Study Types
A researcher asks a randomly selected group of residents how often they use public transit. What type of data collection is this?
A sample survey. The researcher asks selected residents questions and does not assign a treatment.
In a trial of a new study app, one group uses the app and another uses the usual study materials. What is the group using the usual materials called?
The control group or control condition.
A school uses a random-number generator to select students from its full enrollment list for a survey. Is this random selection or random assignment?
Random selection. Students are randomly chosen from the enrollment list.
A survey finds that people who drink more water tend to report fewer headaches. Can researchers conclude that drinking more water caused fewer headaches?
No. The survey establishes an association, not causation. Other factors could explain the relationship.
A company wants to compare two versions of a vocabulary app. Name the two treatments that could be compared.
One treatment is version A and the other is version B of the vocabulary app.
Researchers record the current caffeine intake and sleep duration of students without changing either behavior. What type of study is this?
An observational study. Researchers measure existing behaviors without imposing a treatment.
Researchers assign a sugar pill that looks like the test pill to participants in a medicine study. What is this pill called, and what purpose does it serve?
A placebo. It provides a comparison that resembles the treatment and can help control for expectations.
A research team uses a coin toss to place participating volunteers into either a reminder-text group or a no-text group. What kind of randomization is this?
Random assignment. Coin tosses allocate participating volunteers to conditions.
In an observational study, researchers measure participants’ habits but do not do what central action that defines an experiment?
Researchers do not assign or impose treatments on subjects.
A school wants to test whether a short daily stretch break affects students’ afternoon alertness. Identify a measurable response variable.
Accept a measurable outcome, such as a standardized afternoon alertness rating, a brief attention-task score, or the number of correct responses on a common task.
A team enrolls participants, records their current screen time, and checks in every six months for three years to track sleep quality. What type of observational study is this?
A prospective observational study. Researchers collect information going forward and follow participants over time.
A researcher pairs students with similar baseline reading scores. Within each pair, a random method assigns one student to a new tutoring program and the other to standard support. What design feature is being used?
Matched pairs. Similar students are paired, and treatment conditions are compared within each pair.
A random sample of local households is surveyed about their heating costs. What does the random selection support: generalizing to the population or claiming a cause-and-effect relationship?
Generalizing to the population of local households, provided the sample and survey methods are appropriate. Random selection alone does not establish causation.
People who attend a neighborhood park more often also report better health. Name one plausible confounding variable that could help explain this association.
Accept a reasonable example, such as income, neighborhood walkability, access to health care, age, or baseline health. The variable could be related to both park use and health.
A team has 60 volunteers to compare two revision strategies. Describe a fair method to randomly assign participants to the two groups.
Label the volunteers, use a random-number generator or shuffled assignments to place half in each group, and keep the assignment process impartial. With 60 volunteers, 30 could be assigned to each group.
Researchers use archived clinic records from the past decade to compare past medication use with later diagnoses. Is this prospective or retrospective, and why?
Retrospective observational study. It uses existing records about past exposures and outcomes.
A fitness study includes participants with very different starting fitness levels. Researchers form fitness-level groups first, then randomly assign each person within each group to one of two exercise programs. What design strategy is this, and why use it?
Blocking. Grouping participants by starting fitness level before random assignment helps account for differences in initial fitness when comparing programs.
Participants are randomly assigned to either a new language-learning program or the usual program. What type of conclusion can random assignment help support, assuming the experiment is well conducted?
A cause-and-effect conclusion may be supported because random assignment helps balance other factors between groups. The conclusion applies to the study conditions and participants; generalizing broadly requires an appropriate sample.
A study finds that people who take a certain vitamin have fewer colds. People who take the vitamin may also be more likely to get regular medical care. Identify the potential confounder and explain how it could affect the observed relationship
Regular medical care is a possible confounder. It may be associated with taking the vitamin and may also influence the likelihood of preventing, detecting, or treating colds, making the vitamin’s effect difficult to isolate.
Design a matched-pairs experiment comparing two brands of noise-canceling headphones. State how participants are paired or how each participant serves as their own pair, and how the order of treatments is assigned.
Accept either of these approaches:
Pair participants with similar characteristics, then randomly assign one person in each pair to each headphone brand.
Have each participant test both brands; randomly assign the order, use the same listening conditions, and compare each person’s ratings or performance across brands. A good answer identifies a response measure, such as noise-reduction rating or performance on a standardized listening task.
A city selects residents at random and asks them to report their weekly exercise. The researchers do not assign exercise routines or alter anyone’s behavior. Identify the data-collection method and explain why it is not an experiment.
A sample survey. Researchers ask selected residents questions; they do not assign or impose an exercise treatment.
A team wants to test whether a new note-taking method improves quiz performance. Name the treatment, control condition, response variable, and one way to blind the study. Explain how replication would strengthen the experiment.
Treatment: use the new note-taking method. Control: use the usual method or an appropriate comparison method. Response variable: quiz performance, such as score on a common quiz. Blinding option: have someone who does not know students’ assigned methods score the quizzes, or use anonymous scoring. Replication: include enough participants and repeat the comparison across multiple students or classes to make results less dependent on a few individuals.
A study randomly selects 120 employees from a company roster, then randomly assigns those selected employees to one of two break schedules. State what each random process contributes to the study.
Random selection helps make the selected employees representative of the company roster and supports generalizing to that population. Random assignment helps balance potential confounders across break-schedule groups and supports a cause-and-effect conclusion about the schedules.
A researcher proposes randomly assigning teenagers to sleep fewer hours to test the effect of sleep on grades. Explain why an observational study may be more ethical and practical, and state the kind of conclusion it could support.
Assigning sleep restriction could pose risks and interfere with participants’ well-being; it may also be difficult to control in everyday life. An observational study can examine whether sleep and grades are associated, but it cannot by itself establish that sleep duration caused differences in grades.
Sketch a plan for an experiment testing whether a new plant food increases tomato yield. Include treatment and control groups, random assignment, a response variable, and a blinding method if practical. If blinding is impractical, explain one way to reduce bias.
A strong plan includes:
Treatment group: plants receive the new plant food at a specified dose and schedule.
Control group: similar plants receive no plant food or a suitable standard comparison.
Randomly assign comparable plants, ideally within blocks such as location or starting size.
Response variable: tomato yield measured consistently, such as total mass or number of tomatoes per plant over the same growing period.
If practical, use coded containers so the person measuring yield does not know which plants received the treatment. If full blinding is impractical, standardize care and measurement procedures and use an independent assessor.