Reading...
Play button
Play button
Use LEFT and RIGHT arrow keys to navigate between flashcards;
Use UP and DOWN arrow keys to flip the card;
H to show hint;
A reads text to speech;
78 Cards in this Set
- Front
- Back
|
What is inferential statistics?
|
They are logical deductions about events that cannot be observed directly.
Ex. mean anxiety of Hispanic students in the US |
|
|
What are descriptive statistics?
|
Just like the title, it's just for description.
Ex. mean anxiety scores of Hispanic students at UTPA |
|
|
What are the three PROPERTY scales?
|
Magnitude, Equal Intervals, and Absolute Zero
|
|
|
What is the definition of magnitude?
|
Magnitude is more, less, or equal.
|
|
|
What is the definition of equal intervals?
|
Equal intervals is the difference has the same meaning.
|
|
|
What is the definition of absolute zero?
|
Where the targeted property does not exist.
|
|
|
What are the four TYPES of scales?
|
Nominal, Ordinal, Interval, and Ratio.
|
|
|
What is the definition of nominal and give an example?
|
Nominal is when numbers are attached to categories.
Ex. Ethnicity: 0-African 1-Asian 2-Caucasian 3-Hispanic |
|
|
What is the definition of ordinal and give an example?
|
Ordinal is when you rank individuals or objects.
Ex. Ranking by height, anxiety scores. 0-No anxiety 1-Moderate anxiety 2-High anxiety |
|
|
What is the definition of interval and give an example?
|
Interval is when you have magnitude and equal intervals.
Ex. Fahrenheit, Celsius |
|
|
What is the definition of ratio and give an example?
|
Ratio has magnitude, equal intervals, but they also have an absolute zero.
Ex. Length, velocity, Kelvin |
|
|
What is a frequency distribution?
|
It's where you plot points on a histogram (rectangular diagram where it's close together.)
|
|
|
What does percentile mean?
|
It means specific scores/points within a distribution.
|
|
|
What does percentile rank mean? (Give the formula).
|
The percent of scores that fall below a particular score. The score of interest is Xi. The total number of cases is N. How many cases fall below the score of interest B?
Xi = B / N * 100 |
|
|
What is mean, mode, and average?
|
Mean: average of the total divided by the amount of numbers.
Mode: the most frequent score. Median: the score that falls in the middle |
|
|
What is standard deviation? What is the letter associated with population and sample?
|
It is an approximation of the average deviation around the mean. o is the standard deviation for a population and s is the standard deviation for a sample.
|
|
|
What is the relationship between a standard deviation and variance?
|
The variance is the squared deviation. o2 and s2.
|
|
|
What is the z score?
|
Z score is the difference between a score and the mean divided by the standard deviation.
|
|
|
What is the formula for z score and give the meaning for each letter.
|
x - x / s is the formula.
x is the observed score mean and s is the standard deviation. x with the line above it is the mean average. Z scores have a mean of 0 and a standard deviation of 1. |
|
|
What is the formula for a T score and give the definition for the formula?
|
Formula: T = 10z + 50
T scores have a mean of 50 and a standard deviation of 10. |
|
|
What does quartile mean?
|
Quartile is a percentile scale divided into equal fourths.
|
|
|
What does interquartile range mean?
|
You subtract the 75th percentile - by the 25th percentile.
Ex: Q1- 45 Q2- 68 Q3- 88 So this means that you'll subtract 88-45 and get your answer of 43, that will be your interquartile range. |
|
|
What does norm mean?
|
It is the performances by defined groups on particular tests.
|
|
|
What is a norm-referenced test?
|
Where you compare each person with a norm (Ex. mean,s percentile ranks, Z scores).
|
|
|
What is a criterion-referenced test?
|
Where you quantify the specific amount of skill that the test taker can demonstrate. The criterion is like a cutoff score or target score.
|
|
|
What are the two types of bivariate analysis?
|
Univariate distribution and Bivariate distributions.
|
|
|
What is a Univariate distribution?
|
Only one variable (Ex. frequency distributions.)
|
|
|
What is a Bivariate distribution?
|
two variables for each individual.
(Ex. Anger and Depression) |
|
|
What is a scatter diagram?
|
A visual inspection by drawing where you plot points on a diagram.
|
|
|
The difference between linear relationship and non-linear relationship?
|
A linear relationship is a straight line in the scatter diagram. A non-linear has no straight line.
|
|
|
What does the correlation coefficient describe?
|
It describes the direction and magnitude of a bivariate relationship.
|
|
|
Where does the correlation coefficient range from?
|
It ranges from -1.0 to +1.0
|
|
|
What are the three types of correlations in a scatter diagram?
|
Positive, Negative, and Zero Correlations.
|
|
|
What is the Pearson Product Moment Correlation?
|
It is two continuous variables, represented by r.
|
|
|
What is Spearman's rho?
|
It's another form of a correlation coefficient (Spearman's rho for two ranks).
|
|
|
What is statistical significance?
|
The findings that are not due by chance (represented by p less than 0.05).
|
|
|
What is the regression analysis?
|
Predictions about scores on one variable from knowledge of score on another variable.
|
|
|
What is the regression line?
|
It is the best fitting straight line through a set of points in a scatter diagram.
|
|
|
What is the regression equation?
|
It is the equation for the best fitting line. Represented by Y' = a + bX
a is the intercept, value of Y when x=0 b is the regression coefficient (slope) |
|
|
What does residual mean?
|
Y - Y'
The difference between the observed score (Y) and predicted score (Y'). |
|
|
What is the standard error of estimate?
|
It's relatively small and is the standard deviation of the residuals. (The smaller the better predictions.)
|
|
|
What is the coefficient of determination?
|
It is the coefficient squared (r2). Can only range from 0 to 1. It is the proportion of the variance in one variable explained by the other variable.
|
|
|
What is the coefficient of alienation?
|
It is the measure of non-association between two variables (the higher, the less associated). It ranges from 0 to 1. The equation is square root of 1- r squared.
|
|
|
What is the restricted range problem?
|
It is the restriction on the range of variability on either variable will reduce the observed correlation.
|
|
|
What is the third variable problem?
|
A variable that we cannot measure because it's hidden by the two variables we are measuring.
Ex. Anger causes depression, but financial problems could be the third variable. |
|
|
What is the correlation-causation issue?
|
Where two variables are correlated with each other, but we don't know which one caused what.
|
|
|
What are the two types of Multivariate Analysis?
|
Multiple Regression and Factor Analysis.
|
|
|
What does multiple regression mean?
|
It considers the relationship among combinations of three or more variables.
|
|
|
What does factor analysis?
|
It is used to study interrelationships between sets of variables when the constructs in question are hypothetical.
|
|
|
What is the classical test (score) theory?
|
It is the difference between the true score and the observed score results from measurement error.
The equation is X (Observed Score) = T (True Score) + E (Error) |
|
|
What is the standard error of measurement (SEM)?
|
It's where the classical test theory uses the standard deviation of errors as the basic measure of error.
|
|
|
What is Domain sampling model?
|
It considers the problem s created by using a limited number of items to represent a larger and more complicated construct.
|
|
|
What is Item response theory?
|
It is on a computer and is used to focus the rage of item difficulty that helps assess an individual's ability level.
|
|
|
What are sources of error?
|
An observed score may differ from a true score. Test-Retest method, Parallel forms, and internal consistency are examples to stray from error.
|
|
|
What is the test retest method?
|
It estimates the reliability are used to evaluate the error associated with administering a test at two different times.
|
|
|
What does carryover effects mean?
|
In systematic carryover everyone's score improves by exactly 5 points. In practice effects skills improve.
|
|
|
What is parallel forms method?
|
It compares two equivalent forms of a test that measure the same attribute.
|
|
|
What is the split-half method?
|
A test is given and divided into halves that are scored separately. The results of one half of the test are then compared with the results of the other.
|
|
|
What is the Spearman Brown Formula?
|
corrected r = 2r / 1 + r
r is the estimated correlation between the two halves of the test if each and the total number of items. |
|
|
What is the KR20 Formula?
|
KR20 = r = N / N-1 (S2 - Epq/ S2)
KR20 is the reliability estimate (r) N is the number of items on the test S2 is the variance of the total test score p is the proportion of the people getting each item correct. q is the proportion of people getting each item incorrect. Epq is the sum of the products of p times q for each item on the test. |
|
|
What is the kappa statistic?
|
Best method because it measures of agreement between two judges who each rate a set of objects using nominal scales.
|
|
|
How reliable is reliable?
|
It depends on the use of the test. It has been suggested that reliability in the range of .70 and .80 are good enough for most purpose in basic research.
|
|
|
What is the definition of validity?
|
It is the agreement between a test score or measure and the quality it is believed to measure.
|
|
|
What are the four types of evidence for inferences made about a test score?
|
Content-related, Criterion-related, Construct-related, and face evidence.
|
|
|
What is face validity?
|
Where it looks valid (this is the weakest and most often not considered an evidence).
|
|
|
What is content-related?
|
It is how well the test covers what it was designed to cover and generally made by an expert judgement.
|
|
|
What are the two new concepts of content-related?
|
Construct under-representation (failure to capture important components), and construct-irrelevant variance (when scores are influenced by factors).
|
|
|
What is criterion-related validity?
|
It compares a test to an external standard or criterion.
|
|
|
What are the two types of criterion-related validity?
|
Predictive validity (will the test predict future performance) and Concurrent validity (where the criterion is available at the same time).
Ex Predictive: Will SAT scores predict good college GPA? Ex Concurrent: Learning disability test |
|
|
What is validity coefficient?
|
Represented by r and it is the correlation between a test and a criterion.
|
|
|
What is validity coefficient squared?
|
It is the percentage of variance in the criterion a test predicts.
|
|
|
What is discriminant validity?
|
A test has low correlations with other tests believed to measure unrelated constructs.
|
|
|
What is construct-related validity?
|
Evidence that the construct of interest is really being assessed when no criterion is available.
|
|
|
What are the two types of construct-related validity?
|
Convergent validity (a test has high correlations with other tests) and discriminant validity (a test has low correlations with other tests believed to measure unrelated constructs.)
|
|
|
What is restricted range on predictor and criterion?
|
A predictors with a restricted range tends to fail t predict variability in the criterion.
|
|
|
What is generalization?
|
Predictive validity
|
|
|
What is differential prediction?
|
It is predictive relationships between SAT & GPA may differ across different ethnic groups and need separate studies.
|
|
|
What is the relationship between reliability and validity?
|
It is possible that a reliable test is not valid, but it is impossible that a unreliable test is valid.
|
