1

 Critically Appraising the Evidence:
 Basic Statistics

2

 Clinical Statistics Calculator (Excel)
 Statistics for:
 General
 Samples and Populations
 Means and Medians
 Normal and Skewed Distributions
 Variability, Variance, and Standard Deviation
 Confidence Intervals
 pvalues
 Generalized 2x2 Clinical Table
 Practice Exercises

3

 If available, find the best evidence in secondary sources where analysis
has already occurred.
 If not preassessed, use critical appraisal worksheets to help you
through the process.

4

 Understanding the Limitations of the Author’s Analyses and
Interpretations of the Data
 Assessing Internal Validity
 Assessing External Validity
 Identifying Potential Confounding Variables

5

 View movie as:
 QuickTime (.mov)
 Flash (.swf)
 Doubleclick on video for fullscreen mode.

6

 Population
 The entire group of all potential patients
 Sample
 The patients included in the study
 Representative of the population
 Clinical samples consist of treatment groups
 Nonclinical samples consist of control groups

7

 Often called arithmetic average or just average
 Sum of all of the data points divided by the number of data points

8

 If there are n data points lined up in numerical order the median is the
one in the exact middle or the average of the 2 middle points if there
is an even number of data points.
In other words it is the n^{th}/2 data point if n is odd
and the ((n1)^{th}/2 + (n+1)^{th}/2)/2 data point if n
is even.

9

 Line up all of the data points in increasing order.
 The one in the middle is the median.
 If there is no clear single midpoint (i.e. there is an even number of
data points), the median is halfway between the two middle points.
 So if 0, 1, 2, 4 were our data set, 1.5 would be the median.

10

 Normal
 Symmetric, bellshaped distribution where the frequency of data within
an interval is greater the nearer it is to the mean
 Skewed
 Asymmetric distribution
 More data to one side of the mean than the other

11

 Additional outliers
 usually increase/decrease mean more than median
 Skewed data
 Mean shifted toward the tail (i.e. the side where the data points are
more spread out)
 Median unaffected by distribution/shape
 Always the middle value regardless of where the other points lie

12

 Mean (i.e. Arithmetic Average)
 Used when data is approximately normally distributed
 Median
 Sometimes used with skewed data due to its robustness
 Preferred on Likert scales (survey data) since values are ranked but
their differences are not clearly quantifiable.
 After all, what number does “agree” minus
“disagree” equal? Regardless of how one quantifies these
differences, the median remains the same, but the mean depends on the
scale.

13

 either:
 the data is not normal and is skewed one way or another
 there is at least one outlier with a lot of leverage

14

 Variability / Dispersion
 How data points are distributed
 Variance
 Sum of the squares of the differences between each data point and the
sample mean, divided by the total number of data points
 Standard Deviation
 Square root of the variance

15

 Range in which one would expect the values of the outcome variables to
fall into when replicating the experiment given a quantifiable
probability of error.

16

 Probability of obtaining a result as extreme as the observed one if the
data were a result of chance.
 Small pvalues mean the result could not likely be the result of chance.
 Often researches use pvalues such as 0.10, 0.05, 0.01, and smaller.

17


18

 Critical Appraisal Practice Exercises

19

 EBM Glossary
 Critical Appraisal Practice Exercises

20

