MEASURE Or Ass0CIAMDN)

to opase ate oj asticulan

dusease expeRienied ay people to
dsease

astualioO

) RisKs 2 odds
De dusig
a cdyund pernel
Qdds duaing
dhat plhiod
lomypase cesease
2) Relai Ve Y5k
devolop pert pxohab1ty blw
erpoeol & no aposed groups

dsenne w cpcad poru

Y1Sk= pIohabtty Dt
rloetve yske
Prob
 PosItive
a550Sliso

dhe
3) Atnbut able
3 )Sk

Midoue Tate
expDsed gro
Risk
lezpos
NO 1SK

tve assouation
-Ve ussDuation
Risk

duyease
hevy
y thos e wviuals eposeo
YiSK talor
 obeevaoal

) analytal
txpeunNal studly

Here, yoYOatiD
Changuy he e son NOnt
stucly that (5
epesiMental

Des Ciupt wo ely

* humited
dis ease a opulatoo
They duLJea sese
desbo desea by
 oher
dlisoq)twe Stucly 15 done
the cl1sease

yoes
SUNplegt methods
the Desentoto
YepOAt a pati ent
wse,ADR,
Ctc

2) ase se ie5 it s the collerlo

eports

B>
Xalso knouo) paoatoy sues

Cotho
oup
 Mgpohess
stuly
2) ohost tudy
) case Coniso!
Londsol siudy
to

selot
hospiias mai chi
CUDOSU0 theagn

uderetatn

Popalato
-nLpsed
Case

( coeiscl
also as lasyjitudnl
nidunee tudi cs

gpes o} choxt srudy (anBN)

cohort Shlanthis
hus

m Jhe

2) Re)05pcLtwe toi0at siucy (o) -(olUaw)

typ,ha has scod
Jhe VeSi gaton
DIseused
eseoosed

DSeasec
not chao
 ) ConbunaiuoNal Sucly
lanb ualion both

ExPERIMENSiALERVENION SIUDY

torisgleol CivumtaniS

(Yteia
5ab1sty
be a nievealisO

2)d shoud lo Canto|

qoup
patiup nis Shouud be

groups
ciesged kunan a auna studly

anumw ) invttypes
In 3)-vivo.
2) Inviv0
 xpesuDetal study
types

DRandlomized Umbaol hinls

RCT preqiams
by mou dhlrapies

mest lho

Steps
PYoto (o |

selt YegerLnee populatico

5elot eipemental ))

Ronalomiadioo
 Hesvertio)

ollow up

Assest

ypes

3) g055 oveY design

Rautdomigeal cOtvalea als

2) Non -

he
wethod that is

Yndlom
ERRORS

dyucdl a
y beiny
obs eTvel value 2tashe

Sypes
) Runolom 0

* asO Knoo as

hu Yalue

jaub
and jauel
data
 arclan
meQsusamartl e30 '- due to

human mst akes

.erso dhet

ohole pepellctuin
Brros hat do

ot O r beecause

devae
Bias
Jenclcneuy
duslio ton a tiue value.

*ypes Bias

) lnHowndg bias
2) HervieueY bias

3) ngormatu
 4) Arialgthcal has
6) seloutum has

VALI DÍY

edabiluty
Types
valdady valeeluty
estobbsb
melod

Jho to whih

akauat y 7epreperd the

dae Lonept
3) ue valuhly
 Screting 4ppeass
to be valid

Jhe dayee to chech atest

Consi Stent SOres

Los1ss ently measwl

Tyesi ypes

) 1e5t 1etest eiabllu

2)
3) nta rateo
4) ntes ctes

selabity
nshmOnt
)stabily
1esut wohen uoed
lt uously" also knoun is test-Yetest
 stma

Suml

) aquwalene t tw osIslGny
lohuub diyhere Yans procwths
SLOAl or Sune palint
Also hnow nter aten

ielabelay
gakny vclabty,

4) no: of ebsevatoo
15 a vasabie yeladoel to clwease

ohb
dot1buted w

ýrous to

etc
55o(atod
be

bl alohol ung cane es