exp1_analysis_main.Rmd

---
title: 'Experiment 1: Main Analyses'
date: "`r Sys.Date()`"
output:
  github_document:
    toc: yes
    toc_depth: 3
  pdf_document:
    toc: yes
    toc_depth: 3
  html_document:
    toc: yes
    toc_depth: '3'
    df_print: paged
editor_options:
  markdown:
    wrap: 72
---

```{r setup, include=FALSE}
knitr::opts_chunk$set(echo = TRUE)
library(tidyverse)
options(dplyr.summarise.inform = FALSE)
library(magrittr)
library(lme4)
library(lmerTest)
library(broom.mixed)
library(insight)
library(kableExtra)
```

# Setup

Variable names:

-   Experiment: exp1\_
-   Data (\_d\_)
    -   d = main df
    -   count = sums of response types
    -   FF = First + Full Name conditions only
-   Models (\_m\_)
    -   cond = effect of Condition (Last vs First+Full)
    -   nameGender = effects of Condition (First vs Full) and Name
        Gender Rating
    -   FF = dummy coded with First + Full Name conditions as 0, Last
        Name condition as 1
    -   L = dummy coded with Last Name condition as 0, First + Full Name
        conditions as 1

Load data and select columns used in model. See data/exp1_data_about.txt
for more details.

```{r load-data}
exp1_d <- read.csv("../data/exp1_data.csv",
                   stringsAsFactors = TRUE) %>%
  rename("Participant" = "SubjID", "Item" = "NameShown") %>%
  select(
    Participant, SubjGenderMale, Condition, GenderRating,
    Item, He, She, Other
  )
str(exp1_d)
```

Center gender rating for names: Original scale from 1 to 7, with 1 as
most masculine and 7 as most feminine. Mean-centered with higher still
as more feminine.

```{r center-gender-rating}
exp1_d %<>% mutate(GenderRatingCentered = scale(GenderRating, scale = FALSE))
```

Set contrasts for name conditions.

```{r contrast-coding}
contrasts(exp1_d$Condition) <- cbind(
  "last vs first/full" = c(.33, .33, -0.66),
  "first vs full"      = c(-.5, .5, 0)
)
contrasts(exp1_d$Condition)
```

Subset for gender rating effects (First and Full conditions only).

```{r subset-FF}
exp1_d_FF <- exp1_d %>% filter(Condition != "last")
exp1_d_FF$Condition %<>% droplevels()

contrasts(exp1_d_FF$Condition) <- cbind(
  "first vs full" = c(-.5, .5)
) # add contrast back
contrasts(exp1_d_FF$Condition)
```

# Data Summary

Responses by condition.

```{r count-responses}
exp1_d %<>% mutate(ResponseAll = case_when(
  He    == 1 ~ "He",
  She   == 1 ~ "She",
  Other == 1 ~ "Other"
))

exp1_d_count <- exp1_d %>%
  group_by(Condition, ResponseAll) %>%
  summarise(n = n()) %>%
  pivot_wider(
    names_from = ResponseAll,
    values_from = n
  ) %>%
  mutate(
    She_HeOther = She / (He + Other),
    She_He = She / He
  ) %>%
  select(She, He, Other, She_HeOther, She_He)

kable(exp1_d_count, digits = 3, align = "c")
```

-   First name condition has second-most *she* responses
-   Full name condition has most *she* responses
-   Last name condition has fewest *she* responses

# Model 1: Condition

Effect of Condition (first name, last name, full name) on likelihood of
a *she* response, as opposed to a *he* or *other* response. Participant
and Item are included as random intercepts, with items defined as the
unique first, last and first + last name combinations. Because the
condition manipulations were fully between-subject and between-item,
fitting a random slope model was not possible.

```{r model-condition}
exp1_m_cond <- glmer(
  She ~ Condition + (1 | Participant) + (1 | Item),
  data = exp1_d, family = binomial
)
summary(exp1_m_cond)
```

-   Fewer *she* responses overall

-   First+Full have more *she* responses than Last. Full has more *she*
    responses than First (n.s. but matches ratios).

## Odds Ratios: Intercept

```{r OR-intercept}
exp(get_intercept(exp1_m_cond))
exp(-get_intercept(exp1_m_cond))
```

0.24x less likely to use to use *she* overall (or: 4.17x more likely to
use *he* or *other* overall), p\<.001

## Odds Ratios: Last vs First+Full

```{r OR-L-FF}
exp1_m_cond %>%
  tidy() %>%
  filter(term == "Conditionlast vs first/full") %>%
  pull(estimate) %>%
  exp()
```

16.85x more likely to use *she* in First + Full compared to Last (or:
16.85 times more likely to use *he* and *other* in Last than in First +
Full), p\<.001

## Odds Ratios: Last Only

Dummy code with Last Name as 0, so that intercept is the Last Name
condition only.

```{r dummy-code-L}
exp1_d %<>% mutate(Condition_Last = case_when(
  Condition == "first" ~ 1,
  Condition == "full"  ~ 1,
  Condition == "last"  ~ 0
))
exp1_d$Condition_Last %<>% as.factor()
```

```{r model-L}
exp1_m_L <- glmer(
  She ~ Condition_Last + (1 | Participant) + (1 | Item),
  data = exp1_d, family = binomial
)
summary(exp1_m_L)
```

```{r OR-L}
exp(get_intercept(exp1_m_L))
exp(-get_intercept(exp1_m_L))
```

0.04x times less likely to use *she* in the Last Name condition (or:
26.91x more likely to use *he* and *other* in the Last Name condition),
p\<.001

## Odds Ratios: First and Full Only

Dummy code with First and Full Name as 0, so that intercept is average
for these two conditions.

```{r dummy-code-FF}
exp1_d %<>% mutate(Condition_FF = case_when(
  Condition == "first" ~ 0,
  Condition == "full"  ~ 0,
  Condition == "last"  ~ 1
))
exp1_d$Condition_FF %<>% as.factor()
```

```{r model-FF}
exp1_m_FF <- glmer(
  She ~ Condition_FF + (1 | Participant) + (1 | Item),
  data = exp1_d, family = binomial
)
summary(exp1_m_FF)
```

```{r OR-FF}
exp(get_intercept(exp1_m_FF))
exp(-get_intercept(exp1_m_FF))
```

0.70x times less likely to use *she* in the First and Full Name
conditions (or: 1.42x more likely to use *he* and *other* in the First
and Full Name conditions), p=.26

# Model 2: Condition \* Name Gender

Effects of Condition (first name, full name) and the first name's Gender
Rating (centered, positive=more feminine) on the likelihood of a *she*
response, as opposed to a *he* or *other* response. In Experiment 1, the
Last Name condition does not include any instances of the gendered first
name, so only the First and Full Name conditions are analyzed here.
Participant and Item are again included as random intercepts.

```{r model-gender-rating}
exp1_m_nameGender <- glmer(
  She ~ Condition * GenderRatingCentered + (1 | Participant) + (1 | Item),
  exp1_d_FF,
  family = binomial
)
summary(exp1_m_nameGender)
```

-   More *she* responses as first names become more feminine.

-   Difference between First and Full is now significant (as compared to
    condition-only model).