Intro to Stats Flashcards
population
complete collection of all indivs to be studied
sample
subcollection of members selected from a population
variables
characteristics about the population you want to learn about or measure
i.e height, weight, BP
parameter
mathematical description of some characteristic/variable for a population
i.e population mean
statistic
mathematical description of some characteristic/variable for a sample
same as parameter just with a sample instead
descriptive stats
used to summarize and describe characteristics of data numerically and graphically
i.e sample mean AND standard deviation for age of ER pts
inferential stats
used to make conclusions/inferences about populations
types of data
- categorical- aka qualitative, consists of names or labels
- quantitative- aka numerical, numbers representing counts or measurements
subtypes of quantitative
- discrete- number of possible values is finite/countable
- continuous- numerical, infinitely many possible values w/ some continuous scale
measurement
another way to classify data to use scales of measurement
aka assignment of values to outcomes
scales of measurement
nominal
ordinal
interval
ratio
nominal level
names/labels/categories that CANNOT be ordered
least precise level of measurement
i.e f/m gender, y/n responses, survey, vitamins
categories only
ordinal level
can be ordered
gaps b/t data cannot be determined/are meaningless
i.e pain scale
categories with some order
interval level
ordered
gaps b/t data ARE meaningful
NO natural zero
i.e temperature scale, pH values
ordinal + meaning aka differences but no natural starting pt
ratio level
interval + natural zero aka 0= none
most precise scale of measurement
i.e. heart beat, blood pressure
difference and natural starting pt
experiment
apply some treatment then observe its effects on subjects
subjects called experimental units
gold standard
random assignment of subjects
observational study
observing and measuring specific characteristics w/o modifying subjects
aka no manipulation
probability sampling
selecting members from a pop so that each has a known chance of being selected
simple random sample
indiv chosen randomly and entirely by chance aka** same probability** of being chosen at any stage during sampling process
via random generator on computer
systematic sampling
select some starting point and then select every kth element in pop
stratified sampling
subdivide pop into at least two different subgroups that share same characteristics
then draw sample/stratum from each subgroup
cluster sampling
divide pop area into sections/clusters
randomly select some of clusters
choose all members from selected clusters
multistage sampling
collect data by using combo of basic sampling methods
pollsters select a sample in diff stages- each stage might use diff methods of sampling
non-probability sampling
often biased so results not necessarily valid
voluntary response sample or self selected
write in or call ins
convenience sampling
results that are easy to get
key concepts of data presentation
organize and summarize with frequency distribution to understand nature of data
frequency distribution/table
shows how data is partitioned among several categories/classes
list categories w/ number of data values in each
can be relative (%) or cumulative
histogram
graph w/ bars of equal width drawn adj to each other/without gaps
x axis = classes of quantitative data values
y axis= frequencies w/ heights of bars corresponding to frequency value
can be relative (%) too
frequency polygon
line segments connected to points directly above class midpoint values
can be relative too (proportions of %)
ogive
line graph w/ cumulative freq