impractical.

Population vs. Sample:

       +-------------------------------------------------------+
       | POPULATION: The entire group of individuals or objects|
       |             under investigation (e.g. All 1,200       |
       |             students in a school district)            |
       |                                                       |
       |            +----------------------------+             |
       |            | SAMPLE: A smaller subset   |             |
       |            |         selected from the  |             |
       |            |         population to      |             |
       |            |         represent it       |             |
       |            |         (e.g. 60 randomly  |             |
       |            |          chosen students)  |             |
       |            +----------------------------+             |
       +-------------------------------------------------------+

A census is a study that attempts to gather data from every single individual in an entire population. Because a census requires immense time and financial resources, researchers collect data from a representative sample.

Representative vs. Biased Samples

A sample is representative if its characteristics closely mirror those of the entire population. A sample is biased if certain members of the population are systematically favored or excluded over others, producing inaccurate, skewed conclusions.

Types of Sampling Methods:

1. Simple Random Sample (Unbiased / Fair):
   Every single individual in the population has an equal chance of being selected.
   Techniques: Drawing names from a hat, using a computer random number generator.

2. Systematic Sample:
   Selecting individuals according to a predetermined rule or interval.
   Example: Surveying every 10th student who enters the school building.
   (Unbiased if the list or sequence does not follow a hidden pattern).

3. Convenience Sample (Biased):
   Choosing individuals who are easiest to reach or readily available.
   Example: Asking only your 4 closest friends sitting at your lunch table.

4. Voluntary Response Sample (Biased):
   Allowing individuals to decide entirely on their own whether to participate.
   Example: An online poll on a website. Typically over-represents people with strong
   or extreme negative opinions.

Sources of Bias in Surveys

Beyond sampling methods, bias can infect statistical studies through survey design and administration:
1. Question Wording Bias: Phrasing questions in a leading or emotionally loaded way that nudges respondents toward a specific answer.
Example of leading question: "Don't you agree that homework is stressful and should be abolished?" Neutral question: "Do you favor or oppose reducing the amount of homework assigned each week?"
2. Non-Response Bias: When a large percentage of selected individuals refuse or fail to answer.

3. Response Bias: When respondents give untruthful or socially desirable answers rather than accurate facts.

Exercises

Exercise 1. Identify the population and the sample in each study: (a) A quality control inspector tests 50 light bulbs out of a production batch of 10,000 to determine average lifespan; (b) To determine student opinion on a new dress code, the principal surveys 80 randomly chosen students from the total enrollment of 650 students.

Exercise 2. State whether each sampling method is likely to produce an unbiased (representative) sample or a biased sample. Justify your reasoning: (a) A surveyor asks people exiting a movie theater about their favorite leisure activity; (b) To select a sample of students, a teacher places the name of every student in the school into a drum, mixes them thoroughly, and draws 40 names; (c) A radio host asks listeners to call in to vote on whether a proposed tax increase should pass.

Exercise 3. Critique the survey question: "Since public parks provide essential fresh air and recreation, should the city increase park funding?" Rewrite it to eliminate wording bias.

Solutions

Solution to Exercise 1. For (a), the population is all 10,000 light bulbs produced in the factory batch; the sample is the 50 light bulbs selected and tested. For (b), the population is all 650 students enrolled at the school; the sample is the 80 randomly selected students surveyed.

Solution to Exercise 2. For (a), this is a convenience sample that is heavily biased because people leaving a movie theater are already predisposed toward watching movies as their leisure activity, excluding people engaged in other hobbies. For (b), this is an unbiased simple random sample because every student in the school has an equal probability of having their name drawn from the thoroughly mixed drum. For (c), this is a voluntary response sample that is biased because only listeners with strong opinions take the time to call the station, while the majority with moderate views remain unrepresented.

Solution to Exercise 3. The question exhibits wording bias because the opening clause ("Since public parks provide essential fresh air and recreation...") leads respondents by emphasizing positive attributes to elicit agreement. An unbiased rewrite is: "Do you support or oppose increasing city funding for public parks?"