“Where are We?”: Developing Behavioral Indicators of Relationship Development (BIRD) for Quantifying Relationship Progression

• Situationship: A romantic and sexual relationship with no clear label or commitment.

• Intermediate relationship: A relationship between a one-night stand and a long-term partnership, such as a situationship or “friends with benefits” (FWB).

• Friends with benefits (FWB): Repeated sex within a friendship, without romantic love.

• Act nomination procedure: Participants list concrete behaviors typical of a category (here, relationship stages), so the items come from people’s reports rather than from a theory.

• Principal components analysis (PCA): A statistical method that reduces many questionnaire items to a few underlying dimensions.

• Sternberg’s Triangular Love Scale (TLS-15): A questionnaire measuring love as intimacy, passion, and commitment. It relies on self-evaluation.

• Sociosexuality: How strongly a person prefers casual, uncommitted sex. It was measured with the SOI-R.

• Dark triad: Three socially undesirable traits: narcissism, Machiavellianism (manipulativeness), and psychopathy.

• Commitment device: The idea that love helps people stay committed by reducing interest in alternative partners.

• Convergent validity: Evidence that a new measure correlates with a related, established measure as theory predicts.

• Measurement invariance: Whether a scale works the same way across groups (e.g., men vs. women). It is tested at configural, metric, and scalar levels.

• Cross-sectional design: Data collected at one point in time, so it cannot show cause and effect.

1. Can relationship progression be captured through observable behaviors and characteristics? The first step was to identify the acts typical of different relationship forms, including intermediate ones, where little empirical data existed.

1. Can people in current or past relationships be reliably placed along a continuum from one-night stand to marriage using these acts? Do the acts reflect a developmental continuum, with intermediate relationships as part of it?

2. Is such a behavior-based measure associated with self-reported love (Triangular Love Scale), sociosexuality, dark triad traits, and self-denoted relationship type?

The authors expected a strong positive link with love, an increase across relationship types from one-night stand to long-term, and negative links with sociosexuality and dark triad traits.

Study design.

• Sample 1 (act nomination): 30 students in university cafeterias described observable behaviors for 10 relationship stages, from an extended one-night stand to marriage or cohabitation. This produced 327 acts. After the authors removed duplicates and misunderstandings, 75 items remained.

• Sample 2 (main study): Norwegian students aged 18-49 (222 men, 427 women, 9 “other”). The abstract reports 662 participants. 366 were in a current relationship and 198 reported a previous one. 92 had never had a relationship.

• Participants rated how well each behavior described their relationship (1-5). They also completed the SOI-R, the Dark Triad Dirty Dozen, and (if currently in a relationship) the TLS-15.

Scale development.

• PCAs on current and previous relationship items each showed one dominant component. Current: 35.9% of variance, with five components above eigenvalue 1. Previous: 40.6%, with three components above 1.

• Following the theory of a single continuum, the authors constrained analyses to one factor.

• Twelve items were among the top loaders in all four subgroups (current vs. previous; men vs. women), with loadings of 0.75-0.89. These became the 12-item BIRD. Examples include gives gifts, loves each other, refer to each other as “we,” are exclusive, keep belongings at each other’s place, public displays of affection, and make weekend plans together.

• Internal consistency was excellent (α = .97 overall; .95-.97 across subgroups).

Measurement invariance.

• Configural fit was mixed (RMSEA = 0.100, CFI = 0.939, SRMR = 0.087). Three items differed across groups: “talks about the future,” “meets family,” and “honest about personal topics.”

• A 9-item version fit slightly better, but dropping those items seemed to hurt content validity. Results were nearly identical for 9 and 12 items, so the authors report the 12-item version.

• Metric invariance was supported (χ²(24) = 35.44, p = .062). Scalar invariance could not be established because the model did not converge. The authors therefore limit their analyses to associations, not group-level comparisons.

Associations with other measures.

• Love (TLS-15): BIRD correlated very strongly with the total score (r = .90), commitment (.90), intimacy (.87), and passion (.82).

• Sociosexuality: BIRD correlated negatively with the SOI total (r = −.45) and with its desire (−.46), behavior (−.34), and attitudes (−.14) components.

• Dark triad: The correlation was small but significant (total r = −.13). Only narcissism showed a significant link (r = −.12).

Differences across relationship types. Welch’s ANOVAs were significant for both previous relationships (F(7, 21.4) = 122.84) and current ones (F(7, 15.9) = 180.14).

• Married/cohabiting and boyfriend/girlfriend relationships scored highest. In current relationships, the gap between these two was moderate (g = 0.49), while comparisons with the other types gave very large effects.

• Scores rose gradually from one-night stands and FWB, through situationships, to committed relationships.

• A permanent sexual partner with commitment and a situationship with intention to be together did not differ significantly. Effect sizes were small-to-moderate for previous relationships and moderate-to-large for current ones. Both scored above FWB and situationships without intention.

• Some groups were tiny (one-night stand: current n = 2, previous n = 3; FWB: current n = 11, previous n = 5). Effect sizes involving them were unstable and need great caution.

Interpretation.

• BIRD captures what people do (e.g., meeting family, gift-giving, exclusivity) rather than how they feel. It works as a single dimension across sex and across current and past relationships.

• The very high correlation with the love scale is read as convergent validity between related but distinct constructs, not redundancy. BIRD indexes the behavioral stage a relationship has reached, not its quality or felt commitment.

• A relationship may show behavioral progression while felt commitment stays ambiguous. Conversely, strong feelings can exist without much behavioral consolidation, as in an early long-distance relationship.

• Lower sociosexuality among people in long-term relationships supports validity. The small narcissism link fits earlier research on narcissism and long-term stability.

• Intermediate relationships appear to occupy an identifiable position along the same continuum as short- and long-term relationships. They are not a separate category with sharp boundaries. The findings are described as tentative support for this view.

• BIRD was less able to separate categories within the intermediate group.

Added value.

• Because BIRD is behavior-based, it may suit retrospective reports of past relationships better. Feelings after a breakup can be shaped by identity change and emotional distress. (The authors say this argument is theoretical and has not yet been tested.)

Limitations.

• The sample was a non-probability sample of young students, about two-thirds of them women.

• Some behaviors (meeting family, keeping belongings at each other’s homes) may reflect life stage, logistics, or culture rather than commitment. The meaning of “situationship” may also vary across cultures and generations.

• The study was cross-sectional, so causal direction is unknown. It is unclear whether behavioral consolidation strengthens commitment or committed couples enact more of these behaviors.

• Scalar invariance was not established, so group-level comparisons are invalid.

• Small subgroups limit the comparisons.

• Dyadic data were not collected. The authors predict that partners’ behavioral reports will agree more than their emotional ones. Until that is tested, claims that BIRD captures “where the relationship is” rather than “where one perceives it to be” should be treated cautiously.

Strengths. The items came from an empirical act-nomination procedure rather than theory alone. The sample was relatively large and analyzed by sex and by current vs. past relationships. The behavioral focus suits ambiguous relationships.

Recommendations.

• Test BIRD in age-diverse and culturally diverse samples.

• Use longitudinal or experimental designs.

• Collect dyadic data.

• Use BIRD as a complement to subjective measures of relationship quality when studying relationship progression, including intermediate relationships.

Keep reading on Amurdat

Study summary: duty sex, infidelity concerns and mate value

Review: empirical research on incels