Chapter 5: Measuring Concepts

Question 1

When is a measure reliable?

Answer 1

When it’s consistent and precise, it doesn’t fluctuate much

Question 2

Any measurement is comprised of what 2 things?

Answer 2

True Score

Measurement error

Question 3

What is a True Score? What is a Measurement error?

Answer 3

True score- the score on the variable being tested or measured if there was no error
Measurement error- Anything else that affects the true score

Question 4

Would an unreliable measure have high or low measurement error?

Answer 4

High

Question 5

What is a correlation coefficient?

Answer 5

Way to assess reliability, a number that tells us how strongly 2 variables are related to one another, most common is the Pearson Product Moment Correlation Coefficient

Question 6

Summarize the Pearson Correlation method

Answer 6

• Symbolized as r
• Ranges from -1.00 to +1.00
• 0.00 means no relationship
• Stronger relationships = r closer to to +/- 1.00

Question 7

What are the classifications of r in the Pearson Correlation Coefficient

Answer 7

0. 00- no relationship
1. 30- weak relationship
2. 50- moderate relationship
3. 80- strong relationship
4. 00- perfect relationship

Question 8

r=.50, what is the coefficient of determination? What does this mean?

Answer 8

r^2=.25
.25 x 100% = 25%
An r equal to .50 is 25% better than an r equal to 0.00

Question 9

What is the coefficient of alienation of r=.50

Answer 9

(1-.50^2)

.75 x 100% = 75%

Question 10

What are 3 ways to assess reliability?

Answer 10

Test-retest reliability, Internal consistency reliability, Interrater reliability

Question 11

What are the 3 types of internal consistency?

Answer 11

1. Split half reliability- Correlate an individual’s total score on the first half of the test to their total score on the second half of the test
2. Cronbach’s Alpha- Scores on each item are correlated with scores on every other item
3. Item-Total Correlations- Correlate each item with the total score based on all of the items

Question 12

What is interrater reliability?

Answer 12

When multiple raters agree in their observations of the same thing (Cohen’s kappa)

Question 13

What do we consider when determining the accuracy of a measure?

Answer 13

Validity

Question 14

Define construct validity. What are the 6 indicators of it?

Answer 14

Construct validity is how well the operational definitions of variables reflect the true meaning of the variables

1. Face validity
2. Content validity
3. Predictive validity
4. Concurrent validity
5. Convergent validity
6. Discriminant validity

Question 15

Define reactivity

Answer 15

When awareness of being measured changes an individual’s behavior

Question 16

4 kinds of measurement scales that describe the different values that a variable can possess:

Answer 16

Nominal scales
Ordinal scales
Interval scales
Ratio scales

Question 17

Define nominal scales:

Answer 17

are qualitative, they categorize with no numerical properties
For example, gender, family structure, major, etc.
One of these is not better than the other, they are just different

Question 18

Define ordinal scales:

Answer 18

are qualitative, rank order the levels of the variables being tested.
For example, place in a race, class rank, star ratings,

Question 19

Define interval scales:

Answer 19

are quantitative, differences between levels are meaningful and equal in size, but have no absolute 0
For example temperature, measure of intelligence (intelligence can’t be 0)

Question 20

Define ratio scales:

Answer 20

are quantitative, have an absolute zero point that indicates the absence of the variable being measured.
For example, length, weight, or time