Module 1: Intro to Data Flashcards
data
collection of observations
statistics
study of how to best collect, analyze, and draw conclusions from data
data matrix
convenient common way to organize data (chart)
how many types of variables are there?
2: numerical and categorical
numerical
a number that can have calculations done with it
what are the 2 types of numerical variables
continuous and discrete
continuous
any numerical value (fractions and decimals)
discrete
integers only (can count)
categorial
set list of options/categories; can be words or numbers
what are the 2 types of categorial variables
nominal and ordinal
nominal
unordered; one is not less or more than another
ordinal
ordered
associated (dependent) variables
when 2 variables show some connection w/ one another; can be positive or negative
independent variables
when there is no evident connection between 2 variables
explanatory variable _________ response variable (correlation vs causation)
might affect
T or F: an explanatory and response variable means that the relationship between the 2 is casual
false; labelling 2 variables as such does not guarantee that the relationship between them are casual
observational studies
collect data w/o interfering w/ how data arises (ex. surveys); cannot establish causation
experiments
researchers randomly assign participants to various treatments to establish casual connections between explanatory and response variables
prospective study
individuals are identified and information is collected as events unfold
retrospective study
data is collected after events have taken place
research questions refer to a ___ ___
target population
sample
represents a subset of the cases
census
sample of the entire population; complex
anecdotal evidence
data collected in a haphazard fashion that represents only one or 2 cases; may be true, but may only represent extraordinary cases
3 types of sampling bias
non-response, voluntary response, and convenience sampling
non-response
if only a small fraction of the randomly sampled ppl choose to respond to a survey, the sample may NO longer be representative of the population
voluntary response
when the sample consists of people who volunteer to respond b/c they have strong opinions on the issue; NOT representative of the population
convenience sample
individuals who are easily accessible are more likely to be included in the sample
biased
unfairly prejudiced for or against something
exploratory analysis
when you analyze results from a sample
inference
when you generalize based on the results for the sample to infer something abt the population
good statistical methods are based on the notion of ______ _______
implied randomness
T or F: if observational data is not collected in a random framework from a population, it is NOT reliable
true
4 common random sampling techniques
simple, stratified, cluster, and multistage sampling
simple random sampling
randomly selected cases from the population; no implied connection between the points that are selected
stratified sample
strata are made up of similar observations; simple random sample in taken from EACH stratum
cluster sample
clusters are not made up of alike observations; we take a simple random sample of clusters, and then sample all observations in that cluster
multistage sample
take a simple random sample of clusters, and then take a simple random sample of observations from the sampled clusters
when a researchers assignment is randomized it is a ____ ____
randomized experiment
randomized experiments are critical for _____ ______
showing causality
4 principles of experimental design
control, randomize, replicate, and block
factors are ___________
conditions we can impose on the experimental units
blocking variables
characteristics that the experimental units come with (we want to control them)
blocking is like _________
stratifying (except used in experimental settings when randomly assigning
placebo
fake treatment (often used as the control group)
placebo effect
experimental units showing improvement simply b/c they believe they are receiving a special treatment
blinding
when experimental units don’t know whether they are in the control or treatment group
double-blind
both the experimental units and researchers who interact w/ the patients don’t know who is in the control or treatment group