Page 1 of 19
Advances in Social Sciences Research Journal – Vol. 9, No. 11
Publication Date: November 25, 2022
DOI:10.14738/assrj.911.13392. Thackerson, E. K., Jones, I. T., Baker, M. E., & Grice, J. W. (2022). An Investigation of the Emotional Balance Inventory. Advances in
Social Sciences Research Journal, 9(11). 132-150.
Services for Science and Education – United Kingdom
An Investigation of the Emotional Balance Inventory
Emery K. Thackerson
Department of Psychology, Oklahoma State University
Ian T. Jones
Department of Psychology, Oklahoma State University
Megan E. Baker
Department of Psychology, Oklahoma State University
James W. Grice
Department of Psychology, Oklahoma State University
ABSTRACT
Using experience sampling methodology, this study investigated the scaling format
and validity of an alternative affect measure, the Emotional Balance Inventory
(EBI). Signaling occurred twice per day for a week, directing undergraduate
students (n = 79) to report the frequencies of their emotions using either a
dichotomized or Likert scaled format of the EBI. At the conclusion of the daily
ratings, participants completed two measures of psychological well-being. Results
revealed little evidence in the distinction of the response formats, and there was
marginal evidence supporting the monotonic relationship between the affect
measure and psychological well-being measures. External validity and
generalizability are discussed, along with implications for future research utilizing
clinical populations.
Keywords: affect balance, emotion, scale validity, psychological well-being
INTRODUCTION
Numerous measures exist that allow researchers and clinicians to evaluate the frequency,
intensity, and valence of one’s affective experiences. Nevertheless, researchers who wish to
examine the construct of affect balance, which is the balance between positive and negative
affect, may find themselves at a disadvantage. Many widespread affect and mood assessments,
such as the positive and negative affect schedule (PANAS; [60]) and the differential emotions
scale (DES; [27]), were not constructed to measure affect balance. Moreover, other issues
regarding conventional affect measures pose problems for affect balance research. These
problems include moderately valanced inventory items, items that depict arousal rather than
hedonic states, and inventories with an unequal distribution of positive and negative items [15,
56]. Consequently, these inventories may exclude extreme affective states seen in those with
mood disorders [8], and they may lead to mathematical complications in calculating balance
scores. Developed by Schwartz [48], the Emotional Balance Inventory (EBI) is a unique measure
of affect balance that resolves these problems and possesses several methodological and
statistical advantages over standard affect and affect balance measures.
Page 2 of 19
133
Thackerson, E. K., Jones, I. T., Baker, M. E., & Grice, J. W. (2022). An Investigation of the Emotional Balance Inventory. Advances in Social Sciences
Research Journal, 9(11). 132-150.
URL: http://dx.doi.org/10.14738/assrj.911.13392
The EBI yields a quantitatively distinct affect balance ratio. According to the States of Mind
(SOM) model from which the EBI is derived, the ratio of positive to positive plus negative affects
or cognitions (P/[P+N]) defines an individual’s state of mind [44, 45]. Under the initial SOM
model, a functionally optimal state of mind associated with psychological adaptation arose
when the balance ratio equaled .618. Using Boolean algebra from Lefebvre’s [31] theory of self- reflection, Schwartz [47, 48] amended the original SOM model by redefining the optimal
balanced SOM ratio as .813, and he associated it with positive mood and well-being. The
reformulated model also outlined new parameters and setpoints for SOM categories, which
delineate distinct degrees of psychological functioning based upon the affect balance ratio.
Table 1 reports these revised SOM categories. Since its formulation, the EBI has been
successfully used in clinical intervention studies whereby patients’ affect balance ratios were
assessed pre- and post-treatment, with reports of patient’s affect balance moving from the
extreme negative (.34) to normal (0.72) values after treatment [22, 48, 56]. Moreover,
preliminary evidence suggests that the affect balance ratios correlate with other measures of
well-being such as Friedman’s well-being (r =.73) and Scheier and Carver’s life orientation
optimism scale (r = .47 [21]).
The structure and content of the EBI provide several methodological and mathematical
advantages over standard affect and affect balance measures. First, the EBI is theoretically
grounded in the SOM model which offers a precise operationalization of the measure's
construct compared to other affect balance measures. For example, the affective balance
construct in Bradburn's Affect Balance Scale [6] was based on empirical evidence from
Bradburn and Capolvit's [7] findings on life satisfaction, not general affect. Second, the EBI's
ability to differentiate among varying categories of psychological functioning and
psychopathology yields more qualitative information about individuals which has proven
beneficial in clinical assessment and therapeutic interventions [48, 56]. Third, the EBI's
utilization of the P/[P+N] ratio in measuring affect balance offers several advantages in terms
of computational analyses over other affect balance constructs that rely upon either a
difference score (positive affect minus negative affect) or a simple P:N ratio [17, 20]. These
advantages include eliminating the need for data transformations and imputation methods for
undefined values, simplifying statistical procedures, and the affordance for cross-sectional
comparison of results across studies [3, 4]. For example, the affect balance ratio (P/[P+N])
avoids positively skewed distributions and undefined values often found with P:N ratios and
difference scores as the balance ratio contains an upper bound of 1, a lower bound of zero, and
a non-zero denominator [4].
Although the EBI has been successfully implemented and has many benefits, due to the limited
number of studies that have utilized the EBI, more evidence is needed to establish its criterion
and external validity. In addition, the item format of the EBI needs to be examined as many
researchers and psychometricians contend that Likert-type scales are superior in terms of
validity and reliability compared to dichotomous scales [42, 59]. Contrary evidence suggests,
however, marginal or no differences in validity and reliability outcomes between dichotomous
and polychotomous scales [2, 9, 19, 34, 35, 43, 55]. When studies compare these two scaling
formats often, they fail to find any differences in the validity estimates between the two forms,
particularly for criterion and convergent validity estimates [13, 26, 32]. Moreover, the
assumption that a scale’s reliability increases as the number of response options increases has
not always been proven [28, 34, 36]. Factor analysis studies comparing polychotomous and
Page 3 of 19
134
Advances in Social Sciences Research Journal (ASSRJ) Vol. 9, Issue 11, November-2022
Services for Science and Education – United Kingdom
dichotomous scales have also reported adequate or equivalent fit indices for dichotomous
scales [9, 32].
The notion that dichotomous scales are a poor choice for scale development is not as
straightforward as some researchers may claim because dichotomous scales offer several
benefits over Likert-type scales. These benefits include shorter completion time, precision, and
ease of scoring. First, dichotomous inventories allow respondents to answer more items in a
short period, yielding more information compared to Likert-type scales [11]. Thus,
dichotomous scales would likely prove beneficial to those respondents in repeated measure
design studies, as these scales would reduce respondent fatigue. Second, since dichotomous
scales are a forced-choice format, these scales extract exact answers, and there is little
misunderstanding regarding the question posed. Though the level of precision in dichotomous
scales has been a noted critique made by those in favor of Likert-type scales [23, 38], Matell and
Jacoby [34] reported that the number of response options (2 to 19) did not impact the quality
of the data. Moreover, this level of precision has proven useful for emotion researchers
interested in the frequency of affective states [57]. Third, dichotomous scores have been
historically implemented in personality assessments for their ease of scoring, especially for
instruments with clinical applications [12, 52]. Considering the evidence that indicates
marginal differences between polychotomous and dichotomous scales, the benefits that
dichotomous scales pose, and the design of the EBI to deliver the affect balance ratio (P/[P+N]),
it stands to reason the EBI measure may benefit from incorporating a dichotomous item
format.
RESEARCH AIMS AND HYPOTHESES
The primary aim of the current study was to determine if dichotomizing the EBI items would
lead to better psychometric outcomes over the Likert-type format of the inventory. A secondary
aim of this study was to provide more evidence concerning the EBI's criterion validity. To
investigate the validity of the EBI, we correlated the EBI's affect balance ratios with previously
validated measures of well-being, particularly life satisfaction [16] and psychological
flourishing [17]. Since the EBI measures affect balance, we hypothesize that the EBI's affect
balance scores should exhibit a monotonic association with the two psychological well-being
constructs, as previous research has indicated a positive correlation between affect balance and
life satisfaction [40, 64] and psychological flourishing [17, 54]. We moreover employed an
experience sampling design (ESM) to allow for a momentary assessment of affect balance
scores for approximately one week [30]. Daily assessment of affect balance scores allowed for
assessment of the scale's variability and dispersion of the EBI ratios. Experience sampling
designs have several noted advantages over one-time assessments, such as reducing recall bias
and improved ecological validity, and are often utilized to assess affect and mood [1, 50, 58].
METHOD
Sample
Recruitment was conducted through the university’s SONA administration system, where in
exchange for their participation, participants received prorated course credit which was
contingent upon the number of responses given. Participants included 79 undergraduate
students from a midwestern public university (14 men, 65 women, Mage = 21.33, SDage = 5.38).
Two-thirds of the sample identified their ethnicity as Caucasian (67.1%), with the remaining
Page 4 of 19
135
Thackerson, E. K., Jones, I. T., Baker, M. E., & Grice, J. W. (2022). An Investigation of the Emotional Balance Inventory. Advances in Social Sciences
Research Journal, 9(11). 132-150.
URL: http://dx.doi.org/10.14738/assrj.911.13392
participants identifying as African American (6.3%), Hispanic/Latino (6.3%), Asian American
(5.1%), Native American (12.7%), Native Hawaiian (1.3%), and other (1.3%).
Measures
Emotional Balance Inventory (EBI)
Participants completed the 42-item emotional balance inventory (EBI). The EBI assesses
emotional balance measured by dimensions of positive and negative affect. The positive affect
dimension was comprised of 21 items and three subscales of happiness, friendliness, and
vitality. The three subscales were summed to yield the positive (P) score. The negative affect
dimension was comprised of 21 items and three subscales of anger, sadness, and fear. The three
subscales were summed to yield the negative (N) emotion score. Participants' emotional
balance was calculated by dividing the positive emotion score by the sum of the positive and
negative totals; viz., P / (P+N). The emotional balance ratios were categorized by the SOM levels
of psychological functioning [46, 47]: Extreme Negative (.000 - .437), Conflicted (.438 - .563),
Subnormal/Coping (.564 - .672), Normal (.673 - .766), Optimal (.767 - .844), Deep Optimal (.845
- .900), and Extreme Positive (.901 – 1.00).
Participants were administered either the polychotomous or dichotomous version of the EBI
(see Appendix). To report the frequency of an emotion, participants given the polychotomous
version responded using a five-point Likert scale ranging from 0 (never) to 4 (almost always).
Example polychotomous EBI items included asking participants, “how frequently have you
experienced feeling proud within the past three to four hours?” and “how frequently have you
experienced feeling hopeful within the past three to four hours?” The dichotomous EBI utilized a
forced-choice format, and participants responded to the frequency of the EBI items using a yes
or no format, which indicated either the presence or absence of the emotion. An example
dichotomized EBI question asked participants, “have you experienced feeling exuberant within
the past three to four hours?”
Depending upon the participants’ response, ten situational contexts were presented as a list to
participants, and they selected the context in which the emotion occurred. Potential situational
contexts included education, work, leisure and physical activities, interpersonal and romantic
relationships, basic needs, spiritual or religious activities, social or political involvement, and
loneliness. Those in the dichotomous group reporting yes to the exhibited emotion were given
the list of contexts. If participants reported feeling the emotion frequently (3) or almost always
(4), representing higher ends of the polychotomous scale, then the list of situational contexts
followed the preceding EBI question. The situational contexts were not analyzed in the results
below.
Psychological Well-Being Measures
Participants completed the five-item Satisfaction with Life Scale (SWLS; [16]) that assessed the
individuals' cognitive judgments of life satisfaction. Based upon a seven-point Likert scale (from
1, strongly disagree, to 7, strongly agree), scores on the SWLS can range from 5 to 35. Total
scores were obtained by summing all items from the measure, with lower scores indicating
dissatisfaction with one's life and higher scores indicating satisfaction. Previous research by
Pavot & Diener [41] demonstrated high internal reliability of SWLS, with reports of Cronbach's
alpha ranging from 0.79 to 0.89. The Flourishing Scale (FS; [17]) measured individuals'
propensity to engage in social-psychological functioning, such as having positive relationships.
Page 5 of 19
136
Advances in Social Sciences Research Journal (ASSRJ) Vol. 9, Issue 11, November-2022
Services for Science and Education – United Kingdom
Comprised of eight items, the Flourishing Scale is a seven-point Likert scale anchored by 1
(strongly disagree) and 7 (strongly agree), with scores ranging from 8 to 56. Lower levels of
social-psychological functioning were represented by low scores obtained on the FS, and higher
scores represented higher levels of social-psychological functioning. Diener and colleagues [17]
reported high reliability, with a reported Cronbach's alpha of .87 and temporal reliability of .71.
Procedure
Participants enrolled in the SONA system and were directed to the Qualtrics online server
(http://www.qualtrics.com). After signing an electronic consent form, participants were given
a demographics questionnaire. Next, participants provided basic contact information, such as
their email address and phone number; then, participants created a unique code to ensure the
anonymity of their data collected during the week. The phone numbers collected were
converted to email addresses, allowing text message reminders to be sent to the participants'
phones. Participants were randomly assigned to either take the dichotomous or polychotomous
version of the EBI. Using the signal-contingent approach [62], twice per day, over seven days,
participants were emailed and texted at random times between the waking hours of 8:00 a.m.
8:00 p.m., prompting them with a message containing a link to take their respective EBI
measures whenever it was convenient for them. Previous research restricted response
collection 20 to 30 minutes after signaling occurred [10, 14]; however, this response range was
arbitrarily set. Responses were included in the data analyses if the response occurred on the
same day as the signal was sent and the response from the second signal was at least three or
four hours apart from the first response. If the second response did not have a timestamp of at
least three or four hours apart from the first recorded response, then the response was treated
as a missing case. Moreno et al. [37] pilot findings suggested that a time frame of three to four
hours between each signal is ideal time window for college students. Each participant received
a total of fourteen signals. The fourteenth signal prompted participants to report their
emotional balance and psychological well-being, whereas signals 1 through 13 prompted
participants to report only their emotional balance. Of the participants in the sample, two- thirds (n = 54) completed the psychological well-being measures, and the participants who
failed to complete the measures appropriately were treated as missing cases in the analyses
below.
RESULTS
Distributional Comparisons
Frequencies of missing observations are reported in Table 2. The frequency of missing
observations was calculated by dividing the number of missing observations by the total
number of observations. The frequency of missing data was estimated to be 34%, which is
similar to missing response data reported in other studies using ESM procedures [51]. The
impact of using the dichotomous or polychotomous response format was assessed by first
comparing the distributions of EBI ratios from the two scaling methods. As can be seen in Table
3, descriptive statistics revealed highly similar mean and median values for the ratios for the
two scaling methods. The standard deviations for the ratios, however, were typically higher for
the dichotomous scoring method compared to the polychotomous method across the fourteen
time points. Examination of the ratios, grouped into the seven categories of functioning and
tallied across all participants and all fourteen time points, revealed the nature of this difference
in variability; specifically, the dichotomous scaling format yield greater proportions of
“extreme negative” and “extreme positive” ratios compared to the polychotomous format. As
Page 6 of 19
137
Thackerson, E. K., Jones, I. T., Baker, M. E., & Grice, J. W. (2022). An Investigation of the Emotional Balance Inventory. Advances in Social Sciences
Research Journal, 9(11). 132-150.
URL: http://dx.doi.org/10.14738/assrj.911.13392
can be seen in Figure 1, the dichotomous format yielded approximately 19% more EBI ratios in
the two extreme categories than the polychotomous format. A randomization test moreover
indicated that the difference between the two distributions in the figure was not plausibly the
result of physical chance (c-value < .001; see [24]).
Given these between-group differences in variability, the within-person standard deviations for
the fourteen daily EBI ratios were also computed and examined. Figure 2 shows two cases in
which the variability of ratios across the fourteen time points was dramatically different. As can
be seen, Case #3, who completed the polychotomous version of the EBI, showed very little
variability in their emotion balance ratios across time (SD = .08). By comparison, Case #32
completed the dichotomous version of the EBI and varied greatly in their emotional balance
scores across the fourteen days (SD = .28). Across all participants the standard deviations of the
ratios were generally higher for those who used the dichotomous scale (MSD = .26, MdnSD = .25)
compared to those who used the polychotomous scale (MSD = .14, MdnSD = .15). As a simple non- parametric test of these individual differences, the standard deviation for each person in the
dichotomous group (n = 36) was compared to the standard deviation of each person in the
polychotomous group (n = 39) with the expectation that the former value would be greater than
the latter value (viz., dichotomous standard deviation > polychotomous standard deviation).
Examination of all possible pairs of individuals from the two groups revealed that 1064 of the
1404 comparisons (75.78%) matched expectation; in other words, most of the participants who
completed the EBI using the dichotomous scaling format revealed greater variability in their
emotional balance ratios across the fourteen days when compared to individuals who used the
polychotomous format. A randomization test indicated this high percent difference was not
plausibly the result of physical chance (c-value < .001).
Psychological Well-Being
For all participants scores for the Satisfaction with Life Scale (SWLS; M = 24.07, Mdn = 25.00, s
= 7.59) and Flourishing Scale (FS; M = 45.29, Mdn = 46.00, s = 7.48) were skewed toward the
low ends of their respective possible scale ranges. Scores from these two measures of
psychological well-being were expected to be monotonically and positively associated with the
EBI ratios. Avoiding assumptions of continuous quantitative measurement and distributional
normality, Spearman rank-order correlations between the well-being scores and balance ratios
were computed for each of the fourteen time points and reported in Table 4. As can be seen, the
life satisfaction and flourishing scales were generally more highly correlated with balance
ratios for those participants who completed the polychotomous version of the EBI (MdnSWLS r =
.419; MdnFS r = .545) compared to those who completed the dichotomous version (MdnSWLS r =
.274; MdnFS r = .106). The fourteen balance ratios were then aggregated into medians within
each participant and correlated with the measures of life satisfaction and psychological
flourishing. Figure 3 shows the four resulting scatterplots.
With regard to the polychotomous version of the EBI, the Spearman correlations were
moderate in magnitude for the SWLS (r = .45) and FS (r = .70) variables. By comparison, the
correlations for the dichotomous version of the EBI were notably lower (r = .33 for life
satisfaction; r = .20 for flourishing). Only the scatterplot comprised of the EBI ratios and
flourishing scale scores demonstrated a compelling monotonic relationship. Lastly, given the
differences in variability in balance ratios noted above between the dichotomous and
polychotomous groups, the individual standard deviations computed across the fourteen time
Page 7 of 19
138
Advances in Social Sciences Research Journal (ASSRJ) Vol. 9, Issue 11, November-2022
Services for Science and Education – United Kingdom
points were correlated with the life satisfaction and flourishing scale scores. The Spearman
correlation values for these four analyses were low, not exceeding .16 in absolute value.
DISCUSSION
The current study sought to investigate if dichotomizing the EBI scaling format would produce
better psychometric outcomes over the polychotomous EBI format. We also aimed to provide
additional evidence concerning the EBI’s criterion validity. Polychotomous scales, such as
Likert scales, have become a popular psychometric scoring paradigm used in research [5], as
previous evidence suggests that, on average, Likert scales have better reliability and validity
estimates compared to dichotomous scales [42, 53]. However, other studies have failed to
confirm this finding, showing little to no evidence of the distinction between these two scaling
formats [2, 9, 19, 34, 36]. Generally, our findings suggest somewhat marginal evidence to
support any substantial distributional differences between the scaling formats since both scales
had highly similar mean and median values. Nonetheless, the polychotomous scale had higher
correlation coefficients with the life satisfaction and flourishing scales, suggesting that the
polychotomous EBI led to improved psychometric validity compared to the dichotomous
format.
Overall, the distributional pattern of the data between the dichotomous and polychotomous
scaling formats indicated that the EBI ratios for the dichotomous group clustered at the tail
ends of the distribution. In contrast, the EBI ratios for the polychotomous group fell towards
the middle of the distribution. The binary nature of the dichotomous scale illustrates the
extremity of these EBI ratios found within the dichotomous group, as Weijters et al. [61] found
that increasing the number of item response options led to a decrease in an extreme response
style. These slight differences that appeared in the distributional pattern were also
demonstrated by the differences in variability expressed by the standard deviations.
Researchers have suggested that decreasing the number of response options available will
reduce the variability of the scale and increasing the number of response options leads to an
increase in the reliability of a scale [33, 53]. Our results from a simple non-parametric test
support the notion that reducing the number of response options impacts the variability of the
scale, as indicated by the higher standard deviations found within the dichotomous group.
However, the fluctuations observed among the standard deviations for both scaling formats
may likely be attributed to the study’s design. Due to the inherent nature of experience
sampling methodology, which assesses the subjective nature of one’s experiences, this allows
researchers to measure the variability of an experience, as researchers have used respondents’
standard deviation as a measure of intraindividual affect variability [18, 39]. Thus, we cannot
conclude that the polychotomous scale led to greater reliability than the dichotomous scale.
Regarding the EBI’s criterion validity, the results from Spearman’s correlation coefficients
reveal a weak monotonic relationship between the EBI ratios and life satisfaction and the EBI
ratios and psychological flourishing. Albeit the small monotonic association between the EBI
ratios and life satisfaction, our findings corroborate previous evidence suggesting a positive
association between SOM ratios and life satisfaction [63]. The weak monotonic association
between the EBI ratios and the psychological well-being measures is plausibly due to the
weaker correlation coefficients observed among those in the dichotomous group. Thus, the
polychotomous EBI’s correlation coefficients were greater in magnitude than the dichotomous
Page 9 of 19
140
Advances in Social Sciences Research Journal (ASSRJ) Vol. 9, Issue 11, November-2022
Services for Science and Education – United Kingdom
References
1. aan het Rot, Hogenelst, K., & Schoevers, R. A. (2012). Mood disorders in everyday life: A systematic review of
experience sampling and ecological momentary assessment studies. Clinical Psychology Review, 32(6), 510–523.
https://doi.org/10.1016/j.cpr.2012.05.007
2. Alwin, D. F. (2007). Margins of error: A study of reliability in survey measurement. John Wiley & Sons.
3. Amsel, & Fichten, C. S. (1998). Recommendations for self-statement inventories: Use of valence, end points,
frequency, and relative frequency. Cognitive Therapy and Research, 22(3), 255–277.
https://doi.org/10.1023/A:1018744826190
4. Amsel, & Fichten, C. S. (1990). Ratio versus frequency scores: Focus of attention and the balance between
positive and negative thoughts. Cognitive Therapy and Research, 14(3), 257–277.
https://doi.org/10.1007/BF01183996
5. Bishop, & Herron, R. L. (2015). Use and Misuse of the Likert Item Responses and Other Ordinal Measures.
International Journal of Exercise Science, 8(3), 297–302.
6. Bradburn, N. M. (1969). The Structure of Psychological Well-Being. Oxford: Aldine.
7. Bradburn, N. M., & Caplovitz, D. (1965). Reports on Happiness. A Pilot Study of Behavior Related to Mental
Health. Chicago, IL: Aldine Publishing Company.
8. Bylsma, Morris, B. H., & Rottenberg, J. (2008). A meta-analysis of emotional reactivity in major depressive
disorder. Clinical Psychology Review, 28(4), 676–691. https://doi.org/10.1016/j.cpr.2007.10.001
9. Capik, & Gozum, S. (2015). Psychometric Features of an Assessment Instrument with Likert and Dichotomous
Response Formats. Public Health Nursing (Boston, Mass.), 32(1), 81–86. https://doi.org/10.1111/phn.12156
10. Cerin, Szabo, A., & Williams, C. (2001). Is the Experience Sampling Method (ESM) appropriate for studying
pre-competitive emotions? Psychology of Sport and Exercise, 2(1), 27–45. https://doi.org/10.1016/S1469-
0292(00)00009-1
11. Clark, & Watson, D. (1995). Constructing Validity: Basic Issues in Objective Scale Development. Psychological
Assessment, 7(3), 309–319. https://doi.org/10.1037/1040-3590.7.3.309
12. Comrey (1988). Factor-Analytic Methods of Scale Development in Personality and Clinical Psychology.
Journal of Consulting and Clinical Psychology, 56(5), 754–761. https://doi.org/10.1037/0022-006X.56.5.754
13. Cox, Courrégé, S. C., Feder, A. H., & Weed, N. C. (2017). Effects of augmenting response options of the MMPI-2-
RF: An extension of previous findings. Cogent Psychology, 4(1), 1323988–.
https://doi.org/10.1080/23311908.2017.1323988
14. Csikszentmihalyi, M., & Larson, R. (1987). Validity and reliability of the experience-sampling method The
Journal of Nervous and Mental Disease, 175
15. Diener, E., & Emmons, R. A. (1984). The independence of positive and negative affect. Journal of personality
and social psychology, 47(5).
16. Diener, Emmons, R. A., Larsen, R. J., & Griffin, S. (1985). The Satisfaction With Life Scale. Journal of Personality
Assessment, 49(1), 71–75. https://doi.org/10.1207/s15327752jpa4901_13
17. Diener, Wirtz, D., Tov, W., Kim-Prieto, C., Choi, D., Oishi, S., & Biswas-Diener, R. (2010). New Well-being
Measures: Short Scales to Assess Flourishing and Positive and Negative Feelings. Social Indicators Research,
97(2), 143–156. https://doi.org/10.1007/s11205-009-9493-y
18. Eid, & Diener, E. (1999). Intraindividual Variability in Affect: Reliability, Validity, and Personality Correlates.
Journal of Personality and Social Psychology, 76(4), 662–676. https://doi.org/10.1037/0022-3514.76.4.662
19. Finn, J.A., Ben-Porath, Y.S., & Tellegen, A. (2015). Dichotomous versus polytomous response options in
psychopathology assessment: Method or meaningful variance? Psychological Assessment, 27(1).
https://doi.org/10.1037/pas0000044
20. Fredrickson, & Losada, M. F. (2005). Positive Affect and the Complex Dynamics of Human Flourishing. The
American Psychologist, 60(7), 678–686. https://doi.org/10.1037/0003-066X.60.7.678
Page 10 of 19
141
Thackerson, E. K., Jones, I. T., Baker, M. E., & Grice, J. W. (2022). An Investigation of the Emotional Balance Inventory. Advances in Social Sciences
Research Journal, 9(11). 132-150.
URL: http://dx.doi.org/10.14738/assrj.911.13392
21. Friedman, P. H. (1997). Correlational analyses of Emotional Balance Inventory, Self-Image Balance Inventory,
and Optimism Balance Inventory. Unpublished raw data.
22. Friedman. (1999). The balanced states of mind model: Optimal cognitive balance and the consequences of
excess positivity. ProQuest Dissertations Publishing.
23. Greenwald, H. J., & O'Connell, S. M. (1970). Comparison of dichotomous and Likert formats. Psychological
Reports, 27(2), 481-482.
24. Grice, J. W. (2021). Drawing inferences from randomization tests. Personality and Individual Differences, 179,
https://doi.org/10.1016/j.paid.2021.110931.
25. Henrich, Heine, S. J., & Norenzayan, A. (2010). The weirdest people in the world? The Behavioral and Brain
Sciences, 33(2-3), 45–67. https://doi.org/10.1017/S0140525X0999152X
26. Hilbert, S., Küchenhoff, H., Sarubin, N., Nakagawa, T. T., & Bühner, M. (2016). The influence of the response
format in a personality questionnaire: An analysis of a dichotomous, a Likert-type, and a visual analogue scale.
TPM-Testing, Psychometrics, Methodology in Applied Psychology.
27. Izard C. E., Dougherty F. E., Bloxom B. M. and Kotsch N. E. (1974) The Differential Emotions Scale: A method
of measuring the subjective experience of discrete emotions. Vanderbilt Univ. Press, Nashville, Tenn.
28. Jacoby, & Matell, M. S. (1971). Three-Point Likert Scales Are Good Enough. Journal of Marketing Research,
8(4), 495–500. https://doi.org/10.1177/002224377100800414
29. Jones, I., O’Lansen, C., Baker, M., Thackerson, E., Horvath, S., Maupin, K., McDaniel-Johnson, H., Grice, J.(2020).
The effects and implications of mood on moral judgements. Advances in Social Sciences Research Journal, 7(10), 1-
12. https://doi.org/10.14738/assrj.710.9070
30. Larson, & Csikszentmihalyi, M. (2014). The Experience Sampling Method. In Flow and the Foundations of
Positive Psychology (pp. 21–34). Springer Netherlands. https://doi.org/10.1007/978-94-017-9088-8_2
31. Lefebvre, V. A. (1990). The fundamental structures of human reflection. In H. Wheeler (Ed.), The structure of
human reflexion (pp. 5–69). New York: Peter Lang.
32. Lemmens, Valkenburg, P. M., & Gentile, D. A. (2015). The Internet Gaming Disorder Scale. Psychological
Assessment, 27(2), 567–582. https://doi.org/10.1037/pas0000062
33. Lozano, García-Cueto, E., & Muñiz, J. (2008). Effect of the Number of Response Categories on the Reliability
and Validity of Rating Scales. Methodology, 4(2), 73–79. https://doi.org/10.1027/1614-2241.4.2.73
34. Matell, & Jacoby, J. (1971). Is There an Optimal Number of Alternatives for Likert Scale Items? Study I:
Reliability and Validity. Educational and Psychological Measurement, 31(3), 657–674.
https://doi.org/10.1177/001316447103100307
35. Maurer, & Pierce, H. R. (1998). A Comparison of Likert Scale and Traditional Measures of Self-Efficacy. Journal
of Applied Psychology, 83(2), 324–329. https://doi.org/10.1037/0021-9010.83.2.32
36. Maurer, & Andrews, K. D. (2000). Traditional, Likert, and Simplified Measures of Self-Efficacy. Educational
and Psychological Measurement, 60(6), 965–973. https://doi.org/10.1177/00131640021970899
37. Moreno, Jelenchick, L., Koff, R., Eikoff, J., Diermyer, C., & Christakis, D. A. (2012). Internet use and multitasking
among older adolescents: An experience sampling approach. Computers in Human Behavior, 28(4), 1097–1102.
https://doi.org/10.1016/j.chb.2012.01.016
38. Munshi, J. (2014). A method for constructing Likert scales. Available at SSRN 2419366.
39. Palmier-Claus, Myin-Germeys, I., Barkus, E., Bentley, L., Udachina, A., Delespaul, P. A. E. G., Lewis, S. W., &
Dunn, G. (2011). Experience sampling research in individuals with mental illness: reflections and guidance. Acta
Psychiatrica Scandinavica, 123(1), 12–20. https://doi.org/10.1111/j.1600-0447.2010.01596.x
40. Pavot, Diener, E., Colvin, C. R., & Sandvik, E. (1991). Further Validation of the Satisfaction With Life Scale:
Evidence for the Cross-Method Convergence of Well-Being Measures. Journal of Personality Assessment, 57(1),
149–161. https://doi.org/10.1207/s15327752jpa5701_17
Page 13 of 19
144
Advances in Social Sciences Research Journal (ASSRJ) Vol. 9, Issue 11, November-2022
Services for Science and Education – United Kingdom
TABLES
Table 1: Revised States of Mind Categories and Levels of Psychological Functioning
SOM Category Range
Extreme Negative .000 - .437
Conflicted .438 - .563
Subnormal/Coping .564 - .672
Normal .673 - .766
Optimal .767 - .844
Deep Optimal .845 - .900
Extreme Positive .901 - 1.00
Page 14 of 19
145
Thackerson, E. K., Jones, I. T., Baker, M. E., & Grice, J. W. (2022). An Investigation of the Emotional Balance Inventory. Advances in Social Sciences
Research Journal, 9(11). 132-150.
URL: http://dx.doi.org/10.14738/assrj.911.13392
Table 2: Frequency of Missing Observations
Time Point
Observation
s
Total
Observation
s
No. of
Missing
Observation
s
Frequency
% Missing
Observation
s
EBI
Ratio
s
Total No. of
Observation
s
No. of
Missing
Observation
s
Frequency
% of
Missing
Observation
s
Timepoint 1 69 10 3.79% EBI
Ratio
1
65 14 4.65%
Timepoint 2 71 8 3.03% EBI
Ratio
2
71 8 2.66%
Timepoint 3 63 16 6.06% EBI
Ratio
3
60 19 6.31%
Timepoint 4 71 8 3.03% EBI
Ratio
4
68 11 3.65%
Timepoint 5 66 13 4.92% EBI
Ratio
5
64 15 4.98%
Timepoint 6
Timepoint 7
Timepoint 8
Timepoint 9
Timepoint
10
Timepoint
11
Timepoint
12
Timepoint
13
Timepoint
14
Total
65
54
63
56
57
52
53
49
53
842
14
25
16
23
22
27
26
30
26
264
5.30%
9.47%
6.06%
8.71%
8.33%
10.23%
9.85%
11.36%
9.85%
99.99%
EBI
Ratio
6
EBI
Ratio
7
EBI
Ratio
8
EBI
Ratio
9
EBI
Ratio
10
EBI
Ratio
11
EBI
Ratio
12
EBI
Ratio
13
EBI
Ratio
14
Total
62
52
62
51
56
48
50
46
50
805
17
27
17
28
23
31
29
33
29
301
5.65%
8.97%
5.65%
9.30%
7.64%
10.30%
9.64%
10.96%
9.64%
100.00%
Note. The frequency of the missing data for the sample (n = 79) was estimated to be 34%
Page 16 of 19
147
Thackerson, E. K., Jones, I. T., Baker, M. E., & Grice, J. W. (2022). An Investigation of the Emotional Balance Inventory. Advances in Social Sciences
Research Journal, 9(11). 132-150.
URL: http://dx.doi.org/10.14738/assrj.911.13392
Table 4: Spearman’s Rank Correlation Coefficients Between the EBI Ratios and Well-Being
Measures
EBI Ratios
Dichotomous Polychotomous
SWLS
(ρ)
FS
(ρ)
SWLS
(ρ)
FS
(ρ)
Ratio 1 0.21 0.08 0.46 0.62
Ratio 2 0.52 0.26 0.28 0.49
Ratio 3 0.21 0.07 0.54 0.52
Ratio 4 -0.08 -0.18 0.42 0.54
Ratio 5 0.35 0.24 0.15 0.35
Ratio 6 0.33 0.08 0.37 0.55
Ratio 7 0.34 0.40 0.42 0.57
Ratio 8 0.35 0.35 0.29 0.49
Ratio 9 0.21 0.31 -0.07 0.11
Ratio 10 0.41 0.37 0.55 0.63
Ratio 11 -0.18 0.02 0.51 0.65
Ratio 12 0.15 0.03 0.25 0.45
Ratio 13 0.29 0.13 0.56 0.62
Ratio 14 0.26 0.07 0.51 0.72
FIGURES
Figure 1. Ratio proportions for dichotomous and Likert scale formats
Note. Values represent proportions for the dichotomous or polychotomous EBI ratios classified
into the seven categories of functioning.
Page 17 of 19
148
Advances in Social Sciences Research Journal (ASSRJ) Vol. 9, Issue 11, November-2022
Services for Science and Education – United Kingdom
Figure 2. EBI ratios for two participants with contrasting variability
Note. Two respondents’ EBI ratios across one week are graphically depicted. The red line
represents the polychotomous EBI ratios (Case #3) and the black line represents the
dichotomous EBI ratios (Case #32).
Figure 3. Scatterplots comparing median EBI ratios with life satisfaction (SWLS) and
psychological flourishing (FS) for dichotomous and polychotomous rating scales
Page 18 of 19
149
Thackerson, E. K., Jones, I. T., Baker, M. E., & Grice, J. W. (2022). An Investigation of the Emotional Balance Inventory. Advances in Social Sciences
Research Journal, 9(11). 132-150.
URL: http://dx.doi.org/10.14738/assrj.911.13392
APPENDIX
Polychotomous Emotional Balance Inventory
Directions: Please consider how frequently you have experienced each of the listed feelings to
follow. For each feeling, indicate the frequency of the feeling using the following options: never,
rarely, sometimes, frequently, or almost always. Depending upon the answers you provide, you
may be asked additional questions.
Item Stem –
How
frequently
have you
experience
d feeling
______
within the
past three
to four
hours?
Emotions presented in the item stem, _______
Response
choices:
o Never o
Rarely o
Sometimes
o
Frequently
o Almost
Always
Proud Anxious Interested Grateful Ashamed Loving Irritable
Hopefu
l
Aroused Devoted Nervous Hostile Happy Enraged
Afraid Dread Frustrate
d
Romantic Exuberan
t
Friendly Regretful
Panicky Lively Annoyed Tense Inspired Engage
d
Amused
Guilty Enthusiasti
c
Terrified Despair Joyful Hateful Affectionat
e
Sad Passionate Ecstatic Depresse
d
Vigorous Angry Tormented
Note. The presented polychotomous emotional balance inventory above were adapted to meet
the experience sampling methodology design.
Page 19 of 19
150
Advances in Social Sciences Research Journal (ASSRJ) Vol. 9, Issue 11, November-2022
Services for Science and Education – United Kingdom
Dichotomous Emotional Balance Inventory
Directions: Please consider whether or not you have experienced each of the listed feelings to
follow. For each feeling, click either yes or no, indicating that you either experienced the feeling
or not. Depending upon the answers you provide, you may be asked additional questions.
Item Stem –
Have you
experience
d feeling
______
within the
past three
to four
hours?
Feelings/emotions presented in the item stem, _______
Response
choices:
o Yes o No
Proud Anxious Interested Grateful Ashamed Loving Irritable
Hopefu
l
Aroused Devoted Nervous Hostile Happy Enraged
Afraid Dread Frustrate
d
Romantic Exuberan
t
Friendly Regretful
Panicky Lively Annoyed Tense Inspired Engage
d
Amused
Guilty Enthusiasti
c
Terrified Despair Joyful Hateful Affectionat
e
Sad Passionate Ecstatic Depresse
d
Vigorous Angry Tormented