1 Flashcards
First exam for civil service
T’ang Dynasty
Systematized the test content
Ming Dynasty
Mentally retarded indvdls
Distinguishing idiots
Jean Esquirol
Treatment of retarded individuals
Edouardo Seguin
1st RPsy to teach statistical analysis of experimental results
James Mckeen Cattell
Mental test
James Mckeen Cattell
1st truly psych test of mental ability developer
Alfred Binet
Put in place a system of assessments for admission to civil service
Manuel Carreon
Dissertation on PH Studies in Mental Mesurements
Sinfroso Padilla
Research on trends in psych testing in educational, training and research
Carlota
Panukat ng Ugali’t Pagkatao
Virgilio Enriquez
Annadaisy Carlota
Panukat ng Pagkataong Pilipino
Stable characteristics of an individual
True characteristics
Chance features of the indvl or the situation
Random measurement of error
Tools used to estimate or infer the extent to which an observed score deviates from the true score
Standar Error of Estimates
Random Measurement Error Formula
X = T + E
X = Raw score
T = True score
E = Error
In a reliable test, the value of _ should be close to 0 and the value of _ should be close to the actual test score X
E ; T
Test Development (CoCoTAR)
Conceptualization
Construction
Tryout
Anaysis
Revision
Used to evaluate error associated w administering a test at 2 diff times
Test-retest Method
This type of analysis is of value only when we measure “traits” or characteristics that don’t change iver time
Test-retest
For test-retest, ideally, it should be
6 months or more
When interval between testing is greater than 6months, referred as
Coefficient of stability
Compares 2 forms of tests measuring the same attribute
Parallel/Alternate Forms
2 forms, diff items, same level of difficulty
Parallel / Alternate Forms
Test divided in half then the results of one half of the test are then compared with the results of the other
Split-Half Method
Spearman-Brown Formula w/c allows u to estimate the correlation between 2 halves would have been if each half had been the length of the whole test
Correlation Formula
Degree of correlation among all the items on the scale
Inter-item Consistency
Useful in assessing the homogeneity of the test
Inter-item Consistency
Methods to obtain estimates of internal consistency
- Cronback Alpha / Coefficient Apha
- KR 20
Estimate internal consistency of non-dichotomous items (no right or wrong answers)
Coefficient Alpha / Cronbach Alpha
Coefficient Alpha / Cronbach Alpha typically range from
0 to 1 ; negative values are theoretically impossible
As reliability of our test increases, the highest possible value of the _ increases
Validity coefficient
4 Types of Validity
Face
Content
Criterion
Construct
Make inferences about the broader domain of knowledge or skills
Content Validity
Test to predict performance on another measure
Criterion Validity
The measure of interest being the criterion
Criterion-related Validity
2 Types of Validity Evidence
- Predictive Validity
- Concurrent Validity
Index of the degree to which a test score predicts some criterion measure
Predictive Validity
Index of the degree to which a test score is related to some criterion measure obtained at the SAME TIME
Concurrent Validity
Whether a test measures what it indends to measure. Personality dimensions of personality traits
Construct Validity
2 Types of Construct Validity
Congruent
Discriminant/Divergent
Error in rating tendency to be lenient in scoring
Leniency Error
Systematic reluctance to give ratings either positive or negative
Central Tendency Error
Give higher rating from failure to discriminate mong distinct aspects of ratee’s behavior
Halo Effect
Group of statistics that can be calculated for indvl test items; Helps explaining why a test shows a certain level of reliability and validity; Useful if test is unreliable or fail to demonstrate relationship w criterion mesures
Item analysis
2 Commonly used technique of item analysis
Item Difficulty
Item Discrimination
Appropriate for maximal performance test - Achievement & aptitude
Item Difficulty
Optimal item Difficulty
.5-.7
High p value such as .9…
Most ppl got it right
P = .00 - .20
Very Difficult
P = .21 - .80
Moderately Difficult
P = .81 - 1.00
Very easy
Indicates the extent to which different types of eople answer an item differently
Item discrimination
Separates the bright and poor pupils
Discrimination index
D is the proportion obtained by comparing performance of 2 subgroups of test takers (use in maximal peromance testing)
Extreme Group Method
Multifaceted capacity that manifests itself in different ways across all life span
Intelligence
Intelligence is INHERITED; Belived the most intelligent is equipped with best sensory abilities
Francis Galton
Intelligence as the degree of using experience to solve present and future problems
Henry Goddard
