Use LEFT and RIGHT arrow keys to navigate between flashcards;
Use UP and DOWN arrow keys to flip the card;
H to show hint;
A reads text to speech;
61 Cards in this Set
- Front
- Back
what is statistics
|
the science of collecting numerical data and drawing conclusions from them
|
|
what are individuals
|
the people place or things being studied in an experiment
|
|
"what are other names for ""individuals"""
|
subject, entities, participants, elements, units(usually used in experiments)
|
|
what are variables?
|
characteristics of the individual
|
|
what are some examples of variables?
|
age, height, BP, weight
|
|
define: quantitative
|
something that can be measured with math
|
|
"define: categorical
|
"something that falls into a category
|
|
another name for categorical"
|
qualitative"
|
|
define: population
|
ALL individuals of interest in a study
|
|
Define: sample
|
a part of the population that is representative of the whole population
|
|
Define: Parameter
|
"a numerical characteristic of the population
ie. proportion of all langara students that are female" |
|
Define: Statistic
|
"a numerical characteristic of the sample
ie proportion of n= 100 that (X)" |
|
what are some forms of tabular summary
|
"frequency table
relative frequency table (see distribution) percent frequency table" |
|
what are some forms of graphical summary
|
"bar charts
pie charts" |
|
Define: raw data
|
bunch of data that has not been organized or grouped
|
|
Define: ungrouped frequency distribution
|
shows sample and frequency of individual data (no ranges)
|
|
Define grouped frequency distribution
|
shows sample and frequency of data (ranges)
|
|
what do you look for in a graph
|
"shape - bell, bimodal, multimodal
center - where the center is spread - how varied is the data outliers - extreme" |
|
described L / R skew
|
"right tail = right skew
left tail = left skew" |
|
describe bimodal
|
"two peaks
symmetrical" |
|
describe complex multimodal distribution
|
multiple peaks and kinda all over the place
|
|
what are some bad sampling examples (non-random sample)
|
"convenience sample
voluntary response sample judgement sample" |
|
what is convenience sampling
|
selecting people in vicinity to be part of study
|
|
what is voluntary response sample
|
asking people to be part of study (usually results in highly opinionated people being the only source of data)
|
|
what is judgement sample
|
choosing specifically who is going to be part of study
|
|
what is a result of bad sampling
|
introduces bias into study b/c sample is not representative of population
|
|
what are some good random sampling methods
|
"simple random
systematic stratified cluster" |
|
what is simple random sampling
|
"each member of population has equal opportunity to be part of study
requires sampling frame randomly selected" |
|
systematic sampling
|
"members selected by system
sample size =n population = N every n/Nth person selected" |
|
what is a sampling frame
|
a list of population in which you can randomly select from to make sample
|
|
what is stratified sampling
|
population split into group by characteristic, then randomly chosen from within group
|
|
what is cluster sampling
|
"population randomly split into group,then chosen randomly from within groups"
|
|
what are measurement instruments
|
variables to be collected
|
|
what is the difference between observational and experimental study
|
"observational: only watch. no manipulation of subjects
experiments: treatments are given" |
|
what is a treatment
|
"any experimental condition
all factors combined" |
|
what are experimental units
|
"individuals in experiment- if human = subjects"
|
|
what are factors
|
explanatory (independent) variable that causes effect
|
|
what are lurking variables
|
variables that aren't being studied but can have effect on results
|
|
what are confounding variables
|
variables that cannot be differentiated in causing effect (ie studying vs intelligence for good mark)
|
|
what is the nature of experiments and what do they usually do
|
"comparitive;
compares treatment to control, placebo, another treatment" |
|
what is a method to remove bias
|
"randomize
double blind study- patients and investigators dont know who is in which group" |
|
what is matched pairs design
|
"choosing pair of subjects same in some characteristic
- can be same person (2 treatments 1 person), in this case pair is themselves" |
|
what is a response variable
|
the outcome measured of the explanatory (independent( variable.
|
|
can a placebo be control
|
yes b/c the placebo isn't actually doing anything
|
|
what is undercoverage bias
|
where not everyone in the population has an opportunity to participate in study
|
|
what is nonresponse bias
|
not everyone responding or can't be contacted for response
|
|
what is undercoverage bias
|
where groups of population are left out of selection of sample
|
|
what is wording effect
|
when wording of question is confusing or is bias for one side
|
|
what is bivariate data
|
set of data with two variables
|
|
what are the three types of bivariate data
|
"one cat, one quant
two cat two quant" |
|
what is the best graph to compare two quantitative variables
|
scatter plot (time plot if one variable is time)
|
|
best graph for two categorical variables
|
side by side bar chart
|
|
best graph for displaying one cat and one quant data
|
side by side box n whisker
|
|
what is the formula for variance
|
s^2 = (1/n-1)x SUMOF(Xi - Ave)^2
|
|
what are the units for variance
|
"s^2 for sample
σ^2 for population" |
|
what are the units for standard deviation
|
"s for sample
σ for population" |
|
What is block design experiments
|
pre divided groups to test hypotheses about different groups
|
|
when is the mean more right than median
|
right skewed
|
|
when is the mean more left than median
|
left skewed
|
|
which is more affected by outliers? mean / median
|
mean because outliers are more in the calculation
|
|
|
|