R语言回归、anova方差分析、相关性分析 《精品购物指南》调研数据可视化|附代码数据

##变量赋值
colnames(data)=c(
  "编号",
  "性别",
  "年龄",
  "婚姻状况",
  "受教育程度",
  "职业或身份",
  "个人月平均收入",
  "报摊购买",
  "个人订阅",
  "单位订阅",
  "赠阅",
  "借阅他人",
  "其他",
  "上班途中",
  "下班途中",
  "午休时间",
  "逛衔购物时",
  "不定时",
  "对《精品购物指南》的零购情况",
  "家人是否会提醒购买《精品》",
  "是第几选择",
  "家人是否阅读《精品》",
  "",
  "",
  "一直购买(或订阅)本报(习惯)",
  "豪华版内容吸引人",
  "普通版文章吸引人",
  "查找广告信息",
  "参加读者乐园版的活动",
  "逛商场购物需要",
  "价格便宜",
  "没有什么原因想起来就买",
  "其他",
  "住所",
  "工作场所",
  "车站或乘车路途",
  "娱乐场所",
  "其他场所",
  "多少人阅读同一份报纸",
  "阅读《精品》的时间",
  "每天读报时间",
  "对我消费有指导",
  "提高我的生活品位和档次",
  "信息实用性强",
  "关注社会新闻",
  "信息量大",
  "信息质量高",
  "报道领域全面",
  "内容贴近市场",
  "寻找有用的信息和广告",
  "放松自我",
  "通俗易懂",
  "广告信息丰富",
  "版面编排合理",
  "售价合理",
  "订阅或零购方便",
  "其他",
  "平均每月阅读几期《精品》",
  "首先欣赏豪华版",
  "先看标题，再拣有意思的看",
  "我只看自己喜欢的固定栏目",
  "看完喜欢的文章，再浏览其他内容",
  "只查找对自己有用的信息",
  "无目的地翻阅报纸",
  "如果未能看到某一期《精品》",
  "能够阅读完《精品》内容",
  "封面要闻",
  "百姓生活新闻",
  "北京都市新闻",
  "消费新闻",
  "读者乐园",
  "体坛新闻",
  "关注足球",
  "篮球时空",
  "绿茵评说",
  "精品回顾",
  "文化资讯",
  "影视介绍",
  "读书生活",
  "音乐欣赏",
  "外企专递",
  "留学必备",
  "充电课堂",
  "人才聚焦",
  "处世情感",
  "健康",
  "美食",
  "休闲旅游",
  "保险",
  "律师",
  "百货新品",
  "休闲宠物",
  "户外休闲",
  "美食",
  "家庭保健",
  "超市SHOPPER"
  ,"新品试验"
  ,"样品透视"
  ,"汽车时代"
  ,"电脑"
  ,"市场行情"
  ,"通讯网络"
  ,"楼市了望"
  ,"房产金融"
  ,"政策扫描"
  ,"家具世界"
  ,"家装热点"
  ,"选材指南"
  ,"饰品快递"
  ,"非常男人"
  ,"特别女人"
  ,"新新人类"
  ,"网络家庭"
  ,"扮美家居"
  ,"今日妈咪"
  ,"选题大众化"
  ,"信息实用性强"
  ,"可读性强"
  ,"帮助我消费选择"
  ,"趣味性强"
  ,"报道领域全面"
  ,"帮助我了解市场行情"
  ,"放松自我"
  ,"文字优美"
  ,"观点新颖"
  ,"版式活泼"
  ,"品位高雅"
  ,"风格突出"
  ,"广告内容丰富"
  ,"其他"
  ,"招商展览"
  ,"电脑"
  ,"通讯"
  ,"汽车"
  ,"房地产"
  ,"家电"
  ,"食品酒类"
  ,"旅游娱乐"
  ,"商场饭店"
  ,"服装服饰"
  ,"美容用品"
  ,"保健品"
  ,"医疗器械"
  ,"家居用品"
  ,"航空订票"
  ,"招生"
  ,"人才招聘"
  ,"公益广告"
  ,"文化用品"
  ,"房屋祖赁"
  ,"金融证券"
  ,"发行广告"
  ,"其他"
  ,"广告对于消费是否有帮助"
  ,"广告数量适中"
  ,"广告信息丰富"
  ,"广告信息及时"
  ,"广告设计新颖"
  ,

data=complete.cases(data)

  geom_point() + 
    geom_smooth(method=method, ...)

for(i in 1:ncol(data))datacor[,i]=as.numeric(data[,i])

#数据归一化

data=scale(datacor)

cor.test(datacor$"性别",  
         datacor$"阅读《精品》的时间")
##  
##  Pearson's product-moment correlation  
##  
## data:  datacor$性别 and datacor$"阅读《精品》的时间"  
## t = 0.63616, df = 1995, p-value = 0.5247  
## alternative hypothesis: true correlation is not equal to 0  
## 95 percent confidence interval:  
##  -0.02964101  0.05806894  
## sample estimates:  
##        cor  
## 0.01424136

cor.test(datacor$"婚姻状况",  
         datacor$"阅读《精品》的时间")
##  
##  Pearson's product-moment correlation  
##  
## data:  datacor$婚姻状况 and datacor$"阅读《精品》的时间"  
## t = -1.7215, df = 1995, p-value = 0.08531  
## alternative hypothesis: true correlation is not equal to 0  
## 95 percent confidence interval:  
##  -0.082238962  0.005358859  
## sample estimates:  
##         cor  
## -0.03851404

cor.test(datacor$"受教育程度",  
         datacor$"阅读《精品》的时间")

##  
##  Pearson's product-moment correlation  
##  
## data:  datacor$受教育程度 and datacor$"阅读《精品》的时间"  
## t = -0.71111, df = 1995, p-value = 0.4771  
## alternative hypothesis: true correlation is not equal to 0  
## 95 percent confidence interval:  
##  -0.05974084  0.02796468  
## sample estimates:  
##        cor  
## -0.0159187

which(abs(cormatrix)>0.6,arr.ind = T)

##                              row col  
## 编号                           1   1  
## 性别                           2   2  
## 年龄                           3   3

summary(model)

library(car)  
qqPlot(model2,labels = row.names(datacor))

dwtest(model2)

##  
##  Durbin-Watson test  
##  
## data:  model2  
## DW = 2.0242, p-value = 0.7604  
## alternative hypothesis: true autocorrelation is greater than 0