Statistics Experiment Vocab
Terms
undefined, object
copy deck
- Actively imposing some treatment in order to observe response
- Experiment
- Part of the population that we examine in order to gather information
- Sample
- Entire group of individuals that we want info about
- Population
- Studying a part in order to get info about the entire group
- Sampling
- Applying the information gained from a sample to the entire population the sammple is supposed to rep
- Statistical Inference
- Method used to choose the sample from the population
- Sample Design
- Two variables' effects on a response variable cannot be distinguished from each other
- Confounded
- A sample created by those who volunteer themselves, responding to general appeal
- Voluntary Response Sample
- Choosing individuals that are easiest to reach
- Convenience Sampling
- A study that systematically favors certain outcomes
- Biased
- Set of individuals chose from the population where every set has an equal chance of being selected
- Simple Random Sample SRS
- Label/Assign values to each individual, then select random labels
- Choosing an SRS
- Gives each member of the population known chance to be selected (greater than zero)
- Probability sample
- Divides population into groups of familiar individuals (strata) then chooses SRS from the strata to do the test
- Stratified Random Sample
- Way of narrowing down a sample to the specific individual you want to study
- Multistage Sample
- When some groups are left out of the sampling choosing
- Undercoverage
- When someone refuses to cooperate in survey
- Nonresponse
- When interviewer can cause problems in gathering data (wording effects)
- Response Bias
- Recalling events from the past
- Telescoping
- List of individuals from which the sample is actually selected
- Sampling Frame
- observes individuals and measures variables of interest (no attempt to influence response)
- observational study
- units are human beings
- Subjects
- Specific experimental condition applied to units
- Treatment
- Explanatory (independent)variables in the experiment
- Factors
- Combining specific values of each factor when conducing experiments that study the joint effects of several factors
- Level
- Dummy treatment with no physical effect
- Placebo
- When patients respond to the dummy treatment
- Placebo effect
- Group of patients receiving sham treatment
- Control Group
- when the experimenter systematically creates similar treatment groups based on statuses
- Matching
- Using chance to divide units into groups
- Randomization
- When all experiemtnal units are allocated at random among all treatments
- Completely randomized
- Effect too large to be of chance
- statistically significant
- repetition of experiment to reduce liklihood of chance
- replication
- when experimenter does not treat all subjects the exact same way
- hidden bias
- where neither of teh subjects or people having contact with them know which treatment is which
- double-blind
- when an experiment does not duplicate the true conditions they study
- Lack of realism
- random assignment of units to treatments separately in each block
- block design
- group of units similar in ways affecting their response to treatment
- block
- block design consisting of two units closely matched
- matched pairs design