Qualities of Good Test.pdf

no

1. Qualities of Good Test

2. Test
Atestinsimplewordis
“Astandardizedsituationthatprovidesan
individualwithascore”
Itmeans:
•Anexamination
•Anykindofdeviceorprocedurefor
measuringability,achievements,and
interestetc.

3. Test
•Amethodtodetermineastudent's
abilitytocompletecertaintasksor
demonstratemasteryofaskillor
knowledgeofcontent.
•Sometypeswouldbemultiple-choice
tests,oraweeklyspellingtest.
•Itiscommonlyusedinterchangeablywith
assessment,orevenevaluation,itcanbe
distinguishedbythefactthatatestis
oneformofanassessment.
•Itisatechniqueofobtaininginformation
neededforevaluationpurposes.(Tests,
Quizzes,MeasuringInstruments)are
devicesusedtoobtainsuchinformation.

4. Test
•Agoodtestshouldhaveapositiveeffecton
learningandteaching.
•Atestshouldaimatspecifyingareasof
difficulties.
•Thewholeclassorindividualstudentsmay
experiencesothatadditionalpracticeand
correctiveexercisesshouldbegiven.
•Agoodtestshouldalsomeasurestudents’
performancewithoutsetting“traps”for
them.
•Theteachershoulddevelopitwellto
provideanopportunityforstudentsto
showtheirabilitytoperformcertain
actions.

5. Test
Thetestshouldenabletheteacherstofindoutwhichpartsofthe
coursecausedifficultyfortheclass.
Teacherscanevaluatetheeffectivenessofthesyllabusaswellasthe
methodsandmaterialstheyareusing.
Awrittentestisaclassicexampleofformalassessmentwhere
studentsshouldbeawareofbeingtestedforareason.

6. Qualities of
Good Test
Some qualities of good test are:
•Validity
•Reliability
•Usability
•Item analysis for achievement
test
•Item discrimination
•Item difficulty
•Building item bank

7. Validity
“Validityistheextenttowhichatestmeasureswhatitclaims
tomeasure.”
Validityisthedegreetowhichatestmeasureswhatitis
supposedtomeasured.
Example
Meterisusedforthepurposeofmeasuringlengthwhilescaleis
usedforthepurposeofmeasuringweight.Ifmeterisusedfor
measuringlength,itwillbevalidbecauseitissupposedto
measurelengthandifitisusedtomeasureweight,itwillnotbe
valid.

8. Validity
Whatismeasuredandhowconsistentlyitismeasured.
•EbelandFrisbie(1991)—“Thetermvalidity,whenappliedtoa
setoftestscores,referstotheconsistency(accuracy)with
whichthescoresmeasureaparticularcognitiveabilityof
interest.”
•C.V.Good(1973)—“Theextenttowhichatestorother
measuringinstrumentfulfilsthepurposeforwhichitisused.”
•Freeman(1962)defines,“Anindexofvalidityshowsthe
degreetowhichatestmeasureswhatitpurportstomeasure,
whencomparedwithacceptedcriteria.”
•Lindquist(1942)hassaid,“validityofatestmaybedefinedas
theaccuracywithwhichitmeasuresthatwhichitisintended
tomeasure,orasthedegreetowhichitapproaches
infallibilityinmeasuringwhatitpurportstomeasure.”
•Validityofanevaluationdeviceisthedegreetowhichit
measureswhatitisintendedtomeasure.

9. Validity
•Validityisalwaysconcernedwiththespecificuseofthe
resultsandthesoundnessofourproposedinterpretation.
•Itisnotalsonecessarythatatestwhichisreliablemayalso
bevalid.
Forexample:
•Theresultsofavocabularytestmaybehighlyvalidtotest
vocabularybutmaynotbethatmuchvalidtotest
compositionabilityofthestudent.

10. Types of Validity
•Content Validity
•Construct Validity
•Concurrent Validity
•Predictive Validity
•Face Validity

11. Content
Validity
Contentvalidity
isthedegreeto
whichatest
measuresan
intended
contentarea.
Atestwithgood
validitycovers
allthecontent.
Contentvalidity
isofprime
importancefor
achievement
test.
Itismeasured
by expert
judgement.

12. Construct Validity
•Constructvalidityisthedegreetowhichatestmeasuresan
intendedhypotheticalconstruct.
•Aconstructisameasuringoftraitsuchaslearning,anxiety,
creativity,curiosity,scientificattitude,intelligencewhich
explainsbehavior.
•Aconstructvaliditycannotbeseen,onlyitseffectcanbe
observede.g.,intelligencecannotbeseenbutwecan
observethatsomestudentslearnfasterthanothers.

13. Concurrent
Validity
•Concurrentvalidityisthedegree
towhichthescoresonatestare
relatedtothescoresonanother
alreadyestablishedtest
administeredatthesametime.

14. Predictive Validity
PREDICTIVE VALIDITY IS THE DEGREE
TO WHICH A TEST CAN PREDICT HOW
WELL AN INDIVIDUAL WILL DO IN
FUTURE.
PREDICTIVE VALIDITY IS OF PRIME
IMPORTANCE FOR TESTS WHICH ARE
USED FOR THE PURPOSES OF
ADMISSION OR SELECTION OF
STUDENTS.

15. Face Validity
Facevalidityreferstowhethertheinstrumentlooksasthoughitismeasuringthe
appropriateconstruct.
Facevalidityshouldnotbeconsideredstrongevidenceforaninstrument’svalidity
For Example
•Concept=pain level
•Measure=verbal rating scale “rate your pain from 1 to 10”.
Face validity is sometimes considered a subtype of content validity.

16. Reliability

17. Reliability
•Reliabilityisthedegreetowhichatestconsistently
measureswhateveritmeasures.
•Areliabletestgivesthesamescoreswhenadministered
andreadministeredwhileanunreliabletestdoesnotgive
thesamescore.
•Example
•Ifanintelligencetestwasunreliable,thenastudent
scoringanIQof120todaymightscoresanIQof140
tomorrowand95dayaftertomorrow.
•IfthetestwasreliableandstudentIQwas110,then
wewouldnotexpectfluctuationinscoresfromtesting
totesting.Ascoreof105wouldnotbeunusualbuta
scoreof145wouldbeveryunlikely.

18. Reliability
Reliabilityisexpressednumericallyasacoefficient.
Ahighcoefficientindicateshighreliabilityandvice
versa.Thevalueofcoefficientwouldbe1.00.
Avalidtestisalwaysreliable,butareliabletestis
notnecessarilyvalid.
Reliabilityisthedegreetowhichanassessmenttool
producesstableandconsistentresults.
Thereliabilityofanassessmenttoolistheextentto
whichitconsistentlyandaccuratelymeasures
learning.

19. Types of
Reliability
Test-Retest
Reliability
Equivalent
Forms
Reliability
Split Half
Reliability
Kuder
Richardson
Method

20. Test-Retest Reliability
•Test-retestreliabilityisameasureof
reliabilityobtainedbyadministeringthe
sametesttwiceoveraperiodtoagroupof
individuals.
•ThescoresfromTime1andTime2canthen
becorrelatedinordertoevaluatethetest
forstabilityovertime.
Example
•Atestdesignedtoassessstudentlearningin
psychologycouldbegiventoagroupof
studentstwice,withthesecond
administrationperhapscomingaweekafter
thefirst.Theobtainedcorrelation
coefficientwouldindicatethereliabilityof
thetest.

21. Equivalent Forms Reliability
•Equivalentformsreliabilityisameasureofreliabilityobtainedby
administeringdifferentversionsofanassessmenttool(bothversionsmust
containitemsthatreviewthesameconstruct,skill,knowledgebase,etc.)to
thesamegroupofindividuals.
•Thescoresfromthetwoversionscanthenbecorrelatedinorderto
evaluatethereliabilityofresultsacrossalternateversions.
Example
•Ifyouwantedtoevaluatethereliabilityofacriticalthinkingassessment,
youmightcreatealargesetofitemsthatallpertaintocriticalthinkingand
thenrandomlysplitthequestionsupintotwosets,whichwouldrepresent
theequivalentforms.

22. Split Half
Reliability
•Split-halfreliabilityisdetermined
byestablishingtherelationship
betweenthescoresontwo
equivalenthalvesofatest
administeredtoagroupatone
time.
•Inthistype,testitemsaredivided
intotwoequalhalves.
•Example
•Supposeatestconsistsof20
items,itcanbedividedintotwo
equalhalvesasunder:
•Odd numbers:
1,3,5,7,9,11,13,15,17,19
•Even numbers:
2,4,6,8,10,12,14,16,18,20

23. Kuder
Richardson
Method
Kuder-Richardson method
estimates reliability.
It measures reliability from a
single administered of a test
with the help of formulas.
It measures the internal
consistency of a test.

24. Factors Influencing
Reliability
•Lengthofatest
•Spreadofscores
•Difficultyofatest
•Objectivity

26. Usability

27. Usability
•“Theextenttowhichthe
productcanbeusedby
specifieduserstoachieve
specifiedgoals with
effectiveness,efficiency,and
satisfactioninaspecified
contextofuse.”
•Itaimedatfindingusability
problemsintheuser’s
interface/System,providing
qualitativeandquantitative
dataaboutpeopleandtheir
behaviororattitudes.
“A usability defectis
a product
characteristicthat
makes itdifficultor
unpleasant forusers
to accomplish tasks
supported by the
product.”
“Kahn andPrail”

28. USABILITY
EVALUATION
•There are two types of
user evaluations are
carried out:
•Formative evaluations
•Summative evaluations
“Whenthecooktastesthesoup,that’s
formative;whenthegueststastethesoup,
that’ssummative.”
RobertStakes

29. Formative VsSummative
FormativeAssessment
Conductedduring design as part of an interactive process, allowing
the designers to see how their target user/audience responds to the
interface. These methods provide qualitative data.
Heuristic Evaluation, User Interface Inspections, Thinking-aloud Testing, Mixed
Usability walkthroughs, and Cognitive Walkthroughs are some methods that can be
used for formative evaluation.
SummativeAssessment
Conducted after completion of products, used in product
comparisons, to identify problems prior to a redesign.
Summative studies are designed to yield quantitative data
that can be tested for statistical significance.

30. Characteristics of
Usability

31. Time for Administration
Theshortertestarefavoredbuttheyhavelowreliability.Asafeprocedureis
toallotasmuchastimeasisnecessarytoobtainvalidandreliableresults.
Somewherebetween20to60minutesoftestingtimeisprobablyagood
guide.

32. Ease of
Administration
Atestwillbeeasytoadminister
when
•Directionswillbesimple
andclear.
•Sub-testswillbefew.
•Timeoftestwillbe
suitable.
Thecomplicateddirections,
severalsub-testandshorttime
adverselyaffectthevalidityand
reliabilityoftest.

33. Ease of Scoring
•Thosetestsare
favoredthatoffer
easeandeconomyof
scoringwithout
sacrificingscoring
accuracy.

34. Ease of
Interpretation
Whenthetestresultsare
presentedtothepupilsor
parents,easeofinterpretation
andapplicationareespecially
important.Ifresultsare
correctlyinterpreted,they
contributetoimportant
educationaldecisions.

35. Availability of
Equivalent Forms
Equivalentformsoftestmeasure
thesameaspectofbehaviorby
usingtestitemsthatarealikein
contents,difficultylevelandother
characteristics.Theequivalent
formsoftestareoftendesirable.
ForExample
•Ifateachermayfeelthatscore
ofstudentinanachievement
testisverylow,hemayeasily
checkbyadministeringthe
equivalentform.

36. Thank You