Account-Making Following Relationship Dissolution: Exploring Sex as a Moderator in Public and Private Breakup Accounts

ABSTRACT

This study examines sex differences in emotion word use in retrospective private and public relationship dissolution accounts through two studies. Study 1 (n = 423) found men’s use of negative emotion words in private breakup accounts correlated with their use of negative emotion words in their public accounts; however, women’s emotion words in their private breakup accounts did not correlate with their use of emotion words in their public accounts. Study 2 (n = 284) replicated Study 1 with a broader sample and revealed the same pattern of negative emotion word use. These studies suggest that after a breakup women and men may process emotions differently in their public and private breakup accounts depending on their audience. The findings offer insights into how relationship dissolution model processes, evolutionary mating strategies, and self-presentation may influence linguistic variations accounting for sex differences.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1. This study examines sex differences; however, we acknowledge the clear conceptual difference between sex and gender and its implications on identity, expression, and communication. Nevertheless, the conceptual difference is a contemporary distinction. Prior scholarship often conflated and misused sex and gender (e.g., sex used as gender, gender used as sex, and interchangeably used them as synonymous). We understand that parsing out these differences leads to unique social constructivist and evolutionary perspectives. As such, we choose to use the terms women and men throughout our article for consistency and ease of readership.

2. Tausczik and Pennebaker (2010) conceptualized and operationalized emotionality or emotion categories for LIWC. Tausczik and Pennebaker (2010) described the following process of word categorization. For instance, to establish emotion word categorization, human judges first evaluate which words represent that category. Emotion words are subjective categories that then require further validation to glean from dictionaries, thesauruses, questionnaires, and lists established by Pennebaker’s research assistants. Finally, groups of judges independently rate whether each word candidate appropriately fulfills an overall word category.

3. Somewhat surprisingly, individuals expected that (on average) more than one or two people would read their private journals. In reviewing individual responses, it is clear that some participants expected that dozens or even hundreds of people might get ahold of and read their private journal. However, since there was still a significant difference between expected private and public audience size, the data still tap the constructs (private vs. public) needed to test the hypotheses. In addition, in the second study participants perceived larger audiences for both their public and private accounts than individuals in the first study. This is likely due to sampling differences: The first study recruited college students, and the second study used Mturk, which may have included users who had a strong online presence. Indeed, several participants in Study 2 commented that they would expect several thousand people to read their blog post or even private journal due to their large online followings.

4. Participants’ mean score on the breakup adjustment scale (i.e., “How difficult has it been for you to make an emotional adjustment to this breakup?”) created by Koenig Kellas, Bean, Cunningham, and Cheng (2008) was entered in the first step of the regression. However, adjustment was not a significant predictor of negative or positive emotion word use, so it was removed from final analyses for parsimony.

5. The triangulation offered a holistic analysis at two levels: word frequency with LIWC (Tables 3 and 4) and holistic account with thematic analysis (Appendix). By coding for overarching emotion at two levels in the accounts (both negative and positive), we captured the emotional tenor generating consistency and similar findings.

Additional information

Funding

This research was partially supported by an appointment to the Intelligence Community Postdoctoral Research Fellowship Program at The University of Texas at Austin, administered by Oak Ridge Institute for Science and Education through an interagency agreement between the U.S. Department of Energy and the Office of the Director of National Intelligence.