Professional Documents
Culture Documents
Relationships
Author Note
We have no known conflicts of interest to disclose. All data and analysis code have been
made publicly available at the “APA Journal Articles: Data and Related Resources” Open
Science Framework repository and can be accessed at https://osf.io/3p69f/. This study’s design
and its analysis were not pre-registered. A pre-print of this manuscript can be accessed at
https://psyarxiv.com/zavkf.
College of Health Solutions, Arizona State University, Phoenix, AZ 85004, United States. Email:
sacloona@asu.edu
2
THE LOVEBIRD SCALE
Abstract
People often use the term “lovebirds” to describe the ongoing loving behaviors and interactions
between two romantic partners, but what specific relationship processes distinguish flourishing
“lovebird” relationships from other adult committed long-term relationships? The present study
aimed to address this knowledge gap through the development and validation of a new measure
called the Lovebird Scale. To do this, we conducted three studies using data collected from 996
English-speaking, U.S. adults who reported being in a romantic relationship lasting at least 6
months (M = 11.2 years). In Study 1, we generated an item pool and conducted an exploratory
factor analysis to determine the best performing items to retain. In Study 2, confirmatory factor
analyses revealed a three-factor model (Mutuality, Romance, and Disconnect) nested within a
higher order structure and these data provide preliminary evidence for the convergent,
discriminant, and incremental validity properties of the Lovebird Scale. We also explored the
relation between the Lovebird Scale and affective state, finding that lovebird relationships were
associated with greater positive affect. We discuss how the Lovebird Scale relates to previous
research on relationship functioning and how it can be used to further our understanding of
relationship flourishing.
Statement of Relevance
barometer for relationship quality, but this does not necessarily indicate relationship flourishing.
The Lovebird Scale aims to address this knowledge gap by measuring the various loving
behaviors, interactions, and cognitions that occur within exceptionally high-quality relationships.
The scale advances the science of love by operationalizing “optimal” romantic relationships,
going beyond broad ratings of relationship satisfaction to distinguish between “lovebirds” and
Relationships
Romantic relationships are considered one of the most important types of adult
relationships, given their significant influence on both physical and mental health outcomes. For
levels of subjective well-being (Dush & Amato, 2005), fewer depressive symptoms (Holt-
Lunstad et al., 2008), engaging in fewer risky behaviors (e.g., binge drinking, substance use)
(Braithwaite et al., 2010), and lower morbidity and mortality (Robles et al., 2014). However, just
being in a long-term committed relationship does not guarantee these benefits; rather, they are
largely dependent on the quality of the relationship. In their study, Holt-Lunstad et al. (2008)
found that higher marital quality was linked to lower ambulatory blood pressure and stress, as
well as less depression and higher subjective well-being. Critically, single individuals in their
study had lower blood pressure than those in unhappy marriages, a finding that highlights the
importance of relationship quality for achieving the health benefits associated with committed
relationships (Holt-Lunstad et al., 2008). Another study by Roberson et al. (2018) found that
people in stable marriages with high conflict reported lower life satisfaction and higher frequency
of depressive symptoms over a 30-year period than those who were married and eventually
divorced. Building and maintaining high relationship quality over time appears to be key for
presence (or absence) of negative processes that diminish relationship quality, such as conflict
and dysfunction (Fincham & Beach, 2010). Fincham and Beach (2010) outline the various
reasons for this focus, one of the main reasons being that negative processes tend to have a
larger impact than positive ones. For instance, the seminal study by Gottman (1994) found it
takes five positive behaviors to counterbalance the impact of one negative behavior on
(2017) observed that the relative effect of negative relationship processes is much stronger than
positive processes, a finding that has been consistently reported in both the relationship and
processes has helped illustrate the importance of romantic relationships, this paints an
incomplete picture of relationship functioning. Instead, Fincham and Beach (2010, p. 7) argue
that “relationship health is not merely the absence of relationship dysfunction, just as the
absence of a physical illness is not sufficient to define physical health;” rather, the absence of
“relatively free of pain but is also relatively free of the positive benefits of relationships”. While
instead of a desire to maintain the relationship (i.e., approach commitment) (Strachman &
Gable, 2006). Differentiating between indifferent relationships and those that are well-
functioning is key to understanding how to facilitate optimal relationship functioning rather than
To address these issues, recent work in the field of relationship science has emphasized
high levels of positive emotion and psychosocial functioning (Keyes, 2002); similarly,
relationship flourishing aims to capture the positive relationship processes that contribute to “a
sense that their life as a couple is a life well lived” (Fincham & Beach, 2010, p. 7). These
processes can include, but are not limited to, emotional connection, partner support,
forgiveness, acceptance, trust, respect, positive affect, satisfaction, commitment, and love
(Fincham & Beach, 2010). Over the past decade, scholars have developed several models to
further our understanding of relationship flourishing. For example, the Strong Relationality
Model of Relationship Flourishing (SRM) is centered on the idea that “Ethical Responsiveness”
(i.e., viewing one’s partner as an “Other” versus an object) motivates partners to engage in pro-
5
THE LOVEBIRD SCALE
relationship behaviors called “Responsible Actions” (e.g., gratitude, support, affection), which
(Galovan & Schramm, 2018). Wood et al. (2022) provide empirical support for the SRM in their
longitudinal study, finding that perceived partner support mediates the impact of individual
stress on gratitude-recognition (i.e., “Responsible Actions”) 12 months later, which then results
measurement of relationship flourishing (Fincham & Beach, 2010; Fowers et al., 2016). Instead,
the emphasis on negative relationship processes that, until recently, has dominated the
literature has largely influenced the way we assess relationship quality. Specifically, existing
bipolar dimension ranging from extremely satisfied to extremely dissatisfied (Fincham & Beach,
2010; Fincham & Rogge, 2010). This overly narrow view of relationship quality fails to capture
the various processes that contribute to positive relationship functioning and further, relationship
flourishing. Additionally, previous research has shown there is more to relationship quality than
just perceptions of satisfaction, particularly within high quality relationships (Fletcher et al.,
2000; Fowers et al., 2016; Galovan et al., 2021). Thus, describing the specific behaviors,
interactions, and cognitions that contribute to the development and maintenance of long-lasting,
The present research aims to add to the growing field of positive relationship science
through the development and validation of a novel self-report measure of high-quality romantic
relationships called the Lovebird Scale. The Lovebird Scale aims to go beyond global,
beliefs, and behaviors that contribute to relationship flourishing and differentiate these
relationships from other long-term committed relationships. We use the term “lovebirds” as it is
often used colloquially to describe extremely affectionate and long-lasting romantic couples who
6
THE LOVEBIRD SCALE
are “in love” with each other, as opposed to just having loving feelings towards one another.
Whereas “soulmate” and “other half” tend to reflect an individual’s beliefs and expectations
about a specific partner or relationships in general (Franiuk et al., 2002), “lovebirds” is a more
accurate description of the feelings of closeness and warmth in everyday behaviors, cognitions,
and interactions that occur between partners that reinforce and further deepen their love for
each other.
Positive affect is central to the concept of flourishing, but few studies have specifically
examined positive affectivity within flourishing relationships (Fincham & Beach, 2010). Thus, a
secondary goal of this study was to explore the affective experiences of individuals in lovebird
relationships using the newly developed Lovebird Scale. We expect the Lovebird Scale to be
associated with more positive affect and less negative affect. Further, we expect the Lovebird
Scale to predict positive and negative affect above and beyond existing measures of
relationship quality. Through this research, we hope to further our understanding of what
constitutes long-lasting, high-quality lovebird relationships, as well as what makes them distinct
Study 1
Methods
Item Generation
Items on the initial Lovebird Scale were developed based on qualitative interviews of
long-term romantic couples who described themselves as being “lovebirds.” These couples
were asked to describe their relationship and why they considered themselves to be “lovebirds.”
Common themes that arose from these interviews included wanting to support each other’s
goals and aspirations, respecting each other’s opinions, prioritizing each other’s happiness,
appreciating the “little things” their partner does for them, not being afraid to be open and honest
with their partner, physical intimacy (e.g., affection, sex), and loving every aspect of their
partner. The initial set of “lovebird” items were generated based on these themes and quotes
7
THE LOVEBIRD SCALE
taken from the interviews. To help distinguish lovebird relationships from other types of
those in which partners may be satisfied with their relationship but would not consider
themselves to be deeply “in love” with each other and may be more motivated to stay together
for reasons other than their love for each other (e.g., financial reasons, children/dependents).
This first iteration of the Lovebird Scale contained 74 items, which were then reviewed
by a panel of relationship experts who excluded items that were redundant or were ambiguously
written. This left 49 items that described both lovebird (e.g., “All of life’s ups and downs seem
pretty insignificant compared to the love that we share,” and “We share a seamless continuum
of compassionate and erotic love”) and committed relationships (e.g., “I have some fundamental
doubts about my partner,” and “Sometimes I think my partner and I come from different
planets”). These items formed the initial Lovebird Scale that was tested in Study 1. Participants
were instructed to read each statement and choose the most appropriate response using a 7-
point Likert scale ranging from Strongly disagree to Strongly agree, keeping their current partner
A total of 552 participants completed Study 1. Twelve participants were excluded prior to
analyses if they had missing data, reported a relationship length less than six months, or
(RAS). We chose to use the RAS to determine eligibility for analyses because it has
& Baier, 1999). Low relationship satisfaction was defined as RAS scores more than three
standard deviations below the sample mean (i.e., RAS < 11.08). The final analyzed sample
used for the EFA included N = 540. This resulted in an observation-to-item ratio of
approximately 11, which meets previous recommendations for EFAs (Watkins, 2018). On
8
THE LOVEBIRD SCALE
average, participants were 39.9 years old (SD = 13.2), 60.6% female, 80.0% White, and 85.0%
heterosexual. 67.6% participants were married and 91.3% currently lived with their partner.
Participants reported being in their current relationship for an average of 11.4 years (SD = 10.8
Procedure
Recruitment and data collection for Study 1 occurred in June 2021. Participants were
recruited via CloudResearch (Litman et al., 2017). Participants needed to be at least 18 years
old and currently involved in a romantic relationship for at least six months. Individuals who met
the eligibility criteria and passed CloudResearch data security measures (i.e., ReCAPTCHA)
were invited to complete an online Qualtrics survey (15-20 minutes) about romantic
relationships. All participants provided electronic consent and were offered a monetary incentive
for their participation. An institutional review board at a small, private Southeastern university
Measures
questionnaire included questions regarding age, gender, ethnicity, sexual orientation, and
Relationship Assessment Scale (RAS), a brief 7-item scale designed to assess global
relationship satisfaction, with higher scores indicating higher relationship satisfaction (Hendrick,
1988). Example items include “How well does your partner meet your needs?” and “How much
partners was measured using the Mutual Psychological Development Questionnaire (MPDQ)
(Genero et al., 1992). The MPDQ is a 22-item self-report measure in which participants rate
9
THE LOVEBIRD SCALE
their own experience as well as perceptions of their partner’s experience when discussing
or bored) using a 6-point Likert scale ranging from Never to All the time, with higher scores
indicating greater perceived mutuality. Example items include “be receptive,” or “try to
understand” for the self-subscale, and “respect point of view,” and “see the humor in things” for
the partner subscale. The MPDQ has exhibited excellent internal consistency (α = 0.87), test-
retest reliability (α = 0.90), and construct validity in previous research (Genero et al., 1992).
relationship prototypes describing “lovebird” (“You are in love with your partner and feel that you
have found the ‘the love of your life’, a real soul mate…”) and “committed” (“There are many
things that you like and even admire about your partner, but there are also some things that
bother you a lot, and they are not likely to change…”) relationships. Each prototype consisted of
approximately the same word count and semantic structure (M = 165.5 words). Participants
used a sliding scale ranging from 1 to 100 to indicate how accurately each vignette described
Data Analysis
Data were analyzed using R Studio (version 2022.12.0+353) (Posit Team, 2022; R Core
Team, 2021). An exploratory factor analysis (EFA) was performed to determine the underlying
factor structure of the initial 49-item Lovebird Scale. The EFA was conducted using principal
axis factoring (PAF) estimation with promax rotation because we expected the identified factors
to be correlated with each other (Watkins, 2018). We used several metrics to determine the
factorability of the data, including inter-item correlations (<0.80), the Kaiser-Meyer-Olkin (KMO)
Measure of Sampling Adequacy (>0.50), and the Bartlett’s Test of Sphericity (p < .05) (Knekta
et al., 2019). To determine the appropriate number of factors to be extracted from the data, we
examined eigenvalues (>1.0), cumulative percentage of variance account for (>50%), and Scree
plots. Factor loadings were used to determine which items loaded onto each factor, the number
10
THE LOVEBIRD SCALE
of factors to retain, and which items should be considered for elimination. We then used the
following criteria to determine which items should be eliminated: (a) items with factor loadings
less than 0.40; (b) items with cross-loadings greater than 0.20; (c) items with communalities less
than 0.20; and (d) conceptual fit (Knekta et al., 2019). Cronbach’s alphas were also calculated
to assess internal consistency, with values ≥ 0.70 considered acceptable (Hair et al., 2010).
Correlations with the RAS and MPDQ were used to assess the preliminary convergent validity
of the Lovebird Scale. Statistical significance for all correlations was set at the p = .05 level.
Results
Descriptive statistics for all Study 1 variables (means, standard deviations, range, and
reliabilities) can be found in Table 2. Results from the KMO Measure of Sampling Adequacy
(0.97) and the Bartlett’s Test of Sphericity (χ2(1128) = 16423.08, p < .001) suggested that an
EFA was appropriate for the data. All 49 items had at least one inter-item correlation greater
than 0.30 and no inter-item correlation pairs were greater than 0.80, further supporting the
factorability of the data. A Scree plot identified five factors with eigenvalues greater than 1.0
(range = 1.04 to 19.81). Thus, we tested a five-factor EFA solution, which explained 43% of the
variance in the data. However, the fifth factor had high loadings for items that cross-loaded onto
other factors. We then tested a four-factor EFA solution, which explained 45% of the variance in
the data, followed by a three-factor EFA solution, which explained 38% of the variance in the
data. Upon further examination, the three-factor solution yielded three distinct factors that had at
least three items with factor loadings greater than 0.40 that also did not highly load onto the
other factors. Based on this, we decided to proceed with a three-factor solution. From here, we
identified several items that should be considered for elimination due to low factor loadings or
multiple high cross-loadings (Items 3, 7, 8, 9, 11, 21, 23, 25, 42, and 43). We removed these
items in a stepwise fashion and then reevaluated the three-factor structure. At this point, we
identified several other items that should be eliminated due to low factor loadings or multiple
11
THE LOVEBIRD SCALE
cross-loadings (Items 43, 39, 2, 40, 16, 29, 35, 28, and 15), leaving 31 items. The final three-
factor solution explained 46% of the variance in the items (Table 3).
After reviewing the content of each factor, we named the three factors as follows: Factor
1 was labeled “Mutuality,” Factor 2 was labeled “Romance,” and Factor 3 was labeled
“Disconnect.” The Mutuality subscale captures the pro-relationship behaviors and social
interactions that signal trust, acceptance, respect, and support. The Romance subscale reflects
the behavioral and cognitive processes that facilitate feelings of love and passion between
partners, such as physical intimacy and relational savoring. The Disconnect subscale
represents the opposite of lovebird relationships – that is, its items reflect indifference towards
the relationship, a lack of authenticity, and relational boredom. All three factors demonstrated
high internal consistency (α = 0.86 to 0.93). Unsurprisingly, Mutuality and Romance were
positively correlated with each other (r = .72, p < .001) and negatively correlated with
We then tested the preliminary convergent validity of the Lovebird Scale by examining
correlations between the three subscales, the relationship prototypes, and existing measures of
relationship quality (RAS and MPDQ). Mutuality was highly correlated with the RAS (r =.76, p
< .001) as well as both MPDQ subscales (MPDQ Self: r = .57, p < .001; MPDQ Partner: r = .74,
p < .001). Romance was also strongly correlated with the RAS (r = .69, p < .001) and both
MPDQ subscales (MPDQ Self: r = .57, p < .001; MPDQ Partner: r = .60, p < .001). Disconnect
was negatively correlated with all three measures (RAS: r = -.71, p < .001; MPDQ Self: r = -.69,
p < .001; MPDQ Partner: r = -.65, p < .001). The lovebird relationship prototype was positively
correlated with Mutuality (r = .72, p < .001) and Romance (r = .68, p < .001) and negatively
correlated with Disconnect (r = -.57, p < .001), while the opposite trend was observed in
correlations with the committed relationship prototype (Mutuality: r = -.28, p < .001; Romance: r
= -.35, p < .001; Disconnect: r = .48, p < .001). A summary of these correlations can be found in
Table 4.
12
THE LOVEBIRD SCALE
Study 1 Discussion
factor structure of the initial 49-item Lovebird Scale. Items were generated based on qualitative
interviews conducted with romantic couples who self-identified as “lovebirds,” and were then
refined by a panel of experts. Both empirical (eigenvalues, factor loadings, communalities) and
theoretical (conceptual fit, previous research) perspectives were considered when evaluating
the overall fit of the factor structure. 18 items were excluded from further consideration for
inclusion in the Lovebird Scale, resulting in a 31-item scale with three distinct factors: Mutuality,
Romance, and Disconnect. All three factors demonstrated high internal consistency and were
strongly correlated with existing measures of relationship quality, providing preliminary evidence
for the psychometric properties of the Lovebird Scale. A more stringent test of the factor
structure of the Lovebird Scale was conducted in Study 2 using confirmatory factor analyses
(CFAs).
Study 2
Methods
A total of 223 participants completed Study 2. Eight participants were removed prior to
analyses based on the exclusion criteria described in Study 1, resulting in a final analyzed
sample of N = 215. On average, participants were 38.7 years old (SD = 11.6), 67% female,
84.7% White, and 86% heterosexual. 61.4% participants were married and 91.6% currently
lived with their partner. Participants reported being in their current relationship for an average of
10.5 years (SD = 12.7 years). A summary of sample characteristics for Study 2 can be found in
Table 1.
Procedure
Recruitment and data collection for Study 2 occurred in October 2021. Like Study 1,
participants were recruited via CloudResearch (Litman et al., 2017) and were offered a
13
THE LOVEBIRD SCALE
monetary incentive for their participation. Individuals who met the eligibility criteria (i.e., at least
18 years old, currently involved in a romantic relationship for at least six months, had not
participated in Study 1) and passed CloudResearch data security measures (i.e., ReCAPTCHA)
were invited to complete an online Qualtrics survey (15-20 minutes) about romantic
and approved all aspects of this study. All participants provided electronic consent prior to
participation.
Measures
In addition to the Lovebird Scale, RAS, MPDQ, and relationship prototypes, participants
completed the Perceived Relationship Quality Components Inventory (PRQC) and the
Relationship Quality Scale (RQS). Participants also completed the demographic questionnaire
used in Study 1.
relationship quality: Satisfaction, Commitment, Intimacy, Trust, Passion, and Love. The PRQC
uses a 7-point Likert scale ranging from Not at all to Extremely, with higher scores indicating
higher levels of each relationship quality component. Example items include “How satisfied are
you with your relationship?” and “How connected are you to your partner?” The PRQC has also
Relationship Quality Scale. The RQS is a brief, 9-item measure of relationship quality
that was developed using a diverse sample of individuals representing over 60 countries.
Participants are asked to rate their degree of agreement with various statements regarding their
current relationship using a 5-point Likert scale ranging from Strongly disagree to Strongly
agree, with higher scores indicating higher relationship quality. Example items include “This is
the relationship I have always dreamed of,” and “I think of my partner as my soulmate.” The
RQS has demonstrated high internal consistency in previous research (Chonody et al., 2018).
14
THE LOVEBIRD SCALE
Data Analysis
Data were analyzed using R Studio (version 2022.12.0+353) (Posit Team, 2022; R Core
Team, 2021). A series of confirmatory factor analyses (CFAs) with robust maximum likelihood
estimation (MLR) were conducted using the R package lavaan (Rosseel, 2012). We first tested
the model proposed by the EFA in Study 1, followed by a one-factor model and an orthogonal
model to examine the degree of improvement in model fit resulting from the proposed factor
structure. The variance standardization method was used to fix factor variances to 1 and allow
factor loadings to be freely estimated. Consistent with previous recommendations, overall model
fit was evaluated using several different model fit indices: root-mean-square error of
approximation (RMSEA) ≤ 0.06, comparative fit index (CFI) ≥ 0.90, standardized root-mean-
square residual (SRMR) ≤ 0.08, and chi-square/degrees of freedom ratio (χ2/df) ≤ 3 (Hu &
Bentler, 1999; Kline, 2016; Knekta et al., 2019). We then consulted modification indices to
determine if there were any empirically and/or theoretically reasonable modifications that could
be made to improve overall model fit. Once the best fitting model was identified, we then
examined a higher-order model in which the subscales of the Lovebird Scale were nested within
a second-order factor representing lovebird relationships globally. Cronbach’s alphas were also
calculated to assess internal consistency, with values ≥ 0.70 considered acceptable (Hair et al.,
2010). We tested the convergent validity of the Lovebird Scale by examining Pearson
correlations with existing measures of relationship quality (RAS, MPDQ, PRQC, RQS).
Statistical significance for all correlations was set at the p = .05 level.
Results
Descriptive statistics for all Study 2 variables (means, standard deviations, range, and
reliabilities) can be found in Table 5. Results from the CFA on the EFA model were obtained
(χ2(431) = 875.9, p < .001; χ2/df = 2.03; CFI = .857; SRMR = .062; RMSEA = .069, 95% CI
[.063, .075], p < .001). The EFA model demonstrated significantly better model fit compared to
15
THE LOVEBIRD SCALE
an orthogonal model and single-factor model, supporting the use of a multidimensional model
with correlated factors. Based on modification indices, we made several revisions to the original
EFA model in an incremental fashion: item 12 (“My partner and I have recurring problems that
we can’t get past”) was moved from Mutuality to Disconnect, item 37 (“My partner and I go
through life savoring moments together”) was moved from Romance to Mutuality, and item 13
(“We are each other’s best friend”) was moved from Mutuality to Romance; item 14 (“My partner
and I know how to make each other laugh, even on our bad days”) was removed from the model
due to high cross-loadings on Mutuality and Romance; and items 4, 19, 26, and 27 were
removed from the model due to low loadings on their assigned factors. The model with the
remaining 26 items produced acceptable model fit statistics (χ2(296) = 485.6, p < .001; χ2/df =
1.64; CFI = .925; SRMR = .055; RMSEA = .055, 95% CI [.047, .062], p = .162). Standardized
regression weights ranged from 0.556 to 0.882 and were all statistically significant (p <.001).
Additionally, all item variances were positive (i.e., no Heywood cases) and statistically
significant (p <.001).
Given the high inter-factor correlations, we then tested a higher-order model, in which
the three factors were nested within one higher order factor representing lovebird relationships
(Figure 1). Model fit statistics for the higher-order model were equivalent to those produced by
the first-order model. The higher-order Lovebird factor accounted for a large portion of the
variance among the first-order factors, with R2 values ranging from 0.746 to 0.871. Based on the
CFAs, we retained 26 items for the Lovebird Scale (see Appendix): 10 items for Mutuality, eight
items for Romance, and eight items for Disconnect. The three subscales exhibited high internal
consistency, with Cronbach alphas ranging from 0.87 to 0.92. The higher-order Lovebird factor
also had high internal consistency with a Cronbach’s alpha of 0.95. A summary of model fit
statistics for the EFA, one-factor, orthogonal, and higher order models can be found in Table 6.
Convergent validity for the revised Lovebird Scale was tested using correlations with the
relationship prototypes and existing measures of relationship quality (RAS, MPDQ, PRQC, and
16
THE LOVEBIRD SCALE
RQS). All correlations were statistically significant and in the expected direction. Replicating
findings from Study 1, Mutuality was positively correlated with the RAS and both MPDQ
subscales (RAS: r = .83, p < .001; MPDQ Self: r = .69, p < .001; MPDQ Partner: r = .79, p
< .001), as was Romance (RAS: r = .67, p < .001; MPDQ Self: r = .63, p < .001; MPDQ Partner:
r = .64, p < .001). Like Study 1, Disconnect was negatively correlated with the RAS and MPDQ
subscales (RAS: r = -75, p < .001; MPDQ Self: r = -.68, p < .001; MPDQ Partner: r = -.69, p
< .001). Mutuality was positively correlated with all six PRQC subscales (r = .46 to .84, p
< .001), especially the Relationship Satisfaction subscale. Romance was also positively
correlated with the six PRQC subscales (r = .55 to .81, p < .001), with the largest correlation
being with the Love subscale. Disconnect was negatively correlated with the PRQC subscales
(r = -.45 to -.71, p < .001). All three subscales were strongly correlated with the RQS, a measure
of positive relationship functioning (Mutuality: r = .86, p < .001; Romance: r = .79, p < .001;
Disconnect: r = -.78, p < .001). The lovebird relationship prototype was positively correlated
with composite lovebird scores (r = .79, p < .001), Mutuality (r = .75, p < .001), and Romance (r
= .73, p < .001) and negatively correlated with Disconnect (r = -.68 p < .001), while the opposite
trend was observed in correlations with the committed relationship prototype (Composite
lovebird scores: r = -.54, p < .001; Mutuality: r = -.41, p < .001; Romance: r = -.44, p < .001;
measures of relationship quality. Composite lovebird scores were created by reverse scoring
items on the Disconnect subscale and then calculating the average of all three subscales. We
predicted this composite lovebird score to be highly positively correlated with existing measures
of relationship quality, particularly the RQS. As expected, composite lovebird scores were
significantly correlated with RAS scores (r = .84, p < .001), both MPDQ subscales (MPDQ Self:
r = .75, p < .001; MPDQ Partner: r = .79, p < .001), all six PRQC subscales (r = .56 to .80, p
17
THE LOVEBIRD SCALE
< .001), and the RQS (r = .90, p < .001). A summary of correlations between all Study 2
Study 2 Discussion
Scale in a new sample. The three-factor, 31-item model proposed by the EFA exhibited less
than satisfactory model fit. Based on modification indices, we made several revisions to the
model. Three items (Items 12, 13, and 37) were moved to different factors and five items (Items
4, 14, 19, 26, and 27) were removed due to high cross-loadings or low loadings on the assigned
factor, resulting in a final 26-item scale. To account for high intercorrelations between factors,
we then tested a second-order CFA model with a single higher-order factor representing all
three subscales of the Lovebird Scale. The final higher-order model demonstrated good model
fit, as well as acceptable psychometric properties. Finally, the three subscales and higher order
Lovebird factor were significantly correlated with existing measures of relationship quality, with
effect sizes ranging from medium to large. Because we made several modifications to the
model, additional cross-validation analyses were warranted to ensure the reliability, validity, and
replicability of the Lovebird Scale. Thus, we aimed to cross-validate the revised Lovebird Scale
Study 3
Methods
A total of 252 participants completed Study 2. Eleven participants were removed prior to
analyses based on the exclusion criteria described in Study 1, resulting in a final analyzed
sample of N = 241. On average, participants were 39.1 years old (SD = 10.2), 69.7% female,
83% White, and 86.7% heterosexual. Seventy percent of participants were married and 92.9%
currently lived with their partner. Participants reported being in their current relationship for an
18
THE LOVEBIRD SCALE
average of 11.7 years (SD = 11.2 years). A summary of sample characteristics for Study 3 can
be found in Table 1.
Procedure
Recruitment and data collection for Study 3 occurred in December 2021. Like Studies 1
and 2, participants were recruited via CloudResearch (Litman et al., 2017) and were offered a
monetary incentive for their participation. Individuals who met the eligibility criteria (i.e., at least
18 years old, currently involved in a romantic relationship for at least six months, had not
ReCAPTCHA) were invited to complete an online Qualtrics survey (15-20 minutes) about
reviewed and approved all aspects of this study. All participants provided electronic consent
prior to participation.
Measures
In addition to the Lovebird Scale, RAS, MPDQ, and relationship prototypes, participants
also completed the Dyadic Adjustment Scale (DAS-32), Adult Attachment Scale (AAS), and the
Positive and Negative Affect Schedule (PANAS), as well as several other measures related to
emotional abilities not discussed in the present paper. Since the survey for Study 3 was longer
than the first two surveys, we also included a six-item attention check measure (Infrequency
Scale). Participants also completed the demographic questionnaire used in Studies 1 and 2.
relationship quality in cohabiting or married couples (Spanier, 1976). The DAS-32 consists of
ordinal, Likert, and dichotomous scales, with total scores ranging from 0 to 151. The DAS-32
has also demonstrated acceptable internal consistency and construct validity in previous
Adult Attachment Scale. Anxious and avoidant attachment were measured using the
revised 18-item Adult Attachment Scale (AAS). Statements describe varying degrees of comfort
19
THE LOVEBIRD SCALE
with closeness and intimacy, as well as fear of rejection, within close relationships. Participants
are asked to read and rate how characteristic each statement is of them using a 5-point Likert
scale ranging from 1 (Not at all characteristic of me) to 5 (Very characteristic of me) (Collins,
1996). The revised AAS has demonstrated acceptable reliability in previous research (Collins,
1996).
Positive and Negative Affect Schedule. Affective state was measured using the
Positive and Negative Affect Schedule (PANAS), a 20-item self-report measure containing 10
positively valenced affect words (e.g., “Enthusiastic”, “Attentive”) and 10 negatively valanced
affect words (e.g., “Irritable”, “Upset”) (Merz et al., 2013). Participants are asked to indicate the
extent to which they have felt each affective state over the past week using a 5-point Likert
scale ranging from 1 (Very slightly or not at all) to 5 (Extremely). The PANAS has also
Infrequency Scale. A brief, 6-item Infrequency Scale was also included in the
attention check measures in online surveys to ensure participants are paying attention to the
questions and are answering honestly (Berinsky et al., 2014). The Infrequency Scale is an
includes items such as, “I enjoy visiting London, Wisconsin,” and, “I once rode my bicycle from
New York City to San Diego.” In this study, the Infrequency Scale was embedded within the
Adult Attachment Scale since the scale points matched those of the Infrequency items.
Infrequency items were listed after every three AAS items. Participants who scored over the
Data Analysis
Data were analyzed using R Studio (version 2022.12.0+353) (Posit Team, 2022; R Core
Team, 2021). Confirmatory factor analyses (CFAs) with robust maximum likelihood estimation
(MLR) were conducted using the R package lavaan (Rosseel, 2012). The variance
20
THE LOVEBIRD SCALE
standardization method was used to fix factor variances to 1 and allow factor loadings to be
freely estimated. Consistent with previous recommendations, overall model fit was evaluated
using several different model fit indices: root-mean-square error of approximation (RMSEA) ≤
0.06, comparative fit index (CFI) ≥ 0.90, standardized root-mean-square residual (SRMR) ≤
0.08, and chi-square/degrees of freedom ratio (χ2/df) ≤ 3 (Hu & Bentler, 1999; Kline, 2016;
Knekta et al., 2019). Cronbach’s alphas were also calculated to assess internal consistency,
correlations with existing measures of relationship quality (RAS, MPDQ, PRQC, RQS). We used
the Fronell-Larcker Criterion to assess discriminant validity, which states that the square root of
the average variance extracted of each latent variable must be greater than the correlation
between the latent variable and other constructs (i.e., attachment style). Finally, we tested the
incremental validity of the Lovebird Scale by estimating a series of regression models where
both the Lovebird Scale and the RAS or DAS-32 were entered as simultaneous predictors of
affective state (PANAS). Statistical significance for all correlations was set at the p = .05 level.
Results
Descriptive statistics for all Study 3 variables (means, standard deviations, range, and
reliabilities) can be found in Table 8. The higher-order model demonstrated acceptable model fit
in our cross-validation analyses, although slightly lower than those seen in Study 2 (χ2(296) =
616.64, p < .001; χ2/df = 2.08; CFI = .885; SRMR = .061; RMSEA = .067, 95% CI [.060, .074], p
< .001). Standardized regression weights ranged from 0.469 to 0.833 and were all statistically
significant (p < .001). The higher-order Lovebird factor accounted for a large portion of the
variance among the first-order factors, with R2 values ranging from 0.818 to 0.845. The three
subscales exhibited high internal consistency, with Cronbach alphas ranging from 0.85 to 0.92.
21
THE LOVEBIRD SCALE
The higher-order Lovebird factor also had high internal consistency with a Cronbach’s alpha of
0.94.
Like Studies 1 and 2, we tested the convergent validity of the Lovebird Scale by
examining correlations with the relationship prototypes and existing measures of relationship
quality (RAS, MPDQ, and DAS-32). All correlations were statistically significant and in the
expected direction. Mutuality and Romance were positively correlated with RAS scores
(Mutuality: r = .82, p < .001; Romance: r = .70, p < .001), while Disconnect was negatively
correlated with these measures (RAS: r = -.78, p < .001; MPDQ Self: r = -.61, p < .001; MPDQ
Partner: r = -.70, p < .001). Mutuality and Romance produced similar patterns of correlations
with the DAS-32, with both subscales having the largest correlation with the DAS Satisfaction
subscale (Mutuality: r = .81, p < .001; Romance: r = .68, p < .001) and the smallest correlation
with the DAS Consensus subscale (Mutuality: r = .47, p < .001; Romance: r = .41, p < .001).
Disconnect was negatively correlated with the DAS-32 (r = -.46 to -.79, p < .001). Correlations
between composite lovebird scores and existing relationship quality measures ranged from .50
to .86, with the smallest correlation being with the DAS Consensus subscale and largest
correlation being with the RAS and DAS Satisfaction subscale. The lovebird relationship
prototype was positively correlated with composite lovebird scores (r = .79, p < .001), Mutuality
(r = .76, p < .001), and Romance (r = .74, p < .001) and negatively correlated with Disconnect (r
= -.71, p < .001), while the opposite trend was observed in correlations with the committed
relationship prototype (Composite lovebird scores: r = -.37, p < .001; Mutuality: r = -.24, p
< .001; Romance: r = -.35, p < .001; Disconnect: r = .42, p < .001).
computed the AVE for the three subscales and the higher-order Lovebird factor, which ranged
from 0.43 to 0.55. We then calculated the square root of each AVE and compared them with
correlations with the AAS Anxiety and Avoidant subscales (Table 5). We observed small, but
statistically significant negative correlations between Mutuality, Romance, and the AAS
22
THE LOVEBIRD SCALE
subscales (r = -.13 to -.23, p < .05), while Disconnect was positively correlated with both AAS
subscales (r = .17 to .34, p < .01). Composite lovebird scores were negatively correlated with
both AAS subscales (AAS Anxiety: r = -.20, p < .01; AAS Avoidant: r = -.28, p < .001).
Importantly, these correlations were smaller than the square roots of the AVE for each latent
variable, providing initial evidence for the discriminant validity of the Lovebird Scale.
Lovebird Scale and positive and negative affect (PANAS). We expected composite lovebird
scores, Mutuality, and Romance to be associated with higher positive affect and lower negative
affect, while Disconnect would exhibit the opposite pattern. As we predicted, Mutuality and
Romance were positively correlated with the PANAS Positive subscale (Mutuality: r = .20, p
< .01; Romance: r = .34, p < .001), as were composite lovebird scores (r = .25, p < .001).
Surprisingly, neither Mutuality nor Romance were significantly correlated with the PANAS
Negative subscale, but composite lovebird scores were, although this correlation was small (r =
-.14, p < .05). Disconnect was negatively correlated with the PANAS Positive subscale (r = -.18,
p < .01) and positively correlated with the PANAS Negative subscale (r = .21, p < .01), as
expected. A summary of correlations between all Study 3 variables can be found in Table 9.
Finally, we tested the incremental validity of the Lovebird Scale by regressing PANAS
Positive and Negative scores on composite lovebird scores while controlling for the RAS and
DAS-32. We also controlled for sex (female), relationship status (married), and age in each set
of models. As seen in Table 10, the Lovebird Scale emerged as a significant and unique
predictor of PANAS Positive scores above and beyond the RAS and DAS-32, but not PANAS
Negative scores. In both sets of models, composite lovebird scores were positively related to
positive affect (RAS model: β = 2.86, p = .007; DAS-32 model: β = 2.28, p = .005), such that
higher lovebird scores were associated with higher levels of positive affect. Composite lovebird
scores were not significantly related to negative affect when controlling for the RAS and DAS-32
Study 3 Discussion
The purpose of Study 3 was to cross-validate the factor structure from Study 2 in an
independent sample. Using the same methods as Study 2, CFA confirmed the existence of
three subscales (Mutuality, Romance, Disconnect) that together form one higher order factor
representing lovebird relationships. Although model fit indices for the higher-order model in this
sample were slightly less than the recommended cut offs, they were still in an acceptable range
(Hu & Bentler, 1999). The Lovebird Scale exhibited high internal consistency, with Cronbach’s
alphas ranging from 0.85 to 0.95. Evidence for the convergent validity of the Lovebird Scale was
demonstrated through large correlations with existing measures of relationship quality, which
included the RAS, MPDQ, and DAS-32. Mutuality and Romance, as well as the composite
lovebird factor, were highly positively correlated with these measures, while Disconnect was
negatively correlated with them, as expected. The Lovebird Scale also exhibited good
discriminant validity, as evidenced by small correlations with the AAS Anxiety and Avoidant
subscales, which were smaller than the square root of the AVE for each of the latent constructs
Preliminary evidence for the incremental validity of the Lovebird Scale was
scores, while controlling for existing measures of relationship quality (RAS and DAS-32).
Composite lovebird scores emerged as a significant and unique predictor of positive affect
above and beyond both the RAS and DAS-32. Further, neither the RAS nor DAS were
significant predictors of positive affect when composite lovebird scores were entered into the
model simultaneously. This suggests that composite lovebird scores may offer additional insight
into positive affectivity within high-quality romantic relationships. We did not observe the same
pattern when examining the regression models predicting negative affect. Specifically,
composite lovebird scores were not a significant predictor of negative affect when controlling for
RAS and DAS-32 scores. While RAS scores were also not related to negative affect when
24
THE LOVEBIRD SCALE
entered alongside composite lovebird scores, DAS-32 scores were still negatively related to
negative affect even when composite lovebird scores were included in the model. One possible
explanation for this is the scope of the Lovebird Scale compared to the DAS-32. The DAS was
originally developed to distinguish between married and divorced couples, hence the overall
emphasis on the (lack of) communication within the relationship (Fincham & Rogge, 2010).
Thus, the DAS may be better at tapping into the negative affectivity within relationships than the
Lovebird Scale.
General Discussion
Positive relationship science has garnered increasing interest in recent years, with more
2010; Fowers et al., 2016). Yet, few existing measures of relationship quality capture the
specific processes that contribute to relationship flourishing (Fowers et al., 2016). The present
study aimed to address this gap in the literature through the development and validation of the
designed to assess the various behavioral, social, cognitive, and affective processes that occur
within long-lasting, high-quality “lovebird” relationships that may not be adequately captured by
existing unidimensional measures (Fincham & Rogge, 2010; Fletcher et al., 2000). The
furthering our understanding of relationship flourishing and the key factors that contribute to
Three studies were conducted to develop and validated the Lovebird Scale using data
collected from over 850 individuals in long-term committed romantic relationships. Results
supported a higher-order factor structure with three subscales nested within one overarching
“lovebird” factor. The three subscales we identified were labeled Mutuality, Romance, and
Disconnect. Cronbach’s alphas for the three subscales and composite lovebird factor surpassed
25
THE LOVEBIRD SCALE
standard criteria (≥ 0.70) across all three studies, supporting the internal consistency of the
scale. The Lovebird Scale also exhibited good convergent validity across Studies 1 through 3,
as well as satisfactory discriminant and incremental validity in Study 3. Below we describe the
content of each subscale and how each relate to previous research on positive relationship
processes.
Mutuality
Mutuality within intimate relationships has previously been defined as the “modes of
social interaction that facilitate participation in and growth through the relationship” (Genero et
al., 1992, p. 37). In line with this definition, our Mutuality subscale captures the behaviors and
interactions that signal trust, respect, acceptance, and support between partners. Specifically,
lovebirds are respectful and considerate of each other when sharing their feelings and opinions,
instead of being afraid that their partner may react with contempt or judgment (e.g., “My partner
and I respect each other’s opinions, even when we don’t agree with each other,” and “I can talk
correlated with the MPDQ Partner subscale across all three studies, which measures the
degree to which an individual believes their partner is receptive and open when they are
perception of their partner being capable of mutuality, rather than just an introspective reflection
Mutuality also signals dependability and responsiveness of one’s partner, both of which
are important for building trust within one’s relationship (Coan et al., 2013; Weigel, 2010). This
is illustrated through large, positive correlations between the Mutuality subscale and the PRQC
indicates an individual’s confidence in the ability to confide in and depend on their partner,
without worry, jealousy, or suspicion (Campbell & Stanton, 2019; Kito, 2016). Couples with
higher levels of trust tend to attribute their partner’s behaviors and motives during a conflict
26
THE LOVEBIRD SCALE
situation more positively, which can help promote trust and relationship maintenance over time
(Miller & Rempel, 2004; Rempel et al., 2001). Higher levels of mutuality within lovebird
relationships may also facilitate authenticity (“I don’t have to sacrifice aspects of myself to keep
my partner happy”) and goal sharing (“I support my partner in their goals and aspirations, and
they do the same for me”), both of which have been linked to trust and intimacy within close
relationships (Fowers & Owenz, 2010; Wickham, 2013). Further, both authenticity and goal-
sharing have been cited as key positive relationship processes associated with relationship
Romance
passion and intimacy components of love (Sternberg, 1986). Sternberg (1986, p. 124) describes
romantic lovers as those who are “not only drawn physically to each other but are also bonded
emotionally,” much like those depicted in classic works of literature like Romeo and Juliet. In line
with this, the Romance subscale captures the behavioral and cognitive processes that facilitate
love and passion between partners, such as physical intimacy and relational savoring. Research
has generally focused on emotional intimacy as a key predictor of relationship quality; however,
our findings emphasize the importance of physical intimacy in promoting lovebird relationships
as well. We found two aspects of physical intimacy especially important for fostering lovebird
relationships – sexual intimacy (e.g., “Our sex life is deeply satisfying”) and physical touch (e.g.,
that the Romance subscale was strongly correlated with global assessment of relationship
quality, as well as more specific components of relationship quality such as emotional intimacy,
passion, and love. Indeed, sexual intimacy plays a vital role in the quality and maintenance of
romantic relationships (Muise et al., 2018), and previous research has also shown that
affectionate touch is positively associated with relationship satisfaction, perceived intimacy, and
couple’s positive affectivity (Debrot et al., 2013; Hesse & Mikkelson, 2017; Jakubiak & Feeney,
27
THE LOVEBIRD SCALE
2016; Wagner et al., 2020). Based on previous findings (Yoo et al., 2014), these results suggest
that physical intimacy may help foster lovebird relationships by increasing emotional intimacy,
The Romance subscale also measures instances of relational savoring within lovebird
appreciate positive experiences; relational savoring, on the other hand, focuses specifically on
savoring experiences that involve a romantic partner (Borelli et al., 2015; Bryant, 2021). Prior
work suggests that relational savoring may be an important interpersonal resource for romantic
partners, as it is associated with several indicators of both individual and relational well-being
including more positive emotions, fewer negative emotions, relationship quality, and dyadic
adjustment, and has been shown to buffer the negative effects of relationship stressors on
relationship satisfaction, especially for those in highly satisfied couples (Borelli et al., 2015;
Borelli et al., 2014; Borelli et al., 2020; Bryant, 2021; Costa-Ramalho et al., 2015; Lenger &
Gordon, 2019). Convergent correlations revealed a large positive correlation between the
Romance subscale and the RQS, highlighting the role of relational savoring in promoting
positive relationship processes. Additionally, Romance was positively associated with the PRQC
Love and Intimacy subscales, which suggests that relational savoring may help foster and
intensify feelings of love towards and closeness with one’s partner within lovebird relationships.
Disconnect
Factor analyses identified a subset of items that represented the opposite of lovebird
relationships, which we called Disconnect. The Disconnect subscale is comprised of items that
reflect feelings of ambivalence and indifference within the relationship, all of which are
characteristic of “numbed” relationships (Fincham & Beach, 2010). A lack of positive relationship
inconsistent feelings towards one’s partner and the relationship (e.g., “Although I love my
partner, I would not say that I am currently ‘in love’”) (Holt-Lunstad & Uchino, 2019; Ross et al.,
28
THE LOVEBIRD SCALE
2019). One indicator of ambivalence is extradyadic desire, which refers to a desire for an
attractive alternative to one’s current partner (Zoppolat et al., 2022) – this is captured by items
such as “I am easily attracted to others when I am away from home.” Such ambivalence may
individual employs various behavioral (e.g., not spending time with one’s partner; “I feel like I
need space after we spend a lot of time together”) and/or emotional strategies (e.g., avoiding
self-disclosure; “I am more myself when I am along than when I am with my partner”) to create
distance between oneself and their partner (Callaci et al., 2020; Holt-Lunstad & Uchino, 2019).
Large negative correlations with the DAS Cohesion and PRQC Intimacy and Love subscales
may reflect how ambivalent relationship partners may experience less intimacy and fewer
The Disconnect subscale illustrates how the absence of positive relationship processes
may lead to disaffection, which refers to the gradual decline in love and increase in feelings of
indifference towards one’s partner (Abbasi & Alghamdi, 2017). Disaffection is often initiated by
disaffection: disillusionment (i.e., qualities once perceived positively are progressively viewed
negatively), tunnel vision (i.e., focus on a partner’s negative qualities), and indifference (Abbasi
& Alghamdi, 2017). Items such as “Sometimes I think my partner and I come from different
planets” and “My partner and I have recurring problems that we can’t get past” reflect these
various stages of relationship disaffection. The Disconnect subscale was moderately negatively
associated with global measures of relationship quality, consistent with previous findings
(Abbasi & Alghamdi, 2017; Barry et al., 2008). Indeed, disaffection is a leading cause of
relationship distress and frequently cited reason for seeking marriage counseling (Abbasi &
Alghamdi, 2017). Negative correlations between the Disconnect subscale and the PRQC
Intimacy and Trust subscales further illustrates the deterioration of both physical and emotional
Positive affect has been cited as a central feature of relationship flourishing (Fincham &
Beach, 2010). Thus, a secondary goal of this study was to examine the affective state of
relationships to be associated with higher levels of positive affect and lower levels of negative
affect. As predicted, composite lovebird scores, as well as the two lovebird subscales, were
associated with higher levels of positive affect, suggesting that individuals in lovebird
relationships tend to experience more positive affectivity within a given week. Notably, the
Romance subscale was more strongly associated with positive affect than the Mutuality
subscale. In the context of lovebird relationships, taking time to think about and enjoy one’s
partner appears to be an important mechanism for increasing positive affect, which makes
sense given that savoring is a self-regulatory process used to generate positive emotions
(Borelli et al., 2020; Costa-Ramalho et al., 2015; Lenger & Gordon, 2019). Physical intimacy has
also been linked to emotional well-being, including higher levels of positive affect and lower
levels of negative affect (Debrot et al., 2017). A daily diary study by Debrot et al. (2017) found
that sex was associated with higher levels of same-day positive affect and predicted next-day
positive affect, whereas positive affect did not increase the likelihood of having sex. Thus,
physical intimacy may enhance feelings of closeness between partners, which in turn promotes
affect. While surprising, there are several possible explanations for this finding. First, people in
lovebird relationships may value and validate whatever the partner is feeling whether it is
positive or negative, and there is no particular emphasis on making sure the other person does
not experience negative affect. Second, even though they may experience negative affect, the
higher levels of positive affect reported in lovebird relationships may “cancel out” the negative
emotions to some degree. In fact, Gottman (1994) observed that divorce is more likely if couples
30
THE LOVEBIRD SCALE
displayed a ratio of positive to negative emotions of less than 5-to-1 during a conflict task.
Couples will inevitably experience negative emotions during conflict; however, highly satisfied
couples, whom Gottman and Gottman (2017) call “masters of relationships,” do not enter these
states as often and when they do enter them, they are able to escape more easily. Third, it is
possible that the time to recovery for a given level of negative affect may be shorter in lovebird
relationships. Although there is evidence to suggest that social sharing can lead to momentary
increases in negative affect (Rimé et al., 2020), a recent study by Rauers and Riediger (2022)
found that sharing daily hassles predicted momentary and long-term increases in relationship
closeness in romantic partners. However, since there was no significant association in the
current sample, we cannot rule out the possibility that the association was missed with this
sample size. Future research should examine the types of loving and other positive feelings that
characterize lovebird versus committed ambivalent relationships (e.g., being in love as well as
being happy, satisfied, warm, etc.). Longitudinal or daily experience studies can also track this
ratio over time to determine how it relates to temporal changes in relationship quality.
Finally, we found that Disconnect was associated with lower levels of positive affect and
higher levels of negative affect. This is not surprising, given that aspects of “numbed”
relationships such as ambivalence and relational distancing have been associated with greater
negative affect in previous research (Harasymchuk & Fehr, 2012; Holt-Lunstad & Uchino, 2019;
Ross et al., 2019). Moreover, indifferent relationships have been characterized by a lack of
positivity within the relationship, which may explain the negative association between
Disconnect and positive affect (Harasymchuk & Fehr, 2012; Holt-Lunstad & Uchino, 2019; Ross
et al., 2019). Further, Gottman’s research also revealed that some couples do not engage in
negative, hostile behavior during conflict; rather, they behave in a more detached way,
experiencing very little positive emotion during conflict interactions. Though these relationships
may last longer than the “disastrous” relationships, the lack of positive affect will eventually
related to but conceptually distinct from relationship quality (Li & Chan, 2012). Attachment style
is closely linked to various cognitive, emotional, and behavioral indicators of relationship quality
and functioning, including satisfaction, closeness, emotional experience, and conflict resolution
behaviors (Li & Chan, 2012). Composite lovebird scores, as well as the Mutuality and Romance
subscales, were associated with less anxious and avoidant attachment. This finding is
expected, as lovebird couples tend to exhibit more constructive interactions (e.g., ability to
discuss problems or insecurities), greater feelings of connectedness, and less general conflict
(Li & Chan, 2012). Instead, we would expect individuals in lovebird relationships to be more
securely attached. However, since we did not explicitly measure secure attachment in our study,
On the other hand, the Disconnect subscale was positively associated with both anxious
and avoidant attachment. A meta-analysis by Li and Chan (2012) found that anxious attachment
was strongly associated with the presence of negative relationship processes like general
conflict and destructive interactions, while avoidant attachment was strongly associated with the
absence of positive relationship indicators like connectedness and support behaviors. Items
such as “My partner and I have recurring problems that we can’t get past” reflect destructive
interactions associated with anxious attachment, while items like “I feel like I need space after
we spend a lot of time together” represent the lack of closeness seen characteristic of an
avoidant attachment style. Additionally, the correlation between the Disconnect subscale and
anxious attachment was larger than the correlation with avoidant attachment. Joel et al. (2018)
found that anxious attachment was associated with the endorsement of many reasons for
wanting to both stay and leave one’s relationship (e.g., “We stay together for external reasons
such as marriage vows and children, more than because of our enjoyment of being together”).
Nearly half of their sample endorsed motivation to both stay and leave the relationship,
32
THE LOVEBIRD SCALE
highlighting the prevalence of stay/leave decision conflict among people in relationships. These
findings may help explain the stronger association between Disconnect and anxious attachment
There are several limitations of the present study that should be addressed. First, the
cross-sectional nature of this research limits the generalizability of these findings. Self-report
questionnaires are subject to various biases including response bias and social desirability.
Further, since data were only collected at one point in time, the findings may not reflect
relationship quality over time and may be influenced by temporal fluctuations in relationship
quality for a given level of negative affect. Experience sampling techniques (e.g., daily diaries)
longitudinal study design would be beneficial for understanding how lovebird and committed
relationships are developed and maintained over time. It is also important to note that this study
only captures the experiences of one relationship partner, and there may be discrepancies
between how partners view their relationship. Future research should have both partners
complete the scale. This could highlight potential relationship issues if one partner believes they
are lovebirds, but the other does not. In-laboratory behavior observations may help illustrate
how lovebird couples interact with each other and how the factors identified in the present
Another limitation to consider the study sample. The use of an online recruitment
platform such as CloudResearch allowed us to recruit individuals from across the U.S. but
despite this, our sample was still primarily White, female, educated (i.e., bachelor’s degree or
higher), and heterosexual. Future research should take steps to include a more diverse sample
sexualities, and cultures. In addition, given the use of online recruitment methods, there is
always the possibility that “bots” completed the survey (i.e., automated responses), which can
33
THE LOVEBIRD SCALE
impact the quality of the data (Litman et al., 2017). While multiple steps were taken to address
this limitation in the current research (i.e., use of data security measures), future research
should aim to replicate these findings in different populations and use other forms of data
There are some methodological limitations that should also be mentioned. It is possible
that the results from the CFAs conducted in Studies 2 and 3 may have been underpowered due
to the small sample size. While the sample sizes in this study met previous recommendations
for factor analysis sample sizes (i.e., sample sizes greater than 200), other recommendations
suggest using a ratio of 10+ per item (Knekta et al., 2019; Watkins, 2018). Additionally, because
we excluded individuals with low relationship quality from analyses, the non-normal distribution
of the data could have affected the analyses. Future research should routinely examine the
model fit of these measurements in independent samples and propose modifications as needed
to improve the reliability and validity of these scales. Another methodological limitation that
should be considered is the limited number of measures we used to establish the construct
validity of the Lovebird Scale. For example, to avoid overexerting participants taking the length
survey in Study 3, we only measured anxious and avoidant attachment but not secure
attachment. Though it can be assumed that lovebirds are securely attached, future studies
should also use measures of secure attachment to confirm this. Additionally, future research
should also incorporate more measures of love (e.g., compassionate, passionate) and other
measures relevant to the specific facets of lovebird relationships to further examine the
Conclusions
Across three studies, we present preliminary evidence for the psychometric properties of
the Lovebird Scale, a novel assessment of relationship quality that aims to capture the ongoing
behaviors, interactions, and cognitions that foster flourishing couple relationships. Previous
scales have largely focused on broad, general assessments that operationalize relationship
34
THE LOVEBIRD SCALE
quality as the presence or absence of relationship problems, rather than factors that contribute
to relationship flourishing. The scale advances the science of love by operationalizing the
satisfaction to distinguish between “lovebirds” and other types of adult committed relationships.
35
THE LOVEBIRD SCALE
References
Abbasi, I. S., & Alghamdi, N. G. (2017). Polarized couples in therapy: Recognizing indifference
as the opposite of love. Journal of Sex & Marital Therapy, 43(1), 40-48.
https://doi.org/10.1080/0092623X.2015.1113596
Barry, R. A., Lawrence, E., & Langer, A. (2008). Conceptualization and assessment of
https://doi.org/10.1111/j.1475-6811.2008.00200.x
Berinsky, A. J., Margolis, M. F., & Sances, M. W. (2014). Separating the shirkers from the
Borelli, J. L., Rasmussen, H. F., Burkhart, M. L., & Sbarra, D. A. (2015). Relational savoring in
1108. https://doi.org/10.1177%2F0265407514558960
Borelli, J. L., Sbarra, D. A., Snavely, J. E., McMakin, D. L., Coffey, J. K., Ruiz, S. K., Wang, B.
A., & Chung, S. Y. (2014). With or without you: Preliminary evidence that attachment
https://doi.org/10.1037/a0037780
Borelli, J. L., Smiley, P. A., Kerr, M. L., Hong, K., Hecht, H. K., Blackard, M. B., Falasiri, E.,
doi.org/10.1037/pst0000284
Braithwaite, S. R., Delevi, R., & Fincham, F. D. (2010). Romantic relationships and the physical
https://doi.org/https://doi.org/10.1111/j.1475-6811.2010.01248.x
36
THE LOVEBIRD SCALE
Braithwaite, S. R., & Holt-Lunstad, J. (2017). Romantic relationships and mental health. Current
https://doi.org/https://doi.org/10.1016/j.copsyc.2016.04.001
Bryant, F. B. (2021). Current progress and future directions for theory and research on savoring.
Callaci, M., Péloquin, K., Barry, R. A., & Tremblay, N. (2020). A dyadic analysis of attachment
https://doi.org/10.1111/jmft.12422
Campbell, L., & Stanton, S. C. (2019). Adult attachment and trust in romantic relationships.
https://doi.org/10.1016/j.copsyc.2018.08.004
Chonody, J. M., Gabb, J., Killian, M., & Dunk-West, P. (2018). Measuring relationship quality in
https://doi.org/10.1177/1049731516631120
Coan, J. A., Kasle, S., Jackson, A., Schaefer, H. S., & Davidson, R. J. (2013). Mutuality and the
social regulation of neural threat responding. Attachment & Human Development, 15(3),
303-315. https://doi.org/10.1080/14616734.2013.782656
Collins, N. L. (1996). Working models of attachment: Implications for explanation, emotion, and
https://doi.org/10.1037/0022-3514.71.4.810
Costa-Ramalho, S., Marques-Pinto, A., Ribeiro, M. T., & Pereira, C. R. (2015). Savoring positive
events in couple life: Impacts on relationship quality and dyadic adjustment. Family
Debrot, A., Meuwly, N., Muise, A., Impett, E. A., & Schoebi, D. (2017). More than just sex:
Affection mediates the association between sexual activity and well-being. Personality
37
THE LOVEBIRD SCALE
https://doi.org/10.1177%2F0146167216684124
Debrot, A., Schoebi, D., Perrez, M., & Horn, A. B. (2013). Touch as an interpersonal emotion
regulation process in couples’ daily lives: The mediating role of psychological intimacy.
https://doi.org/10.1177%2F0146167213497592
Dush, C. M. K., & Amato, P. R. (2005). Consequences of relationship status and quality for
https://doi.org/https://doi.org/10.1177/0265407505056438
Fincham, F. D., & Beach, S. R. H. (2010). Of memes and marriage: Toward a positive
https://doi.org/10.1111/j.1756-2589.2010.00033.x
Fincham, F. D., & Rogge, R. (2010). Understanding relationship quality: Theoretical challenges
and new tools for assessment. Journal of Family Theory & Review, 2, 227-242.
https://doi.org/10.1111/j.1756-2589.2010.00059.x
Fletcher, G. J. O., Simpson, J. A., & Thomas, G. (2000). The measurement of perceived
Fowers, B. J., Laurenceau, J.-P., Penfield, R. D., Cohen, L. M., Lang, S. F., Owenz, M. B., &
of the relationship flourishing scale. Journal of Family Psychology, 30, 997-1007. https://
doi.org/10.1037/fam0000263
Fowers, B. J., & Owenz, M. B. (2010). A eudaimonic theory of marital quality. Journal of Family
2589.2010.00065.x
38
THE LOVEBIRD SCALE
Franiuk, R., Cohen, D., & Pomerantz, E. M. (2002). Implicit theories of relationships:
Implications for relationship satisfaction and longevity. Personal Relationships, 9(4), 345-
367. https://doi.org/10.1111/1475-6811.09401
Galovan, A. M., Carroll, J. S., Schramm, D. G., Leonhardt, N. D., Zuluaga, J., McKenadel, S. E.
M., & Oleksuik, M. R. (2021). Satisfaction or connectivity?: Implications from the strong
Galovan, A. M., & Schramm, D. G. (2018). Strong relationality and ethical responsiveness: A
framework and conceptual model for family science. Journal of Family Theory & Review,
Genero, N. P., Miller, J. B., Surrey, J., & Baldwin, L. M. (1992). Measuring perceived mutuality
Gottman, J. (1994). What predicts divorce? The relationship between marital processes and
Gottman, J., & Gottman, J. (2017). The natural principles of love. Journal of Family Theory &
Hair, J. F., Black, B., Black, W. C., Babin, B. J., & Anderson, R. E. (2010). Multivariate data
6811.2011.01346.x
Holt-Lunstad, J., Birmingham, W., & Jones, B. Q. (2008). Is there something unique about
marriage? The relative impact of marital status, relationship quality, and network social
Holt-Lunstad, J., & Uchino, B. N. (2019). Social ambivalence and disease (sad): A theoretical
https://doi.org/10.1177%2F1745691619861392
Hu, L.-t., & Bentler, P. M. (1999). Cutoff criteria for fit indexes in covariance structure analysis:
Conventional criteria versus new alternatives. Structural Equation Modeling, 6(1), 1-55.
https://doi.org/10.1080/10705519909540118
Jakubiak, B. K., & Feeney, B. C. (2016). A sense of security: Touch promotes state attachment
https://doi.org/10.1177%2F1948550616646427
Joel, S., MacDonald, G., & Page-Gould, E. (2018). Wanting to stay and wanting to go:
https://doi.org/10.1177/1948550617722834
Keyes, C. L. M. (2002). The mental health continuum: From languishing to flourishing in life.
https://doi.org/https://doi.org/10.2307/3090197
Kito, M. (2016). Shared and unique prototype features of relationship quality concepts. Personal
Kline, R. B. (2016). Principles and practice of structural equation modeling (4th ed.). The
Guilford Press.
40
THE LOVEBIRD SCALE
Knekta, E., Runyon, C., & Eddy, S. (2019). One size doesn't fit all: Using factor analysis to
gather validity evidence when using surveys in your research. CBE Life Sciences
Lenger, K. A., & Gordon, C. L. (2019). To have and to savor: Examining associations between
savoring and relationship satisfaction. Couple and Family Psychology: Research and
Li, T., & Chan, D. K. S. (2012). How anxious and avoidant attachment affect romantic
Litman, L., Robinson, J., & Abberbock, T. (2017). Turkprime.Com: A versatile crowdsourcing
data acquisition platform for the behavioral sciences. Behavior Research Methods, 49,
433-442. https://doi.org/10.3758/s13428-016-0727-z
Merz, E. L., Malcarne, V. L., Roesch, S. C., Ko, C. M., Emerson, M., Roma, V. G., & Sadler, G.
original and short forms in an african american community sample. Journal of Affective
Miller, P. J. E., & Rempel, J. K. (2004). Trust and partner-enhancing attributions in close
10.1177/0146167203262803
Muise, A., Maxwell, J. A., & Impett, E. A. (2018). What theories and methods from relationship
research can contribute to sex research. The Journal of Sex Research, 55(4-5), 540-
562. https://doi.org/10.1080/00224499.2017.1421608
Rauers, A., & Riediger, M. (2022). Ease of mind or ties that bind? Costs and benefits of
19485506221112252.
41
THE LOVEBIRD SCALE
Rempel, J. K., Ross, M., & Holmes, J. G. (2001). Trust and communicated attributions in close
10.1037/0022-3514.81.1.57
Rimé, B., Bouchat, P., Paquot, L., & Giglio, L. (2020). Intrapersonal, interpersonal, and social
sharing outcomes of the social sharing of emotion. Current Opinion in Psychology, 31,
127-134. https://doi.org/10.1016/j.copsyc.2019.08.024
Roberson, P. N. E., Norona, J. C., Lenger, K. A., & Olmstead, S. B. (2018). How do relationship
symptoms, and life satisfaction across 30 years. Journal of Child and Family Studies, 27,
2171-2184. https://doi.org/https://doi.org/10.1007/s10826-018-1052-1
Robles, T. F., Slatcher, R. B., Trombello, J. M., & McGinn, M. M. (2014). Marital quality and
https://doi.org/https://doi.org/10.1037/a0031859
Ross, K. M., Rook, K., Winczewski, L., Collins, N., & Schetter, C. D. (2019). Close relationships
and health: The interactive effect of positive and negative aspects. Social and
Rosseel, Y. (2012). Lavaan: An r package for structural equation modeling. Journal of Statistical
Spanier, G. B. (1976). Measuring dyadic adjustment: New scales for assessing the quality of
marriage and similar dyads. Journal of Marriage and the Family, 38, 15.
https://doi.org/10.2307/350547
https://doi.org/https://doi.org/10.1037/0033-295X.93.2.119
Strachman, A., & Gable, S. L. (2006). Approach and avoidance relationship commitment.
9026-9
42
THE LOVEBIRD SCALE
Team, P. (2022). Rstudio: Integrated development environment for r. In Posit Software, PBC.
http://www.posit.co/
Vaughn, M., & Baier, M. E. M. (1999). Reliability and validity of the relationship assessment
https://doi.org/10.1080/019261899262023
Wagner, S. A., Mattson, R. E., Davila, J., Johnson, M. D., & Cameron, N. M. (2020). Touch me
just enough: The intersection of adult attachment, intimate touch, and marital
https://doi.org/10.1177%2F0265407520910791
Watkins, M. W. (2018). Exploratory factor analysis: A guide to best practice. Journal of Black
6811.2010.01303.x
https://doi.org/https://doi.org/10.1016/j.jesp.2013.04.001
Wood, N. D., Fife, S. T., Parnell, K. J., & Ross, D. B. (2022). Answering the ethical call of the
other: A test of the strong relationality model of relationship flourishing. Journal of Marital
Yoo, H., Bartle-Haring, S., Day, R. D., & Gangamma, R. (2014). Couple communication,
emotional and sexual intimacy, and relationship satisfaction. Journal of Sex & Marital
Zoppolat, G., Faure, R., Alonso-Ferres, M., & Righetti, F. (2022). Mixed and conflicted: The role
81-99. https://doi.org/10.1037/emo0001055
44
THE LOVEBIRD SCALE
Appendix
Instructions: You have been asked to complete this questionnaire because you are currently
involved in a committed long-term romantic relationship. Please answer the following questions
about your relationship with your partner using the scale below.
1 2 3 4 5 6 7
Strongly Disagree Disagree Neither Agree Agree Strongly
disagree somewhat agree nor somewhat agree
disagree
2. When I hear certain songs, I think of how much I love my partner. (r)
4. My partner never intentionally insults me, puts me down, or makes me feel bad. (m)
6. We stay together for external reasons such as marriage vows and children, more than
8. Sometimes when I’m alone I find myself thinking about how much I love my partner. (r)
10. My partner accepts every part of me, even the things I dislike about myself. (m)
11. The more time we spend together the more I enjoy my partner’s company. (r)
12. Although I love my partner, I would not say that I am currently “in love”. (d)
13. My partner and I respect each other’s opinions, even when we don’t agree with each
other. (m)
15. I feel like I need space after we spend a lot of time together. (d)
45
THE LOVEBIRD SCALE
17. I often find myself thinking about special things I can do to make my partner happy. (r)
18. I am more myself when I am alone than when I am with my partner. (d)
19. I support my partner in their goals and aspirations, and they do the same for me. (m)
21. There are things about my partner that I wish I could change. (d)
22. I can talk to my partner about anything, even if it is a difficult conversation. (m)
24. My partner and I have recurring problems that we can’t get past. (d)
25. I don’t have to sacrifice aspects of myself to keep my partner happy. (m)
Composite lovebird score: Reverse score items on the Disconnect subscale and then take the
Table 1
participants in Studies 1-3 were 39.9(13.2), 38.7(11.6), and 39.1(10.2) years old, respectively.
Average length of participants’ current relationship in Studies 1-3 was 11.4(10.8), 10.5(12.7),
Table 2
Table 3
subscale; LB-D = Lovebird Scale – Disconnect subscale. Factor loadings were estimated using
principal axis factoring with a Promax rotation. Factor loadings less the 0.20 were omitted from
Table 4
Variable 1 2 3 4 5 6 7
1. LB-M --
2. LB-R .72 --
3. LB-D -.59 -.56 --
4. RAS .76 .69 -.71 --
5. MPDQ-S .57 .57 -.69 .64 --
6. MPDQ-P .74 .60 -.65 .72 .69 --
7. LB-P .72 .68 -.57 .81 .52 .65 --
8. COM-P -.28 -.35 .48 -.30 -.31 -.32 -.26
Note. N = 540. All correlations were statistically significant at the p < .001 level. LB-M =
Lovebird Scale – Mutuality subscale; LB-R = Lovebird Scale – Romance subscale; LB-D =
Table 5
Table 6
because they were equivalent to those produced by the higher order model. EFA = exploratory
factor analysis; CFI = comparative fit index; SRMR = standardized root mean square residual;
Table 7
Variable 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
1. LB-M --
2. LB-R .77 --
3. LB-D -.71 -.69 --
4. LB .93 .88 -.89 --
5. RAS .83 .67 -.75 .84 --
6. MPDQ-S .69 .63 -.68 .75 .63 --
7. MPDQ-P .79 .64 -.69 .79 .73 .76 --
8. PRQC-RS .84 .72 -.71 .84 .89 .59 .72 --
9. PRQC-C .60 .62 -.58 .66 .62 .46 .45 .64 --
10. PRQC-I .74 .79 -.67 .80 .79 .60 .65 .82 .55 --
11. PRQC-T .74 .55 -.56 .69 .73 .49 .59 .74 .49 .59 --
12. PRQC-P .46 .67 -.45 .56 .51 .44 .47 .58 .35 .72 .36 --
13. PRQC-L .73 .81 -.66 .80 .69 .59 .59 .72 .75 .74 .57 .52 --
14. RQS .86 .79 -.78 .90 .87 .67 .74 .87 .69 .79 .72 .54 .80 --
15. LB-P .75 .73 -.68 .79 .77 .60 .62 .77 .56 .74 .54 .54 .75 .79 --
16. COM-P -.41 -.44 .60 -.54 -.46 -.44 -.45 -.45 -.27 -.45 -.32 -.42 -.40 -.49 -.40
Note. N = 215. All correlations were statistically significant at the p < .001 level. LB-M = Lovebird Scale – Mutuality subscale; LB-R =
Lovebird Scale – Romance subscale; LB-D = Lovebird Scale – Disconnect subscale; LB = Lovebird Scale – composite score; RAS =
Relationship Assessment Scale; MPDQ-S = Mutual Psychological Development Questionnaire – Self subscale; MPDQ-P = Mutual
Psychological Development Questionnaire – Partner subscale; PRQC-RS = Perceived Relationship Quality Components Inventory –
Relationship Satisfaction subscale; PRQC-C= Perceived Relationship Quality Components Inventory – Commitment subscale;
PRQC-I = Perceived Relationship Quality Components Inventory – Intimacy subscale; PRQC-T = Perceived Relationship Quality
Components Inventory – Trust subscale; PRQC-P= Perceived Relationship Quality Components Inventory – Passion subscale;
54
THE LOVEBIRD SCALE
PRQC-L = Perceived Relationship Quality Components Inventory – Love subscale; RQS = Relationship Quality Scale; LB-P =
Table 8
Table 9
Variable 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
1. LB-M --
2. LB-R .72*** --
3. LB-D -.70*** -.66*** --
4. LB .92*** .86*** -.89*** --
5. RAS .82*** .70*** -.78*** .86*** --
6. MPDQ-S .60*** .65*** -.61*** .69*** .58*** --
7. MPDQ-P .73*** .64*** -.70*** .78*** .67*** .74*** --
8. DAS-CN .47*** .41*** -.46*** .50*** .42*** .39*** .43*** --
9. DAS-AE .51*** .53*** -.46*** .56*** .48*** .46*** .53*** .73*** --
10. DAS-S .81*** .68*** -.79*** .86*** .89*** .64*** .73*** .47*** .51*** --
11. DAS-CH .61*** .59*** -.56*** .66*** .60*** .54*** .64*** .38*** .41*** .63*** --
12. DAS .70*** .63*** -.67*** .75*** .69*** .59*** .67*** .91*** .80*** .77*** .66*** --
13. AAS-ANX -.23** -.13* .34*** -.28*** -.28*** -.35*** -.35*** -.18** -.20** -.34*** -.31*** -.29*** --
14. AAS-AVD -.18** -.17** .17** -.20** -.18** -.34*** -.27*** -.16* -.18** -.22*** -.27*** -.24*** .46*** --
15. PANAS-P .20** .34*** -.18** .25*** .21*** .38*** .26*** .11 .18** .22*** .25*** .20** -.36*** -.44*** --
16. PANAS-N -.08 -.05 .21** -.14* -.13 -.25*** -.20** -.20** -.19** -.22*** -.14* -.23*** .45*** .46*** -.31*** --
17. LB-P .76*** .74*** -.71*** .79*** .84*** .48*** .61*** .41*** .47*** .81*** .51*** .64*** -.32*** -.21*** .23*** -.11 --
18. COM-P -.24*** -.35*** .42*** -.37*** -.32*** -.30*** -.28*** -.24*** -.24*** -.32*** -.28*** -.32*** .18** -.06 -.14* .05 -.32***
Note. N = 241. LB-M = Lovebird Scale – Mutuality subscale; LB-R = Lovebird Scale – Romance subscale; LB-D = Lovebird Scale –
Disconnect subscale; LB = Lovebird Scale – composite score; RAS = Relationship Assessment Scale; MPDQ-S = Mutual
Psychological Development Questionnaire – Self subscale; MPDQ-P = Mutual Psychological Development Questionnaire – Partner
subscale; DAS-CN = Dyadic Adjustment Scale – Consensus subscale; DAS-AE = Dyadic Adjustment Scale – Affectionate
Expression subscale; DAS-S = Dyadic Adjustment Scale – Satisfaction subscale; DAS-CH = Dyadic Adjustment Scale – Cohesion
subscale; DAS = Dyadic Adjustment Scale – Total; AAS-ANX = Adult Attachment Scale – Anxious subscale; AAS-AVD = Adult
Attachment Scale – Avoidant subscale; PANAS-P = Positive and Negative Affect Schedule – Positive subscale; PANAS-N = Positive
and Negative Affect Schedule – Negative subscale; LB-P = Lovebird relationship prototype; COM-P = Committed relationship
prototype.
57
THE LOVEBIRD SCALE
Table 10
Regressing PANAS Positive and Negative on Composite Lovebird Scores, RAS, and DAS-32
Variable β SE 95% CI p
PANAS Positive
Lovebird Scale – Composite score 2.861 1.043 [0.806, 4.917] .007
RAS -0.109 0.186 [-0.474, 0.257] .558
PANAS Negative
Lovebird Scale – Composite score -0.907 1.059 [-2.994, 1.179] .392
RAS -0.039 0.190 [-0.413, 0.334] .835
PANAS Positive
Lovebird Scale – Composite score 2.278 0.808 [0.686, 3.870] .005
DAS 0.003 0.034 [-0.064, 0.070] .937
PANAS Negative
Lovebird Scale – Composite score 0.189 0.780 [-1.349, 1.727] .809
DAS -0.092 0.033 [-0.156, -0.027] .006
Note. N = 241. Each set of models also controlled for sex (female), relationship status (married),
and age. RAS = Relationship Assessment Scale; DAS = Dyadic Adjustment Scale – Total;
Figure 1