adding up the frequency as you go along - create a running total
y axis shows cumulative frequency
x axis shows data values/groups
if using grouped data always plot the highest value in the group
join the points with a curved line
to find the median, lower or upper quartiles go half, quarter or 3/4 the way up the side respectivley and when you meet the curve go down to the x axis to find the value you are looking for
to estimate the number of values greater or less than something go along the x axis then up and across to the cumulative frequency using same method as above
2 of 6
Box Plots
show the spread of a set of data
first line shows the lowest value
second line shows lower quartile
third line shows the median
fourth line shows upper quartile
fifth line shows the highest value
middle 3 lines joined to make a box
both extreme values joined with one line
can be interpreted by saying which set of data has highest/lowest averages
comparing spreads - a larger spread means the data is less consistent and there is more variation
3 of 6
Stem and Leaf
data is ordered by size
the number of tens is written in front of the line and the number of units behind it
must include a key
easy to find the median and quartiles
4 of 6
Choosing methods
Mean and Standard deviation
uses all data BUT affected by extreme values
Median and Interquartile range
doesnt use all data BUT not affected by extreme values
Mode and Range
easy to find BUT there may not be a mode and range is only extreme values
Pictograms - used for qualitative data, time consuming, visually attractive
Pie Charts - used for qualitative and quantitative data, shows proportion of total but not frequency
Histograms
visualises continuous and grouped discrete data, area represents frequency- use FD
Cumulative frequency
continuous and grouped discrete data, easily fine median and quartiles
Box Plots
quantitative data, clearly shows min, max, quartiles and median but not frequency
Stem and Leaf
visualises quantative data, organises raw data
5 of 6
Standard Deviation
widley used statistical analysis like the rang or interquartile range
higher the value the more the data is spread out
compares every value to the mean
takes an average and works out how far away it is from the mean
Comments
No comments have yet been made