Chapter 5 Flashcards

1
Q

Reliability

A

How consistent is this type of measurement?

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

True Score

A

Is the measured value close to the true score? (Accuracy)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

Measurement Error

A

Is there inconsistency in the measurement/does it have high variability? (Precision)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

Validity

A

Is this the right way to measure this?

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

Reactivity

A

Subject’s awareness of how the measurement works

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

Measurement Scales

Nominal

A
  1. Label (Qualitative)
  2. No Rank or Order

Lowest

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

Measurement Scales

Ordinal

A
  1. Labels
  2. Rank or Order
  3. Distance Random

Higher than nominal

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

Measurement Scales

Interval

A
  1. Numerical
  2. Rank or Order
  3. No “True” 0
  4. Distance NOT Random

Higher than ordinal

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

Measurement

Ratio

A
  1. Numerical
  2. Distance NOT Random
  3. “0” Value

Highest

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

6 Ways to Determine Validity

Face Validity

A

Does it seem right?

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

6 Ways to Determine Validity

Content Validity

A

Does it seem right based on how other people do it?

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

6 Ways to Determine Validity

Predictive

A

Can it predict behavior?

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

6 Ways to Determine Validity

Concurrent Validity

A

Does it correlate with how the person is currently behaving?

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

6 Ways to Determine Validity

Convergent

A

Does it correlate with another type of measurement of the same thing?

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

6 Ways to Determine Validity

Discrimination

A

Does it NOT correlate with something that isn’t related?

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

Pearson’s R

A

How close a correlation is to a straight line.

17
Q

Pearson’s R

Strong Positive Correlation

A

1 (Perfect Straight Line)

18
Q

Pearson’s R

Strong Negative Correlation

A

-1 (Perfect Straight Line)

19
Q

Ways to Test Reliability

Split-Half Reliability

A
  1. Split Questions/Items in half
  2. Do the results correlate?
20
Q

Ways to Test Reliability

Cronchbach’s Alpha

A
  1. Split-Half Test
  2. Tests with every possible set of halves
21
Q

Ways to Test Reliability

Item-Total Correlation

A

Measurement from each question or item is compared to the measurement for everything

22
Q

Ways to Test Reliability

Interrater

A

Person making observations & recording data
2 or more raters observe same thing –> consistent ratings?