1.1 - Statistics, Data and Graphs Flashcards
population
complete set of units that possess some common characteristics
sample (2)
- smaller but hopefully representative set of units from populations used to find true properties about that population
- bigger sample size generally = better estimates
variable
characteristic/ property measured on units (explain a response variable from one/more explanatory variables
types of numerical (quantitative) data (2)
- continuous
- discrete
types of categorical (qualitative) data (2)
- ordinal
- nominal
continuous data
values within a range, can be measured (e.g. size)
discrete data
fixed values, integer can be counted (e.g. number of chromosomes)
ordinal data
n factor levels with implicit order (e.g. size class; small, medium, large)
nominal data
n factor levels without implicit order (e.g. eye colour: blue, grey, green)
categorical data graphics (2)
- 1 variable:
- pie chart
- bar graph - 2/more variables:
- grouped bar graph
- stacked histogram
numerical data graphics
- 1 variable:
- histogram (distribution of numerical variable)
- bar graph
- box-and-whisker plot - 2/more variables:
- scatterplot
graphics rules (3)
- graphs need to have clear axes labels with units
- graphs need to have clear figure legends
- important to think about ranges of axes
random sampling
each member of a population has an equal and independent chance of being selected
sampling bias
systematic discrepancy between estimates and true population characteristics