Sampling Methods
A researcher randomly chooses 50 students so that every possible group of 50 has an equal chance of being selected. What sampling method was used?
What is a simple random sample (SRS)?
A gym posts an optional survey on its website asking members whether they plan to renew. Members choose whether to participate. What type of bias is the main concern?
What is voluntary response bias?
This process uses chance to place experimental units into treatment groups and helps create groups that are similar with respect to extraneous variables.
What is random assignment?
Researchers assign participants to study either with music or without music. This variable, whether or not the participant listens to music, is what type of variable?
What is the explanatory variable?
Which process allows researchers to generalize results from a sample to a larger population?
What is random selection?
A company randomizes a list of customers and then surveys every 500th customer. This is the sampling method...
What is a systematic random sample?
A random sample of 500 students receives a survey, but only 120 respond. The students who don't respond may systematically differ from those who do.
What is nonresponse bias?
Participants are randomly assigned to one of three website layouts with no blocking or pairing. What experimental design is this?
What is a completely randomized design?
In that same experiment, researchers record the number of words recalled. What type of variable is this?
What is the response variable?
Which process allows researchers to establish a cause-and-effect relationship between an explanatory and response variable?
What is random assignment?
A school separates students into freshmen, sophomores, juniors, and seniors and takes an SRS from each grade.
What is a stratified random sample?
A school surveys students about technology use but only contacts students who own smartphones.
What is undercoverage?
Athletes are first grouped according to previous sprint-training experience and then randomly assigned to one of two stretching routines within each group.
What is a randomized block design?
Researchers compare three different doses of a medication: 0 mg, 25 mg, and 50 mg. What are the three doses called?
What are treatments?
Researchers randomly assign volunteers to a caffeine or no-caffeine treatment, but the volunteers were not randomly selected from a population. Assuming a significant difference is found, what kind of conclusion is supported?
What is a cause-and-effect conclusion for the experimental units, but not generalization to a larger population?
A city randomly selects 12 neighborhoods and surveys every household within the selected neighborhoods, this is an example of?
What is a cluster random sample?
A survey asks, “Don't you agree that our school's unfair phone policy should be changed?”
What is response bias caused by leading question wording?
Every participant tries both energy drinks, with the order randomly assigned, and their reaction times are compared.
What is a matched pairs design?
Researchers discover that people who drink more coffee also tend to sleep less. They cannot determine whether coffee or another related factor explains differences in test performance. What statistical issue does this describe?
What is confounding?
Researchers take an SRS of students from a district but simply observe whether students use TikTok and compare their grades. Can the results be generalized? Can researchers conclude TikTok causes differences in grades?
What is: yes to generalization to the sampled population, but no to cause-and-effect?
Explain the key difference between stratified and cluster sampling.
What is: stratified sampling randomly selects individuals from every group, while cluster sampling randomly selects entire groups and samples everyone in the selected groups?
An online news site asks readers, “Should school start later?” Anyone visiting the website can vote. Identify the main source of bias AND explain why.
What is voluntary response bias because individuals choose themselves to participate, so people with stronger opinions may be more likely to respond?
Why might researchers use a randomized block design instead of a completely randomized design?
What is to account for a variable known to affect the response, reducing variation between experimental units and making treatment comparisons more precise?
A researcher studies whether tutoring improves SAT scores. Students decide whether to attend tutoring. Students who choose tutoring also tend to study more independently. Why can't the researcher conclude that tutoring caused higher SAT scores?
What is: there was no random assignment, and studying independently is a potential confounding variable?
A district randomly selects 200 middle-school students and randomly assigns them to either a new homework platform or the existing system. The new-platform group has a statistically significant increase in completion. What conclusions are justified?
What is: the platform caused increased homework completion for the students in the experiment, AND the results can be generalized to middle-school students in the district?