The longitudinal influence of supportive messages on stress reactivity and general well-being for LGBTQ+ recipients of hate speech: Comparing the relative effects of verbal person-centered and autonomy support

ABSTRACT

LBGTQ+ individuals (N = 50) engaged in a 10-minute discussion with a close network member about an experience they had with hate speech that was targeted at their sexual orientation. The relative effects of two predominant social support theoretical frameworks, verbal person centeredness and autonomy support, were compared. Discussions were rated by trained raters for the presence of each type of support. Results support both theoretical frameworks as predicting decreases in stress across the discussion as well as increases in reported general well-being over one month. Theoretical implications and methodological comparison are discussed.

KEYWORDS:

• LGBTQ+
• social support
• verbal person centeredness
• autonomy support

Acknowledgments

The authors would like to thank Annie G. Bequette, Rachael Zeiger, and Justin Shan for their helpful contributions to study recruitment.

Correction Statement

This article has been corrected with minor changes. These changes do not impact the academic content of the article.

Notes

1 Due to technical errors, two of the videos recorded no audio, whereas two other videos had shortened audio. The videos with no audio were not included in the analyses, but the videos with shortened audio were retained and rated.

2 Rating half the video allows for a more wholistic view of a conversation and behaviors that occurred within the conversation. It also increases the capacity for raters to rate reliably, as dividing the video in half reduces the cognitive complexity from having to consider, in a gestalt fashion, how behavior has occurred across 10 min to, instead, 5 min. This approach also acknowledges that support can change during the course of a conversation, so whereas some of the conversation may have included high levels of support, other portions may have entailed less support. Coding in halves allowed such intricacies to be more clearly captured in ratings, while also reducing raters’ cognitive fatigue and mitigating issues with recall.

3 To address concerns that arise with the use of difference scores, the models were also estimated controlling for the baseline values of the dependent variables. Person-centered support [b = −0.15 SE(.05)] and autonomy support [b = −0.26 SE(.08)] were still both significantly associated at the p < .05 level with time 2 sAA levels when controlling for baseline sAA, hate speech severity, experimental location, and age of the participant. Findings were also unchanged for more effective autonomy support and general well-being at time 2 when controlling for time 1 well-being and the other covariates [b = 0.46 SE(.22), p<.05]. The relationship between person centered support and time 2 general well-being, however, did not achieve statistical significance [b = 0.22 SE(.14), p = .10].

Additional information

Funding

This work was supported by the Waterhouse Family Institute for the Study of Communication and Society [grant number 17180045].