Transcript Session 2 -IRB, Purpose, and Method

NIH and IRB
Purpose and Method
M.Ed. 6085
Session 2
NIH Certification
• The National Institute of Health certifies
researchers to ensure they understand
protections dealing with human subjects.
• You must complete your NIH certification prior
to beginning your project (and as a part of this
class).
• http://faculty.weber.edu/kristinhadley/med6085
Institutional Review Board for use of
Human Subjects in Research
• http://departments.weber.edu/meduc/irb/default.htm
• IRB application
– http://departments.weber.edu/meduc/irb/hsr_appl.rtf
• Once you have a signed proposal, complete the
IRB application. Give a hard copy to Dr. Gowans
and email her an electronic copy.
– You must include a copy of your signed proposal title
page and a copy of your NIH Certification with your
hard copy IRB application.
Literature Review Rubric
• http://faculty.weber.edu/kristinhadley/med6085/
• DOI Lookup site
– http://www.crossref.org/guestquery/
NATURE OF THE PROBLEM
Problem Statement: Premise I, Premise II, Interaction problem (III)
Literature Review: supporting details
THE
MASTER’S
PROJECT
PROPOSAL
Premise I - details, analyze, synthesize
Premise II – details, analyze, synthesize
Problem (III) that arises due to I & II
Previous work to address III – include method, instruments
Weaknesses or “holes” in literature
Summary: bring it all together and lead
reader to the need for the study
PURPOSE
Restate problem
The purpose of this research is . . .
The specific objectives are 1. . . . 2 . . .
METHOD
How will you accomplish the objectives stated in
the purpose through your study?
Participants
Instrumentation
Procedure
Data Analysis
Purpose
• In the introductory paragraph, briefly review the
problem or issue
• State the purpose of the proposed study
– May be broken down into objectives which could be
stated as questions or intended outcomes OR
– Could be written as hypotheses
– Example:
• The purpose of this study is to . . . . Specifically, the research
will answer the following questions:
1. How does increased . . . . impact student achievement in 6th
grade social studies?
2. Does increased . . . .. help students feel more connected in the
classroom?
• Should be no more than 1, sometimes 2 pages
Purpose activity
• Purpose activity
Method
• In the introductory paragraph, discuss the study
type and how it will meet the purpose of the study.
• Identify the major tasks that will be completed in
order to achieve the objectives stated in the
purpose. Under each category, provide a detailed
description of the tasks. Tasks could include
–
–
–
–
Identification of participants - specific
Description of measurement instruments
Step by step procedures – specific
Data analysis plan – to answer research questions
DETAILS !!! SPECIFIC ! SPECIFIC !!! DETAILS !!
Data Analysis Plan
• This subsection describes how the data will be analyzed
for your project
• Quantitative: statistical analysis
• Qualitative: how will you present the findings?
Your data analysis plan should enable you to answer your
research question
Statistics Review
(Quantitative Studies)
• Descriptive statistics: describing an outcome
with numbers
– Measures of Central Tendency
• Mean: the average ( X )
• Mode: the most common
• Median: the middle number when the data is put in order
from least to greatest
– When should you use which measure?
More Descriptive Statistics
• Measures of Variability
– Standard Deviation (SD): a measure of how
spread out the data are; roughly, the average of
how far each data point is from the mean
– Range: difference between the lowest data point
and the highest data point
– Interquartile Range: rank order the data, split it
in half and in half again, subtract the median of
the bottom half from the median of the top half
More Descriptive Statistics
• Measures of Association
– Correlation coefficient (r ) : a number between -1
and 1 that describes the relationship between two
data sets
• r=0 if there is no relationship
• r=1 if there is a perfect positive relationship (as one goes
up, the other goes up a perfectly predictable amount)
• r=-1 if there is a perfect negative relationship (as one goes
up, the other goes down a perfectly predictable amount)
• Most correlation coefficients are somewhere in between
– Square the correlation coefficient to show how much
(%) of the second variable can be attributed to
differences in the first variable. This is called the
coefficient of determination (R2).
Association does not mean Causation!
Inferential Statistics
• What is the probability that the difference found
between these samples would have occurred if
there was really no difference in the total
populations?
t-tests
• What is the probability that the differences
between TWO groups has occurred by
chance alone?
The way it is reported:
t(49) = 1.34, p<.05
Degrees of freedom
(typically n-1)
Value
calculated
by the t-test
Probability that this
difference is due to chance
alone
It is likely that there is a real difference
Analysis of Variance (ANOVA)
• What is the probability that the differences between more
than two groups has occurred by chance alone?
The way it is reported:
F(3,53) = 26.26, p<.001
(number of groups -1,
roughly the number of
subjects)
Value
calculated
by the
ANOVA
Probability that this
difference is due to chance
alone
Analysis of Variance (ANOVA)
• ANOVA doesn’t indicate where the differences
occur, just that there is a difference
• Researchers must then pair the means to find
the differences
Analysis of Covariance (ANCOVA)
• Like ANOVA but some covariate (something that is in
common between the two groups) is statistically held
constant when the comparison is calculated.
• For example: comparing the achievement level of
different schools with SES held constant
Chi-Square
• Comparisons when data can’t be averaged
• Nonparametric: without assumptions about
the shape of the data distribution
The way it is reported:
Χ2 (2, N=120) = 12.39, p=.002
(number of groups -1,
number of subjects)
Value
calculated
by the
statistic
Probability that this
difference is due to chance
alone
Regression Analysis
• Method used to develop a predictive equation based
on the relationship between two variables
• Multiple regression is when two or more variables
are used to predict another variable using an
equation
• Confidence interval: accuracy band around the
predicted scores.
Statistical Significance
When a difference is found that appears unlikely to
have occurred by chance, that difference is identified
as being statistically significant. It does not mean the
difference is important, crucial, or practically
significant.
Effect size: a standard measure of the size of the
difference
Standardized mean difference effect size: difference between means
divided by the standard deviation
Data Analysis Plan
• Think – how will the data enable me to answer
my research question?
• Evaluate the data in such a way that you can
answer your question with confidence.