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;
25 Cards in this Set
- Front
- Back
- 3rd side (hint)
|
instruments
|
form the basis of measurement; allow use of quantification to move beyond subjective opinion, provide degree of precision, allow us to quantify abstract/intangible concepts
|
271
|
|
|
it's important to select instruments that are...
|
1) good indicators of what the programs are attempting to accomplish
2) psychometrically strong |
272
|
|
|
behavioral outcomes can be used as...
|
a measure
|
273
|
|
|
the "best" measure is
|
the one closest to the intent of the program intervention
|
275
|
|
|
reliability
|
consistent and dependable measure with accuracy; reported on a scale between 0-1
|
275
|
|
|
scale
|
measures a solitary concept and commonly attempts to assess intensity or amount
|
276
|
|
|
index
|
creation of a new variable that is the sum of other variables that measure a single construct
|
276
|
|
|
subscales
|
evaluate a different construct
|
276
|
|
|
internal consistency
|
how well an individual item of a scale correlates with the scale as a whole
|
277
|
|
|
reliability coefficient also called
|
alpha
|
277
|
|
|
split-half technique
|
way to determine internal consistency, by dividing scale in half and seeing how well they correlate
|
278
|
|
|
parallel or alternative versions of a scale
|
another way to determine internal consistency, the higher the two scales correlate the stronger the reliability
|
278
|
|
|
test-retest
|
another form of reliability
|
279
|
|
|
inter-rater reliability
|
how well the judgments, observations or interviews of two or more data collectors correlate; minimum correlation of .70 needed for adequate reliability
|
281
|
|
|
validity
|
when an instrument closely corresponds to the concept it was designed to measure
|
282
|
|
|
two most important dimensions for evaluating instruments
|
reliability and validity
|
275
|
|
|
content validity
|
when the entire range of a concept is represented in the scale; not sufficient to establish true validity
|
282
|
|
|
face validity
|
when an instrument appears to measure the concept it supposed to measure; not sufficient to establish true validity
|
282
|
|
|
criterion validity
|
the instrument can be validated by an external criterion
|
282
|
|
|
concurrent validity
|
when a new scale correlates well with an established valid scale
|
283
|
|
|
construct validity
|
concerned with the theoretical relationship of the scale to other variables
|
283
|
|
|
known-groups technique
|
used to establish validity, by administering instrument to two very different groups to see if statistically significant differences are found
|
283
|
|
|
factor analysis
|
statistical procedure that aids in understanding or confirming the structure of a scaled construct; produces factor loading (kind of like a correlation)
|
284
|
|
|
instruments should be:
|
-relevant and appropriate to the client group
-easy to administer -useful and easy to interpret -reliable and valid -sensitive to change -relatively inexpensive |
285
|
|
|
steps in questionnaire design
|
1) consider what is needed to evaluate the program
2) choose administration mode (eg, self-report, interviewed in person/telephone, web, etc) 3) create pool of items for possible use 4) refine data collection instrument (eg, question sequencing, difficulty level, private info, memory, length and appearance, cultural sensitivity) |
288
|
