Unit 3 Mini Quiz Flashcards
What are the three ways we use random numbers?
- To simulate the likelihood of an event occurring
- To choose a sample that is representative of the population and avoid bias
- To assign subjects to treatments to evenly distribute variability and reduce confounding variables
How do you use a table of random digits?
- Make a key showing what the digits represent, whether you will use single, double, or triple digits, and which will be ignored
- Decide when trial ends
- Clearly label successes and where trial ends
- Explain what you’re counting
What is random sampling?
When we use chance to select a sample
What is a multistage sample?
Sample that combines several sampling methods
What is response bias? How to avoid it?
Influence (question wording, surveyor appearance/behavior) that may make respondent give more favorable answer. Ask questions that respondents can answer comfortably and honestly/ use neutral wording.
What is the difference between response bias and nonresponse bias?
Response bias occurs when the respondent’s answer is influenced by the questioning method, while nonresponse bias occurs when people who don’t respond may have different opinions than those who did
Standard sampling method?
SRS (Simple Random Sample). Every group of n individuals has equal chance of being sample.
What is wrong with using volunteers in a survey?
Volunteers may not be like rest of population. Voluntary bias, may feel particularly strongly about survey question.
What is wrong with using volunteers in an experiment?
Not much. We’re not looking for a sample like the population, we’re looking for effectiveness of treatment
What type of study would find relationship between Verbal and Math SAT?
You could run a regression of the scores. Retrospective study.
What’s the difference between a retrospective and prospective study?
A retrospective study looks back at a group’s history, while a prospective study watches a group over time and records data
Why do you have to stratify?
You don’t have to, but it might be good if a simple random sample won’t represent the population well
Can you stratify in an experiment?
NO. You can block.
Give example of when you would block?
Impact of different leather preservers on seats at an airport. If one treatment is applied to all the seats in the sun, and the other is applied to all the seats in the dark, the results will be affected by the light. Instead, you can block and randomly assign treatments to each group of seats
What four things do you need in an experimental design? (trick)
NEED only 3: control, randomization, replication…BUT use blocking when appropriate