Define the Groups Before Comparing Their Means
In “Choosing One-Sided or Two-Sided for Two Means,” you learned that the research question determines the direction of an alternative hypothesis. Before writing that hypothesis, make sure the two means in the comparison are clearly defined. A reader should be able to tell exactly which population each mean describes and what is being measured.
For a two-sample comparison, each parameter describes a population mean for a quantitative variable. The groups might be defined by different locations, methods, treatments, or other characteristics. The important point is to define each mean in context—not simply to write \(\mu_1\) and \(\mu_2\) and expect the reader to know what they represent.
A useful definition answers four questions: Who or what is in the population? Which group or condition identifies it? What quantitative variable is measured? What are the units? When the study has a specific time period or setting, include that too. These details make the parameter definitions meaningful and prevent the group labels from being confused.
A Clear Definition Has a Group, a Variable, and a Scope
A concise definition can follow this pattern: “Let \(\mu_1\) be the true mean [quantitative variable, with units] for all [units in population 1] in [setting or time period].” Write a parallel statement for \(\mu_2\). If the group is defined by a treatment or method, name it directly.
The two definitions should use the same quantitative variable and units. If one parameter is mean travel time in minutes and the other is mean travel distance in kilometers, their difference would not be a meaningful comparison of the same quantity. Similarly, “mean score” is incomplete if it is unclear which assessment or which people are included.
Use \(\mu_1\) and \(\mu_2\) consistently. The subscripts are labels, not a ranking: group 1 is not necessarily larger, better, earlier, or more important than group 2. Choose an order that is easy to follow, state it, and keep it through the hypotheses and any later interpretation.
When you write \(\mu_1-\mu_2\), the second group’s mean is subtracted from the first group’s mean. A positive difference means the population mean for group 1 is greater than the population mean for group 2; a negative difference means it is smaller. That sign only has a clear interpretation if the definitions of both means and the order of subtraction are explicit.
Population Means Are Not Sample Means
A population mean is a parameter: it describes the target population, whether or not every member has been measured. A sample mean, written \(\bar{x}\), summarizes the observed values in a sample. The sample mean can provide information about the corresponding population mean, but it does not define that parameter.
For example, if a sample of students has an average quiz score of 82 points, that value is a sample statistic. It does not mean \(\mu\) should be defined as “the average score of the students in the sample.” The population parameter might instead be the true mean quiz score for all students in a specified grade who take that quiz during a specified term. The sample is a subset used to learn about that broader group.
Be precise about the target population. “Customers” might mean customers at one store on a particular day, all customers at a chain’s local branches, or all customers who use a service in a year. Those are different populations and can have different population means. A definition that names the intended group makes the scope of the comparison visible.
Say whether the population consists of people, products, locations, or another clearly identified kind of unit.
State what distinguishes group 1 from group 2, such as a location, method, or treatment condition.
Use the same quantitative variable and units in both definitions.
Include a relevant setting or time period, then keep group 1 and group 2 in the same order when writing the difference.
Worked Examples
Worked Example: Comparing Two Library Locations
A city library system wants to compare how long, in minutes, visitors spend using public computers at its central branch and its east branch. The target is visitors who use a public computer at either branch during a specified spring term. Define both population means and state the order central minus east.
Identify the units and variable: The units are visitors who use a public computer. The quantitative variable is the number of minutes a visitor spends using a public computer during a visit.
Define the first parameter: Let \(\mu_C\) be the true mean number of minutes spent using a public computer per visit by all visitors who use a public computer at the central branch during the spring term.
Define the second parameter: Let \(\mu_E\) be the true mean number of minutes spent using a public computer per visit by all visitors who use a public computer at the east branch during the same spring term.
State the comparison: The requested order is \(\mu_C-\mu_E\), measured in minutes per visit. A positive value would mean that the central-branch population mean is greater than the east-branch population mean.
These definitions identify the relevant visitors and term, not just the people who happened to be observed. They also make clear that the comparison concerns the same measure at both branches.
Worked Example: Comparing Two Seedling Treatments
A horticulture class grows seedlings of the same plant species using two watering schedules: Schedule A and Schedule B. The students want to compare the mean height, in centimeters, of seedlings after six weeks. Define the two population means using Schedule A minus Schedule B.
Identify the units and variable: The units are seedlings grown under the class’s specified conditions. The quantitative variable is each seedling’s height after six weeks, measured in centimeters.
Define the Schedule A parameter: Let \(\mu_A\) be the true mean height after six weeks, in centimeters, for all seedlings of this species grown under the class’s specified conditions using Schedule A.
Define the Schedule B parameter: Let \(\mu_B\) be the true mean height after six weeks, in centimeters, for all seedlings of this species grown under the same conditions using Schedule B.
State the comparison: The difference is \(\mu_A-\mu_B\). It is measured in centimeters, and its sign describes the mean height under Schedule A relative to the mean height under Schedule B.
The definition includes the six-week measurement point and the growing conditions because both help specify which population means the class intends to compare. It does not define \(\mu_A\) as the average height in the particular sample of seedlings.
Worked Example: Comparing Delivery Times on Two Routes
A neighborhood courier service is interested in delivery times on Route P and Route Q. For deliveries made on weekday mornings in a particular service area, it wants to compare the true mean time from dispatch to delivery, measured in minutes. Define both means in the order Route Q minus Route P.
Identify the units and variable: The units are deliveries made on weekday mornings in the specified service area. The quantitative variable is elapsed time from dispatch to delivery, in minutes.
Define the Route Q parameter: Let \(\mu_Q\) be the true mean dispatch-to-delivery time, in minutes, for all deliveries made on Route Q on weekday mornings in the specified service area.
Define the Route P parameter: Let \(\mu_P\) be the true mean dispatch-to-delivery time, in minutes, for all deliveries made on Route P on weekday mornings in the same service area.
State the comparison: The specified difference is \(\mu_Q-\mu_P\), measured in minutes. Here the Route P mean is subtracted from the Route Q mean. A positive value means the true mean delivery time on Route Q is greater than the true mean delivery time on Route P.
This example shows why writing down the subtraction order matters. The parameter names make it clear which route comes first; the positive sign has meaning only when read using that order.
Worked Example: Defining Means for Two Phone Settings
A student technology club compares battery use under two phone display settings during a standardized video task. It wants to compare the mean percentage of battery used over the task, with the target being phones of a specified model running the task under the club’s conditions. Define the means as Setting 1 minus Setting 2.
Identify the units and variable: The units are phones of the specified model tested under the stated conditions. The quantitative variable is the percentage of battery used during the standardized video task.
Define the first parameter: Let \(\mu_1\) be the true mean percentage of battery used during the task by phones of the specified model using Display Setting 1 under the club’s conditions.
Define the second parameter: Let \(\mu_2\) be the true mean percentage of battery used during the same task by phones of that model using Display Setting 2 under the same conditions.
State the comparison: The requested difference is \(\mu_1-\mu_2\), measured in percentage points of battery used. The second setting’s mean is subtracted from the first setting’s mean. A positive difference means the mean percentage used under Setting 1 is greater than the mean under Setting 2.
“Percentage points” describes the difference between two percentages. Both means are defined for the same phone model, task, and testing conditions, so their difference compares like with like.
Use the Definitions to Make Hypotheses Read Clearly
In “Hypotheses for Comparing Two Means,” the null hypothesis of no difference was written as \(H_0:\mu_1-\mu_2=0\). The meaning of that statement depends on the contextual definitions. If \(\mu_1\) is the mean computer-use time at the central branch and \(\mu_2\) is the mean at the east branch, the null says those two population means are equal for the defined visitors and term.
The alternative also relies on those definitions. For instance, \(H_a:\mu_1-\mu_2<0\) says the population mean for group 1 is less than the population mean for group 2. It does not mean that every individual in group 1 has a smaller measurement. It describes a comparison of population means.
If the parameter definitions are vague, even correctly written symbols can leave the hypothesis unclear. “Let \(\mu_1\) and \(\mu_2\) be the means” does not say whose means they are or what quantity is being averaged. Define both parameters first, then use their symbols consistently in the hypotheses.
Common Mistakes and AP Exam Tips
- Defining only the difference: Writing “\(\mu_1-\mu_2\) is the difference in means” does not identify either population. Define \(\mu_1\) and \(\mu_2\) separately before describing the difference.
- Leaving out the measured variable: “The mean for Group A” is incomplete. Say what is averaged, such as delivery time in minutes or seedling height in centimeters.
- Leaving the target group too broad: “All customers” may not match the question. Add the relevant store, service, time period, or eligibility description when needed.
- Defining a parameter as a sample average: A sample mean describes the observed sample. Define the population mean for the intended population; use the sample mean as a statistic that provides information about it.
- Using different variables or units: Both parameters in a two-mean comparison should describe the same quantitative variable in the same units. A difference between unlike quantities is not the intended comparison.
- Reversing the subtraction order in words: In \(\mu_1-\mu_2\), the second group’s mean is subtracted from the first group’s mean. State the order explicitly and check that any interpretation of a positive or negative difference follows it.
- Assuming the subscripts rank the groups: The labels 1 and 2 are identifiers, not claims about which mean is larger. Choose the labels for clarity and keep them consistent.
A full-credit definition names both groups, the quantitative variable, and the relevant scope, with units where they clarify the measure. A full-credit comparison also states the subtraction order. This precision gives the hypotheses and later conclusions an unambiguous meaning in context.
Check Your Understanding
For each situation, write contextual definitions for the two population means and state the requested order of subtraction.
- A school compares the number of minutes students in two lunch periods spend waiting in the cafeteria line. What population and variable details should appear in the definitions?
- A park department compares the mean width, in centimeters, of seedlings grown in two greenhouse locations. Define both parameters and state what the units are.
- A repair shop compares service times for two repair procedures. Explain how to distinguish the population mean parameter from the mean service time in a sample.
- Let \(\mu_A\) be the mean travel distance for buses using Route A and \(\mu_B\) the mean distance for buses using Route B. In \(\mu_A-\mu_B\), which mean is subtracted from which?
- Why should two population means in a comparison refer to the same quantitative variable and units?