
The basis of psychological testing is...
Inference

What are test used for when it comes to psychology?
A test is a measurement device or technique used to quantify behavior or aid in the understanding or prediction of behavior.

Item
Is a specific stimulus to which a person responds overtly

Scales and Raw scores
Scales relate raw scores on test items to some theoretical or empirical distribution.

What are test good for?
 Measure current behavior
 Predict future behavior
 Infer hidden behavior

Test can measure...
 Statestemporal; change
 Traitsfixed


Types of Test...
 Ability testmeasure previous learning
 Aptitude testmeasure potential for acquiring a particular skill
 Intelligence testmeasures intelligence

Intelligence refers to...
 A person's general ability to...
 solve problems
 adapt to changing circumstances
 think abstractly
 profit from experience

First test...
Chinese civil service test

In 1883...
The American government established the American civil service commission

James Cattell
Developed mental test

Seguin Form Board
Developed to educate and evaluate the mentally handicapped.

Standardization sample
A group given the test under standard conditions.

Representative sample
A sample that represents the group we are using comparisons.

Mental Age
How you perfomed compared to others who took the test.

Personality test
 Stimulus
 Response
 Projective
 Ambiguous
 FaceValid

The Rorschach
 Inkblot
 Could not find meaningful standardized responses
 Still widely used
 Full of Shit...lol

MMPIMinnesota Multiphasic Personality Inverntory
uses empirical methods (factor analysis) to determine of a test response

What do statistics provide?
 Concise descriptions of lots of quantitative information.
 demographic breakdown
 prevalance of behavior

To Infer things...
 Deduce things you cant observe directly.
 Measure something in a small group of ppl. the sample
 Infer the results apply to a larger group of ppl.  the population

Measurements
using rules to assign numbers to objects

Scales have 3 properties...
Magnitudemoreness
Equal intervalsdifference between any 2 points on the scale is the same
Absolute zero it is possible to have none of the quality measured

Types of Scales
Nominalnot really scales, assigning random designator numbers to ppl or things
Ordinalhave moreness but not equal intervals or absolute zero. E.g. putting ppl in order shortest to tallest
Interval have magniude and equal intervals but not necessarily absolute zero. E.g. temperature
Ratio have magnitude, equal intervals and absolute zero

Frequency Distribution
Displays how many times (frequency) each score (distribution) was obtained on a scale.
 xaxis=each score, from lowest to highest
 yaxis= # of times each score was obtained

Normal Distribution
Increased sample size; bellshaped curve

Skew
 pos.=tails to the right
 neg.=tails to the left
IQ scores have a pos. skew

Class interval
The distance between two consecutive measurements in your distribution.

Percentiles
Indicate the particular score below which a defined percentage of scores falls.

Quartile system
Divide the percentage scale into four groups

Decile system
Divide the scale into tenths

Stanines
Divide the scores into standard 9ths
this creates 3 groups= the lower, middle and high group

Interquartile range
The middle 50% of scores (25%75% percentiles)

Mean
The average score in a distribution
mean= (sum of all scores) / (# of scores)

Median
The point at which 1/2 the scores are above and 1/2 are below

Mode
The "bump"
most common score

Variance*
The variation of the scores around the mean

Standard Deviation*
The average deviation around the mean
calculation: SD=Square root of the variance

Zscores*
Indicate how far from the mean a score is
universal expression of SD
 scores above the mean are + zscores
 scores below the mean are  zscores
between 3.0 and +3.0

McCall's T; TScores
The mean did not equal 0, as it is with zscores it equaled 50
standard deviation did not equal 1, as it is with zscores it equaled 10

Transformation
Mathematical translation; transformations standardize the distribution.

Norms
 Average performance by standardization sample
 The standardization sample sets the norms

NormReferenced Tests
 Compare individuals to a normative group
 E.g. class exam scores

CriterionReferenced Tests
Compare an individual to a criterion on a specific skill, tasks, or knowledge

Variable
 A measure that can have multiple values
 E.g. weight, test score, etc.

CoVary
As on changes, another changes

You can measure changes using...
 Correlation
 Regression
 Multiple regression

Scatter Plots
A picture of the relationship between 2 variables
One variable increases on the horizontal, or xaxis
One variable increases on the vertical, or yaxis

The Regression Line
Help us make predictions between scores of two variables
Like a slope line on a graph
 steep= x may predict y
 flat= y may predict x
a= ybx

The Residual
The difference between the actual score and the predicted score
the best fitting line minimizes the residual

Correlation
correlation is like regression, except scores from each variable are standardized

Correlation coefficient
Describes the magnitude and direction of relationship between 2 variables

Types of correlations
 Pos. variables 1 and 2 go up or doen together
 Neg.  variables 1 and 2 move in opposite directions
 No correlation

Pearson's r
The statistics used to measure correlation

Null Hypothesis
Assume that there is no relationship between 2 variables

Pearson's r is used for...
2 countinuous variables

Dichotomous variables
yes/no; true/false; pass/fail; etc...
All dichotomous variables are categorical
No Pearson's r for dichotomous variables

Biserial correlation
Between one continuous and one dichotomous variable

Standard error of estimate
 The standard deviation of the residuals
 Predicts fit of the regression line
 Smaller is better

Coefficient of determination
 The correlation coefficient squared (r^{2})
 Indicates how much of the variation in Y is due to X


Cross Validation
Using the regression equation from one group of subject to predict performance in a different group of subjects

Shrinkage
The amount of decrease in predictability from crossvalidating

Restricted Range
If the variability of a variable is extremely restricted, significant correlations may be difficult to find even if they are there

Factor analysis
Creates factors: groups of related variables

Reliability
Is all about error.

Error
 Difference in true ability and measurement of ability.
 The inevitable inaccuracy of our measurements.
 How much our test do not reflect reality.

Psychological testing tries to...
 find the magnitude of error for each test
 try to minimize error for all tests

Tests that are relatively free from error are...
reliable

Yardstick
measure tangible things

Rubber yardsstick
 measure intangible things
 one that may over or under estimate the measurement

Test score theory
 Influences how we think about and calculate reliability
 assumes each person has a true score
 We do not report the true score
 we report the observed score

The distribution of random error is...
Bellshaped

Narrow distribution
Is accurate

Wide distributioin
Is inaccurate

Methods that measure reliability
 Timesampling
 testretest method
 Itemsampling
 parallel forms method
 splithalf method

TestRetest Method
 Administer the test at 2 different times
 compare each testtakers scores
 Caveats:
 stable traits
 carryover effects

Internal Consistency (IC)
when the variance is equal on different parts of the test

2 types of test to calculate IC
Dichotomous test & Likert scales

The reliability statistic
α alpha

The variance of the true scores
σT^{2}

The variance of the observed scores
σx^{2}

Domain Sampling Model
 Uses a sample of items, not the entire domain of items
 pick several items from your original test
 give that small test
 record observed scores
 estimate true scores fro these observed scores
 More items = more reliability (generally)

Adaptive test
Test that adapts to the test taker

Reliability of a Difference score (RDS)
 Subtract one score from another
 same test given at 2 points in time
 2 subscores from one test

RDS rule
Make the comparison in Zunits

Problem with RDS
 The error of a difference score is inflated
 It absorbs error from both scores in the difference score.

Potential solution to RDS
Calculate RDS if you know... α of each test and r between these test

BEWARE! RDS
Even tests with high α and r may have low RDS

Interrater Reliability
The similarity of rater 1 and rater 2's measurements

Agreement for Interrater reliability
 K = Pr (a)Pr (e)
 1Pr (e)

