misc.qmd

### Anthropometric models of mating success

In the anthropometric control model (@sfig-anthro), strength was significantly associated with higher numbers of lifetime and past year sex partners and an earlier age at first sex, and the interaction with sex was not significant.

::: {#sfig-anthro}
```{r}
#| fig-width: 12
#| fig-height: 8

plot_anthro <- forestplot(
  m_agefirst,
  m_lifetime,
  m_pastyear,
  m_partnered,
  intercept = F,
  facet = F,
  dodgewidth = .5,
  modelnames = mnames1,
  varnames = vnames, 
  size = 1.2, 
  linewidth = 1
) + theme_minimal(17) +
  guides(colour = guide_legend(reverse = T), shape = guide_legend(reverse = T))

ggsave("Figures/plot_anthro.pdf", plot_anthro, width = 14, height = 7)

plot_anthro
```

Anthropometric models of mating success (Pilot). Coefficient plot from generalized linear models. Variables labelled (S) have been centered at the mean and standardized by 2 SD.
:::



::: {#sfig-lifetime}
```{r}
#| fig-height: 10
forestplot2(m_lifetime, msoc1, mheal1, mphys1, mhor1, title = "Lifetime Number of Sexual Partners")
```

Models of lifetime number of sex partners.
:::


::: {#sfig-year}
```{r}
#| fig-height: 10
forestplot2(m_pastyear, msoc2,mheal2, mphys2, mhor2, title = "Past Year Number of Sexual Partners")
```

Models past year sex partners.
:::


::: {#sfig-agefirstsex}
```{r}
#| fig-height: 10
forestplot2(m_agefirst, msoc3, mheal3, mphys3, mhor3, title = "Age at First Sex")
```

Models of age at first sex
:::

::: {#sfig-partnered}
```{r}
#| fig-height: 10
forestplot2(m_partnered, msoc4, mheal4, mphys4, mhor4, title = "Currently Partnered")
```

Models of partnered status
:::


```{r}
forestplot2 <- function(..., title=""){
  models <- list(...)
  models <- map(models, \(m) {m$df.residual = Inf; return(m)})
  models <- c(models, list(
    intercept = F,
    facet = F,
    dodgewidth = .8,
    modelnames = mnames,
    varnames = vnames,
    size = 0.7,
    linewidth = 0.7
  ))

  p <- do.call(forestplot, models)

  p +
    guides(colour = guide_legend(reverse = T), shape = guide_legend(reverse = T)) +
    labs(title = title) +
    theme_minimal() +
    theme(axis.text.y = element_text(size = 9), plot.title = element_text(size = 11))
}
```

| Outcome            | Strength coefficient | Strength X Sex coefficient |
|--------------------|:--------------------:|:--------------------------:|
| Partnered status   |         0.99         |            0.90 (0.67)            |
| Lifetime partners  |         0.99         |            0.90 (0.66)            |
| Past year partners |         0.80         |            0.30 (0.21)            |

: Power to detect our predicted effects. Strength coefficients (effect sizes) were set to 3/4 of those we found in our pilot study, interaction coefficients were set to the same magnitude (but opposite sign) of the strength coefficients (value in parentheses is power to detect 3/4 of that effect), sample size was set to that of our unobserved data, and $\alpha = 0.05$. {#tbl-power}