Transcript Statistics Guide

Slide 2
What do you
want to do?
Description
Evaluate
Differences
Examine
Relationships
Return to Slide 2
Purpose of the Evaluation--Description
Description
Individuals
Samples
Populations
Return to Slide 3
Describe Individuals
What level of
Measurement
Nominal
Ordinal
Scale
Category of
Membership
Rank Order
Percentiles,
z scores
Return to Slide 3
Description of Samples
What Level of
Measurement
Nominal
Percentages or
Proportions, Mode
Scale
Central Tendency
Mean, Median
Data Variability
Range, Variance,
Standard Deviation
Return to Slide 3
Description of Populations
What Level of
Measurement?
Nominal
Scale
Confidence
Interval for
Proportions
Confidence
interval for the
mean
Purpose of the Analysis: Evaluate Differences
How many
samples?
One
Two
Three or
more
Return to Slide 2
Purpose of the Analysis – Evaluate differences
Evaluate
Differences
One Sample
Two Samples
Three or
more
Samples
Return to Slide 8
Purpose of the Analysis – Evaluate differences for One Sample
Evaluate
Differences (One
Sample)
For nominal data:
Data: Chi Square
Goodness of Fit
For scale data:
One-sample t test
Return to Slide 8
Purpose of the Analysis – Evaluate Differences for Two Samples
Evaluate Differences (Two Samples)
Independent Samples?
Dependent or
Correlated Samples?
Return to Slide 10
Difference Between Two Independent Samples
Level of Measurement
Nominal
Chi-square test
Ordinal
Scale
Mann-Whitney U
Independent
Samples t test
Return to Slide 10
Difference Between Two DEependent Samples
Level of Measurement
Ordinal
Scale
Wilcoxon T test
Dependent Samples t
test
Return to slide 8
Three or More Samples
How many
variables?
Two or More
One
Dependent
Samples
Independent Samples
For Ordinal Data:
Kruskal-Wallis
One-way ANOVA
For Scale Data:
ANOVA
One-Way
ANOVA
Independent
Samples
Factorial
ANOVA
Return to Slide 2
Purpose of the Analysis—Examine Relationships
Examine
Relationships
Nominal Data
Ordinal Data
Scale
Chi-Square Test
of Association
Spearman RankOrder
Correlation
Pearson Product
Moment
Correlation
Mixed Data
Return to Slide 14
Examine Relationships: Mixed Data Types
What type of mixed data?
Nominal and
Ordinal
Two Categories
For Independent
Sample: MannWhitney U Test.
For Dependent
Samples: Wilcoxon T
Test
Nominal and
Scale
Three or
More
Categories
Kruskal-Wallis
One-Way
ANOVA
Independent
Samples t test;
Dependent
Samples t test
Ordinal and
Scale
For Independent
Samples: One-Way
ANOVA or Factorial
ANOVA
For Dependent
Samples: RepeatedMeasures ANOVA