SampDistrVocab

=** VOCABULARY LIST: SAMPLING DISTRIBUTION A TO Z **=

BIAS - experiment favors a certain outcome BLOCK - a group of experimental units/subjects that are known before the experiment to be similar in some way that is expected to affect the response to the treatments. BLOCK DESIGN - block the groups that are similar in characteristics that would affect the response and carry out the treatment within each block. (same rationale as a stratified random sample) CENSUS - a periodic count of the population. CLUSTER SAMPLE - split into groups-choose __one__ group to randomly select the sample- COMPLETELY RANDOMIZED DESIGN - all experimental units allocated at random to all treatments. CONFOUNDING VARIABLE- a confounding variable is one that affects the response variable and also is related to group membership.A variable that affects the response variable and is not related to group membership (that is, the variable would be expected to even out across groups) is not a confounding variable. CONTROL GROUP - the group that remains unchanged. the "placebo" group. Used in order to control lurking variables. CONVENIENCE SAMPLE - easiest to reach. Common force of voluntary response DESIGN - method used to get sample. DOUBLE BLIND - neither the subject nor the people conducting the experiment know which treatment a subject is given. EXPERIMENT -actively __impose__ a treatment in order to observe a response. EXPERIMENTAL UNIT - subject being tested (examples: people, patients, rats) EXPLANATORY VARIABLE - also called factors. the x-axis: independent variable. HOMOGENOUS - in stratified sampling, the strata should contain individuals that are very similar with regard to some external factor that will affect the response variable (ie: male and female, or black vs white skin). INTERVIEWER BIAS - wording of the question influences the response. MATCHED PAIRS - each subject will receive two, and only two, treatments (special form of blocking) NON-RESPONSE - no response by the subject given to the question possibly due to wording or the interviewer. OBSERVATIONAL STUDY - observe and measure interest. No ability to influence response. PLACEBO - "sham" treatment or placebo (medical). Also known as the contros variable. Used to control the "lurking variables." POPULATION - the ENTIRE group RANDOM ASSIGNMENT - use of chance to divide experimental units into groups. Second major principle of design experiment. RANDOMIZATION - use of chance to divide experimental units into groups. Random sampling from a population helps ensure (but does not guarantee) that individuals in the sample are representative of the population -Without random samples, results cannot be generalized to the population REPLICATION - having enough subjects in the study in order to carry out the experiment. RESPONSE BIAS - question is embarrassing to give a response to. RESPONSE VARIABLE - the dependent variable altered by the explanatory variable. the y-axis. (example: explanatory: aspirin (x) vs response: heart attacks (y) SAMPLE - a part of the population SIMPLE RANDOM SAMPLE - (SRS) method by which every sample of a given size has equal ability of being selected. number the population, use random number generator to get the sample SIMULATION - imitation of chance behavior based on a model that reflects the experiment. SINGLE BLIND - the subject does not know which treatment he/she is given. STRATIFIED SAMPLE - split into groups with a commonality- choose randomly from __each__ group. SYSTEMATIC SAMPLE - have random starting pointing and add a constant number to get the next sample. TABLE OF RANDOM DIGITS - string of digits 0-9 where each number is equally likely to be chosen and entries are independent. TREATMENT - condition applied to the experimental units (examples: drugs) UNDER-COVERAGE - targeted towards a certain group. (also called selection bias) VOLUNTARY RESPONSE - People choose themselves for experiment. very biased. (call in's/survey)

=**Sampling Distributions Home**= =**Practice Problems for thine pleasure.**= =**Things to Know, pretty colors, and a nice bonny picture or two.**= =**​Applets of Wonder.**=