Jealousy as a Function of Rival Characteristics: Gender differences in reactions to rival attractiveness and dominance

class: center, middle, inverse, title-slide

# Jealousy as a Function of Rival Characteristics: Gender differences in reactions to rival attractiveness and dominance
### Thomas V. Pollet (@tvpollet) & Tamsin K. Saxton, Northumbria University
### 2021-03-27 | <a href="http://tvpollet.github.io/disclaimer">disclaimer</a>

---

## Outline of talk today.

* Intro.

* Two replication studies (skimpy on the details) and meta-analyses (paper [here](https://tvpollet.github.io/pdfs/Pollet_Saxton_2020_PSPB.pdf)).

* (Quick and dirty) statistical power simulation on main and interaction effects.

* Summary + further research.

???
References at end of slides.

---
## Replication crisis.

* Replication crisis (e.g., Chambers, 2017; Ritchie, 2020)

* Replication should become a routine activity (e.g., Zwaan et al. 2018)

* Also great teaching tool... .

???
Especially when combined with interesting topic.

---
## Jealousy

* One of the core topics studied by evolutionary psychologists (e.g., Buss, 1992)

* Topic which features heavily in textbooks (e.g., Buss, 1999; Cartwright, 2001)

* Focus on sex differences,... .

* 'Classic finding': gender difference in responses to emotional versus sexual infidelity.

???
Due to asymmetries in cost, evolution has shaped responses.

---
## Dijkstra & Buunk (1998)

* Extension from previous work.

* Vignette-based scenario: imagining their partner being approached by a potential other-sex romantic rival.

* Men’s jealousy: particularly responsive to the dominance of the male rival.

* Women’s jealousy: particularly responsive to the attractiveness of the female rival.

* Sex difference predicted based on evolutionary theory regarding the relative importance of dominance and attractiveness to men’s and women’s appeal as a romantic partner.

---
## Dijkstra & Buunk (1998): A closer look.

* 152 students completed a vignette study. (Between-subject design).

* **Key outcome**: Jealousy (1-5 rating)

* Description: _"You are at a party with your girlfriend \[boyfriend], and you are talking with some of your friends. You notice your girlfriend \[boyfriend] across the room talking to a man \[woman] you do not know. You can see from his \[her] face that he \[she] is very interested in your girlfriend \[boyfriend]. He \[She] is listening closely to what she \[he] is saying, and you notice that he \[she] casually touches her \[his] hand. You notice that he \[she] is flirting with her \[him]. After a minute, your girlfriend \[boyfriend] also begins to act flirtatiously. You can tell from the way she \[he] is looking at him \[her] that she \[he] likes him \[her] a great deal. They are completely absorbed in each other.”_

* Manipulated attractiveness via photograph.

---
## Dijkstra & Buunk (1998): Dominance manipulation (our version).

* _“You find out that your girlfriend is flirting with Jonathan, the man in this photo. Jonathan is a student at \[Name of University where study was conducted] and is about the same age as you. Jonathan is also a teaching assistant and teaches courses to undergraduates. He is also president of a \[Name of University where study was conducted] activities club that numbers about 600 members. Jonathan knows what he wants and is a good judge of character. Jonathan also often takes the initiative to do something new, and he has a lot of influence on other people. At parties, he always livens things up.”_

* The low-dominance version read as follows: _“You find out that your girlfriend is flirting with Jonathan, the man in the photo. Jonathan is a student at \[Name of University where study was conducted] and is about the same age as you. Jonathan attends classes regularly and is one of the 600 members of an activities club at \[Name of University where study was conducted]. Jonathan does not always know what he wants, and he often fails to understand what is going on in other people’s minds. Jonathan often waits for others to take the initiative and is rather compliant. At parties, he usually stays in the background.”_

???
For (heterosexual) women the name and gender of the rival were altered (_"Olivia"_ rather than _"Jonathan"_ (original: 'Hans')).

---
## Dijkstra & Buunk results.

* A three-way ANOVA: a significant Gender\*Attractiveness\*Dominance interaction.

* **Key evidence**: two further significant interaction tests in ANOVA (Gender\*Attractiveness, Gender\*Dominance).

* Participant gender interacted with the attractiveness of the rival: Women responded with more jealousy to an attractive rival, as opposed to an unattractive one, compared to men (interaction: `$\eta^2_p$` = .033, based on our own calculations).

* The dominance of the rival affected men to a greater degree than it did women (interaction: `$\eta^2_p$` = .026, based on our own calculations).

---
## Graphical summary: Dijkstra & Buunk (1998 : 1163)

---
## Study 1.

* Matched facial stimuli closely on the original ratings. Images from DeBruine & Jones (2017).

* 339 self-identified heterosexual participants (target 380, 225 women) collected online + campus (2 Bachelor students).

* Majority current students (55%) and in a relationship (66%). The mean age was 22.48 years (_SD_ = 3.75 years, range = 18 - 57 years)

* Focus on same tests as Dijkstra & Buunk (1998). More on the [OSF](https://osf.io/zytdx/). Figures, manipulation checks, robustness tests, preregistration, etc.

???
Around 75% online / 25% campus

---
## Study 1: Results.

* 2 x 2 x 2 interaction (Gender\*Attractiveness\*Dominance) on ratings of jealousy was not statistically significant (*F*(1, 331) = 0.04, *p* = .849, `$\eta^2_g$` < .01).

* Evidence for the hypothesized Gender\*Attractiveness interaction effect, *F*(1, 331) = 6.55, *p* = .011, `$\eta^2_g$` = .02. For women, an attractive rival, as opposed to an unattractive rival, elicited jealousy to a greater degree than it did for men.

* No support for a Gender\*Dominance interaction on jealousy, *F*(1, 331) = 1.44, *p* = .231, `$\eta^2_g$` < .01.

---
## Study 2.

* Minor tweaks (Wording vignette + Faces from Radboud Faces Database, Langner et al., 2010 -- close as possible to differences in Dijkstra & Buunk 1998).

* Run online via Prolific (N=404) and online opportunity sampling (N=52). We merged both samples for analyses (N=456, 278 women).

* Majority were current students (81%) and in a relationship (61%), restricted to self-identified heterosexual participants. Mean age was 23.34 years (_SD_ = 4.10 years, range = 18 - 56 years).

Again, additional tests, manipulation checks, preregistration and more information on [OSF](https://osf.io/wd7zv/) or in [paper](https://tvpollet.github.io/pdfs/Pollet_Saxton_2020_PSPB.pdf).

---
## Study 2: Results

* In line with Study 1, but unlike Dijkstra and Buunk (1998), the 2 x 2 x 2 interaction was not statistically significant, *F*(1, 448) = 0.42, *p* = .518, `$\eta^2_g$` < .01.

* Contrary to both Study 1 and the original paper, there was no statistical evidence for the hypothesized Gender\*Attractiveness interaction on jealousy, *F*(1, 448) = 1.23, *p* = .268, `$\eta^2_g$` < .01.

* There was also no support for a Gender\*Dominance interaction, *F*(1, 448) = 0.15, *p* = .694, `$\eta^2_g$` < .01.

---
## Synthesis...

???
Incidentally combined Study 1 and Study 2 supported the interaction effect between gender and attractiveness.

---
## Meta-analysis strategy.

* Searched Web of Science and located 198 papers that used the term ‘jealousy’, plus either ‘partner’ or ‘rival’, plus either ‘trait’ or ‘characteristic’ or ‘attribute’ or ‘quality’ or ‘feature’ (and variants of those words, such as ‘traits’).

* Obtained additional 27 candidate papers via Google Scholar (cited Dijkstra & Buunk, 1998 or related).

* Final: 22 samples included.

5 exclusively heterosexual (but others could have done so implicitly) -- 0 focussed exclusively on non-heterosexual participants; 17 mean age was <26 years, 3 samples mean age > 26, 2 did not provide participant ages; 15 used student (or majority student) participants and 6 did not (1 unspecified); 4 samples were collected within the USA while the remainder were based outside the USA.

More information in [the paper](https://tvpollet.github.io/pdfs/Pollet_Saxton_2020_PSPB.pdf) or on [OSF](https://osf.io/wd7zv/).

---
## Meta-analysis: Attractiveness

---
## Meta-analysis: Attractiveness

???
_k_ = 22 samples encompassing 5,899 participants

---
## Meta-analysis: Dominance

---
## Meta-analysis: Dominance

???
(_k_ = 13 samples encompassing 4,038 participants)
---
## Statistical Power

Underappreciated, interaction effects require much larger sample size (e.g., Brookes et al., 2004).

Note this is an illustration for [a different project](https://psyarxiv.com/bvgxy/). (Continuous scenario, assumptions met)

Using 'Simglm' (Lebeau, 2020), set up a simulation:

let's assume `$\beta_{3}$` = 0.15 (_Pearson r_ = .2, Peterson & Brown 2005)

Choice of `$\beta_{1}$` and `$\beta_{2}$` = 0.387 (i.e., `$0.15 = 0.387 \times 0.387$`).

Equation:

`$$\hat y = \beta_1x_{1} +  \beta_2x_{2} + \beta_3x_{1}x_{2} + \epsilon$$`

---
## Power simulation: Main effect.

---
## Power simulation: Interactive effect

---
## Summary

* Glass Half Full?

* Interaction effects require ample statistical power

* More work needed? Study on traits of rivals across social situations.

---
## Any Questions?

Thanks to reviewers, corresponding authors for meta-analysis, students, and our participants.

[http://tvpollet.github.io](http://tvpollet.github.io)

Twitter: @tvpollet

---
## References (errors = blame RefManageR)

<cite>Brookes, S. T, E. Whitely, M. Egger, et al.
(2004).
&ldquo;Subgroup Analyses in Randomized Trials: Risks of Subgroup-Specific Analyses;: Power and Sample Size for the Interaction Test&rdquo;.
In: Journal of Clinical Epidemiology 57.3, pp. 229&ndash;236.
ISSN: 0895-4356.
DOI: <a href="https://doi.org/10.1016/j.jclinepi.2003.08.009">10.1016/j.jclinepi.2003.08.009</a>.</cite>

<cite>Buss, D. M.
(1999).
Evolutionary Psychology: The New Science of the Mind.
Needham Heights, MA: Allyn &amp; Bacon.
ISBN: 0-205-19358-7.</cite>

<cite>Buss, D. M.
(2000).
The Dangerous Passion: Why Jealousy Is as Essential as Love and Sex.
New York, NY: Simon &amp; Schuster.</cite>

<cite>Buss, D. M, R. J. Larsen, D. Westen, et al.
(1992).
&ldquo;Sex Differences in Jealousy: Evolution, Physiology, and Psychology&rdquo;.
In: Psychological Science 3.4, pp. 251&ndash;255.
ISSN: 0956-7976.
DOI: <a href="https://doi.org/10.1111/j.1467-9280.1992.tb00038.x">10.1111/j.1467-9280.1992.tb00038.x</a>.</cite>

<cite>Cartwright, J.
(2001).
Evolution and Human Behavior: Darwinian Perspectives on Human Nature.
New York, NY: Taylor &amp; Francis.
ISBN: 0-262-53170-4.</cite>

---
## More refs 1.

<cite>Chambers, C.
(2017).
The Seven Deadly Sins of Psychology A Manifesto for Reforming the Culture of Scientific Practice.
Princeton, NJ: Princeton University Press.</cite>

<cite>DeBruine, L. and B. Jones
(2017).
&ldquo;Face Research Lab London Set&rdquo;.

DOI: <a href="https://doi.org/10.6084/m9.figshare.5047666.v3">10.6084/m9.figshare.5047666.v3</a>.</cite>

<cite>Dijkstra, P. and B. P. Buunk
(1998).
&ldquo;Jealousy as a Function of Rival Characteristics: An Evolutionary Perspective&rdquo;.
In: Personality and Social Psychology Bulletin 24.11, pp. 1158&ndash;1166.
ISSN: 0146-1672.
DOI: <a href="https://doi.org/10.1177/01461672982411003">10.1177/01461672982411003</a>.</cite>

<cite>Langner, O, R. Dotsch, G. Bijlstra, et al.
(2010).
&ldquo;Presentation and Validation of the Radboud Faces Database&rdquo;.
In: Cognition &amp; Emotion 24.8, pp. 1377&ndash;1388.
ISSN: 0269-9931.
DOI: <a href="https://doi.org/10.1080/02699930903485076">10.1080/02699930903485076</a>.</cite>

<cite>Peterson, R. A. and S. P. Brown
(2005).
&ldquo;On the Use of Beta Coefficients in Meta-Analysis.&rdquo;
In: Journal of Applied Psychology 90.1, pp. 175&ndash;181.
ISSN: 1939-1854.
DOI: <a href="https://doi.org/10.1037/0021-9010.90.1.175">10.1037/0021-9010.90.1.175</a>.</cite>

---
## More refs 2.

<cite>Pollet, T. V. and T. K. Saxton
(2020).
&ldquo;Jealousy as a Function of Rival Characteristics: Two Large Replication Studies and Meta-Analyses Support Gender Differences in Reactions to Rival Attractiveness but Not Dominance&rdquo;.
In: Personality and Social Psychology Bulletin 46.10, pp. 1428&ndash;1443.
DOI: <a href="https://doi.org/10.1177/0146167220904512">10.1177/0146167220904512</a>.</cite>

<cite>Ritchie, S.
(2020).
Science Fictions: Exposing Fraud, Bias, Negligence and Hype in Science.
New York, NY: Random House.
ISBN: 1-4735-6425-5.</cite>

<cite>Zwaan, R. A, A. Etz, R. E. Lucas, et al.
(2018).
&ldquo;Making Replication Mainstream&rdquo;.
In: Behavioral and Brain Sciences 41, p. e120.
ISSN: 0140-525X.
DOI: <a href="https://doi.org/10.1017/S0140525X17001972">10.1017/S0140525X17001972</a>.</cite>