Date of Award


Document Type


Degree Name

Doctor of Philosophy (PhD)



First Advisor

Carol T. Miller


Although romantic relationships are an important source of self-esteem, individuals vary in the degree to which romantic relationships determine their self-esteem. For individuals with relationship-contingent self-esteem (RCSE), self-esteem is based on the quality of perceived romantic relationship functioning. In contrast, global self-esteem is derived from a variety of domains, not specifically relationship quality. The present study investigated the moderating effects of RCSE and global self-esteem on the effects of relationship specific or relationship-unrelated threats and self-affirmations. Individuals with low global self-esteem react to threats by distancing themselves from their romantic partners. For those low in RCSE, this should occur only when the threats are relationship specific, whereas those low in global self-esteem distance themselves regardless of what type of threat they experience. Exposure to self-affirmations limits this defensive distancing in individuals with low global self-esteem. Prior studies examining the effects of induced self-threats and self-affirmations on perceived relationship functioning have been limited by reliance on college student samples, whose relationships are often shorter in duration than older adults. The current studies examined self-esteem and RCSE within the context of older participants in longer romantic relationships.

Specifically, Study 1 examined how self-threats and self-affirmations interacted with dispositional levels of RCSE and self-esteem to predict romantic relationship outcomes. Participants wrote about past experiences to induce self-affirmations or self-threats, which were either relational (i.e., about their current relationship) or non-relational (i.e., about their personal lives), and then completed measures of relationship functioning. Contrary to predictions, there were no significant interactions between either RCSE or self-esteem and the experimental conditions. There was a significant main effect for self-esteem, such that participants with low self-esteem reported less commitment, closeness, and satisfaction as compared to participants with high self-esteem. In Study 2, participants completed one of two writing prompts: a prompt to induce high RCSE, or a control prompt. Participants then wrote about past experiences to induce self-threats that were either relational or non-relational. Contrary to hypotheses, inducing high levels of RCSE and exposing participants to a relational self-threat did not result in decreased relationship commitment, closeness, or satisfaction. The results of both Study 1 and Study 2 are contrary to previous research; potential explanations for this discrepancy and implications are discussed.



Number of Pages

73 p.