Study Guides
(238,085)

Canada
(114,909)

University of Toronto St. George
(7,975)

CSB345H1
(9)

William Navarre
(9)

Quiz

# Session 6 - More Than One Treatment Per Subject. Tests of Repeated Measures.docx

Unlock Document

University of Toronto St. George

Cell and Systems Biology

CSB345H1

William Navarre

Fall

Description

HMB325H © Lisa| Page 113
S E S S I O N 6 : T E S T S O F R E P E AT E D M E A S U R E S ( C H A P T E R 9 )
LEARNING OBJECTIVES
1. identify independent & dependent groups
2. calculate & interpret a paired t test analysis
3. describe the assumptions for the paired t test
4. calculate & interpret the power of a paired t test
5. know when a repeated measures analysis of variance (ANOVA) should be used
6. calculate & interpret McNemar’s test
1. assumption for independent group doesn’t hold
2. each subject receives 2 or more treatments
3. the subject often serves as their own control
4. when the independent group assumption doesn’t hold, use – paired t test
5. the study power for a paired t test – McNemar’s test
6. outcomes is tested grip strength
7. if you use an unpaired t test, you are ignoring that there are not 14 individuals, but only
7
1. in this study, each person serves as their own control
2. variability is decreased
THE PAIRED t TEST
8. should use a paired t test to acknowledge that the data is paired
9. H :0the new training technique is not affective
1. shouldn’t expect difference in grip strength before & after
2. null = 0 difference (same as unpaired t test)
10. average of the differences = overall difference of improvement = 20
11. degrees of freedom = 7 individuals – 1 = 6
12. the differences follow a normal distribution bc (mean = median) therefore, an
assumption of the paired t test is the same as the unpaired t test: the data must follow
a normal distribution
13. negative numbers are possible (ex. difference is negative)
tdf mean−null = mean−null
SED SD
√n
1. the paired t test addresses circumstances when the data being assess are not
independent – the independent groups assumption is not met
2. the most common scenario is a before-after study – the outcome of interest is
measured twice in the same indiv

More
Less
Related notes for CSB345H1