Ch 13 Scaling Flashcards

1
Q

scale

A

a number of items that combine to form a composite score on a variable

use when a single-item measuring instrument is inadequate

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

four advantages of scales

A

improved validity

improved reliability

increased level of measurement

increased efficiency in data handling

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

5 steps to developing a scale

A

develop or locate many potential scale items, far more than will appear in the final scale

eliminate items that are redundant, ambiguous, or, for some other reason, inappropriate for the scale

pretest the remaining items for validity, reliability, or other measurement

eliminate items that don’t pass

repeat steps 3 and 4 to reduce the scale to items required

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

three sources of scale items

A

researcher’s imagination

group of people (judges) who are knowledgeable in topic

the people who are the focus of the research project

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

good scale items have several characteristics which are…

A

validity

range of variation

unidimensionality - measures one variable

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

5 different scaling formats

A

likert scales

thurstone scales

semantic differential scales

guttman scales

multidimensional scales

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

likert scale

A

consists of a series of statements, with each statement followed by a number of ordered response alternatives

ex responses with strongly agree to strongly disagree

it’s an example of a summated rating scale where a person’s score is determined by summing the number of questions answered in a certain way

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

nondiscrimination items

A

those that are responded to in a similar fashion by both people who score high and those who score low

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

discriminatory power score (DP score) can be used to identify ___

A

nondiscriminating items

Compute total score for each respondent

Determine quartiles

For each item, classify upper and lower quartiles by response value

Compute weighted totals

Compute weighted means

Subtract lower quartile weighted mean from upper quartile weighted mean

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

advantages to likert scale

A

offers respondents a range of choices

data considered to be ordinal which lets us use more powerful statistical procedures

fairly straightforward to construct

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

disadvantages to likert scale

A

be careful in interpreting a single score because its a summary of a lot of info

when we summarize data we lose some info

the summary score might hide info abut patterns of variation in responses or about possible multidimensionality of the scale

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

thurstone scales

A

constructed so they use equal-appearing intervals, that it is assumed that the distance between any two adjacent points on the scale is the same

Items range from 1 (most unfavorable) to 11, (most favorable). Respondents select the item(s) with which they most agree

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

Pro to Thurstone scale

A

Provides interval-level data

People can respond to the items more quickly

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

Con to Thurstone scale

A

They are costly and difficult to construct

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

Semantic differential scales

A

presents the respondent with a stimulus, such as a person or event, to be rated on a scale between a series of polar opposite adjectives

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

pros to semantic differential scales

A

requires only 4 to 8 adjective pairs for each dimension to reach reliabilities of .80 or better

much easier and less time-consuming to construct

adjective pairs are more adaptable

17
Q

cons to semantic differential scales

A

generate ordinal data

18
Q

Guttman scales

A

items have an inherently progressive nature relating to the intensity of the variable being measured.

There is only one pattern of responses that will yield any given score on the scale.

uses reproducibility

19
Q

reproducibility

A

the ability of each individual’s composite score to predict exactly the items agreed and disagreed with

R = 1 - # of errors / (# of items) x (# of subjects)

20
Q

multidimensional scales

A

measure variables composed of more than a single dimension.

Locates responses in two, three or N-dimensional space

21
Q

research for intimate partner violence

A

scale based on equality, fairness and care

22
Q

Response bias

A

is the tendency for people’s answers to questions to be influenced by things other than their true feelings, beliefs, and behaviors.

sources are:
response set

response pattern anxiety

social desirability effect

23
Q

response set

A

some people tend to be either yes or no, tending to agree or disagree with statements regardless of their content

24
Q

response pattern anxiety

A

some people become anxious if they have to repeat the same response all the time, and they change their response to avoid doing so

25
Q

social desirability effect

A

people’s tendency to give socially acceptable, popular answers to present themselves in a good light

26
Q

5 ways to reduce response bias

A

designing statements so positive statements are not always an expression of the same attitude

present sensitive issues in a neutral and non-judgmental context

ordering of questions - can be asked in a hierarchical order, beginning with the less sensitive to more sensitive

use an interspersed pattern for the items, where socially acceptable items are interspersed with less acceptable items

funneling - for ex asking questions about their city, then community, then neighbors, then family

27
Q

scaling is the best example of a strong linkage between practice and research. true or false?

A

true

28
Q

human service practitioners use scales to assess specific problems for intervention. true or false?

A

true

29
Q

practitioner uses scales to document the progress of intervention. true or false?

A

true

30
Q

human service organizations rely on scaling as part of their overall program quality and effectiveness-monitoring process. true or false?

A

true

31
Q

rapid assessment instruments (RAI)

A

Short, easy to administer, easy to complete

Simple, fast scoring procedure

RAIs such as the MAST found superior in assessing severity and nature of problems

Improved reliability of assessment