STAT150 Lecture Notes - Lecture 2: Bar Chart, Contingency Table, Standard Deviation
18 May 2018
School
Department
Course
Professor
STAT150 LECTURE – 7/3/18
WK2: SUMMARISING AND DISPLAYING DATA
GRAPHS AND SUMMARIES FOR CATEGORICAL DATA
o The type of summary used depends on the classification of the variable/s of interest
Pie Charts show the proportion (area) of the pie taken up by each category.
Bar Charts require a bar for each category.
Graphs need;
o A title
o Clearly labelled axis
o An explanatory comment
o To be clear and uncluttered
FREQUENCY TABLES
To summarise one categorical variable, we need to construct a table showing the;
o Variable name
o Name of each category
o The count/proportion
o Percentage of observations in each category
CONTINGENCY TABLES
o Sometimes it is of interest to determine whether one variable is contingent on (or
associated with) another variable
o Suaise fo idiidual feuey tales dot help to ase uestios like that
o In this situation we need to create a contingency table
CLUSTERED BAR CHART
A clustered bar chart shows that opinion does seem to be comparable. These bars easily
identify comparable variables and the difference of data.
Bar/Pie
Charts &
CLustered
bar charts
Frequency
tables &
contingency
tables
find more resources at oneclass.com
find more resources at oneclass.com
Document Summary
Graphs and summaries for categorical data: the type of summary used depends on the classification of the variable/s of interest. Pie charts show the proportion (area) of the pie taken up by each category. Bar charts require a bar for each category. Graphs need: a title, clearly labelled axis, an explanatory comment, to be clear and uncluttered tables. To summarise one categorical variable, we need to construct a table showing the: variable name, name of each category, the count/proportion, percentage of observations in each category. Contingency tables: sometimes it is of interest to determine whether one variable is contingent on (or associated with) another variable, su(cid:373)(cid:373)a(cid:396)ise f(cid:396)o(cid:373) i(cid:374)di(cid:448)idual f(cid:396)e(cid:395)ue(cid:374)(cid:272)y ta(cid:271)les do(cid:374)(cid:859)t help to a(cid:374)s(cid:449)e(cid:396) (cid:395)uestio(cid:374)s like that. In this situation we need to create a contingency table. A clustered bar chart shows that opinion does seem to be comparable. These bars easily identify comparable variables and the difference of data.
