Rafael Perera Basic teaching advice Know your audience Know your audience Create a knowledge gap Give a map of the main concepts Decide which ones to focus on Use plenty of examples Let them do the workthinking ID: 932282
Download Presentation The PPT/PDF document "How to Teach S tatistics in EBM" is the property of its rightful owner. Permission is granted to download and print the materials on this web site for personal, non-commercial use only, and to display it on your personal computer provided you do not modify the materials and that you retain all copyright notices contained in the materials. By downloading content from our website, you accept the terms of this agreement.
Slide1
How to Teach Statistics in EBM
Rafael Perera
Slide2Basic teaching advice
Know your audience
Know your audience!
Create a knowledge gap
Give a map of the main concepts
Decide which ones to focus on
Use plenty of examples
Let them do the work/thinking
Slide3Main Concepts
Bias
and Measurement error
P values and Confidence Intervals
Which Statistical tests are needed
and when
Correlation and Association
Models / Regression and alternatives for
Adjustment
Survival
Analysis
Meta Analysis
Statistics
for Diagnostic
Studies
Slide4There is a time and place…
Slide5Fundamental Equation of Error
Measure = Truth +
Bias
+
Random Error
Use
good study design
Use
large numbers
Researcher
Critically Appraise
Design
Confidence
Intervals
and
P-values
Reader
Slide6true result
Bias
low high
Random error
high low
Bias
versus
Random error
Slide7Bias and Measurement error
Groups of 3-4 people
1 – subject
2 – measurers
Measurers – measure (twice)
and record the head size of the subject. Keep measurements hidden.
Slide8Bias and Measurement error
Intra-Observer
variability
Measurement error
Same answer
Varied by < 0.5 cm
Varied by < 1cm
Varied by < 2 cm
Varied by >2 cm
Slide9Bias and Measurement error
Inter-Observer
variability
Measurement error
Same answer
Varied by < 0.5 cm
Varied by < 1cm
Varied by < 2 cm
Varied by >2 cm
Slide10Bias and Measurement error
Bias
Included ears?
Included nose?
Which part of the head?
Other?
Slide11Does it matter?
In paediatric practice following meningitis, a head circumference that increases by 7mm in a day will result in urgent head imaging
In obstetrics measurements of the fundal height can vary by up to 5cm (the difference between having a baby delivered early due to IUGR or not when opposite occur)
The question is can you reproduce the test in your setting and will it perform as well in your setting
Slide12Measuring Random error
Most things don’t work!
Slide13Two methods of assessing the role of random error
P-values
(Hypothesis Testing)
use statistical test to examine the
‘
null
’
hypothesis
if p<0.05 then result is
statistically significant
Confidence Intervals
(Estimation)
estimates the range of values that is likely to include the true value
Relationship between p-values and confidence intervalsIf the ‘no effect’ value falls outside the CI then the result is statistically significant
Slide14The Steps in Testing a Hypothesis
State the
null hypothesis
H
0
Choose the
test statistic
that
summarizes
the data
Based on H
0
calculate the probability of
getting the
value of the
test statistic
Interpret the
P-value
Slide15Some Statistical tests
Comparing groups
T-tests (1 or 2 groups, normally distributed)
Chi-squared (2 or more groups, categorical or binary data)
Mann-Whitney U (2 groups, non-normal data)
Log-rank test (2 groups, survival data)
ANOVA (multiple groups, normally distributed)
…
Tips:
Understand what the hypothesis being tested is
Use the p-value to assess the level of evidence against it
(Experienced) Assess if the test was adequate for the question and data analysed
Slide16Slide17Hand outs
Incidence/ Prevalence and CI
Survival analysis
Regression models / Adjustment
Linear association / Correlation
Confounding / Odds Ratios / Logistic
Regression
Diagnostic Tests
Meta-analysis
Slide18Slide19Reading confidence intervals
Slide20Clinically significant
Vitamin X shortens a 5 day cold
Would you take it twice per day if it shortened the cold by:
Clinically significant
Vitamin X shortens a 5 day cold
Would you take it twice per day if it shortened the cold by:
50%
Clinically significant
Vitamin X shortens a 5 day cold
Would you take it twice per day if it shortened the cold by:
50%
20%
Clinically significant
Vitamin X shortens a 5 day cold
Would you take it twice per day if it shortened the cold by:
50%
20%
10%
Clinically significant
Vitamin X shortens a 5 day cold
Would you take it twice per day if it shortened the cold by:
50%
20%
10%
5%
Clinically significant
Vitamin X shortens a 5 day cold
Would you take it twice per day if it shortened the cold by:
50%
20%
10%
5%
1%
Slide26(a)
(b)
(c)
(d)
Minimum clinical
Important difference
No difference
Which are clinically significant?
0 10 20
Slide27Thank you
Slide28Extras
Slide29Different types of measurements use different types of statistics
Dichotomous:
Male,female OR infected, non-infected
Categorical:
Red, green, blue OR
Ordinal:
Nil, +, ++ of glucose
Interval:
temperature
STATISTICS
Proportion, Risk
Mode, ProportionsMode, Median? Mean, Median
Slide30> t
wo samples
Independence between two or
more
variables
Parametric
Non parametric
Between means for
continuous data
Between
distributions
Hypothesis
testing and
a
ssessing
d
ifferences
Parametric
ANOVA
Sign test for related
samples
Rank sum test for
independent samples
Kruskal Wallis
T test difference for
related samples
Non parametric
T test for independent
samples
McNemar
’
s test for
related groups
Between
one observed
variable and a theoretical
distribution
X
2
test for goodness
of
fit
X
2
test for
independence
Two samples
One
sample
vs. H
0
One
sample
vs. H
0
Z score
equal
proportions
Z score
Between proportions for
categorical data
Flowchart of Statistical Tests for Hypothesis Testing
Slide31Flowchart of Statistical Tests for Hypothesis Testing
Between distributions
Between one observed variable and a theoretical distribution
Independence between two or more variables
c
2
test for goodness of fit
c
2
test for independence
McNemar’s test for related groups
Flowchart of Statistical Tests for Hypothesis Testing
Between means for continuous data
Two samples
t-test independent samples
Rank sum test for independent samples
Sign test for related samples
t-test difference for related samples
ANOVA
Kruskal – Wallis
Parametric
> two samples
Non Parametric
Parametric
Non Parametric
Slide33Flowchart of Statistical Tests for Hypothesis Testing
Between proportions for categorical data
One sample vs. H0
Two samples
Z-score
Z-score equal proportions
Summarising proportions
One sample: Risk, Odds
Two samples: Relative risk, Odds ratios, Risk differences