5 Refurbish

R新纪元： ME.qmd 存在的理由

Re-make!

不在之前的2-3000行基础上改

Battle! ¥¥¥¥¥

Jun 16

Table 1: Triple-H –> GaoXueYa

11:06

Statistical Analysis New Direction 统计分析新方向

他们不接受三高。解决： Triple-H –> each Metabolic disease

回归策略：

A. Response: only all 4 ME, not Triple-H

SanGao: strata. regression of ME to lifestyle variables by stratum

add plots

Loading Packages:

library(tidyverse)
library(readxl)
library(ggplot2)

library(moderndive)
library(autoReg)
library(easystats)
library(gtsummary)
library(GWalkR)
library(reshape2) # for heatmap: melt
library(rrtable) #table2doc

Dataset import:

#file.choose()

Raw <- read_excel("/Users/xue/Desktop/Exquisite/MEData/rawData.xlsx")

Survey coding specifications: under ME folder

5.1 Raw Data Exploration

#View(Raw)

原数据有很多_open.开放选项。n ot informative。暂时不考虑处理方式：select(!ends_with("open"))

#glimpse(Raw)

疾病变量 (GaoXueYa) 等在数据中是数值变量。应该是分类变量，

Tidy 问题

glimpse(), gaze() 发现，性别，年龄身高体重均误为字符。

清洗解决：在原excel改错误的数据格式。做法添加正确列： C2 = value(B2)

创造 Raw_clean: 已直接在excel 里加上了 Age, Height, Weight, 所以无需在R里tidy,直接select去掉年龄，身高，体重这些旧的错误的字符变量。另外，去掉开放问题

Raw_clean: New Dataset Created!

Raw_clean <- Raw |> 
  select(!ends_with("open")) |>
  select(!年龄)|>
  select(!身高cm)|>
  select(!体重kg)

再检查：

gaze(Raw_clean)

————————————————————————————————————————————————
      name             levels          stats    
————————————————————————————————————————————————
性别             Mean ± SD             1.7 ± 0.5 
Age              Mean ± SD            28.9 ± 9.1 
Height           Mean ± SD           167.1 ± 8.7 
Weight           Mean ± SD           71.5 ± 29.1 
Race             35 unique values                
Occupation       Mean ± SD             3.5 ± 2.5 
Living           Mean ± SD             2.5 ± 0.7 
Education        Mean ± SD             2.9 ± 0.5 
Monthly_Income   Mean ± SD             2.1 ± 0.9 
Exercise         Mean ± SD             2.2 ± 1.1 
Sleeping         Mean ± SD             5.1 ± 1.2 
Smoking          Mean ± SD             4.5 ± 0.8 
Drinking         Mean ± SD             3.8 ± 0.7 
ME_Attention     Mean ± SD            20.2 ± 3.3 
ME_SatietyCue    Mean ± SD            19.4 ± 3.5 
ME_Awareness     Mean ± SD            10.2 ± 3.5 
ME_Distraction   Mean ± SD            13.5 ± 3.9 
GaoXueYa         Mean ± SD             0.1 ± 0.2 
XueZhiYiChang    Mean ± SD             0.0 ± 0.2 
GaoXueTang       Mean ± SD             0.0 ± 0.2 
Cancer           Mean ± SD             0.0 ± 0.1 
ManXingFeiBu     Mean ± SD             0.0 ± 0.1 
GanZang          Mean ± SD             0.0 ± 0.1 
XinZangBing      Mean ± SD             0.0 ± 0.1 
Stroke           Mean ± SD             0.0 ± 0.1 
Kidney           Mean ± SD             0.0 ± 0.1 
WeiXiaoHuaDao    Mean ± SD             0.1 ± 0.2 
QingGanJingShen  Mean ± SD             0.0 ± 0.1 
JiYi             Mean ± SD             0.0 ± 0.1 
GuanJieYan       Mean ± SD             0.0 ± 0.1 
Asthma           Mean ± SD             0.0 ± 0.0 
YaJianKang       Mean ± SD             0.0 ± 0.2 
YaLi             Mean ± SD           19.3 ± 11.6 
JiaoLv           Mean ± SD           17.7 ± 11.8 
YiYu             Mean ± SD           17.7 ± 12.4 
————————————————————————————————————————————————

数据依然脏： 1.性别被错误视为了数值变量 2.各疾病,状态被视为了数值变量

1.性别 wrong type. Create Gender

Raw_clean: New Dataset Updated!

Raw_clean <-
  Raw_clean |> 
    mutate(Gender =  case_when (性别 == 1 ~ "Male",
                                性别 == 2 ~ "Female"),
           .after = 性别)
# .keep = "unused"

2.疾病变量 wrong type, 需变成factor: as.factor()

暂时先考虑 3个关心的新陈代谢综合症其它病也需要

Raw_clean: New Dataset Updated!

Raw_clean <-
  Raw_clean |> 
    mutate_at(c('XueZhiYiChang','GaoXueYa','GaoXueTang','Gender',
                'Education','Monthly_Income','Exercise', 
                'Sleeping', 'Smoking','Drinking'), 
              as.factor) |>
    select(!性别)

Re-examine

#gaze(Raw_clean)

Step: Dataset Cleaning Concludes

当然，其他condition/status如果需要考虑，也需要 mutate_at(c('cond1','cond2'),as.factor)

6 Exploratory Data Analysis

Explore: Missing Data

6.1 Missing Data: `missingVis`

对干净数据 Raw_clean 进行检查。我自己的包 xue 不用额外载入 xue:: 即可

xue::missingVis(Raw_clean)

无缺失。

6.2 Descriptive Analysis

Important

autoReg::gaze.

1. 不分组，全体描述性统计量

gaze(data = Raw_clean)

—————————————————————————————————————————————————
      name             levels          stats     
—————————————————————————————————————————————————
Gender           Female               878 (65.8%) 
                 Male                 457 (34.2%) 
Age              Mean ± SD             28.9 ± 9.1 
Height           Mean ± SD            167.1 ± 8.7 
Weight           Mean ± SD            71.5 ± 29.1 
Race             35 unique values                 
Occupation       Mean ± SD              3.5 ± 2.5 
Living           Mean ± SD              2.5 ± 0.7 
Education        1                      25 (1.9%) 
                 2                    159 (11.9%) 
                 3                   1049 (78.6%) 
                 4                     102 (7.6%) 
Monthly_Income   1                    411 (30.8%) 
                 2                    453 (33.9%) 
                 3                    403 (30.2%) 
                 4                      68 (5.1%) 
Exercise         1                    415 (31.1%) 
                 2                    413 (30.9%) 
                 3                    341 (25.5%) 
                 4                     126 (9.4%) 
                 5                      40 (3.0%) 
Sleeping         3                    170 (12.7%) 
                 4                    239 (17.9%) 
                 5                    429 (32.1%) 
                 6                    343 (25.7%) 
                 7                    154 (11.5%) 
Smoking          4                    917 (68.7%) 
                 5                    227 (17.0%) 
                 6                     129 (9.7%) 
                 7                      62 (4.6%) 
Drinking         3                    563 (42.2%) 
                 4                    553 (41.4%) 
                 5                    208 (15.6%) 
                 6                      11 (0.8%) 
ME_Attention     Mean ± SD             20.2 ± 3.3 
ME_SatietyCue    Mean ± SD             19.4 ± 3.5 
ME_Awareness     Mean ± SD             10.2 ± 3.5 
ME_Distraction   Mean ± SD             13.5 ± 3.9 
GaoXueYa         0                   1268 (95.0%) 
                 1                      67 (5.0%) 
XueZhiYiChang    0                   1273 (95.4%) 
                 1                      62 (4.6%) 
GaoXueTang       0                   1297 (97.2%) 
                 1                      38 (2.8%) 
Cancer           Mean ± SD              0.0 ± 0.1 
ManXingFeiBu     Mean ± SD              0.0 ± 0.1 
GanZang          Mean ± SD              0.0 ± 0.1 
XinZangBing      Mean ± SD              0.0 ± 0.1 
Stroke           Mean ± SD              0.0 ± 0.1 
Kidney           Mean ± SD              0.0 ± 0.1 
WeiXiaoHuaDao    Mean ± SD              0.1 ± 0.2 
QingGanJingShen  Mean ± SD              0.0 ± 0.1 
JiYi             Mean ± SD              0.0 ± 0.1 
GuanJieYan       Mean ± SD              0.0 ± 0.1 
Asthma           Mean ± SD              0.0 ± 0.0 
YaJianKang       Mean ± SD              0.0 ± 0.2 
YaLi             Mean ± SD            19.3 ± 11.6 
JiaoLv           Mean ± SD            17.7 ± 11.8 
YiYu             Mean ± SD            17.7 ± 12.4 
—————————————————————————————————————————————————

2. 分组，描述性 + 2组差异检验

Important

注意！不采用的结果！看解释

gaze(GaoXueYa~., data = Raw_clean |> select(!Race)) |> myft()

name	levels	0 (N=1268)	1 (N=67)	p
Gender	Female	849 (67%)	29 (43.3%)	<.001
	Male	419 (33%)	38 (56.7%)
Age	Mean ± SD	28.5 ± 8.8	37.1 ± 11.1	<.001
Height	Mean ± SD	167.0 ± 8.7	169.1 ± 7.9	.049
Weight	Mean ± SD	71.1 ± 29.1	78.8 ± 28.6	.036
Occupation	Mean ± SD	3.5 ± 2.5	3.9 ± 1.8	.051
Living	Mean ± SD	2.5 ± 0.7	2.6 ± 0.6	.136
Education	1	24 (1.9%)	1 (1.5%)	.048
	2	144 (11.4%)	15 (22.4%)
	3	1001 (78.9%)	48 (71.6%)
	4	99 (7.8%)	3 (4.5%)
Monthly_Income	1	403 (31.8%)	8 (11.9%)	.001
	2	423 (33.4%)	30 (44.8%)
	3	375 (29.6%)	28 (41.8%)
	4	67 (5.3%)	1 (1.5%)
Exercise	1	408 (32.2%)	7 (10.4%)	.003
	2	390 (30.8%)	23 (34.3%)
	3	315 (24.8%)	26 (38.8%)
	4	118 (9.3%)	8 (11.9%)
	5	37 (2.9%)	3 (4.5%)
Sleeping	3	163 (12.9%)	7 (10.4%)	.438
	4	228 (18%)	11 (16.4%)
	5	402 (31.7%)	27 (40.3%)
	6	325 (25.6%)	18 (26.9%)
	7	150 (11.8%)	4 (6%)
Smoking	4	893 (70.4%)	24 (35.8%)	<.001
	5	208 (16.4%)	19 (28.4%)
	6	112 (8.8%)	17 (25.4%)
	7	55 (4.3%)	7 (10.4%)
Drinking	3	550 (43.4%)	13 (19.4%)	<.001
	4	528 (41.6%)	25 (37.3%)
	5	180 (14.2%)	28 (41.8%)
	6	10 (0.8%)	1 (1.5%)
ME_Attention	Mean ± SD	20.2 ± 3.3	19.5 ± 3.4	.095
ME_SatietyCue	Mean ± SD	19.4 ± 3.5	19.5 ± 3.3	.861
ME_Awareness	Mean ± SD	10.2 ± 3.6	10.5 ± 3.4	.425
ME_Distraction	Mean ± SD	13.5 ± 3.9	14.0 ± 4.1	.348
XueZhiYiChang	0	1222 (96.4%)	51 (76.1%)	<.001
	1	46 (3.6%)	16 (23.9%)
GaoXueTang	0	1236 (97.5%)	61 (91%)	.007
	1	32 (2.5%)	6 (9%)
Cancer	Mean ± SD	0.0 ± 0.1	0.0 ± 0.1	.916
ManXingFeiBu	Mean ± SD	0.0 ± 0.1	0.0 ± 0.1	.882
GanZang	Mean ± SD	0.0 ± 0.1	0.0 ± 0.1	.821
XinZangBing	Mean ± SD	0.0 ± 0.1	0.0 ± 0.2	.121
Stroke	Mean ± SD	0.0 ± 0.1	0.0 ± 0.2	.208
Kidney	Mean ± SD	0.0 ± 0.1	0.1 ± 0.2	.072
WeiXiaoHuaDao	Mean ± SD	0.1 ± 0.2	0.0 ± 0.2	.330
QingGanJingShen	Mean ± SD	0.0 ± 0.1	0.0 ± 0.0	<.001
JiYi	Mean ± SD	0.0 ± 0.1	0.0 ± 0.2	.357
GuanJieYan	Mean ± SD	0.0 ± 0.1	0.0 ± 0.2	.419
Asthma	Mean ± SD	0.0 ± 0.0	0.0 ± 0.0	.157
YaJianKang	Mean ± SD	0.0 ± 0.2	0.0 ± 0.2	.633
YaLi	Mean ± SD	19.1 ± 11.6	22.9 ± 10.6	.009
JiaoLv	Mean ± SD	17.5 ± 11.7	21.5 ± 11.6	.007
YiYu	Mean ± SD	17.5 ± 12.4	21.9 ± 12.1	.005

暂时不考虑Race。处理方式: select(!Race))

数据需要 modify。看上表。

Compare 高血压0 vs. 高血压1，on 4 scales，只有 Attention 合理，其余结果都不好。其中 Satiety 2组相等不好很正常：题目难懂没有实际情境 Awareness, Distraction 更接近反向差异：

6.2.1 原因：调查问卷反向设计

因此：create Raw_clean_modify

策略：没有任何三高的人(Status = 0) & 答了低分，用18减

mutate SanGaoScore

Raw_clean <-
Raw_clean |>
  mutate(SanGaoScore = as.numeric(GaoXueYa) + as.numeric(GaoXueTang) + as.numeric(XueZhiYiChang) - 3, .after = ME_Distraction)

Raw_clean_modify: New Dataset Created!

Raw_clean_modify <-
  Raw_clean |>
    mutate(ME_New_Awareness = case_when(
      # 没三高 低分 变高
      # 没三高 高分 不变
      # 有三高 低分 不变
      # 有三高 高分 不变？或变低
      SanGaoScore == 0 & ME_Awareness <= 7 ~ 18 - ME_Awareness, 
      SanGaoScore == 0 & ME_Awareness > 7 ~ ME_Awareness,
      SanGaoScore > 0 & ME_Awareness <= 7 ~ ME_Awareness, 
      SanGaoScore > 0 & ME_Awareness > 7 ~ ME_Awareness
      ), 
      .after = ME_Awareness) |>
    
    mutate(ME_New_Distraction = case_when( 
      SanGaoScore == 0 & ME_Distraction <= 8 ~ 24 - ME_Distraction,
      SanGaoScore == 0 & ME_Distraction > 8 ~ ME_Distraction,
      SanGaoScore > 0 & ME_Distraction <= 8 ~ ME_Distraction,
      SanGaoScore > 0 & ME_Distraction > 8 ~ ME_Distraction
      ), 
      .after = ME_Distraction)

7 新描述性统计工作流：

7.0.1 工作流： Two-Group Comparison Workflow

有/无高血压对比

gaze(GaoXueYa~., data = Raw_clean_modify |> select(!Race)) |> myft()

name	levels	0 (N=1268)	1 (N=67)	p
Gender	Female	849 (67%)	29 (43.3%)	<.001
	Male	419 (33%)	38 (56.7%)
Age	Mean ± SD	28.5 ± 8.8	37.1 ± 11.1	<.001
Height	Mean ± SD	167.0 ± 8.7	169.1 ± 7.9	.049
Weight	Mean ± SD	71.1 ± 29.1	78.8 ± 28.6	.036
Occupation	Mean ± SD	3.5 ± 2.5	3.9 ± 1.8	.051
Living	Mean ± SD	2.5 ± 0.7	2.6 ± 0.6	.136
Education	1	24 (1.9%)	1 (1.5%)	.048
	2	144 (11.4%)	15 (22.4%)
	3	1001 (78.9%)	48 (71.6%)
	4	99 (7.8%)	3 (4.5%)
Monthly_Income	1	403 (31.8%)	8 (11.9%)	.001
	2	423 (33.4%)	30 (44.8%)
	3	375 (29.6%)	28 (41.8%)
	4	67 (5.3%)	1 (1.5%)
Exercise	1	408 (32.2%)	7 (10.4%)	.003
	2	390 (30.8%)	23 (34.3%)
	3	315 (24.8%)	26 (38.8%)
	4	118 (9.3%)	8 (11.9%)
	5	37 (2.9%)	3 (4.5%)
Sleeping	3	163 (12.9%)	7 (10.4%)	.438
	4	228 (18%)	11 (16.4%)
	5	402 (31.7%)	27 (40.3%)
	6	325 (25.6%)	18 (26.9%)
	7	150 (11.8%)	4 (6%)
Smoking	4	893 (70.4%)	24 (35.8%)	<.001
	5	208 (16.4%)	19 (28.4%)
	6	112 (8.8%)	17 (25.4%)
	7	55 (4.3%)	7 (10.4%)
Drinking	3	550 (43.4%)	13 (19.4%)	<.001
	4	528 (41.6%)	25 (37.3%)
	5	180 (14.2%)	28 (41.8%)
	6	10 (0.8%)	1 (1.5%)
ME_Attention	Mean ± SD	20.2 ± 3.3	19.5 ± 3.4	.095
ME_SatietyCue	Mean ± SD	19.4 ± 3.5	19.5 ± 3.3	.861
ME_Awareness	Mean ± SD	10.2 ± 3.6	10.5 ± 3.4	.425
ME_New_Awareness	Mean ± SD	12.1 ± 2.1	10.5 ± 3.4	<.001
ME_Distraction	Mean ± SD	13.5 ± 3.9	14.0 ± 4.1	.348
ME_New_Distraction	Mean ± SD	14.9 ± 3.0	14.0 ± 4.1	.068
SanGaoScore	Mean ± SD	0.1 ± 0.3	1.3 ± 0.6	<.001
XueZhiYiChang	0	1222 (96.4%)	51 (76.1%)	<.001
	1	46 (3.6%)	16 (23.9%)
GaoXueTang	0	1236 (97.5%)	61 (91%)	.007
	1	32 (2.5%)	6 (9%)
Cancer	Mean ± SD	0.0 ± 0.1	0.0 ± 0.1	.916
ManXingFeiBu	Mean ± SD	0.0 ± 0.1	0.0 ± 0.1	.882
GanZang	Mean ± SD	0.0 ± 0.1	0.0 ± 0.1	.821
XinZangBing	Mean ± SD	0.0 ± 0.1	0.0 ± 0.2	.121
Stroke	Mean ± SD	0.0 ± 0.1	0.0 ± 0.2	.208
Kidney	Mean ± SD	0.0 ± 0.1	0.1 ± 0.2	.072
WeiXiaoHuaDao	Mean ± SD	0.1 ± 0.2	0.0 ± 0.2	.330
QingGanJingShen	Mean ± SD	0.0 ± 0.1	0.0 ± 0.0	<.001
JiYi	Mean ± SD	0.0 ± 0.1	0.0 ± 0.2	.357
GuanJieYan	Mean ± SD	0.0 ± 0.1	0.0 ± 0.2	.419
Asthma	Mean ± SD	0.0 ± 0.0	0.0 ± 0.0	.157
YaJianKang	Mean ± SD	0.0 ± 0.2	0.0 ± 0.2	.633
YaLi	Mean ± SD	19.1 ± 11.6	22.9 ± 10.6	.009
JiaoLv	Mean ± SD	17.5 ± 11.7	21.5 ± 11.6	.007
YiYu	Mean ± SD	17.5 ± 12.4	21.9 ± 12.1	.005

压力抑郁焦虑均显著。

工作流：

gaze(GaoXueYa~., data = Raw_clean_modify |> select(!Race)) |> 
  myft() |>
  table2docx(vanilla = TRUE)

Exported table as Report.docx

表格modify: 0,1 –> 无高血压高血压，送通义千问

有/无血脂异常对比

gaze(XueZhiYiChang~., data = Raw_clean_modify |> select(!Race)) |> myft()

name	levels	0 (N=1273)	1 (N=62)	p
Gender	Female	847 (66.5%)	31 (50%)	.011
	Male	426 (33.5%)	31 (50%)
Age	Mean ± SD	28.6 ± 8.8	35.9 ± 12.8	<.001
Height	Mean ± SD	167.0 ± 8.6	169.0 ± 9.0	.075
Weight	Mean ± SD	71.3 ± 29.0	76.4 ± 29.6	.177
Occupation	Mean ± SD	3.5 ± 2.5	4.3 ± 2.4	.012
Living	Mean ± SD	2.5 ± 0.7	2.5 ± 0.7	.965
Education	1	16 (1.3%)	9 (14.5%)	<.001
	2	149 (11.7%)	10 (16.1%)
	3	1012 (79.5%)	37 (59.7%)
	4	96 (7.5%)	6 (9.7%)
Monthly_Income	1	404 (31.7%)	7 (11.3%)	.003
	2	431 (33.9%)	22 (35.5%)
	3	375 (29.5%)	28 (45.2%)
	4	63 (4.9%)	5 (8.1%)
Exercise	1	408 (32.1%)	7 (11.3%)	.006
	2	389 (30.6%)	24 (38.7%)
	3	323 (25.4%)	18 (29%)
	4	115 (9%)	11 (17.7%)
	5	38 (3%)	2 (3.2%)
Sleeping	3	164 (12.9%)	6 (9.7%)	.631
	4	226 (17.8%)	13 (21%)
	5	407 (32%)	22 (35.5%)
	6	326 (25.6%)	17 (27.4%)
	7	150 (11.8%)	4 (6.5%)
Smoking	4	889 (69.8%)	28 (45.2%)	<.001
	5	209 (16.4%)	18 (29%)
	6	120 (9.4%)	9 (14.5%)
	7	55 (4.3%)	7 (11.3%)
Drinking	3	547 (43%)	16 (25.8%)	.007
	4	526 (41.3%)	27 (43.5%)
	5	190 (14.9%)	18 (29%)
	6	10 (0.8%)	1 (1.6%)
ME_Attention	Mean ± SD	20.2 ± 3.2	18.9 ± 3.8	.002
ME_SatietyCue	Mean ± SD	19.4 ± 3.5	19.5 ± 3.5	.910
ME_Awareness	Mean ± SD	10.1 ± 3.6	11.2 ± 2.7	.004
ME_New_Awareness	Mean ± SD	12.1 ± 2.2	11.2 ± 2.7	.015
ME_Distraction	Mean ± SD	13.4 ± 3.9	15.3 ± 3.2	<.001
ME_New_Distraction	Mean ± SD	14.8 ± 3.0	15.3 ± 3.2	.225
SanGaoScore	Mean ± SD	0.1 ± 0.3	1.4 ± 0.6	<.001
GaoXueYa	0	1222 (96%)	46 (74.2%)	<.001
	1	51 (4%)	16 (25.8%)
GaoXueTang	0	1242 (97.6%)	55 (88.7%)	<.001
	1	31 (2.4%)	7 (11.3%)
Cancer	Mean ± SD	0.0 ± 0.1	0.0 ± 0.2	.190
ManXingFeiBu	Mean ± SD	0.0 ± 0.1	0.0 ± 0.2	.242
GanZang	Mean ± SD	0.0 ± 0.1	0.0 ± 0.2	.355
XinZangBing	Mean ± SD	0.0 ± 0.1	0.0 ± 0.2	.243
Stroke	Mean ± SD	0.0 ± 0.0	0.0 ± 0.2	.099
Kidney	Mean ± SD	0.0 ± 0.1	0.0 ± 0.1	.648
WeiXiaoHuaDao	Mean ± SD	0.0 ± 0.2	0.1 ± 0.3	.370
QingGanJingShen	Mean ± SD	0.0 ± 0.1	0.0 ± 0.1	.648
JiYi	Mean ± SD	0.0 ± 0.1	0.0 ± 0.0	<.001
GuanJieYan	Mean ± SD	0.0 ± 0.1	0.1 ± 0.2	.095
Asthma	Mean ± SD	0.0 ± 0.0	0.0 ± 0.0	.157
YaJianKang	Mean ± SD	0.0 ± 0.2	0.0 ± 0.2	.564
YaLi	Mean ± SD	19.1 ± 11.5	22.8 ± 12.0	.014
JiaoLv	Mean ± SD	17.5 ± 11.8	22.0 ± 11.2	.003
YiYu	Mean ± SD	17.5 ± 12.4	23.1 ± 12.6	<.001

工作流：

gaze(XueZhiYiChang~., data = Raw_clean_modify |> select(!Race)) |> 
  myft() |>
  table2docx(vanilla = TRUE)

Exported table as Report.docx

表格modify: 0,1 –> 无高血压高血压，送通义千问

8 回归(需要改)

GXY_lm <- lm(ME_Attention ~ Gender + Age + Education + Monthly_Income + Exercise + Smoking + Drinking, 
   data = Raw_clean_modify|>filter(GaoXueYa == 1) )

get_regression_table(GXY_lm)

# A tibble: 19 × 7
   term              estimate std_error statistic p_value lower_ci upper_ci
   <chr>                <dbl>     <dbl>     <dbl>   <dbl>    <dbl>    <dbl>
 1 intercept           17.5       4.56      3.84    0        8.33    26.7  
 2 Gender: Male         1.71      1.11      1.54    0.129   -0.518    3.94 
 3 Age                  0         0.045     0.009   0.993   -0.091    0.091
 4 Education: 2         0.001     3.93      0       1       -7.9      7.90 
 5 Education: 3         1.83      3.86      0.475   0.637   -5.92     9.59 
 6 Education: 4         1.89      4.71      0.402   0.69    -7.58    11.4  
 7 Monthly_Income: 2   -0.006     1.79     -0.003   0.997   -3.61     3.60 
 8 Monthly_Income: 3    1.42      1.88      0.756   0.453   -2.36     5.20 
 9 Monthly_Income: 4   -1.14      4.11     -0.278   0.782   -9.40     7.12 
10 Exercise: 2          0.583     1.80      0.324   0.747   -3.04     4.20 
11 Exercise: 3          1.47      1.72      0.856   0.396   -1.99     4.94 
12 Exercise: 4         -1.04      2.13     -0.489   0.627   -5.32     3.24 
13 Exercise: 5         -3.52      2.66     -1.33    0.191   -8.87     1.82 
14 Smoking: 5          -0.797     1.36     -0.585   0.561   -3.54     1.94 
15 Smoking: 6          -1.86      1.53     -1.21    0.231   -4.94     1.23 
16 Smoking: 7          -1.48      1.91     -0.774   0.443   -5.32     2.36 
17 Drinking: 4         -0.473     1.61     -0.294   0.77    -3.71     2.76 
18 Drinking: 5         -1.20      1.67     -0.723   0.473   -4.56     2.15 
19 Drinking: 6          5.87      4.14      1.42    0.163   -2.46    14.2

get_regression_summaries(GXY_lm)

# A tibble: 1 × 9
  r_squared adj_r_squared   mse  rmse sigma statistic p_value    df  nobs
      <dbl>         <dbl> <dbl> <dbl> <dbl>     <dbl>   <dbl> <dbl> <dbl>
1     0.271        -0.003  8.11  2.85  3.36     0.989   0.487    18    67

GXY_lm |> select_parameters() |> summary()


Call:
lm(formula = ME_Attention ~ 1, data = filter(Raw_clean_modify, 
    GaoXueYa == 1))

Residuals:
     Min       1Q   Median       3Q      Max 
-14.5075  -2.5075   0.4925   2.4925   5.4925 

Coefficients:
            Estimate Std. Error t value Pr(>|t|)    
(Intercept)  19.5075     0.4104   47.53   <2e-16 ***
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

Residual standard error: 3.359 on 66 degrees of freedom

9 EDA: Visualization

Variables of Interest

Categorical: Status 高血压，高血糖，血脂异常，性别 - GaoXueYa, GaoXueTang, XueZhiYiChang Gender,

Numeric: ME scales ME_Attention ME_SatietyCue ME_Awareness ME_Distraction

1cat 1num plot: 3*4 = 12 plots

2cat plot (gender + status):

Categorical Variables of Interest: 高血压，高血糖，血脂异常，性别

9.0.1 1 categorical and 1 numerical

9.0.1.1 Boxplot: Mapping a continuous variable to Y, a discrete/categorical variable to X.

GaoXueYa, ME_Attention score is raw, not modified

ggplot(data = Raw_clean_modify,
       mapping = aes(x = GaoXueYa, y = ME_Attention)
) + 
  geom_boxplot()

density: x = ME_Attention, color = GaoXueYa

ggplot(data = Raw_clean_modify,
       mapping = aes(x = ME_Attention, color = GaoXueYa)
) +
  geom_density(linewidth = 1)

ggplot(data = Raw_clean_modify,
       mapping = aes(x = ME_Attention, fill = GaoXueYa)
) +
  geom_density(linewidth = 1, alpha = 0.6)

ggplot(data = Raw_clean_modify,
       mapping = aes(x = ME_Attention, color = GaoXueTang)
) +
  geom_density(linewidth = 1)

Interpretation: 非高血糖患者（红线：0）的ME_ Attention 在高分分数区域（》20）的占比显著高于高血糖患者。

ggplot(data = Raw_clean_modify,
       mapping = aes(x = ME_Attention, color = XueZhiYiChang)
) +
  geom_density(linewidth = 1)

ME 2nd 指标 density: x = ME_New_Awareness, color = GaoXueYa

ggplot(data = Raw_clean_modify,
       mapping = aes(x = ME_New_Awareness, color = GaoXueYa)
) +
  geom_density(linewidth = 1)

ggplot(data = Raw_clean_modify,
       mapping = aes(x = ME_New_Awareness, fill = GaoXueYa)
) +
  geom_density(linewidth = 1, alpha = 0.6)

ggplot(data = Raw_clean_modify,
       mapping = aes(x = ME_New_Awareness, color = GaoXueTang)
) +
  geom_density(linewidth = 1)

Interpretation: 非高血糖患者（红线：0）的ME_ Attention 在高分分数区域（》20）的占比显著高于高血糖患者。

ggplot(data = Raw_clean_modify,
       mapping = aes(x = ME_New_Awareness, color = XueZhiYiChang)
) +
  geom_density(linewidth = 1)

9.0.2 2 categorical variables

高血压性别. 2 categorical variables

adjust Y-axis: position = “fill” to better observe the distribution/proportion of the categorical variables of interest.

ggplot(data = Raw_clean_modify,
       mapping = aes(x = GaoXueYa)
) +
  geom_bar(mapping = aes(fill = Gender),
           position = "fill")

Raw_clean_modify |>
  group_by(Gender) |>
  summarize(ME_Attention_mean = mean(ME_Attention))

# A tibble: 2 × 2
  Gender ME_Attention_mean
  <fct>              <dbl>
1 Female              20.2
2 Male                20.1

#gwalkr(Raw_clean_modify)

9.1 变量间关系探索

9.1.1 Heatmap

注意！

在数据清洁时，指标已经正确调成 factor 了，为了求协方差矩阵，data_lifestyle中指标回调成数值。 mutate_at(c('Monthly_Income','Exercise','Sleeping','Drinking','Smoking'),as.numeric)

9.2 A. 「全」Age + 生活方式变量 + ME得分

data_lifestyle_ME <- 
  Raw_clean_modify |>
  select(Age,Monthly_Income,Exercise,Drinking,Smoking,
         ME_Attention,ME_SatietyCue,ME_New_Awareness,
         ME_New_Distraction) |>
  mutate_at(c('Monthly_Income','Exercise','Drinking','Smoking',
              'ME_Attention','ME_SatietyCue','ME_New_Awareness',
              'ME_New_Distraction'), as.numeric)

# Calculate correlation matrix
cor_matrix <- cor(data_lifestyle_ME)

# Melt the correlation matrix for plotting
melted <- melt(cor_matrix)

# Plotting the correlation heatmap
ggplot(melted, aes(Var1, Var2, fill = value)) +
  geom_tile(color = "white") +
  scale_fill_gradient2(low = "blue", high = "red", mid = "white", midpoint = 0) +
  theme_minimal() +
  theme(axis.text.x = element_text(angle = 45, hjust = 1)) +
  labs(title = "Correlation Heatmap")

Regression 是否有必要对 GaoXueYa 作回归？还是假设检验即可？

9.3 线性回归！ ME scales 作为响应变量

对 ME_scale 作回归

9.3.1 「薛雨晴」生活方式分类变量太碎，分成2类？再看回归结果。因为本来问题也主观，分成这么细也不会帮助解释

model1 <- lm(ME_Attention ~ GaoXueYa + Age + Monthly_Income, data = Raw_clean_modify)

model1 |> get_regression_table()

# A tibble: 6 × 7
  term              estimate std_error statistic p_value lower_ci upper_ci
  <chr>                <dbl>     <dbl>     <dbl>   <dbl>    <dbl>    <dbl>
1 intercept           20.5       0.302    68.0     0       19.9     21.1  
2 GaoXueYa: 1         -0.574     0.421    -1.36    0.172   -1.4      0.251
3 Age                 -0.02      0.011    -1.78    0.076   -0.042    0.002
4 Monthly_Income: 2    0.158     0.245     0.645   0.519   -0.322    0.638
5 Monthly_Income: 3    0.531     0.252     2.11    0.035    0.037    1.02 
6 Monthly_Income: 4    0.612     0.442     1.38    0.167   -0.256    1.48

model1 |> get_regression_summaries()

# A tibble: 1 × 9
  r_squared adj_r_squared   mse  rmse sigma statistic p_value    df  nobs
      <dbl>         <dbl> <dbl> <dbl> <dbl>     <dbl>   <dbl> <dbl> <dbl>
1     0.007         0.004  10.7  3.27  3.28      1.98   0.079     5  1335

9.3.2 easyStats:: 模型选择

(model1_sel <-
model1 |>
  select_parameters() |> summary()
)


Call:
lm(formula = ME_Attention ~ GaoXueYa, data = Raw_clean_modify)

Residuals:
     Min       1Q   Median       3Q      Max 
-15.1956  -2.1956  -0.1956   2.8044   5.4925 

Coefficients:
            Estimate Std. Error t value Pr(>|t|)    
(Intercept) 20.19558    0.09217 219.104   <2e-16 ***
GaoXueYa1   -0.68812    0.41144  -1.672   0.0947 .  
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

Residual standard error: 3.282 on 1333 degrees of freedom
Multiple R-squared:  0.002094,  Adjusted R-squared:  0.001345 
F-statistic: 2.797 on 1 and 1333 DF,  p-value: 0.09467

原回归表现

#report(model1)
model_performance(model1)

# Indices of model performance

AIC      |     AICc |      BIC |    R2 | R2 (adj.) |  RMSE | Sigma
------------------------------------------------------------------
6966.806 | 6966.890 | 7003.183 | 0.007 |     0.004 | 3.271 | 3.278

9.3.3 Export Raw_clean_modify

library(readr)
#write_csv(Raw_clean_modify,"Raw_clean_modify.csv")

5.1 Raw Data Exploration

6 Exploratory Data Analysis

6.1 Missing Data: missingVis

6.2 Descriptive Analysis

6.2.1 原因： 调查问卷反向设计

7 新 描述性统计 工作流：

7.0.1 工作流： Two-Group Comparison Workflow

8 回归(需要改)

9 EDA: Visualization

9.0.1 1 categorical and 1 numerical

9.0.1.1 Boxplot: Mapping a continuous variable to Y, a discrete/categorical variable to X.

9.0.2 2 categorical variables

9.1 变量间关系探索

9.1.1 Heatmap

9.2 A. 「全」Age + 生活方式变量 + ME得分

9.3 线性回归！ ME scales 作为响应变量

9.3.1 「薛雨晴」 生活方式分类变量太碎，分成2类？再看回归结果。因为本来问题也主观，分成这么细也不会帮助解释

9.3.2 easyStats:: 模型选择

9.3.3 Export Raw_clean_modify

6.1 Missing Data: `missingVis`

6.2.1 原因：调查问卷反向设计

7 新描述性统计工作流：

9.3.1 「薛雨晴」生活方式分类变量太碎，分成2类？再看回归结果。因为本来问题也主观，分成这么细也不会帮助解释