practice.Rmd

---
title: "Practice"
author: "Fangru"
date: "2025-04-16"
output:xaringan::moon_reader:
    css: [default, css/Custumed_Style.css,css/zh-CN.css]
    lib_dir: libs
    nature:
      highlightLines: true
      highlightStyle: github
      countIncrementalSlides: false
      seal: true
      ratio: 16:9
params:
  output_dir: "../output"
  
---
# <h1 lang="en">7.2 函数</h1>
<span style="font-size: 30px;">综合练习</span></center><br>

- 定义一个函数用于计算 $d'$ 
- 计算不同Shape情况下(immoralself，moralself，immoralother，moralother)<br> 基于信号检测论(match为信号，mismatch噪音)的 $d'$ 值
<br>

- 1 导入所需要镜像和R包 
```{r 7.1 R package}
# 可以将清华的镜像设置为下载的镜像
# options(repos = c(CRAN = "https://mirrors.tuna.tsinghua.edu.cn/CRAN/"))

# 导入所需R包
library(tidyverse)
```

---
- 2 读取所需要录入的数据
```{r 7.2 load data}
# 此前已经将合并后的实验数据存入"match_raw.csv"，此次可直接读取
df7 <-  bruceR::import(here::here("data", "match","match_raw.csv"))
# 也可以用原始方式导入
# penguin_data <- read.csv('./data/match/match_raw.csv')  # 仅用基础包的做法

```

- 3 选择我们需要的变量
```{r 7.3 select variables,message=FALSE}
# 选择我们需要的变量
df7 <- dplyr::select(df7,
                     Sub, Age, Sex, Hand, #人口统计学
                     Block, Bin, Trial,   # 试次
                     Shape, Label, Match, # 刺激
                     Resp, ACC, RT)       # 反应结果
```

---
- 4 分组并计算 $d'$ 值
```{r 7.4 count Dprime}
#按照
df7 <- dplyr::group_by(df7, Sub, Shape,)
df7
df7 <- dplyr::summarise(df7,
      hit = length(ACC[Match == "match" & ACC == 1]), 
      fa = length(ACC[Match == "mismatch" & ACC == 0]),
      miss = length(ACC[Match == "match" & ACC == 0]),
      cr = length(ACC[Match == "mismatch" & ACC == 1]),
      Dprime = qnorm(
        ifelse(hit / (hit + miss) < 1,
               hit / (hit + miss),
               1 - 1 / (2 * (hit + miss))
              )
        ) 
             - qnorm(
        ifelse(fa / (fa + cr) > 0,
              fa / (fa + cr),
              1 / (2 * (fa + cr))
              )
                    )
      )
df7 <- dplyr::ungroup(df7)
print(df7) #可以逐步打印结果以便检查
```

---
- 5 删除额外变量
```{r 7.5 delete variables}
df7 <- df7[, -which(names(df7) == "hit"|names(df7) == "fa"|names(df7) == "miss"|names(df7) == "cr")]
print(df7)
```

---
- 6 长转宽，将形4个形状为列标题，以显示不同被试在不同形状条件下的$d'$ 值
```{r 7.6}
df7 <- tidyr::pivot_wider(df7, names_from = "Shape", values_from = "Dprime")
print(df7)
```

---
- 7 合并Tidyverse操作
```{r example of total rawdata_matchtask, message=FALSE}
# 用管道操作符合并以上代码
df7_1 <- bruceR::import(here::here("data", "match","match_raw.csv")) %>%
  dplyr::select(Sub, Age, Sex, Hand, #人口统计学
                Block, Bin, Trial,   # 试次
                Shape, Label, Match, # 刺激
                Resp, ACC, RT) %>%  # 反应结果
  tidyr::drop_na() %>%               #删除缺失值
  dplyr::group_by(Sub, Shape,) %>%
  dplyr::summarise(
      hit = length(ACC[Match == "match" & ACC == 1]), 
      fa = length(ACC[Match == "mismatch" & ACC == 0]),
      miss = length(ACC[Match == "match" & ACC == 0]),
      cr = length(ACC[Match == "mismatch" & ACC == 1]),
      Dprime = qnorm(
        ifelse(hit / (hit + miss) < 1,
               hit / (hit + miss),
               1 - 1 / (2 * (hit + miss))
              )
        ) 
             - qnorm(
        ifelse(fa / (fa + cr) > 0,
              fa / (fa + cr),
              1 / (2 * (fa + cr))
              )
                    )
      ) %>%
  dplyr::ungroup() %>%
  dplyr::mutate(hit = NULL, fa = NULL, miss = NULL, cr = NULL) %>%
  tidyr::pivot_wider(names_from = "Shape", values_from = "Dprime")
```

---
layout: false
#练习
1. 读取match数据，对不同shape的击中率进行分组绘图，可使用boxplot观察差异。
- (1) 按照7的过程，处理好数据，使shape为一列，hit为一列
```{r 8.1}
# 用管道操作符合并以上代码
df8_1 <- bruceR::import(here::here("data", "match","match_raw.csv")) %>%
  dplyr::select(Sub, Age, Sex, Hand, #人口统计学
                Block, Bin, Trial,   # 试次
                Shape, Label, Match, # 刺激
                Resp, ACC, RT) %>%  # 反应结果
  tidyr::drop_na() %>%               #删除缺失值
  dplyr::group_by(Sub, Shape,) %>%
  dplyr::summarise(hit = length(ACC[Match == "match" & ACC == 1])) %>%
  dplyr::ungroup() %>%

print(df8_1)
```

---
- (2) 作为x和y绘制箱型图
```{r 8.2 pic_box}
pic_box_1 <- df8_1 %>% 
      ggplot(aes(x = Shape, y = hit)) +
      geom_boxplot(staplewidth = 1) +
      # 绘制箱线图并添加上下边缘线 
      theme_classic()
```

---
2. 在上一题中，如何按照特定的顺序来展现 boxplot，比如按照moralSelf - immoralSelf - moralOther - immoralOther(提示：设置因子)
```{r 8.2 pic_box}
df8_2 <- bruceR::import(here::here("data", "match", "match_raw.csv")) %>%
  dplyr::select(Sub, Age, Sex, Hand,      
                Block, Bin, Trial,        
                Shape, Label, Match,    
                Resp, ACC, RT) %>%      
  tidyr::drop_na() %>%                  
  dplyr::group_by(Sub, Shape) %>%
  dplyr::summarise(hit = length(ACC[Match == "match" & ACC == 1])) %>%
  dplyr::ungroup() %>%
  dplyr::mutate(Shape = factor(Shape, 
                               levels = c("moralSelf", "immoralSelf", "moralOther", "immoralOther")))
                #利用dplyr中的mutate函数，将Shape转化为因子，并利用levels来进行自定义排序

pic_box_2 <- df8_2 %>% 
  ggplot(aes(x = Shape, y = hit)) +
  geom_boxplot(staplewidth = 1) +
  theme_classic()

```