Testing PSY 434 PDF

Preview

Summary

All lecture notes ...

Description

Testing Psych 434 MWF 10-10:50  9/16 Monday  FAPE, LRE & IEP  Detailed testing to see what criteria children met whether they need special educational services- had to provided in LRE.  IEP- Educational and other services like special psychological services  If a state decided not to comply with PL 94-142 the govt couldn't do much about it but they could withhold federal money from school districts - They were not providing that much anyways (ie: New Mexico told the govt to keep their money)  Law Section 504- Civil rights law for disabled people - Its a civil rights crime to discriminate by disability - Now states like NM couldn't discriminate bc it was made illegal - ADHD children weren't eligible for FAPE and IEp before passage of 504 - Now they are eligible for services- It would be illegal to withhold services from children with ADHD bc of 504  Early Intervention Programs (IEP)- The earlier you provide services, the less intense it has to be later on- it will help their conditions  Hypothetical Constructs/ Latent (hidden) Traits: Intelligence- IQ tests which isn't able to be seen heard or touch. Instead we infer from behavior - Collect sample of behavior and what it represents for the latent trait  Intra individual comparison- comparing within the individual Inter individual comparison- comparing between people  True/False Test-  Social Desirability Responding- People want to respond how they think they should/ what people want to hear  How to reduce SDR- make test anonymous

 9/23 Monday  Standard Z scores for std deviation (normal)  Percentiles: Used when tests are given to groups Associated with normal curve When closer to mean on normal curve, the percentiles are very close together and vice versa  Ordinal scale for percentiles on the curve  - Write out what central limit theorem is for extra credit  Grade equivalence: - Interpolation - Extrapolation: extending line on graph beyond the point They are scaled in ordinal units You can’t subtract ordinal data- also can’t tell what std dev is so you can’t assume a child is behind a year If the std dev is one month then it is larger and not in the avg area of the curve False standard of comparison- not knowing std dev and trying to subtract ordinal scale score  Age equiv: chronological age of person not grade level (12-3) 3rd month of being 12 years old  and grade equiv: (3-1) 1st month of first grade ]   9/25 Wednesday  Factor analysis: how do items intercorrelate with each other?  Exploratory factor analysis: - Choose a method and create a matrix - Determine number of factors - Rotate factors - Associations between factors and terms - Associations between factors  

Study for midterm: - History - Transforming scores - Ethics  Eigenvalues- are related to proportions of variance - We develop a Scree Plot from eigenvalues - Scree plot: vertical axis is the eigenvalue / horizontal axis is the number of factors - Get data for plot from the table - Choose the number right before it levels off -   Z scores have a mean of 0 and std dev of 1  T scores have a mean of 50 and std dev of 10  - Page 11-13 participant reactivity - Standardization - Reliability - Validity - Scores characteristics - Scales of measurement Orthangal rotation- factors are independent of each other which is different from oblique rotation where factors are correlated with each other  Eigenvalue- how many factors are in the data set. Determine # of factors by plotting them against # of factors on scree plot  Speeded tests are times and items are generally easy but too many to finish- measuring how many items correct in time allotted  Present info in easily understood way directed to recipient - Individualize test according to person when going over results  Multiple choice- to assess latent trait or hypothetical construct  Percentiles-  Variance = (std dev)^2

Friday 10/11 Regression line or line of prediction - If there is an x score and the relationship of x and y is represented in scatter, what Y score would be predicted? - Look at prediction line.. Not 100% accurate - If the correlation coefficient (+1) was perfect then every single point would fall on the line, meaning the prediction line would be 100% accurate - Linear relationship Pear shape- - Heteroscedasticity: y scores are distributed differently - The strength of the correlation coefficient is strong when close to the line - This varies in a pear shape  Range- measure of variability Ie: variance, std deviation The higher the SD the larger the range, the more variability, the higher the correlation is going to be - Higher variability, higher correlation - Truncated range: restricted range - When we restrict the variance in one set of scores, we restrict or truncate the range - Truncated by preselection - WRAT test: wide range achievement test - Moderator variable:   Monday 10/21  - When we have speeded tests (timed) you shouldn't use an internal reliability coefficient bc they're easy items but assessing how many can get correct   Wednesday 11/6  A: Same trait, same method Monotrait, monomethod RELIABILITY Homotrait, homomethod 

 B: Same trait, different method Homotrait, heteromethod CONVERGENT VALIDITY Monotrait, heteromethod  C: Different trait, same method Heterotrait, homotrait DISCRIMINANT/ DIVERGENT VALIDITY (Same method) Heterotrait, homotrait  We DO NOT want high C count in multitrait multimethod matrix!!!  D: Different trait, different method DISCRIMINANT/ DIVERGENT (Different method) Heterotait, heteromethod  *chart on Sakai  If we have true construct validity, then A will be the highest followed by B next C and then D….  A/B should be high and C/D should be low  11/13 Wednesday - Standard error of estimate= validity - SEM vs SE est - Rxy= + strong   11/22 Friday  - High SES vs low ses - One regression line= no bias - Slopes is how steep the line is - Intercept is where the x and y connect - Under and over-predicting (line B) when using common regression line - Equal slope, but different intercepts= constant bias/ - Different slopes but same intercept= - Determining criterion related bias

Content validity  12/4 Wednesday - Item characteristic curve (ICC) - P level= difficulty in psychometrics or proportion of people who passed - What is the difficulty level of an item for a certain person - Graph of P values vs theta - Theta is a hypothetical construct (for example: how much a student learned in the course) - The slope tells us about the discriminability of an item - A steep slope means that the item is discriminating among the finer levels of theta - Tells us theta (trait), level of difficulty for level of trait (P), and steepness of the slope for discriminability ( how well an item differentiates among different levels) - Icc also tells us that - Correction for guessing- better to skip then try to guess because they subtract # wrong from # right (ie: GRE exam) - level of difficulty vs discriminability - Of everyone passes an item or everyone fails it, there is no discriminability - Best level of discrimination is P value thats in the middle of 0 to 1 = .5 (U shaped curve)  12/9/19 Monday...Review!   Sensitivity: a/a+b Specificity : B/D+b  ICC- parameters Horiz (x) : theta Vertical (y): difficulty level for corresponding level of theta Steepness of curve is discriminability Difficulty of the item.. The P of .5 level corresponds to what level of theta Where is the asymptote (.5 = true/ false item) - Can determine Guessing, discriminability, difficulty **No repeating of past questions but similar items from past exams Factor analysis- construct validity  Divergent validity- low correlations between theoretically unrelated traits Degree of accuracy of a prediction(criterion predicted from the test score) - standard error of estimate How much error in test score- standard error of measurement

  ...