L4

Question 1

What is the construct definition

Answer 1

what do we want to measure?

Question 2

What is construct operationalizatoin?

Answer 2

How can we measure it?

Question 3

Behavioural vs attitudinal?

Answer 3

Behavioural is on behavior and actions.
Attitudinal is on memories, though, and consumer attitudes

Question 4

What are the comparative. scaling techniques?

Answer 4

1. Paired comparison technique. Here, 2 items are paired and compared
2. Rank order scaling. Here, an item is ranked from dislike to like
3. Constant sum Scale. Here, a constant sum or points are given. so 100 points between 3 items

Question 5

what are the non-comparative scaling techniques?

Answer 5

1. Likert scale which is the agreement degree to a single concept.
2. Differential scale which is where two opposing concepts are placed opposite to each other

Question 6

Why are multi-item quesiton scales used instead of single-item sclaeS?

Answer 6

to minimize error which leads to reliability

Question 7

what is the multiple item measurement theory?

Answer 7

Here, the theory is that the observed score from a test consists of 3 scores. The true score that you attained, plus the systematic error, and the random error.
X0=Xt+Xs+Xe
where X0 is the observed score, Xt is the true score, Xs is the systematic error, and Xe is the random error

Question 8

What is reliability and how does that relate to the multiple item measurement theory?

Answer 8

Reliability is the extent to which the measurement instrument gives the same results when the measurement is repeated.
Reliability is when there is no random error, thus when Xe = 0

Question 9

What is validity and how does it relate to the multiple item measurement theory?

Answer 9

validity is the extent to which the measurement instrument measures the construct we intentd to measure.
This happens when there is no systematic error AND no random error.
So when xs = 0 and xe= 0

Question 10

What are the 2 tests for assessing scale reliability

Answer 10

1. test-retest - here, the same test is repeated with the same respondents later
2. internal consistnecy

Question 11

what is internal consistency

Answer 11

this is measured with cronbach’s alpha which is a function of the number of items and mean correlations among items. The more items, the higher the alpha. The higher the mean correlation among items, the higher the alpha.
Threshold is when a > 0.7.

Question 12

alpha.drop

Answer 12

alpha.drop is a function that helps us determine what item can be dropped to change the alpha. If the alpha is bigger when the item is dropped, it means it should be dropped. Whereas if it becomes lower than the general alpha, then it should not be dropped.