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;
46 Cards in this Set
- Front
- Back
scales of measurement |
magnitude, equal intervals, absolute zero |
|
nominal |
the most basic measurement & used for classifying & naming (gender, political affiliation, race) |
|
ordinal |
refers to rank order and are characterized by magnitude (height-short, average, tall) |
|
interval |
possess magnitude and equal intervals (temperature-100 vs 105 degrees) |
|
ratio |
the most advanced and precise measurement scale (time and money) |
|
reliability |
refers to consistency and the extent to which a test eliminates chance and other extraneous errors in its results (individual errors; test error; testing condition error) |
|
measurement error |
the positive or negative bias within an observed score three elements: observed score (X), true score (T), error score (e) X=T-e |
|
correlation |
the extent to which two sets of measures are related (the strength of the relationship)/about the relationship not necessarily about numbers same relationship: -1.0 no relationship: -0.0 reverse order: -1.0 |
|
coefficient of determination |
correlation squared (r) squared explains the shared variance between two variables. if r=.80, r^2=64% |
|
correlation coefficient |
measures the degree between two variables that are linearly related. the sign of each variable indicates whether the 2 variables tend to rank (+ same order/- reverse order) and the magnitude indicates the strength of the relationship. the larger the correlation indicates a stronger relationship (+/- 1.00), but it dependent on what items are being correlated. |
|
types of reliability |
test-retest alternate form internal consistency (spilt-half, interitem reliability) interrater reliability |
|
test-rest reliability |
trait test should be repeatable over time. for example: testing the same test 2x within periods of time btn each administration. it is calculated by finding the correlation of mean of the individual scores for each administration. |
|
alternate form reliability |
the crucial question is whether or not the 2 forms are actually equivalent (eg PRA) |
|
internal consistency reliability types |
items within the test split half interitem reliability |
|
spilt half |
odd & even items share similarities |
|
interitem |
items related to overall test |
|
interrater reliability |
refers to the degree of agreement between two or more independent judges. it is calculated by the ratio btn the judges agreement and possible agreement. |
|
validity |
the extent to which meaningful and appropriate inferences made from the test. reliability is a precursor to validity- a test can only be as valid as it is reliable |
|
types of validity |
content criterion-related construct treatment *face |
|
content validity |
refers to the degree that items represent the domain of interest involves defining the domain & identifying items that refer to that domain |
|
criterion-related validity |
comparing test scores with performance on a criterion measure concurrent: SAT & HS GPA (consider range) predictive: SAT & college GPA |
|
construct validity |
synonymous with the term validity itself; refers to the psychological meaning of the test itself |
|
treatment |
do the results obtained from the test make a difference in tx. if the test results are useful, if they make a difference in the counseling process, then the test could be said to have tx validity. |
|
face validity |
not evidence of validity; refers to the assessment looking like what it is supposed to |
|
converging |
measuring items that are in the same direction (depression/fatigue). measures similar characteristics/substantial correlation |
|
divergent |
towards construct of what you want/not substantially correlated with other tests from which they're supposed to differ (math test vs. clerical speed) |
|
false negative |
occurs when a person obtains a score below the cutoff score, but actually meets criteria (eg suicidality) |
|
false positive |
occurs when a person obtains a score above the cutoff score, but does not meet criteria (eg severe depression vs. suicidality) |
|
sensitivity |
picking up all, including the false negatives (MMPI-2 depression scale) |
|
specificity |
excluding all that do not meet criteria, including false positives (3 suicide questions) |
|
measures of central tendency |
mean, median, mode |
|
mean |
arithmetic average (most frequently used) M= sum of scores/# of individuals |
|
median |
50th percentile (line up all the scores in order and the middle most value) |
|
mode |
the most frequently occuring score |
|
variability |
refers to the variation or range in scores |
|
3 types of score reference |
norm-referenced criterion-referenced self-referenced |
|
norm-referenced |
a comparison of scores with other individuals |
|
criterion-referenced |
comparison with an absolute score est. by an authority |
|
self-referenced |
comparison of scores obtained by the same individual |
|
standard deviation |
most frequently reported measure of variability the larger the SD, the greater the dispersion of scores |
|
range |
refers to the distance btn the lowest & highest score +1 R= (LS-HS)+1 range can never be 0 |
|
normal curve |
normal distribution-3 measures of central tendency are equal SD & Percentiles -/+ 3SD (~.1 or 99th) -/+ 2SD (~2 or 98th) -/+ 1SD (~16 or 84th) |
|
histograms |
used with quantitative variables |
|
bar graphs |
depict nominal data |
|
Positive skew |
majority of scores left (mode, median, mean) |
|
negative skew |
majority of scores right (mean, median, mode) *median is best in both |