week 3 presenting data Flashcards
univariate
one variable
bivariate
2 variables
IV and DV
more than 2 relationships
multivariate
more than 2 variables
complex relationships
noise
not part of testing logic, but interferes with the tested variation and defines reliability of results
where does the IV go on a graph
x-axis
where does the DV go on a graph
y-axis
where does noise go on a graph
error bars
give an example of noise
individual differences
discrete data
integer values, potentially values eg number of cars in car park
no fractions or decimals
continuous data
infinitesimal, real numbers eg time, temp and height
can include fractions or decimals
what is the general rule for freedom
more freedom = more versatile but less organized
pie chart
continuous, ratio-scaled axis presented as a circle
can include discrete data shown as wedges/circular sectors
pro of pie chart
relate parts/wedges to the whole
cons of pie chart
limited to 2 sources of variation - no error bars
requires ratio-scales, but finite data to define beginning and end of circumference
pro of bar chart
highlighting distance from 0 or baseline = constant value
highlight differences across groups or conditions = nominal
visualise groups and conditions
cons of bar chart
no continuos X - one axis needs to be discrete
clutter with many scores
pros of line chart
shows trends and relationship
as change in y…change in x
cons of line chart
covariation - useless when multiple data points per x-value
less efficient for grouping than bars
pros of scattergraph
covariation
visuals 2D measures, bivariate distribution
cons of line graph
lacks structure
occlusion
free floating points - clutter
what is aggregation by frequency
individual data points are grouped and summarized based on how often they occur within specific categories, intervals, or values
steps for aggregation
- define categories
- count occurrence
- summarise data eg frequency table - can work out %
where does frequency go on a graph
y-axis
binning
transform continuous data into discrete data by allocating the continuous data to intervals (bins)