Psych Assessment (Lec) 2.

Retrospective Assessment- looking into

the past events
Level A- for novice
3. Remote Assessment- assessment through
Level B- psychometrician
computers if client is not in physical
Level C- master level proximity (telepsychology)

Draw a person test- test for anything 4. Ecological Momentary Assessment- “in
the moment” evaluation
- study different types of tests
Psychological Psychological
Testing Assessment Process of Assessment

process of Gathering and Collaborative- assessors and assesse work

measuring integration of together as partners
psychology-related psychology-related Therapeutic- self-discovery and new
variables by means data for the purpose understandings
of devices or of making a
procedures designed psychological Dynamic- an interactive approach to
to obtain a sample evaluation assessment that follows a model of
of behavior evaluation, intervention of some sort, and
evaluation
only scoring, inferring from all
administration, data Tools of Psychological Assessment
interpretation
1. The test- a measuring device or procedure
1 test gets one - typically
sample (e.g. individualized  Psychological Test- device or
personality test) procedure designed to measure
can be individually/ variables related to psych
group
o Content- subject matter
tester is not key to assessor is key to
process the process o Format- form, plan, layout,
technician like educated of tools of etc
skills are needed evaluation o Item- a specific stimulus to
which a person responds
Results are from logical problem- overtly and this response is
test scores solving approach being scored or evaluated
there needs to be a o Administration procedure-
report - needs to be one-to-one basis or group
in third person pov administration
o Score- code or summary of
statement that reflects and
Varieties of Assessment evaluation of performance on
a test.
1. Educational Assessment- tests to
evaluate abilities and skills of student
 o Scoring- the process of quantitative and/or qualitative information
assigning scores to regarding those actions.
performances
 Naturalistic observation
o Cut Score- reference point
divided by judgement and 6. Role play- defined as acting an
used to divide a set of data improvised or partially improvised part in a
into classification stimulated situation.
o Psychometric Soundness-
7. Other tools- computer, physiological
technical quality
device
o Psychometrics- science of
psychological measurement Who are the parties

2. The interview- method of gathering 1. Test developer- creates the test

information through communication 2. Test user- buys the test from the
involving reciprocal exchange developers
3. Test taker- takes the test
 Panel interview (board interview)-
more than one interviewer participate Psychological Autopsy- if patay na ang
in the assessment person
 Motivational interview- used by
4. Other parties- orgs, companies, gov
counselors and clinicians to gather
that could sponsor the development
information about some problematic
of the test.
behavior, while simultaneously
attempting to address it Settings
therapeutically.
1. Educational setting
3. Portfolio- sample of one's ability and
 Achievement test – evaluates
accomplishment
accomplishments or the degree of
4. Case history data- refers to records, learning that has taken place.
transcripts, and other accounts in written,  Diagnostic test - used to help narrow
pictorial, or other form that preserve archival down and identify areas of deficit to
information, official and informal accounts, be targeted for intervention.
and other data and items relevant to assessee o Diagnosis – description or
conclusion reached on the
 Case study- report or illustrative
basis of evidence and opinion
account concerning a person or an
 Informal Evaluation-
event that was compiled on the basis
nonsystematic assessment that leads
of case history data.
to the formation of an opinion or
 Groupthink- w/ your own
attitude. e.g. anecdotal records
colleagues
2. Geriatric settings- for the oldies. Usually
5. Behavioral Observation – monitoring
an individualized test to screen for or
actions of others or oneself by visual or
diagnose behavior problems.
electronic means while recording
3. Counseling Setting- with the goal to more suitable for a client not mentally
improve client in terms of adjustment, or capable
some related variable
Where
4. Clinical Setting- improvement of client’s
Test catalogues- brief description of the test
quality of life
Test Manuals- detailed information about
5. Business and Military Setting- wide
the test
range of achievement, aptitude, interest,
motivational, and other tests may be History
employed in the decision to hire as well as in
related decision regarding promotions, job 2200 BCE- Testing program in China for
satisfaction, etc. Civil Service

6. Government and Organizational 1733, Abraham De Moivre – introduced

Credentialing – governmental licensing, the notion of sampling error
certification, or general credentialing of 1859, Charles Darwin- theory of evolution
professionals.
1869, Francis Galton- explored and
7. Academic research settings- conducting quantify individual differences of people
any sort of research typically entails
measurement of some kind, and academian - Classified people
who ever hopes to publish research should according to their
have a sound knowledge of measurement natural gift and to
principles and tools of assessment. ascertain their
deviation from an
8. Other settings (judiciary; program average
evaluation) - Coefficient
How Assessment Conducted correlation

- the test must be appropriate and necessary Karl Pearson- developed product-moment
for the person being evaluated correlation technique

- test user must be prepared and suitably Wilhelm Wundt- founded the first
trained to administer the test properly experimental psychology in Germany

Protocol- form/sheet or booklet on which - Found individual

test takers responses are taken differences as
frustrating source of
Rapport- working relationship between error
examiner and examinee
James McKeen Cattell- coined the term
Alternate Assessment- accommodates to Mental Test
the test takers who are not mentally capable
Charles Spearman- originated the concepts
Accommodation- adaption of a test of test reliability as well as building
procedure to another ta make the assessment mathematical framework for the statistical
technique of factor analysis
Victor Henri- collaborated with Alfred 1939, David Wechsler- Introduced Adult
Binet on papers suggesting on how mental Intelligence Test
tests could be used to measure higher mental
- Intelligence- capacity
processes
of an individual to act
Emil Kraepelin- early experimentation with purposefully, think
the word- association technique as a formal rationally, and to deal
test effectively with his
environment
- Founding founders of
modern psychiatry Lewis Terman, Robert Yerkes, and others
- Classification and developed Army Tests for recruits- Army
diagnosis of mental Alpha- literate; Army beta- illiterate
disorder
Robert Woodworth- was assigned the task
- Dementia praecox
of developing measure of adjustment and
Lighter Witmer- “little known founder of emotional stability that could be
clinical psych” and founded the first psych administered quickly and efficiently to
clinic in US groups of recruits disguised as personal data
sheet
1895, Alfred Binet and Victor Henri-
argued for measurement of abilities Woodworth Psychoneurotic Inventory-
first self-report measure of personality to
1905, Alfred Binet and Theodore Simon-
identify soldiers at risk for shell shock.
first intelligence test designed to help
identify Paris schoolchildren with ID Projective Test- one in which an individual
is assumed to project onto some ambiguous
- Considered
stimulus his or her own unique feelings
standardization
sample Herman Rorschach- developed Roschach
- Representative Inkblot Test
sample- comprises
Henry Murray and Christian Morgan-
individuals similar to
developed Thematic Apperception Test
those for whom the
test is to be used 1943, Minnesota Multiphasic Personality
- 1908- mental age was Inventory was published to use empirical
determined methods to determine the meaning of a test
- Lewis Terman- response
revised Binet’s test
for US use Factor Analysis- method of finding the
- Devised IT to Group minimum number of dimensions (factors) to
intelligence test in account for a large number of variables
response to the US J.R Guilford- made the first serious attempt
military’s screening to use factor analytic technique in the
of recruits for WWI development of a structured personality test.
Raymond Catell- 16PF measurement beyond those specifically
measured by the test or measurement
Beginning of 1980s – several major
branches of applied psychology emerged Four Levels of the Scale of Measurement
such as neuropsych, health psych, forensic
Nominal- involve classification or
psych, and child psych
categorization based on one or more
distinguishing characteristics
Ordinal- classification and rank ordering on
some characteristics
Statistics Refresher
Interval- classification and rank ordering
Scales of Measurement
and contain equal interval between numbers
Descriptive Statistics- description of a
Ratio- classification and rank ordering and
collection of quantitative data
contain intervals between numbers and has a
Inferential Statistics- used to make true 0 point
inferences from observation of a small group
Describing Data
of people
Distribution – defined as a set of test scores
Measurement- act of assigning numbers or
arrayed for recording a data
symbols to characteristics of things
Raw Scores- unmodified accounting of
Scale- set of numbers whose set of
performance that is usually numerical
properties model empirical properties of the
object to which the numbers are assigned Frequency Distribution- all scores are
listed alongside the number of times each
Continuous Scale- used to measure
occurred
continuous variable/ has unlimited possible
numbers Measure of Central Tendency
Discrete Scale- used to measure discrete Mean – average; used in interval and ratio
variable/ countable numbers
Median- middle score of a distribution/ used
Magnitude- property of moreness/ in when there is a skewness because it is more
research, we use cohens d resistant to the skewness
Equal Intervals- the difference between Mode- frequently occurring score in a
two points at any place on the scale has the distribution of scores
same meaning as the difference between two
Measure of Variability
other points that differ by the same number
of scale units Variability- indication how scores in a
distribution are scattered or dispersed
Absolute 0- nothing of the property being
measured exist Measures of Variability- statistics that
describe the amount of variation in a
Error- refers to the collective influence of
distribution
all the factors on a test score or
Range- Equal to the difference between T-Score- scale with a mean set of 50 and a
highest and lowest score sd set of 10
Quartile- dividing points between the four Stanine- test scores on a nine-point standard
quarters in the distribution (34.13, 13.59, scale with a mean of five and deviation of 2
2.14)
Correlation and Inference
Standard Deviation- equal to the square
Correlation Coefficient- number that
root of the average squared deviations about
provides us with index of the strength of the
the mean/ square root of variance
relationship between two things
Variance- spread between numbers in a data
Correlation- an expression of the degree
set
and direction of correspondence between
Normal Curve- gaussian cave. A bell two things
shaped defined curve that is at highest at its
Pearson r- used when two variables being
center
correlated are continuous and linear
Tail- 2-3 standard deviations above and (Pearson Correlation Coefficient/Pearson
below the mean Product-Moment Coefficient of Correlation)
Symmetrical Distributions- right side of Spearman Rho (Rank-Order Correlation
the graph mirror the image of the left side Coefficient/Rank-Difference Correlation
Coefficient)- used if the sample size is small
Skewness- nature and extent to which
and when both sides of measurement are un
symmetry is absent
ordinal
Positive skew- few score fall the high end of
Statistical Foundations
the distribution
Reliability- consistency in measurement
Negative Skew- when relatively few of the
scores fall at the low end of the destribution Classical Test Theory- a score reflect both
the score and error
Kurtosis- steepness if a distribution in its
center Reliability Coefficient- index of reliability
that indicates the true score variance and the
Platykurtic- relatively flat
total variance
Leptokortic- relatively peaked
Error- refers to the component of the
Mesokurtic- somewhere in the middle observed that does not have to with the test
taker’s ability
Standard Scores
Variance- useful in describing test score
Standard Score- standard score that has validity
been converted from one scale to onther
scale True Variance- variance from true
differences
Z-Score- indicates how many standard
deviation is from the mean Error Variance- variance irrelevant sources
Measurement Error- all of the factors Split-Half Reliability (Internal
associated with the process of measuring Consistency)
irrelevant variables
- split the questionnaire to half
Random Error- source of error in
- odd even reliability
measuring variable caused by unpredictable
fluctuations and inconsistencies of other - uses pearson r to test reliability
variables in the measurement process
- if correlating 2 halves- Spearman Brown
Systematic Error- source of error in Formula
measuring variable that is typically constant
or proportionate to what is presumed to be Inter-Item Consistency (Internal
the true value of the variable being measured Consistency)

Sources of Error Variance - useful in assessing homogeneity - 1 trait

only
1. Item Sampling/ Constant
Sampling- variation among items - all items should be correlated and answers
within a test as well as variation to one variable only
among items between tests Kuder-Richardson Formulas
2. Test administration- test taker’s
motivation or attention, KR-20- dichotomous items
environments and test taeker and KR-21- degree of difficulty (speed tests)
examiner’s variable
3. Test scoring interpretation- if Inter-Scorer Reliability (Internal
subjectivity is involved Consistency)

Reliability Estimates - administrators give the score to nonverbal

behaviour
Test-retest reliability (Time Sampling)
Kappa Statistics
- retesting to test the validity of raw data-
should be within 6 months - checks for the agreement of the raters

- only given to check the personality trait Fleiss Kappa- more than 2 raters

-same questions Cohen’s Kapa- 2 raters

Parallel Forms and Alternate Forms Validity- legitimacy. Does the test really
Reliability (Item Sampling) measure what it intends to measure?

- looking for coefficient of equivalence Validation- gathering and evaluating

evidence about validity
- same difficulty but contents are different
Validation studies- insights regarding the
- 2 test measuring one variable validity of test to certain population of test
Alternate forms takers as compared to the norming sample
described in the test manual
- same variable- different difficulty
Local Validation Studies- validation Bias- systematic error that could make test
process to alter format, instructions, sample unfair
described in a test manual
Rating- numerical or verbal judgement that
Types of Validity places a person or an attribute along a
continuum identified by a scale of numerical
1. Face Validity- looks like they measure
or word description known as rating scale
what they are supposed to measure
Rating Error- intention or unintentional
- first impression of the test
misuse of scale
2. Content Validity - does the content
Leniency Error- rater is lenient in scoring
includes measure the variable
Severity error- rater is strict in error
3. Criterion- Related Validity
Central Tendency Error- rater's rating
- how useful is the test to predicting
would tend to cluster in the middle of rating
personality
scale
Concurrent Validity - comparing a test to
Halo Effect- tendency to give high score
an already established test
due to failure to discriminate among
Predictive Validity- predicts future
Fairness- no discrimination, just
behavior. If it meets the criteria
Utility- usefulness of practical value of
4. Construct Validity- to prove if
testing to improve efficiency
something really exists
Factors that Affect a Test's Utility
Construct- idea
1. Psychometric soundness- high Validity
Construct validation techniques
& reliability
A. Evidence of Homogeneity- look into
2. Costs- disadvantages
inter item
3. Benefits – if it justifies the cost
B. Evidence of changes with age- peaks at
some point, decline at some point Utility Analysis- family of technique that
entails a cost-benefit analysis designed to
C. Evidence of Pretest- Post test changes
yield information relevant to a decision
D. Convergent Evidence- correlates with about the utility of a tools of assessments
an established test
How is Utility Analysis Conducted? –, the
E. Discriminant Evidence- prove a type of brogden-cronbach-gleser formula
test using a different kind of test
1. expectancy data- converting a
Psychometrically Sound- a psychological scatterplot to an expectancy table-
test is good indication that a test taker will score
within some interval of scores on a
Validity, Bias, Fairness criterion measure- passing,
acceptable, failing
 2. The Brogden-Cronbach-Gleser 4. Method of Paired Comparisons:
Formula- used to calculate the dolter Presenting pairs of stimuli for
amount of a utility gain resulting comparison.
from the use of a particular selection 5. Guttman Scale: Ordinal-level
instrument under specified measure where items range from
conditions weaker to stronger expressions of
a. Utility gain- estimate of the attitude or belief.
benefit of using a particular
Writing Items- Items should be written
test
considering the scaling method.
b. Productivity Gains-
estimated increase in work Item Pool: A collection of potential test
output items.
Test Construction Item Format: The structure of individual
test items.
Test Development- is the entire process of
creating a test. Types of Item Formats:
Test Conceptualization Selected-Response Format: Test-takers
select a response (e.g., multiple-choice).
Test Construction
Constructed-Response Format: Test-
Test Tryout
takers create or supply an answer (e.g.,
Item Analysis essay).
Test Revision Scoring
Test Construction- Writing test items and Cumulative Scoring: Higher scores indicate
revising existing ones, including formatting higher ability.
and scoring.
Class Scoring/Category Scoring:
Scaling: Setting rules for assigning numbers Responses earn credit toward category
in measurement. placement.
Types of Scales: Age-Based, Grade-Based, Ipsative Scoring: Comparing scores within
Stanine, Unidimensional, Multidimensional. a test.
Scaling Methods Item Analysis- statistical procedure used to
analyze items might become complex and
1. Rating Scale: Grouping of
our treatment of this subject could be viewed
judgments to indicate strength of a
as only introductory
trait.
2. Summative Scale: Final score is
obtained by summing ratings across
items.
3. Likert Scale- scales attitude; usually
reliable
Item-Difficulty Index: Proportion of test-
takers who answered correctly.
Item-Reliability Index: Indicates test's
internal consistency.
Item-Validity Index: Shows how well the
test measures its intended concept.
Item-Discrimination Index: Shows how
well an item discriminates between high and
low scorers.