點數與標示語對評定量尺反應的影響(II)
Date Issued
2000
Date
2000
Author(s)
DOI
892413H002012
Abstract
The purpose of the study is to
investigate the effects of anchor labels and
number of response categories on participant
responses on Chinese rating scales and the
psychometric properties of the scales. Five
types of rating scales with different formats
of labels were constructed. Number of
response categories varied from there to nine.
A total of 29 different types of rating scales
have been developed. The results indicated
that the effects of anchor labels mainly came
from the differences between participant
responses on the positive and the negative
scales. Scale means, scale standard
deviations, and internal consistency
reliability coefficients have been affected.
The magnitude of the effects depended on the
content of the scale or the heterogeneity of
participant responses on the scale. The
scales with homogeneous participant
responses may be more sensitive to the
changes of anchor labels than those with
heterogeneous participant responses. The
effects of the number of response categories
on internal consistency reliability also
depended on content of the scales or the
degree of heterogeneity of participant
responses on the scales. The results of the
present investigation suggested that anchor
labels and number of categories of a rating
scale may influence participant responses and
psychometric properties of the scale.
Researchers better consider the heterogeneity
of participant responses on the scale and
select appropriate scale format.
investigate the effects of anchor labels and
number of response categories on participant
responses on Chinese rating scales and the
psychometric properties of the scales. Five
types of rating scales with different formats
of labels were constructed. Number of
response categories varied from there to nine.
A total of 29 different types of rating scales
have been developed. The results indicated
that the effects of anchor labels mainly came
from the differences between participant
responses on the positive and the negative
scales. Scale means, scale standard
deviations, and internal consistency
reliability coefficients have been affected.
The magnitude of the effects depended on the
content of the scale or the heterogeneity of
participant responses on the scale. The
scales with homogeneous participant
responses may be more sensitive to the
changes of anchor labels than those with
heterogeneous participant responses. The
effects of the number of response categories
on internal consistency reliability also
depended on content of the scales or the
degree of heterogeneity of participant
responses on the scales. The results of the
present investigation suggested that anchor
labels and number of categories of a rating
scale may influence participant responses and
psychometric properties of the scale.
Researchers better consider the heterogeneity
of participant responses on the scale and
select appropriate scale format.
Subjects
rating scales
Likert-type scales
anchor labels
number of response categories
internal
consistency reliability
consistency reliability
Publisher
臺北市:國立臺灣大學心理學系暨研究所
Type
report
File(s)![Thumbnail Image]()
Loading...
Name
892413H002012.pdf
Size
53.48 KB
Format
Adobe PDF
Checksum
(MD5):7cbdc5acb15e0809586d348fa5dc3fbf