ArticlePDF Available

Five point vs. eleven point scales: Does it make a difference to data characteristics?

Authors:
John Dawes at University of South Australia
John Dawes
Download full-text PDF
Read full-text
Download citation

Figures

Figures - uploaded by John Dawes
Author content
All figure content in this area was uploaded by John Dawes
Content may be subject to copyright.
Content uploaded by John Dawes
Author content
All content in this area was uploaded by John Dawes on Dec 18, 2015
Content may be subject to copyright.
1
Five Point vs. Eleven Point Scales:
Does It Make A Difference To Data Characteristics ?
(published in Australasian Journal of Market Research Vol 10 No. 1 2002)
John Dawes
Marketing Science Centre
University of South Australia
Email John.Dawes@marketingscience.info
Keywords: Scales, split-sample experiment.
2
Five Point vs. Eleven Point Scales:
Does It Make A Difference To Data Characteristics ?
Abstract
This study examines whether the number of scale points used in a market research survey affects
the resultant data. It uses two 'split sample' surveys, one using face to face interviews and one
gathered using telephone interviews. In each case, a subset of the sample was administered
questions using a five point scale and another subset was administered an eleven point 'zero to
ten' scale. The results show that the eleven point scale produces data that is essentially the same
as that produced by the five point scale in terms of mean, after allowing for the five point scale to
be re-scaled for comparability. However, the eleven point scale produced data with more
variance (coefficient of variation) than the five point scale. There were some differences between
the scale types in terms of kurtosis and skewness, but these were not systematic.
Background & Previous Research
One of the most ubiquitious tools of the marketing or academic researcher is the use of multiple
category numerical scales. The question of finding the 'optimal' number of scale points has been
discussed from a variety of perspectives. Several studies have used simulations to examine
issues of information recovery and the precision of data (Givon and Shapira 1984; Green and Rao
1970; Lehmann and Hulbert 1972). Others have used survey data. Hulbert (1975) examined the
degree to which respondents used different response categories in a very lengthy questionnaire,
where the respondents were not provided with pre-set response categories, but simply allocated
numbers to each question. He found that the mean number of different responses was between
3
six and ten. This suggests that ten may be an upper limit in terms of the degree to which
respondents can discriminate between questionnaire stimuli, confirming earlier research from the
1950's.
Cox (1980) extensively reviewed previous studies to conclude that scales with two or three
response categories were inadequate - (rebutting Jacoby & Mattell (1971)), but there was little
marginal gain from using more than nine categories. Clarke (2000) found that increasing the
number of scale categories from three to five reduced extreme responses but beyond five
categories there was little effect. Holmes (1974) conducted an experiment comparing the
responses for unstructured versus structured scales and other derivations of scale construction,
but did not test for differences according to the number of response categories. Schertzer &
Kernan (1985) investigated the semantic intensity, ordinality and interval level qualities of
response labels according to the number of labels used (such as three, four, five categories of
response label etc.), however respondents did not rate objects per se, they matched each response
label to a point on a 100 point scale. Alwin (1997) found that eleven point scales performed
better than seven point scales in terms of reliability and validity. Grigg (1980) found that an
eleven point scale produced more dispersion in responses than a seven point scale. Grigg’s result
is closer to the issue examined in this paper, however was conducted twenty years ago and used a
self completion method. More analysis using data gathered via telephone as well as by self
completion would be desirable.
Several studies have also examined whether the inclusion of scale mid points affect responses.
Worcester & Burns (1975) found that descriptors such as tend to agree represented more positive
sentiment on a four point scale compared to a five point scale. Spagna (1984) found that the
omission of a mid point had most effect on increasing the frequencies of the neighbouring scale
4
responses. Si & Cullen (1998) found that mid points did not affect mean response levels. Dawes
(2001) found a scale mid point resulted in fewer positive responses. Garland (1991) found the
opposite - a mid point resulted in fewer negative ratings. So there is mixed evidence on this
issue.
In general there is support for the view that more scale categories reduce the incidence of extreme
responses, and may produce more dispersion in the data. However, as this review demonstrates,
not much is known in descriptive terms about the effect of different numbers of scale response
categories on basic data characteristics such as the relative mean, variance, shape and so on - and
certainly not for data gathered via telephone. Yet these are important issues:
• The mean (average) forms the basis of many evaluations contained in market research
reports. Would the mean (relative to the highest possible score obtainable from the scale
used) have been different if a different number of response categories had been used ?
• The variance (ie standard deviation) is integral to many tests of statistical inference.
For example, if one has a hypothesis that responses may differ according to a certain
grouping variable, ideally support for this would come in the form of little within-group
variance and considerable across-group variance. Does the number of scale points affect
the amount of variance in the data ?
• Some statistical techniques are sensitive to kurtosis. Do scales with different numbers
of response categories produce different degrees of kurtosis ?
• The degree of skewness in the data may affect various statistical analyses. Do scales
with different numbers of response categories produce different degrees of skewness ?
This study sets out to address these four issues using results from two survey studies.
5
Data
Two data sets are used in the analysis. The first comprised a convenience sample of 301
undergraduate students who were interviewed at various campuses of an Australian university.
The topic was the campus cafeteria at that location. Respondents were interviewed face-to-face
at various locations on each campus by IQCA accredited interviewers. A voucher for a free cup
of coffee was offered as an incentive to participate. There was no criteria for inclusion except
that the respondent was a student at that campus.
Of the 301 respondents, 146 were administered a series of questions and asked to respond using a
five point "1 = strongly disagree … 5 = strongly agree" scale. The other 155 were administered
the same questions but were asked to respond on an eleven point zero to ten scale, where zero
meant no agreement and ten denoted 100% agreement. Allocation of respondents to the
respective scale type was random. The response rate for this survey was not recorded but
qualitative information from the interviewers was that it was approximately 50%.
The second data set was part of a telephone survey of 751 people who had used taxicabs in the
past six months. The sample was selected from a random generation based on the electronic
telephone directory. Of the 751 respondents, 151 were administered a series of questions and
asked to respond using a five point "strongly disagree - strongly agree' scale. The other 592 were
administered the same questions but were asked to respond on an eleven point zero to ten scale,
where zero meant no agreement and ten denoted 100% agreement. The difference in sample
sizes for the two groups was due to commercial considerations. The sponsor of the study was
particularly interested in obtaining scores "out of ten" to retain comparability with previous
research, but was prepared to allow a sub group of respondents to be administered the other five
point scale to determine how it might affect results. As with the first study, allocation of
6
respondents to the respective scale was random. The response rate for this survey, based on
automated call management software was 49%. Therefore we have confidence that the results
are likely to be typical of such surveys since they did not suffer from unduly low response rates.
Analysis
The first issue concerns the possible effect on the mean value of the variables. In order to
determine whether the two different scale types had any effect on the level, the mean score for
each question was tabulated. To compare the mean scores for the two scale types, the 5 point
scale data was re-scaled, by recoding the five point scores to a score out of ten as shown in the
diagram below:
0
1
2
3
4
5
6
7
8
9
10
1
3
5
1 and 5, the extreme points on the five point scale were transformed to 0 and 10, the
corresponding extreme points on the eleven point scale. 3 was transformed to the corresponding
mid-point on the eleven point scale, namely 5. 2 and 4 were fixed mid-way between the extreme
point and the mid point. This meant that 2 was transformed to 2.5 and 4 was transformed to 7.5.
Obviously the analysis based on this re-scaling depends on the assumption of equal intervals for
both scales. This presumption is arguable with some writers findings showing a lack of
intervality (eg. Schertzer and Kernan 1985) and others much more supportive (Crask and Fox
1987). The assumption of equal interval qualities was assumed for this analysis, and if the
different scale types produced similar results this assumption would be justified. Table 1 shows
the differences between the eleven point and five point (rescaled) data for both data sets; so all
7
scores are “out of 10”. The Mean Absolute Difference (MAD) is shown at the bottom of the
table, which is the average of differences, not accounting for the sign of the difference. The data
are ordered by mean score.
INSERT TABLE 1 HERE
On average, the use of a five point scale compared to an eleven point scale results in quite a small
difference in mean scores (when re-scaled). The MAD (Mean Absolute Difference) is 0.25 of a
scale point on a zero to ten scale (once the 5 point data has been re-scaled) for both data sets. If
we do take the sign of the differences into account, the eleven point scale produces mean scores
that are around 0.1 of a scale point higher than the re-scaled five point scale. So the difference is
minor.
It is also noteworthy that there are more cases where the use of the eleven point scale results in a
higher comparative score than a lower comparative score. In Table 1, for data set 1 the eleven
point scale produces higher mean scores than the re-scaled five point scale in five out of seven
instances. In data set 2 this occurs in eight out of thirteen instances.
There is also a tendency for this effect to be more marked where the particular question attracted
a lower average score across respondents. We see in Table 1, for data set 1 the difference
between the 11 point score and the 5 point re-scaled score is larger for question 8 (the question
with the lowest mean score). This occurs again in data set 2 (still in Table 1) for the two lower
scoring questions, namely Q 11 and Q7.
Variance
To examine the impact of the different scales on the variance, we report on the coefficient of
variation for each variable. The coefficient of variation is the standard deviation divided by the
8
mean score, then multiplied by 100. It facilitates the comparison of variances for items with
different mean levels.
Table 2 shows the results for both data sets. The results suggest that the use of an eleven point
scale results in a greater amount of variance (relative to the mean score) compared to a five point
scale - note how the "difference" column is mostly positive. The results from data set 2 confirm
this, as shown in the right-hand side "difference" column for data set 2 in Table 2. The rationale
for this result is that the eleven point scale simply provides a wider range of responses, therefore
produces more dispersion in responses (consistent with Grigg 1980). Note that data set 2 overall
has much higher coefficients of variation for most questions than data set 1. This result is not
relevant to the major issue examined in the study, (which is within-data set differences according
to the number of scale points used) but the reason for this result should be clarified. It is simply
due to a much higher degree of variation in perceptions of service quality by users of taxicabs
compared to student users of cafeteria services.
INSERT TABLE 2 HERE
Kurtosis
Kurtosis is the degree to which the data is peaked, or flat. Kurtosis is also 'scale free' so it is
possible to directly compare the results for the five and eleven point scales without any need to
transform the data. Table 3 shows that for data set 1, the eleven point scale data has higher levels
of kurtosis in four out of six cases. In data set 2 the eleven point scale data produces lower levels
of kurtosis in eight out of 13 cases. Therefore, there does not appear to be any systematic
association between the two scale types and the amount of kurtosis, in these two data sets.
INSERT TABLE 3 HERE
9
Skewness
Negative skewness denotes the extent to which there is a "tail" below the mean. A positive skew
means the opposite - a "tail" above the mean. The data from both surveys is all negatively
skewed, with mean scores all above the scale mid-point.
The results for skewness according to scale type are rather mixed. In Table 4, for data set 1 there
is no clear pattern in the difference in skewness between the two types of scales. For the second
data set, there is a tendency for the 11 point scale to produce a more negative skewness figure (a
longer tail below the mean). However, this appears in only one of the data sets. The conclusion
is that there is no systematic relationship between the number of scale points and skewness.
INSERT TABLE 4 HERE
10
Discussion & Conclusions
This study found in two "split sample" experiments that:
• In both cases, the difference in mean scores produced by the eleven point scale was small,
compared to the re-scaled scores produced by the five point scale. The eleven point (zero
to ten) scale produced more instances of slightly higher mean scores than the five point
scale - after allowing for the five point scale to be re-scaled.
• Furthermore, this tendency of the eleven point scale to produce slightly higher
comparative scores was more marked for questions that attracted lower overall mean
scores to begin with.
• An eleven point (zero to ten) scale produced higher levels of variance (ie coefficient of
variation) compared to a five point scale
• There was no systematic relationship between the use of an eleven point vs. a five point
scale in terms of kurtosis
• There was no systematic relationship between the use of an eleven point vs. a five point
scale in terms of skewness
It would appear that the choice of the number of response categories depends on the purpose of
the study. If the major purpose is simply to obtain numerical responses and produce mean scores,
the number of response categories does not make a marked difference to the mean score obtained,
relative to the upper scale limit. In terms of interpretability, a ‘score out of ten’ has some appeal
in terms of managerial interpretability.
11
However, if there is an intent to examine dependence relationships between scale variables using
tools such as regression, for example, a scale with more response categories could be more useful
as it appears to result in more variance in the data.
The results from this study are also 'good news' for those who commission or undertake market
research for clients with historical data they wish to preserve comparability. For example, if an
organisation had years of data that was gathered using a five point scale, it is likely that it could
be converted or transformed to make it comparable with other data that was gathered using more
scale categories. Comparability would be more of an issue with items that received lower scores.
In this study, the rescaling process had the effect of ‘deflating’ the questionnaire variables that
received lower scores using the five point scale. The author plans to conduct more replications of
this result to test the robustness of the findings.
12
References
Alwin, Duane F (1997), "Feeling thermometers vs 7-point scales," Sociological Methods and
Research, 25 (3), 318-51.
Clarke III, Irvine (2000), "Extreme Response Style in Cross-Cultural Research: An Empirical
Investigation," Journal of Social Behavior & Personality, 15 (1), 137-52.
Cox III, Eli P. (1980), "The Optimal Number of Response Alternatives for a Scale: A Review,"
Journal of Marketing Research, 17 (November), 407-22.
Crask, Melvin R. and Richard J. Fox (1987), "An Exploration of the Interval Properties of Three
Commonly Used Marketing Research Scales: A Magnitude Estimation Approach," Journal of
the Market Research Society, 29 (No. 3), 317-39.
Dawes, John G. (2001), "The Impact of Mentioning a Scale Mid-Point in Administering a
Customer Satisfaction Questionnaire Via Telephone," Australasian Journal of Market Research,
9 (No. 1, January), 11-18.
Garland, Ron (1991), "The Mid-Point on a Rating Scale: Is It Desirable?," Marketing Bulletin, 2,
66-70.
Givon, Moshe M. and Zur Shapira (1984), "Response to Rating Scales: A Theoretical Model and
Its Application to the Number of Categories Problem," Journal of Marketing Research, 21
(November), 410-19.
Green, Paul E. and Vithala R. Rao (1970), "Rating Scales and Information Recovery - How
Many Scales and Response Categories to Use?," Journal of Marketing, 34 (July), 33-39.
Grigg, A.O. (1980), "Some Problems Concerning the Use of Rating Scales for Visual
Assessment," Journal of the Market Research Society, 22 (No. 1), 29-43.
Holmes, Cliff (1974), "A Statistical Evaluation of Rating Scales," Journal of the Market
Research Society, 16 (No. 2, April), 87-107.
Hulbert, James (1975), "Information Processing Capacity and Attitude Measurement," Journal of
Marketing Research, 12 (February), 104-06.
Jacoby, Jacob and Michael S. Matell (1971), "Three Point Scales Are Good Enough," Journal of
Marketing Research, 8, 495-500.
Lehmann, Donald R and James Hulbert (1972), "Are Three-Point Scales Always Good
Enough?," Journal of Marketing Research, 9 (November), 444-46.
13
Schertzer, Clinton B. and Jerome B. Kernan (1985), "More on the robustness of Response
Scales," Journal of the Market Research Society, 27 (No. 4, October), 261-82.
Si, Steven and John Cullen (1998), "Response Categories and Potential Cultural Bias: Effects of
an explicit middle point in cross-cultural surveys," Internationl Journal of Organizational
Analysis, 6 (3), 218-31.
Spagna, Gregory J. (1984), "Questionnaires: Which Approach Do You Use?," Journal of
Advertising Research, 24 (No. 1, February/March), 67-70.
Worcester, Robert M. and Timothy R. Burns (1975), "A Statistical Examination of the Relative
Precision of Verbal Scales," Journal of the Market Research Society, 17 (No. 3), 181-97.
14
Table 1: Differences in Means for Data Set 1 & 2
Data set 1
Data set 2
Q no.
11 pt.
scale
5 pt.
(rescaled)
Diff.
Q. no
11 pt.
scale
5 pt.
(rescaled)
Diff.
Q12
7.6
7.8
-0.2
Q8
7.7
7.7
0
Q11
7.8
7.6
+0.2
Q9
7.5
7.6
-0.1
Q6
7.3
7.6
-0.3
Q3
7.3
7.2
+0.1
Q10
7.6
7.5
+0.1
Q10
6.9
6.9
0
Q7
7.7
7.5
+0.2
Q4
6.7
6.2
+0.5
Q9
7.3
7.2
+0.1
Q6
6.4
6.3
+0.1
Q8
6.7
5.9
+0.8
Q5
6.3
6.2
+0.1
Q2
5.9
6.0
-0.1
Q12
5.9
5.6
+0.3
Q13
5.8
5.4
+0.4
Q1
5.7
5.8
-0.1
Q11
5.6
4.9
+0.7
Q7
5.5
4.7
+0.8
MAD
0.25
MAD
0.25
15
Table 2: Differences in Variance for Data Set 1 & 2
Data set 1
Data set 2
Coeff.
Var.
(11 point)
Coeff.
Var.
(5 point)
Difference
(11 - 5)
Coeff.
Var.
(11 point)
Coeff.
Var.
(5 point)
Difference
(11 - 5)
Q8
28.7
30.1
-1.4
Q11
51.9
39.7
+12.2
Q9
27.5
24.0
+3.5
Q7
47.6
39.4
+8.2
Q6
24.9
20.9
+4.0
Q5
42.5
33.5
+9.0
Q10
23.5
22.9
+0.6
Q13
42.2
32.8
+9.4
Q11
22.6
20.0
+2.6
Q2
42.0
27.9
+14.1
Q7
21.0
18.5
+2.5
Q1
40.1
29.3
+10.8
Q12
18.0
17.3
+0.7
Q12
37.4
29.3
+8.1
Q6
36.5
29.3
+7.2
Q4
36.3
35.3
+1.0
Q10
33.7
26.1
+7.6
Q3
30.3
25.8
+4.5
Q9
28.5
28.4
+0.1
Q8
23.6
24.3
-0.7
MAD
2.2
MAD
7.1
16
Table 3 Differences in Kurtosis for Data Set 1 & 2
Data set 1
Data set 2
Kurtosis
(11 point
scale)
Kurtosis
(5 point
scale)
Difference
(11 point -
5 point)
Kurtosis
(11 point
scale)
Kurtosis
(5 point
scale)
Difference
(11 point -
5 point)
Q6
2.1
1.9
+0.1
Q1
-0.3
-0.43
-0.13
Q7
1.6
2.2
-0.6
Q2
-0.52
-0.38
+0.14
Q8
0.1
-0.6
+0.7
Q3
0.12
0.43
+0.31
Q9
1.6
1.9
-0.3
Q4
-0.39
-0.93
-0.54
Q10
2.1
0.8
+1.3
Q5
-0.54
-0.73
-0.19
Q11
3.9
0.5
+3.4
Q6
-0.35
-0.48
-0.13
Q12
2.5
2.6
-0.1
Q7
-0.45
-0.97
-0.52
Q8
0.86
1.23
+0.37
Q9
1.32
0.78
-0.54
Q10
0.53
0.67
0.14
Q11
-0.79
-0.85
-0.06
Q12
-0.09
0.39
+0.48
Q13
0.06
-0.22
-0.28
MAD
0.8
MAD
0.3
17
Table 4 Differences in Skewness for Data Set 1 & 2
Data set 1
Data set 2
Skewness
(11 point)
Skewness
(5 point)
Difference
(11 point -
5 point)
Skewness
(11 point)
Skewness
(5 point)
Difference
(11 point -
5 point)
Q6
-1.0
-1.0
0.0
Q1
-0.42
-0.18
-0.24
Q7
-0.91
-1.0
+0.1
Q2
-0.3
-0.26
-0.04
Q8
-0.54
-0.38
-0.2
Q3
-0.78
-0.85
0.07
Q9
-1.15
-1.28
+0.1
Q4
-0.63
-0.36
-0.27
Q10
-1.11
-0.97
-0.1
Q5
-0.5
-0.32
-0.18
Q11
-1.54
-0.79
-0.8
Q6
-0.46
-0.45
-0.01
Q12
-0.81
-0.96
+0.2
Q7
-0.26
-0.08
-0.18
Q8
-0.92
-1.19
0.27
Q9
-1.14
-1.21
0.07
Q10
-0.9
-0.84
-0.06
Q11
-0.41
-0.04
-0.37
Q12
-0.38
-0.56
0.18
Q13
-0.68
-0.47
-0.21
MAD
0.21
MAD
0.17
... In einzelnen Fällen gab es in der Umfrage aus 2010 eine siebenstufige Antwortmöglichkeit. Diese wurde im Vorfeld in eine fünfstufige Skala umgewandelt (Dawes 2002), um die Daten vergleichbar zu machen. ...
View
... The questionnaire was developed through several rounds of discussions with experts in design and psychology. Scholars specializing in structural equation modeling suggested that a larger-scale questionnaire provides better reliability and validity, as demonstrated by empirical research [80], [81]. Therefore, this study employed a sevenpoint Likert scale (ranging from 1= strongly disagree to 7= strongly agree) and a total of 37 items measuring nine latent variables. ...
Empirical Study on the Factors Affecting User Stickiness of Online Visual Art Platform From the Perspective of User Experience
Article
Full-text available
  • Jan 2023
This paper aims to analyze the factors affecting user stickiness of the online visual art platform by developing a Push-Pull-Mooring (PPM) framework from the perspective of user experience. The study collected 332 valid responses through an online questionnaire and tested the research model using structural equation modeling. The results show that push effects (convenience and perceived security risk), pull effects (function experience, content experience, interactive experience, emotion experience), and mooring factor (relationship inertia) positively impact user stickiness. These findings are helpful for the online visual art platform to understand the influencing factors of user stickiness and adjust strategies from the perspective of user experience. The research contributes to the field of online visual art platform by proposing and verifying hypotheses on user stickiness through the PPM framework and constructing a structural equation model from the perspective of user experience.
View
... Each user who completed the questionnaire received a WeChat bonus of 5 RMB as a reward to express our appreciation for their time and truthful answers to each question. As a large scale performed better than a small scale in terms of reliability and validity in an empirical study [90], all items in Section 3 were measured on a 7-point Likert scale (1: "strongly disapprove", 2: "disapprove", 3: "somewhat disapprove", 4: "fair", 5: "somewhat approve", 6: "approve", and 7: "strongly approve"). ...
Is Everyone an Artist? A Study on User Experience of AI-Based Painting System
Article
Full-text available
  • May 2023
Artificial Intelligence (AI) applications in different fields are developing rapidly, among which AI painting technology, as an emerging technology, has received wide attention from users for its creativity and efficiency. This study aimed to investigate the factors that influence user acceptance of the use of AIBPS by proposing an extended model that combines the Extended Technology Acceptance Model (ETAM) with an AI-based Painting System (AIBPS). A questionnaire was administered to 528 Chinese participants, and validated factor analysis data and Structural Equation Modeling (SEM) were used to test our hypotheses. The findings showed that Hedonic Motivation (HM) and Perceived Trust (PE) had a positive effect (+) on users’ Perceived Usefulness (PU) and Perceived Ease of Use (PEOU), while Previous Experience (PE) and Technical Features (TF) had no effect (−) on users’ Perceived Usefulness (PU). This study provides an important contribution to the literature on AIBPS and the evaluation of systems of the same type, which helps to promote the sustainable development of AI in different domains and provides a possible space for the further extension of TAM, thus helping to improve the user experience of AIBPS. The results of this study provide insights for system developers and enterprises to better motivate users to use AIBPS.
View
... Moreover, the use of a Likert scale as opposed to the original RRS (Treynor et al., 2003) may difficult the comparability between results of both scales. In this sense, Dawes (2002) found that scales that employed different types of point scales produced the same data. ...
An approach to the etiology of effects in targets of workplace bullying
Article
Full-text available
Background: Workplace bullying constitutes a negative interaction, that owing to its frequency and duration, leads targets to an inferior position, having several effects on the targets. Personality interacts with workplace factors in the development of aggression, appearing in some cases as vulnerability, which is expressed in perceptions and behavior and in some cases in coping with rumination. Objectives: This study, on the one hand, aimed to detect the presence of not only the antecedents of such vulnerability but also the protective factors in the personal history of people who experience workplace bullying. On the other hand, it aimed to understand how workplace bullying activates the said vulnerability, so that the association between the antecedents and their effects could be determined. Methods: The study sample consisted of 108 people who had reported being victims of workplace bullying. Most of them were of Spanish origin (84%); women comprised 78% of the study participants, and the average age of this group was 47 years (SD = 9.07). Correlations, hierarchical regressions, multiple mediations, and a moderated mediation model were used to determine the said association, as well as its indirect effects. The results described a relation between antecedents and effects. Maternal cognition predicts memory recall. A high attainment level constitutes a protective factor from psychological discomfort in targets of workplace bullying. Implications: Promoting access to education and the search for elevation attainment, especially in women, will enhance executive function and act as a preventive measure for psychological discomfort.
View
... Strategic costing, strategic pricing and brand valuation is also very low in Nigeria. a-Mean based on five scale of GE=Great Extent-5, M=Moderate-4, SE=Small Extent -3 , VS=Very Small-2, NA=Not at All-1 Converted to a 7-point scale based on Dawes (2002) in order to compare results. Table 4 shows that accountants' participation in the strategic decision making include mostly in some small measures: evaluating options, generating options, developing details about options, taking necessary actions and problem identification and objectives proposition. ...
OPPORTUNITIES AND THREATS POSED BY ARTIFICIAL INTELLIGENCE ON THE ACCOUNTING PROFESSION:EMPHASIS ON THE SUB-SAHARA AFRICAN
Article
Full-text available
  • May 2023
“We stand on the brink of a technological revolution that will fundamentally alter the way we live, work, and relate to one another. In its scale, scope, and complexity, the transformation will be unlike anything humankind has experienced before. We do not yet know just how it will unfold, but one thing is clear: the response to it must be integrated and comprehensive, involving all stakeholders of the global polity, from the public and private sectors to academia and civil society” (Klaus Schwab, 2016). Overview This topic is discussed under the following sub-headings: I. Introduction: Preamble, Concepts of Depreciation and Change in Technology. ii. Definition of Terms: Artificial and Intelligence, Types of intelligence. iii. Definitions, Types, History, Fears and Challenges of Artificial Intelligence. iv. Comparative Analysis of Artificial and Natural Intelligence. v. Current applications of Artificial Intelligence vi. Opportunities provided by Artificial Intelligence. vii. Threats posed by Artificial Intelligence particularly in Sub-Sahara Africa. viii. Conclusion.
View
... Each user who completed the questionnaire received a WeChat bonus of 5 RMB as a reward to express our appreciation for their time and truthful answers to each question. As the large scale performed better than the small scale in terms of reliability and validity in the empirical study [89]. Therefore, all items in Section 3 were measured on a 7-point Likert scale (1 "strongly disapprove", 2 "disapprove", 3 "somewhat disapprove", 4 "fair", 5 "somewhat approve", 6 "approve", 7 "strongly approve"). ...
Everyone's an Artist? Exploring User Acceptance of AI-Based Painting Systems
Preprint
Full-text available
  • Apr 2023
Artificial intelligence (AI) applications in different fields are developing rapidly, among which AI painting technology, as an emerging technology, has received wide attention from users for its creativity and efficiency. This study aimed to investigate the factors that influence user acceptance of the use of AIBPS by proposing an extended model that combines the Extended Technology Acceptance Model (ETAM) with the AI-based Painting System (AIBPS).A questionnaire was administered to 528 Chinese participants, using validated factor analysis data and Structural Equation Modeling (SEM) was used to test the hypotheses. The findings showed that hedonic motivation (HM) and perceived trust (PE) had a positive effect (+) on users' perceived usefulness (PU) and perceived ease of use (PEOU), while previous experience (PE) and technical features (TF) had no effect (-) on users' perceived usefulness (PU). This study provides an important contribution to the literature on AIBPS and the evaluation of systems of the same type, which helps to promote the sustainable development of AI in different domains and provides a possible space for further extension of TAM, thus helping to improve the user experience of AIBPS. The results of the study provide insights for system developers and enterprises to better motivate users to use AIBPS.
View
... In two of the surveys some items were based on a 7-point Likert scale. In these cases, the 7-point Likert scales were transformed to 5 point scales following Dawes (2002), where the 7-point scale endpoints were anchored to the endpoints of the 5 point scales and the intervening scale values were inserted at equal numerical intervals. Items were purposely constructed following standard procedures for psychological research (Oppenheim, 1992;de Leeuw et al., 2008;Dillman et al., 2014;Ajzen, 2016) either on an ad-hoc basis if no scale for measuring a certain construct existed in the literature or as adaptations from publications in fisheries science (e.g., Arlinghaus and Mehner, 2005;Anderson et al., 2007). ...
View
... In set to investigate aesthetic preferences of participants about vertical greening; two architects were asked to answer 11 questions about each 28 photograph and residents were asked to answer one, 5-points rating Likert scale was used to answer each of the 11 questions. The rating scale ranged from 1 "the lowest to ratings" to 5 "the highest ratings", this rating scale was used in several preference studies (Stamps III, 2004;Dawes, 2002;Likert , 1932). Because of the large number of photographs in the visual survey to be rated by one group of participants comfortably (residents or architects); they were divided into two sets (A and B) ...
VERTICAL GREENING AS A STRATEGY FOR URBAN SUSTAINABLE DEVELOPMENT IN CAIRO Perceptions and Aesthetic Preferences of Architects and Residents
Thesis
  • Jan 2014
A rapidly growing literature presents vertical greening as a promising strategy for urban sustainable development with multiple environmental, economic, and psycho-social benefits. The aim of this thesis is to inform and guide eventual future efforts to promote the acceptance and application of vertical greening in Cairo. The research objectives were to investigate the perceptions and preferences of residents and architects about the strategy and their opinions about the perspective of implementing it in Cairo. Survey questionnaires were administrated to 63 male and female residents and 45 male and female architects from the Greater Cairo area. The large majority of participating architects indicated that they had some form of knowledge about vertical greening prior to the survey. But, at least for most of them, knowledge about the strategy appeared to be rather limited or superficial as answers to the questionnaire revealed inaccurate perceptions or misconceptions. In particular, there appeared to be excessive concerns about the amount of maintenance that vertical greening would require, about the cost of vertical greening installation, and about the damage that vertical greening could cause to building façades. Inaccurate perceptions were also evident in respondents’ characterization of the different vertical greening systems. Despite the various concerns expressed by architects and residents about the application of vertical greening, there was a clear majority of positive opinions about the suitability of the strategy for the specific context of Cairo. This suggests that, for most participants, the perceived potential benefits of vertical greening outweighed the concerns about its implementation. This is quite encouraging for the prospect of vertical greening implementation in Cairo. Findings also suggests that a relatively high level of agreement between architects and residents in relation to judgments of aesthetic quality of vertical greening installations. For both groups, aesthetic preferences were found to be related to a number of design characteristics (and more particularly color diversity, texture and complexity in design) and to a number of affective qualities (and more particularly celebration and vitality).
View
... Experts were asked to use the 5-points rating Likert scale to rate each photograph according to the different characteristics and the perceived different healing effect of every photograph. The rating scale ranged from 1 "the lowest to ratings" to 5 "the highest ratings", this rating scale was used in several studies that concerned assessing preference (Stamps III, 2004;Dawes, 2002;Likert , 1932). ...
Ph.D. by Yomna Amr-Vertical Gardens as a tool to improve urban quality of life
Thesis
  • Feb 2020
The output of this thesis contributes to the reinterpretation of” healing public spaces” from restoration and behavioral science to be transformed into the building morphology as vertical gardens in urban spaces. As an alternative strategy to improve general health and enhance urban quality of life. New Cairo faced problems that negatively affected the individual quality of life, limited green areas and affected the well-being. This thesis is proposing a vertical gardens’ framework as a building morphology providing restoration effect in urban public spaces of New Cairo. And it is achieved through 1-formulating a process of application and a morphological model for design and implementation process. Which is built from the theoretical study and analysis of literature and administrating international and local interviews to experts in the field. 2- Investigating the social acceptance towards the application, through survey questionnaires to users in urban spaces of New Cairo. 3- And assessing the restorative effect of the application; using a non-equivalent quasi-experimental approach. Results was presented in a quantitative manner through ANOVA, T tests and mean rating comparative analysis. Findings proved the research hypothesis; that vertical gardens with restorative qualities and healing characteristics could add to the quality of urban life, increase restoration effect and lower stress levels of users in public spaces. In addition to the high social acceptance towards vertical gardens.
View
Prevention of the social isolation of older persons: the impact of community environmental satisfaction on social isolation
Article
Full-text available
  • Jun 2023
Background To explore how to prevent the social isolation of the older adult, this study constructed a model of the influence of community environmental satisfaction on the social isolation of the older adult from the three dimensions of environmental facilities, transportation, and supporting facilities around the community. Methods: The social network scale and environmental satisfaction scale were used to collect the sample data of nine communities in Xi’an, and the maximum likelihood estimation method was used to analyze the data and test the model. Results (1) Environmental facilities, transportation, and community surrounding facilities promoted community environment satisfaction ( R ² = 0.904). Among them, environmental facilities ( β = 0.869) had the greatest impact on community environmental satisfaction, followed by transportation ( β = 0.118), and surrounding facilities ( β = 0.084) had the least impact on community environmental satisfaction. (2) Environmental satisfaction had a direct positive impact on social isolation. Among them, the impact of environmental satisfaction on friend isolation ( R ² = 0.895, β = 0.829) was greater than that on family isolation ( R ² = 0.718, β = 0.747). Conclusion Environmental satisfaction can directly affect the social isolation of the older adult in the community and can be used as an intermediate variable of environmental facilities, transportation, and surrounding facilities in the community so that it can indirectly affect the social isolation of the older adult. The results of this study provide a scientific basis for the design of aging environments in the future.
View
View
View
View
View
View
Are Three-point Scales Always Good Enough?
Article
  • Nov 1972
Recent work by Jacoby and Mattell [6] has suggested that three-point Likert scales are sufficient to meet criteria of test-retest reliability, concurrent validity, and predictive validity. Green and Rao [3], using the criterion of data configuration recovery, concluded that sixor seven-point scales are preferable, and the authors are "skeptical about the ability of large numbers of such scales (three- or two-point scales) to 'make up' for the limited information provided by each scale separately." In a reply to Green and Rao, Benson [1] argued that the frequent applicability and practical convenience of twoor three-point scales are strong points in their favor. Moreover, the focus of marketing research on population averages, rather than individuals, suggests that scales with few categories are adequate. This article delineates the conditions under which a two- or three-point scale may be good enough. BACKGROUND
View
View
The Optimal Number of Response Alternatives for a Scale: A Review
Article
  • Nov 1980
A conceptual framework employing the distinction between stimulus-centered and subject-centered scales is presented as a basis for reviewing 80 years of literature on the optimal number of response alternatives for a scale. Concepts and research from information theory and the absolute judgment paradigm of psychophysics are used. The author reviews the major factors influencing the quality of scaled information, points out areas in particular need of additional research, and makes some recommendations for the applied researcher.
View
Response to Rating Scales: A Theoretical Model and Its Application to the Number of Categories Problem
Article
  • Nov 1984
A stochastic model of response to rating scales, using the beta distribution, is developed and tested against empirical data. The effect of number of items by number of points on the sampling error of a composite scale value estimator is investigated and quantitatively formulated. An application of this formulation to arrive at optimal combinations of number of points and number of items is described.
View
Rating Scales and Information Recovery. How Many Scales and Response Categories to Use?
Article
Managers and researchers concerned with marketing and attitudinal research frequently encounter the problem of determining the number of rating scales to use and the number of response categories to provide for each scale that is used. This article approaches the problem through a numerical simulation designed to measure the sensitivity of "solution recovery" to changes in these variables.
View