SampDistrTTK

=** THINGS TO KNOW **=

=** TWO TYPES OF DESIGN SAMPLES: **= 1) Observational Studies- observe and measure interest. Can't influence response. 2) Experiment- actively IMPOSE a treatment on order to observe a response. FOUR TYPES OF BIAS Voluntary response-call-in's Interviewer- Wording of ?'s Response Bias- ? is embarrassing Undercoverage or Selection bias- certain group

=** FIVE TYPES OF SAMPLING: **= 1) Simple Random Sample- # pop. use a random # generator to get sample. 2) Cluster Random Sample- group, randomly pick ONE group to get the sample. 3) Stratified Sample- grouped by similarity. Pick a sample from EACH group. 4) Systematic Sample- Pick # then add a constant to it to get sample. 5) Convenience Sample- EASIEST to reach. Susceptible to Voluntary Bias.

=** EVERY EXPERIMENT HAS TO HAVE: **= -Randomization -Control -Replication (repeated samples)

=** RANDOM ALLOCATION: **= Use a table of random digits- a string of digits 0-9 in which each number in equally likely to be chosen and entries are independent. Number the population depending on the possible outcomes. Randomly select using a random number generator or Table B in the text book.
 * Try to use all #s 0-9 (10 #s)

Random Digit Table: 11382 20995 92155 78674 92560 50903 02432 74402 05377 12450 17790 01768 49834 05636 64788 34217 25931 03905 43426 24863 07109 33812 53702 20590 19927 95360 69061 21399 64384 88546 16059 80406 13922 86815 94696 44494 21658 16722 04973 40221 34476 78270 20331 67588 78933 71456 93887 00037 28472 69724 72524 90683 09650 99632 56243 86410 22468 61179 98160 87219 66925 27808 45157 46226 43021 52634 23795 10314 24199 82542 07514 06041 37017 44089 49430 33408 81474 10718 96428 65856 31013 35544 75065 29944 38749 65452 85342 25672 51566 26999 00700 53039 96023 93628 47698 12045 45562 18454 26336 76133

=**Randomization in an experimental design:**=

media type="custom" key="5937961"

A block is 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 has same rationale as the stratified random sample. Blocks allow us to reduce the amount of variation to improve the accuracy of our conclusions.
 * TO IMPROVE THE DESIGN: **

=** ELEMENTS OF A SIMULATION: **= - # assignment to the probabilities--> EX: 75% free throws made. 3/4:6/8. Made/Total. 1-6 (F.T. made) 7,8(miss) 9,0(don't use) - description of a trial -stopping rule -execution of simulation (marking of # line) -documentation of results -draw out the simulation!

=** Elements of a Design: **= -State your initial measurement -Stratify where mentioned in the problem -Randomize your samples to their clusters (Show how you randomized, if using a digits table, use 000-100 not 0-100) -Preform the test (Don't forget to have a control group, unless the problem has two groups competing against each-other) -Compare the results of the tests

=Sampling Distributions home= =** Vocabulary of Definite Importance.**= =**Practice Problems for thine pleasure.**= =**​Applets of Wonder.**=