PSYCHOLOGICAL TESTING AND ASSESSMENT QUICKNOTES BY: Kristine Confesor
a.
PSYCHOLOGICAL TESTING AND ASSESSMENT Based on: 2012 Edition book of Kaplan and Saccuzzo
I.
II.
III.
Principles - Introduction Testing - Norms and Basic Statistics for Testing - orrelation and !egression - !eliabilit" - #alidit" - $riting and E%aluating Test Items - Test &dministration Applic!i"ns - Inter%ie'ing Tec(ni)ues - T(eories of Intelligence and Binet Scales $IS*I#+ - T(e $ec(sler Intelligence Scales: $&IS*I#+ $IS*I#+ and $,,SI*III indi%idual Tests Tests of &bilit" in Education Education and - -t(er indi%idual Special Education Standardized d Tests in Education+ Education+ i%il Ser%ice+ Ser%ice+ and - Standardize t(e .ilitar" - &pplications in linical and ounseling Settings Tests - ,ro/ecti%e ,ersonalit" Tests omputers ers and Basic Basic ,s"c(o ,s"c(olog logical ical Scienc Science e and - omput Testing - Testing in ounseling ,s"c(olog" - Testing in ealt( ,s"c(olog" and ealt( are - Testing in Industrial and Business Settings Iss#es -Test Bias -Testing and t(e a' - Et(ics and uture of ,s"c(ological Testing
P$INCIPLES - Introductioneducation ional al test or a set of items that are Ps%c&"l"'icl !es! – educat designed to measure characteristics of human beings that pertain to behaviour. gather erin ing g and and inte integr grat atio ion n of Ps%c&" &"l" l"'i 'ic cll Asse Assess ss(e (en! n! – gath Ps%c psychology-related data for the purpose of making a psychological evaluation that is accomplished through the use of such tools as tests, tests, interv interview iews, s, case case studie studies, s, behavi behaviour oural al observ observati ation, on, and specially designed apparatuses and measurement procedures C"ll)"r C"ll)"r!i*e !i*e + assessor and assesse may work o as “partners” “partners” from initial contract through nal feedback o T&erpe#!ic + therapeutic self-discovery and new understand understandings ings are encouraged encouraged throughou throughoutt the entire assessment process D%n(ic + interactive approach to psychological o assess assessmen mentt that that usuall usually y follow follows s a model model of !" evaluation, #" intervention, $" evaluation + relate raw scores on test items to some dened theoretical or empirical distribution Sc"rin' + process of assigning such evaluative codes or statements to perfor performan mance ce on tests, tests, tasks, tasks, interv interviews iews,, or other other behavi behaviour our samples. any refere reference nce point, point, usuall usually y C#! C#! Sc"r c"re o numerical, divided by %udgment and used to divide a set of data into two or more classications
. /. 0. 1.
T%pes "2 Tes!s !. Indivi Individua duall tests tests – can be be given given to only only one one perso person n at a time time #. 0rou 0roup p test tests s – more more tha than n one one pers person on at at a time time by by a sing single le e1aminer I. bil bility ity 'ests ests – meas measur ure e skills skills in in terms terms of spe speed ed,, accur accuracy acy,, or both a. chiev chieveme ement nt 'est - previ previous ous learn learning ing b. pti ptitu tude de – pote potent ntia iall for for lear learni ning ng or ac2u ac2uir irin ing g specic skill c. Inte Intelli llige genc nce e – person person3s 3s gener general al poten potentia tiall to solve solve problems, adapt to changing circumstances, think abstractly, and prot from e1perience. II. II.
Se!!in's ,. E#c E#c! !i" i"n nll Se!! Se!!in in's 's
*erso erson nalit ality y 'ests est s – m me eas asu ure typi typica call behav ehavio iour ur-- trait raits s, temperaments, and dispositions a. tru truct ctur ured ed ob% ob%ec ecti tive ve"4 "4 prov provid ides es a self self-r -rep epor ortt statement to which the person responds “'rue” or “5alse”, “6es” or “7o” b. *ro% *ro%ec ectiv tive4 e4 prov provid ides es an ambi ambigu guou ous s test test stimu stimulu lus8 s8 response re2uirements are unclear
HISTO$ICAL3 CULTU$AL3 n LEGAL4ETHICAL CONSIDE$ATIONS 9an 9an )yna )ynast sty y4 use use of test test batte batteri ries es two two or more more test tests s used used in con%unction" :ing )ynasty4 national multistage testing program involved local and regional testing centers e2uipped with special testing booths8 series of tests for public o;ce !<==4 >ritish government government copied the ?hinese system for employee employee selection and for its civil service !<<$4 + government established merican ?ivil ervice ?ommission
Scles
&ho are the parties to a 'est( !. 'est )evelo )eveloper pers s and *ublis *ublisher hers s #. 'est +ser $. 'esttaker . ocie ociety ty at lar large
cho chool ol bility bility 'es 'ests ts – iden identi tify fy childr children en w/ spec specia iall needs b. chiev chieveme ement nt 'est 'ests s – evaluate evaluates s accompl accomplish ishmen ments ts or the degree of learning that has taken place c. )iagno )iagnosti stic c 'est 'ests s – tool tool of asses assessme sment nt used used to help help narrow narrow down and identify identify areas of decit to be targeted for intervention Clin Clinic icl l Se!! Se!!in in's 's C"#n C"#nse seli lin' n' Se! Se!!i !in' n's s Geri Geri! !ri ric c Se! Se!!i !in' n's s B#sine B#siness ss n Mili! Mili!r r% % Se!! Se!!in' in's s
C&rles Dr5in higher forms of life evolved partially because of individual di@erences within a species di3erences e1ist in human sensory 6rncis Gl!"n 4 indi%idual di3erences and motor functioning functioning such as reaction reaction time, visual visual acuity, and physical strength. developed ped the produ productct-mom moment ent correl correlati ation on Krl Krl Pers" Pers"n n4 develo techni2ue formulate ated d a genera generall descri descripti ption on of 7il&el( 7il&el( M8 7#n! 7#n!4 formul human abilities with respect to variables such as reaction time, percep perceptio tion, n, and attentio attention n span8 span8 focuse focused d on how people are similar based on 9(es Mceen C!!ell : coined the term mental test based 0alton3s work on individual di@erences in reaction time. lso instrume instrumenta ntall in foundi founding ng the ,s"c(ological o orporation orporation with a goal for the “advancement of psyc psycho holo logy gy and and the the prom promot otio ion n of the the usef useful ul applications of psychology. psychology. as C&rles Sper(n: originating the concept of test reliabilit" as well as building the mathematical framework for the statistics of factor anal"sis ;ic!"r Henri: collaborated with >inet on papers suggesting how mental tests could be used to measure higher mental processes E(il Krepelin: early e1perimenter with the word association techni2ue as a formal test Li'&! 7i!(er: little known founder of clinical psychology and founded founded the %ournal ,s"c(ological linic linic having its rst article entitled as “?linical *sychology”
Al2re Bine! n ;ic!"r Henri 4 !
PSYCHOLOGICAL TESTING AND ASSESSMENT QUICKNOTES BY: Kristine Confesor
Bine! Si("n Scle – !AB=, contained $B items of increasing di;culty and was designed to identify intellectually subnormal individuals8 concept of (en!l 'e was made. D*i 7esc&ler, !A$A4 intelligence was the aggregate or global capacity of the individual to act purposefully, to think rationally, and to deal e@ectively with the environment o Gr"#p In!elli'ence Tes!: came into being in the + in response to the military3s need for an e;cient method of screening the intellectual ability of &orld &ar I recruits S!n2"r Bine! Scle – revision made by 'erman on !A!C 77I- they army re2uested the assistance of 6erkes, * president to create a committee of distinguished psychologists to develop # structured group tests of human abilities4 'hey rmy lpha and the rmy >eta. Ac&ie*e(en! Tes!s: provide multiple choice 2uestions that are standardiDed on a large sample to produce norms against which the results of new e1aminees can be compared.
SCALES O6 MEASU$EMENT Pr"per!ies 15 :agnitude – property of moreness. scale has the property of magnitude if we can say that a particular instance of the attribute represents more, less, or e2ual amounts of the given 2uantity than does another instance. 25 E2ual Intervals – if the di@erence between two points at any place o the scale has the same meaning as the di@erence between two other points that di@er by the same number of scale units. 65 bsolute Fero- when nothing of the property being measured e1ists. T%pe "2 Scle
M'ni!#e
7ominal Grdinal Interval Hatio
7o 6es 6es 6es
E#l In!er*ls 7o 7o 6es 6es
A)s"l#!e < 7o 7o 7o 6es
Mes#re(en! "2 Pers"nli!%
Pers"nli!% Tes!s @,<-,0< – measured presumably stable characteristics or traits that theoretically underlie behaviour. o
o
o
o
o
o
o
o
– relatively enduring dispositions Tri!s tendencies to act, think, or feel in a certain manner in any given circumstance" that distinguish one individual from another. $")er! S 7""5"r!& + measure of ad%ustment and emotional stability that could be administered 2uickly and e;ciently to group of recruits *ersonal )ata heet" 7""5"r!& Pers"nl D! S&ee! – an early structured personality test that assumed that a test response can be taken at face value Pr"ec!i*e Tes!s + an individual is assumed to “pro%ect” onto some ambiguous stimulus his or her own uni2ue needs, fears, hopes, and motivation. $"rsc&c& In)l"! !es!4 highly controversial pro%ective test that provided an ambiguous stimulus and asked the sub%ect what it might be T&e(!ic Appercep!i"n !es!4 pro%ective test that provided ambiguous pictures and asked sub%ects to make up a story Minnes"! M#l!ip&sic Pers"nli!% In*en!"r% @MMPI4 structured personality test that made no assumptions about the meaning of a test response. uch meaning was to be determined by empirical research ,?P64 structured personality test based on the statistical procedure of factor analysis
C#l!#re– socially transmitted behaviour patterns, beliefs, and products of work of a particular population, community, or group of people Henr% S G"r 4 highly instrumental in getting >inet3s test adopted for use in various settings in the + who raised 2uestions about how meaningful such tests are when used with people from various cultural language and backgrounds ;er)l C"((#nic!i"n + language is a key yet sometimes overlooked variable in the assessment process.
6re#enc% Dis!ri)#!i"ns + displays scores on a variable or a measure to reKect how fre2uently each value was obtained. P"si!i*e Se5 + relatively few scores fall at the high end of the distribution e.g. the test was too di;cult" Ne'!i*e Se5 + &hen relatively few of the scores fall at the low end of the distri bution e.g. the test was too easy" Percen!ile $n+ an e1pression of the percentage of people whose score on a test or measure falls below a particular raw score, or a converted score that refers to a percentage of testtakers8 contrast with percentage correct , 7 B8N 9 100 7 percentile rank of i ,7 percentile rank +7 score of interest B 7 number of scores belo' + N7 total number of scores
Percen!iles + specic scores or points within a distribution8 divide the total fre2uency for a set of observations into hundredths.
DESC$IBING DIST$IBUTIONS 15 :ean – arithmetic average score in a distribution 25 tandard )eviation – appro1imation of the average deviation around the mean; s2uare root of variance
=
C#l!#re n Assess(en!
Grdinal4 I 'ests Interval4 'emperature on ? and 5 Hatio4 Jelvin cale8 6ards, peed
σ
65
√
(
2
)
Σ X – X N
Lariance – measure of variability e2ual to the arithmetic mean of the s2uares of the di@erences between the scores in a distribution and their mean
σ ²
=
(
2
)
Σ X – X N
N"r(s n Bsic S!!is!ics 2"r Tes!in'
Tes!s– devices used to translate observations into numbers P#rp"se "2 S!!is!ics Descrip!i*e S!!is!ics + are methods used to provide a o concise description of a collection of 2uantitative information In2eren!il S!!is!ics + used to make inferences from o observations of a small group of people known as a sample to a larger group of individuals known as population.
<5
F core – transforms data into standardiDed units that are easier to interpret8 di@erence between a score and the mean, divided by the standard deviation
PSYCHOLOGICAL TESTING AND ASSESSMENT QUICKNOTES BY: Kristine Confesor
X , – X z =¿
0.
Sc!!erpl"! – useful in revealing curvilinearity in a relationship eyeball guage” of how curved the graphy is
$e'ressi"n + analysis of relationships among variables for the purpose of understanding how one variable may preic! another. Si(ple $e'ressi"n + M predictor variable"8 6 outcome variable"8 results in an e2uation for a regression line. $e'ressi"n Line + line of best t M#l!iple $e'ressi"n + takes into account the intercorrelations among all variables i nvolved8 correlation among predictor scores
¿
S
- orrelation and !egression E1pression of degree and direction of correspondence between two thing. C"ecien! "2 C"rrel!i"n + numerical inde1 that e1presses this relationship4 It tells us the e1tent to whic h M and 6 are co-related. Pers"n r +when the relationship is liner and when the two variables being correlated are c"n!in#"#s C"ecien! "2 De!er(in!i"n + indication of how much variance is shared by the * and the =* variables. Sper(n $&" + rank-order correlation coe;cient when sample siDe is fewer than $B and when both sets of measurements are in ordinal or rank-order form. G$APHIC $EP$ESENTATIONS O6 CO$$ELATION ,. Bi*ri!e is!ri)#!i"n . Sc!!er Di'r( /. Sc!!er'r(
Me!-nl%sis + family of techni2ues used to statistically combine information across studies to produce single estimates of the statistics beig studied C#l!#re n In2erence
- !eliabilit" - #alidit" - $riting and E %aluating Test Items - Test &dministration