hi math

fucking hate this class istg omg

population

entire group we want info about

sample

part of the population we actually examen to gather into about pop.

census

collect data from every individual in a population

observational study

observes individuals and measures variables of interest, but does not attempt to influence responses

experiment

deliberately imposes some treatment on individuals to measure their responses

generalization

• only cane make generalizations about a population if the samples are randomly selected

• not possible to determine casual relationships between variables in an observational study

what are the 6 types of sampling

• convenience

• voluntary response

• simple random sample (SRS)

• stratified random sampling

• cluster

• systematic

convenience sampling

• uses subjects that are readily available

• advantage: easy and less costly to collect

• disadvantage: not representative of the population

voluntary random sample

• obtained by allowing subjects yo decide wether or not to respond

• advantage: easy peasy to collect

• disadvantage: over represents people with strong opinions.

simple random sample (SRS)

• consists of “n” individuals form the population chosen in such a way that every set of “n” individuals has an eqaul chance in the sample selected

• advantage: easy to accomplish and good representation of the population

• disadvantage: none (cost or time maybe)

stratified random sample

• divide the population into groups of similar individuals (strata) then select an SRS within each strata. Combine the SRS’s from each strata to form a full sample.

• advantage: exact info

• disadvantage: not appropriate unless strata is defined

cluster sample

• divide the population into sections (clusters) and then randomly choose a few of those clusters. Every number of the chose cluster becomes your sample.

• advantage: don’t need a list of the entire population

• disadvantage: more variability between samples depending on how clusters are determined

systematic radnom sampling

randomly select an arbitrary (random) starting point and then select every Kth member of the population.

with replacement

an item can be selected more than once

without replacement

an item can NOT be selected more than once

voluntary response bias

when a sample is comprised entirely of volunteers or people who choose to participate

under coverage bias

occurs when groups in the population are left out of the process of choosing a sample

non-response

occurs when an individual chose for a sample can not be contacted or refuses to respond.

response bias definition

bias called by the behavior of the respondent or interviewer

response bias LIT UP

• L - lack of memory

• I - ignorance → people give answers to seem like they know what they’re doing/saying

• T - timing

• U - untruthful answers

• P - phrasing

sampling error

• the difference between a sample result and the population result. RESULTS FROM CHANCE VARIATION

• to minimize → increase sample size

• not your fault

non sampling error

occurs when the sample data is incorrectly collected, recorded or analyzed. usually occurs when the sample is selected n a non-random fashion.

experimental units

things on which the experiment is done

subjects

when the experimental units are human beings

treatment

experimental condition applied to the units.

principles of experimental design

• comparison

• randomization

• control

• replication

comparison (experimental design)

makes sure that design that compares two or more treatments

randomization (experimental design)

Random assignment of experimental units to avoid bias

control (experimental design)

• the control group is treated the same as the other groups in experiment

• reduces variability in the response variable

replication (experimental design)

use enough experimental units in each group so that any difference in the effects of the treatments can be distinguished form chance difference between groups.

factor (experimental terms)

explanatory variable (s)

level (experimental terms)

various groups the factors take

blinding

subjects does not know the treatment they are receiving to eliminate suggestion

double blind

both administrator and subject does not know who will receive the treatment

confounding variable

variables that might affect outcome, but we did not control or account for the in the experiment. affects the response

completely random design

experimental units are assigned to treatment by chance

lurking variable

• another variable drives each of the two variables under investigation, making it look like there is an association between those two variables.

• sampling and observational studies

matched pair (experimental design)

• same individual or two matched indivuals are assigned to receive the treatment and the control

• should be double blind

block design (experimental design)

• random assignment of experimental units to treatments is carried out separately within each block.

infernece

drawing conclusions beyond the data at hand

random selection of individuals

allows inference for the population

random assignment of an experiment

permits inference about cause or effect

randomly assigned/selected

• if both : there is inference about population AND cause and effect

• if only randomly assigned: no inference population, but there is inference about cause and effect

• if randomly selected, only inference about the population

• if none, than no inference .