Theories

Question 1

True ability of the testtaker (true score) + error

Answer 1

Classical test score theory

Question 2

Center of distribution should represent the true score and dispersion around the mean of the distribution should display the distribution of sampling errors

Answer 2

Classical test score theory

Question 3

Seek to estimate the extent to which specific sources of variation (error) under defined conditions are contributing to test score

Answer 3

Domain sampling theory

Question 4

Using a limited number of items to represent a larger and more complicated construct. Also states that as reliability of a test increases as the number of items increases

Answer 4

Domain sampling theory

Question 5

Based on the idea that a person’s test scores vary from testing to testing because of variables in the testing situation; analogous to true score theory

Answer 5

Generalizability theory

Question 6

Influence of particular facets on the test score nga bisag unsa pa na factor (test admin, no. of items) maka lead ra ug same score in different situations

Answer 6

Coefficient of generalizability

Question 7

Used to focus on the range of item difficulty that helps assess an individual’s ability level

Answer 7

Item-response theory (aka latent-trait theory)

Question 8

An IRT model with very specific assumptions about underlying distribution

Answer 8

Rasch model

Question 9

Assumptions in using IRT

Answer 9

• unidimensionality
• local independence (only one construct)
• monotonicity (item mutugma sa construct kay mu increase if piliun sya)

Question 10

Probabilistic relationship between a testtaker’s response to a test item and that testaker’s level on the latent construct being measured expressed in graphic form

Answer 10

Item characteristic curve

Question 11

Understand the range over construct for which an item is most useful in discriminating groups of testtakers

Answer 11

information curve

Question 12

Estimate which an observed score deviates from a true score

Answer 12

Standard error of measurement (the higher the SEM, the lower the reliability)

Question 13

Estimates how repeated measures of a person on the same instrument tend to be distributed around his or her “true” score.

Answer 13

The standard error of measurement (SEM

Question 14

A range or band of test scores that is likely to contain the true score; set or derived from SEM; ideal is 95%

Answer 14

Confidence interval

Question 15

Determine how large a difference should be before it is considered statistically significant

Answer 15

Standard error of difference

Question 16

Assess whether any differences we observe in test scores are real differences or simply differences that occurred due to measurement error

Answer 16

Standard error of difference

Question 17

Difference between SEE and SEM

Answer 17

SEE= error in the prediction of Y from X

SEM=focus kay sa test as a whole and error kay gkan sa observed score or sa test (measurement)