ILR STATS - 3:47:00 PM STATISTIC PRELIM 1 Chapter 2...

03/03/2008 15:47:00
←
← STATISTIC PRELIM 1
←
←
Chapter 2
•
Answering the who, what, when, where, why and how questions provide context for
data values.
o
If you can’t answer the who or the what there is no data
•
The people we experiment on are the subjects or participants, people who answer
questions are the respondents
• Characteristics recorded about each individual are called variables
•
When a variable named categories and answers question about how cases fall into
those categories, names categories they are categorical variables
•
When a measured variable with units answered questions about the quantity of what
is measured they are , numbers act as numeric values, quantitative variables
• Counts are used in two different ways
o
When we count the cases is each category of a categorical data, the categories
are labeled at the What and the Who – summarize the data
o
other times counting focuses on the amount of something – quantitative
• How the data is collected makes the difference between insight and nonsense
WHAT CAN GO WRONG?
o
Don’t label a variable as categorical or quantitative without thinking about the
question you want to answer
o
Just because your variable’s values are number, don’t assume that its
quantitative
o
Always be skeptical
•
Who = cases, What = variables, Why helps us decide how to treat the variables
←

←
Chapter 3 – Categorical Data
• Make a picture with the data
o
Reveal things you are not likely to see in table – think clearly
o
Show the important features of the patterns
o
Best way to tell others
Categorical Data
•
We can organize counts into frequency tables
•
Counts are useful but sometimes it better to use proportions, percentages
•
Relative frequency table displays the percentages (instead of the counts)
o
Can describe the distribution
•
The best data displays observe the area principle, which says that they are occupied
by a part of the graph, should correspond to the magnitude of the value it represents.
•
Bar chart – check categorical data condition
o
Displays the distribution of a categorical variable, showing the sound for each
category next to each other for easy comparison
o
Small spaces between the bars
o
Relative frequency bar chart

Draws attention to the relative proportion of passengers falling into
each of the categories
•
Pie charts – check categorical data condition
o
Quick expression of how the whole group is partitioned into smaller groups
•
Contingency Tables
o
Shows how individuals are distributed along each variable, contingent of the
values of the other variables
o
Margins give totals

o
Margins of a contingency table, the frequency distribution of one of the
variables is called its marginal distribution
o
Conditional distribution are created when they show the distribution of one
variable just for those cases that satisfy the condition on another variable
o
When the distribution of one variable is the same for all categories of another
– the variable is