AP Stats Review Chapter 4

Problem with convenience samples

selects individuals from the pop. who are easy to reach, not representative of the true population

voluntary response sample

allows people to choose to be in a sample by responding to a general invitation; only those with strong opinions respond

what is a simple random sample

a simple random sample is a sample of size n chosen in such a way that every group of n individuals has an equal chance of being selected as the sample

explain random assignment by random number generator

label each subject 1-N, using a random number generator select X unique numbers, ignoring repeats. the numbers correspond to subjects in the sample

what is a stratified random sample

a stratified random sample is selected by doing a simple random sample from each stratum and combining them to one overall sample

why do we stratify?

ensures sample that is more representative of the population and will reduce sampling variability

how do you choose a variable to stratify by?

choose a variable most likely to affect the variable being measured

what is a cluster sample

a sample in which clusters are randomly selected and each member of the selected clusters make up one overall sample

what are the advantages of cluster sampling

more time-saving and easier to do

what's a disadvantage of cluster sampling

more sampling variability

what is undercoverage

occurs when some members of the population cannot be reached or are excluded from being chosen in the sample

what is nonresponse

occurs when an individual chosen for a sample cannot be contacted / refuses to participate

what is response bias

systematic patter of inaccurate answers to a survey question due to poor question wording/leading questions or a desire to appear moral; respondents may not answer truthfully or accurately to the intended meaning of the question

what is inference

a conclusion drawn from the sample to make a prediction about the population

what is sampling variability

different random samples of the population will produce different estimates

what is a margin of error

compensates for variability in random samples of the population; larger samples = lower margin of error

what's the benefit of increasing the sample size

reduces variability in the estimate of a given parameter, giving a more precise estimate closer to the true population value

what is an observational study

a procedure where observers observe individuals and measures variables without attempting to influence responses / impose a treatment

what is confounding

confounding occurs when two variables are associated in such a way that their effects on a response variable are indistinguishable

what is an experiment

an experiment is a procedure in which experimenters deliberately impose a treatment on experimental units to measure another variable's responses

what is a placebo

a placebo is a treatment with no active ingredient but is otherwise like other treatments in the experiment

what is a treatment

a specific condition applied to individuals in the experiment

what is a factor

a manipulated variable that may have effects on the response variable

what is the purpose of a control group?

a control group provides a baseline for comparing the effects of different treatments on a response variable; a control group may be a placebo, old active treatment, or nothing at all

what's the placebo effect

phenomenon where some subjects respond favorably to a treatment, even if inactive

what is a single-blind experiment

an experiment where the subjects/experimental units don't know which treatment they're receiving

what is a double-blind experiment?

an experiment where neither those administering the treatment, nor the subjects know which treatment they're receiving

what is the importance of blindness?

different treatment groups are as similar as possible, controlling for other factors that may influence response

whats the purpose of random assignment

random assignment should roughly distribute different types of subjects equally between the treatment groups; it ensures that the effects of other variables are spread evenly among the treatment groups; produces groups that should be roughly similar in all respects before treatments are applied

what are the benefits of controlling other variables

makes it easier to detect a difference between the effects treatment groups had on a response variable

what is replication

using many experimental units to distinguish a difference in the effects of a treatment; reduces likelihood that observed difference between the treatment groups is due to random assignment

how can we determine if results are convincing

set up a simulation; if the likelihood of observing a difference between the treatment groups is less than 5%, the difference is considered unlikely to happen from the chance variation in the random assignment, and there is statistically significant

what is a randomized block desgin

a randomized block design is an experimental design in which subjects are divided into blocks and then are assigned into treatment groups within each block

what is the benefit of blocking

blocking accounts for variability in responses that are created by different blocks suspected to influence the response; therefore, it is easier to discern a difference between the treatment groups

what is a matched pairs design

an experimental design for comparing two treatments that uses blocks of size 2; in some matched pairs designs, two very similar units are paired and 2 treatments randomly assigned to each while in others, each unit receives both treatments in random order