Contribution
It extends SDT’s need-satisfaction findings from individual well-being to the quality of a specific relationship, with three designs: a pooled analysis of eight unpublished samples (Study 1), an actor–partner analysis of couples (Study 2), and an event-contingent diary of disagreements with a mediation model (Study 3). It is the relationships-motivation work, after La Guardia et al. 2000, that tests all three needs inside one relationship.
Key insight
Within a close relationship the three needs do not weigh alike: “relatedness consistently emerged as the strongest and, in many cases, as the only unique predictor” of relationship outcomes, while competence carried individual well-being (pp. 451, 440–441). And need fulfillment is said to work through the reasons a person stays: people whose needs are met “have more intrinsic reasons for being in the relationship, which, in turn, predicts better relationship well-being” (p. 450).
Assumptions
- Need fulfillment = the Basic Need Satisfaction in Relationships Scale (La Guardia et al. 2000): nine 1–7 items, three per need, e.g. “When I am with my partner, I feel free to be who I am” (p. 438). α: overall .86–.90, autonomy .59–.76, competence .67–.79, relatedness .75–.82 (Study 1).
- Correction for attenuation (Hunter and Schmidt) on every Study 1 estimate, then Fisher-z inverse-variance pooling across samples (p. 438).
- Relationship motivation = the Couple Motivation Questionnaire (Blais et al. 1990), 18 reasons weighted −3 to +3 along the autonomy continuum, α .76 (p. 447).
- All self-report, all correlational, all students (pp. 454, 437).
Key results
Study 1: eight samples, N = 1,918, 81% women, mean age 22.6 (pp. 437–441)
- Overall need fulfillment (Table 2, p. 440), corrected r: satisfaction .72 (k = 8), commitment .57, perceived conflict −.61, defensiveness −.51, understanding .34, avoidant attachment −.56, anxious −.50, self-esteem .42, positive affect .42, negative affect −.43, vitality .42.
- Among the subscales (p. 439): autonomy–relatedness .86, competence–relatedness .84 (corrected); autonomy–competence exceeded the scales’ reliabilities in several samples, so an uncorrected .68 is given; corrected it is .96 (95% CI .88 to 1.0) (p. 441).
- Unique contributions, partial r (Table 3, p. 440): relatedness with satisfaction .65, commitment .59, avoidance −.42; competence with self-esteem .39; autonomy with perceived conflict −.21, and near zero with positive affect (−.01) and vitality (−.00). Relatedness was a stronger predictor than autonomy and competence for every relationship outcome (Meng–Rosenthal–Rubin Z from 4.75 to 39.90, p. 441).
- Repeated with an autonomy–competence composite: “nearly identical” (p. 441). For the nearly 400 men alone, autonomy’s unique associations with negative affect, both attachment dimensions, satisfaction, commitment and defensiveness were not significant (footnote 4, p. 454).
Study 2: 66 couples (pp. 441–446)
- Within-couple correlations significant for overall fulfillment (.38), autonomy (.30), relatedness (.49), not competence (.06) (Table 4, p. 443).
- APIM, overall fulfillment (Table 5, p. 444): actor β with satisfaction .53, partner .31; actor with commitment .43, partner .05 (n.s.); actor with defensiveness −.47, partner −.14.
- Actor × partner relatedness: satisfaction t(62) = 3.19, pr = .37; conflict −2.22; defensiveness −2.12 (pp. 444–445). Held with actor and partner closeness (IOS) controlled (p. 445).
Study 3: 120 diarists, 84% women, 908 disagreements over 10 days (pp. 446–451)
- Overall fulfillment predicted post-disagreement satisfaction (β = .42) and commitment (β = .41); with the three needs entered together, relatedness was the only unique predictor (β = .39, .46) (p. 449).
- Mediation (Figure 4; Table 7, p. 451): fulfillment → relationship motivation β = .62; motivation → satisfaction β = .43 with fulfillment controlled; fulfillment’s β falls from .42 to .15 (Sobel z = 4.33) for satisfaction and from .41 to .02 (z = 6.26) for commitment. With the three needs, only relatedness predicts motivation (β = .62).
- Unchanged controlling for perceived resolution, prior discussion, number and length of disagreements, recording lapse and closeness (pp. 450–451).
Claims
| id | claim | strength | support |
|---|---|---|---|
| C1 | Perceived need fulfillment from a partner correlates with relationship quality, secure attachment and individual well-being | strong (as correlation) | Study 1, 8 samples, N = 1,918, corrected rs .34–.72 |
| C2 | Relatedness is the strongest unique predictor of relationship outcomes; competence of self-esteem | moderate | partial rs, Studies 1 and 3; the subscales intercorrelate .84–.96 corrected, so partials are fragile |
| C3 | One’s partner’s need fulfillment predicts one’s own satisfaction, conflict and defensiveness beyond one’s own | moderate | 66 couples, APIM, Table 5 |
| C4 | Autonomous relationship motivation mediates the link of need fulfillment to post-disagreement satisfaction and commitment | weak | cross-sectional Level-2 mediation, Baron–Kenny with Sobel; predictor and mediator measured at the same time |
| C5 | The three needs are distinct in this relationship measure | weak | contradicted by the paper’s own .96 autonomy–competence correlation (p. 441) |
| C6 | The associations are not item overlap between fulfillment and satisfaction | weak | argued from the range of correlations and item wording (p. 441); not tested |
Concepts
- Need fulfillment (in a relationship) — the degree to which one feels one’s needs are supported when with this partner; relationship-specific, not general (p. 438).
- Relationship motivation — one’s reasons for being in the relationship, scored as a relative-autonomy index (p. 447).
- Actor and partner effects — one’s own and one’s partner’s predictor scores predicting one’s own outcome (APIM, p. 442).
- Autonomy (SDT) — “a sense of volition, agency, and initiative”, not “independence or detachment from others” (p. 435).
Connections
It builds on La Guardia et al. 2000 (LIT-1979), whose need-satisfaction measure it uses and whose finding that relatedness best predicts attachment it extends (pp. 436–437), and on Knee et al. 2005 and Blais et al. 1990 for relationship motivation. It places SDT beside self-expansion, interdependence, communal–exchange and partner-responsiveness accounts (pp. 452–454), claiming that each lacks either autonomy and competence or a distinction of qualities of motivation.
Bearing on the record
- QUESTION-369 (what shows a need is a need). This study is of the kind the question says cannot settle it: needs are identified by their correlation with self-reported well-being and relationship quality. It also shows something the question did not ask: in this measure the three needs are hardly separable (autonomy–competence .96, autonomy–relatedness .86, corrected), so “three needs” is not evidenced here as three dimensions. A note on the question should say so.
- THEORY-052 (need frustration as transdiagnostic). No bearing on its promote_when: satisfaction, not frustration, is measured, and nothing precedes anything.
- THEORY-054. Study 3 reports “a simplex pattern” among the Couple Motivation Questionnaire’s subscales (p. 447) and then collapses them into one weighted index, the practice THEORY-054’s note says keeps integration unmeasured.
- RMT, the registration’s reason. The registration filed this as relationships motivation theory with La Guardia et al. The paper does not use that name; it is one of the studies RMT was later built on, and its mutuality finding (C3) is the “mutuality of autonomy support” Ryan et al. 2022 name as RMT’s concern. Ryan et al. report no RMT meta-analysis.
- QUESTION-131. No bearing: the paper has no notion of frustration among incompatible demands.
- No instruction for machine-learning practice.
Limitations
- Correlational, cross-sectional at the person level, self-report, and mostly women students (Studies 1 and 3: 81% and 84%) (p. 454).
- Corrected correlations reach the ceiling. Autonomy–competence .96 with a CI to 1.0 means the two subscales measure one thing in these data; the partial correlations dividing their variance are then not interpretable as unique effects of distinct needs.
- Content overlap. Relatedness items (“I feel loved and cared about”) and satisfaction items (“My relationship with my partner makes me happy”), and the motivation items (“Because I love the many fun and exciting times I share with my partner”), share content with the outcomes; the paper argues against overlap but does not test it.
- Mediation with the mediator and predictor measured together cannot show the order the model assumes.
- One-tailed Z tests in Study 3: Z = 1.80 reported as p < .05 (p. 449).
- Two confidence intervals in Table 3 do not bracket their estimates as printed (competence with perceived conflict −.12, “−.19 to .05”, and with defensiveness the same; relatedness with self-esteem .04, “−.20 to .09”), likely sign or typesetting errors.
- The communal–exchange paragraph inverts its definitions (p. 453): “The former [communal] carries the expectation of immediate repayment … The latter [exchange] concerns … mutual responsiveness”, the reverse of Mills and Clark, after which the paragraph uses the usual sense.
Open questions
- Does a need measure that separates autonomy from competence in close relationships exist, and do the unique effects survive it?
- Do changes in partner need support precede changes in relationship quality, in a longitudinal or experimental design?