(Validity)
Predictive Validity
The predictions made by the test are confirmed by later behavior
Example: SAT scores are used to predict a college students first year GPA (forecast a future outcome)
A measure of reliability useful for interpreting the test scores of an individual, helps determine the range within which an individuals test score probably falls (precision of the test)
Standard Error of Measurement
The four measures of variability
Range- Highest score minus the lowest score in a set
Inclusive Range- which is the high score minus the low score & adding 1 (inclusive-everyones included)
Standard Deviation- Standard Deviation
Variance- the square of the standard deviation
This type of test measures the effects of learning or a set of experiences (What someone knows)
Example: Standardized testing
Achievement Test
Percentile
A value below which a specified percentage of cases fall
example: If a student scores in the 90th percentile this means that their score was higher than 90% of the other students who took the test
Concurrent Validity
The results of the test are compared with other tests results (could be past or present)
Examples: Comparing student past test scores in a class to the current class test scores
A test used to see how relaible the test would be if not cut into two. Reducing the length of the test in turn reduce the measured reliability, this formula accounts for that
Spearman-Brown formula
Standard Deviation
example: two classes of students taking a test.
Class A: The average score is 75, and the standard deviation is 5. This means most students scored between 70 and 80 (75 +/- 5).
Class B: The average score is also 75, but the standard deviation is 20. This means the scores are more spread out, with some students scoring much higher or lower than 75
This type of test is often referred to as a ability test, this type of test measures the effects of general learning and are used to predict future performance
Example: Career Ability Placement Survey
Aptitude test
The function of a normal curve
The normal curve distributes the scores into six equal equal parts, three above the mean and three below the mean(is a symmetrical distribution of scores with an equal number of scores above and below the midpoint)
Construct Validity
Testing that has validity to the extent it measures some hypothetical construct.
examples:
This type of reliability uses the same instrument on both occasions, you are testing the same group twice
Stability
Correlation Coefficient
statistical index that shows the relationship between two sets of numbers (range from -1.00 to +1.00) -1.00 is a perfect negative correlation, +1.00 is a perfect positive correlation (does not determine or tell you anything about cause and effect
The three types of "referencing" in assessment
Norm Referenced- comparing individuals to others (CPCE, NCE), compainr to others is more important than what you know
Criterion Referenced-comparing an individuals performance to predetermined criterion (cut off score)
examples: Drivers exam, theres an established cut off score
Ipsatively interpreted- compares the results on he test within the individual
examples: comparing the scores of how an individual did on his first exam and second exam
Stanine
converts a distribution of scores into nine points (1 to 9 with 5 being in the middle. "Standard Nine", used to categorize student performance and provide a normalized way to compare scores,
Content Validity
The extent to which a measurement instrument (like a test or questionnaire) accurately and adequately measures the specific content or construct it is designed to assess.
Example: creating a test to measure someone's knowledge of basic arithmetic operations, content validity would ensure that the test questions cover all the essential topics, such as addition, subtraction, multiplication, and division, and do not include irrelevant mathematical
alternate forms of the same test are administered to the same group group and the correlation between the two is calculated (test the tests same skills, just different questions)
equivalence
The Coefficient of determination/ Coefficient of nondetermination
The coefficient of determination represents the proportion of variance in the dependent variable that is explained by the independent variable(s) in a regression model. Conversely, the coefficient of nondetermination represents the proportion of variance that is not explained by the model(represents the error variance)
Tests are typically considered :
Power based or Speed based -
Power based- no time limits or very generous ones
Speed based- timed, emphasis is placed on speed & accuracy.
Examples:
Z Score
Indicates how many standard deviations a data point is away from the mean of the distribution
example: if a test score of 87 has a z-score of 1.75 when the mean is 80 and the standard deviation is 4, it means the score is 1.75 standard deviations above the mean
Face validity
refers to the degree to which a test or measurement appears, on the surface, to measure what it is intended to measure
example: if you create a survey to measure customer satisfaction and it includes questions about product usage and service experiences, people might say it has high face validity because the questions seem to relate directly to what you're trying to measure.
split half method" , you are splitting the test into two halves and correlating the between the two halves
Internal Consistency
Semantic Differential
A scale that asks individuals to report where they are on a dichotomous stage.
Example. A scale with one end that says "very good" and another that says "very bad"
These tests present a relatively unstructured task or stimulus. The person projects thought processes, needs, anxieties.
Example: The Rorschach ink blot test
Projective test
T- Score