Christian Jones MD FACS Department of Surgery Johns Hopkins School of Medicine stats mean typical average median middle number mode most frequent number normal type of distribution equal about a mean bell curve mean median mode ID: 727795
Download Presentation The PPT/PDF document "ABSITE statistics: the absolute basics" 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
ABSITE statistics: the absolute basics
Christian Jones, MD, FACS
Department of Surgery
Johns Hopkins School of MedicineSlide2
stats
mean: typical “average”
median: middle number
mode: most frequent number
normal: type of distribution, equal about a mean, “bell curve”, mean = median = modeSlide3
terms
in
cidence:
in
terval of time
p
revalence:
p
oint in timeSlide4
types of studies
cohort study: prospective, nonrandom assignment to treatment group
case-control study: retrospective, nonrandomSlide5
biases
selection bias: test groups end up being different, e.g., volunteers vs. population
measurement bias: aka system bias, actual measurement is inaccurate, e.g., recall bias
exposure bias: aka treatment bias, treatments not applied equally or as originally planned, e.g., dead patients “withdraw”Slide6
test parameters
sensitivity: how well does the test see the disease?
specificity: how well does the test make sure it really is the disease?
positive predictive value: if the test says yes, how certain can you be that it’s right?
negative predictive value: if the test says no, how certain can you be that it’s right?Slide7
copyright, etc.
This document is entirely the work of
Christian Jones
(
on-call@christianjones.md
) and is hereby released into the public domain, with
no rights reserved
.
This is version 1.1, last modified on 28 January 2016 by
Christian Jones
. Contact the author by email or on
Twitter
(
@jonessurgery) for questions, comments, or concerns.