ADVERB INTENSIFIERS FOR USE IN RATINGS OF ACCEPTABILITY, ADEQUACY, AND
Stephen J. Krsacok and William F. Moroney
Human Factors Program
University of Dayton. Dayton, Ohio
First time users of this material are encouraged to review the description
of the process used to develop the adverb intensifier tables.
Practitioners are encouraged to use the material provided in the tables
or to develop their own. Publication of these tables is not authorized. Those
who are familiar with the table generation process can go to Adverb Intensifiers.pdf.
Survey designers often assume the existence of an underlying linear continuum with equal intervals between anchors when they create a scale. However, this is not necessarily the case when labels, such as “somewhat acceptable,” “completely acceptable” or “absolutely acceptable” are assigned to the intervals on those scales. This study examines numeric ratings assigned by college students to these and similar adverb intensifiers.
This study examined ratings provided by college males, college females, and Army personnel for positive and negative adverb-intensifiers of acceptability (n=50), adequacy (n= 50), and relative goodness (n=41). Subjects in the current study (54 college males & 54 college females) assigned numeric values (from –5 to + 5) to three separate lists of adverb-intensifiers. Ratings provided by college males were compared separately with ratings provided by college females, and those provided by Army personnel (participants in a study by Matthews, Wright, Yudowitch, Geddie, and Palmer, 1978).
The mean ratings, standard deviations, and order of the ratings assigned by college males and college females were essentially the same. Essentially, there were minimal differences due to gender among college students. However, when the ratings of college males and Army personnel were compared, there were considerable significant differences between mean ratings (Krsacok, & Moroney, 2002) . Differences were also noted in the variability of the adverbs used to describe acceptability, adequacy and relative goodness, while the order of the ratings was essentially the same for both groups. These differences were attributed to the both the 22-year interval between these studies and to possible demographic differences between the samples.
Since there were minimal differences in the numeric values assigned by college males and college females to the adverb intensifiers, the data from the college male and college female groups were combined and means and standard deviations were re-calculated
(see Tables 64-66). Three different scale development strategies (Balanced Terms, Two Standard Deviation Separation, and Balanced Numeric Values ) were utilized to develop a total of 63 adverb intensifier scales (Tables 1- 63) with intervals of 5, 6, 7, 8, 9, 10 and 11 descriptors for acceptability, adequacy, and relative goodness. The strategies used in developing these scales follows below.
method compared all descriptors, which used equivalent positive and negative
terms, or parallel wording, in the survey, e.g., “largely acceptable”
and “largely unacceptable.” The descriptors that anchor the ends of the
scale were selected first. To qualify for selection, the absolute value of the
means of the positive and negative descriptors needed to be nearly equal, and
the standard deviations small. The next pair of descriptors (positive and
negative balanced terms) was selected approximately one standard deviation from
the first pair and followed the same evaluation criteria. This process was
continued until five positive/negative descriptor pairs were selected. The term,
“neutral”, was initially selected for the scale midpoint because its
mean was close to zero (mean = -0.01) and it had a small standard deviation
(SD=0.10). However, it was replaced
by the descriptor “neither ___ nor ___”. Dillman (2000) and the authors
believe that this descriptor defines the midpoint more clearly than terms such
as neutral, marginal, and borderline. The
table below lists the eleven descriptors selected using the Balanced Terms
acceptability scale developed using the Balanced Terms Method
Using this method, the most extreme positive adverb-adjective combination was selected based on the mean value and a small standard deviation. A small standard deviation indicates a higher level of respondent agreement on the meaning of the term. Once the most extreme positive term was selected, the next adverb-adjective combination was selected using the following criterion:
mean – standard deviation > smaller mean + standard deviation
This criterion ensures that the distributions of any two adjacent
descriptors minimally overlap on the rating scale continuum. The same method was
followed, starting with the most extreme negative term and working back to the
midpoint. An eleven-point acceptability scale is provided in the table below.
acceptability scale developed using the Balanced Numeric Values Method
Survey designers are invited to use the scales provided in
Tables 1-63 or to use the raw data (Tables 64-66) to develop their own scales.
Those who do so will have the advantage of using data based on a current college
population. Click here to access those tables.
D. A. (2000). Mail and
internet surveys: The tailored design method.
New York: J. Wiley.
S.J. & Moroney, W.F. (2002). Quantification of Adverb intensifiers for use
in ratings of acceptability, adequacy, and relative goodness. Proceedings of
the Human Factors and Ergonomics Society 46th Annual Meeting (pp
1944-1948). Santa Monica, CA: Human Factors and Ergonomics Society
Matthews, J.J., Wright, C. E., Yudowitch, K. L., Geddie, J., & Palmer, R. L. (1978). The perceived favorableness of selected scale anchors and response alternatives. U. S. Army Research Institute for the Behavioral and Social Sciences.