R_homework04

data = read.csv("dat1.csv") # 读入数据

library(agricolae)

cat("###################################################################\n\n")
cat("1.1 试评价不同教育程度(Education)的BMI是否有差别？如果有,哪两组有差别？\n\n")

d_BMI = data$BMI # 提取BMI
d_Education = data$Education # 提取edu

d_BMI_edu = data.frame(data$BMI, data$Education) # 组合

# 单因素方差分析
result1= aov(d_BMI~d_Education, data = d_BMI_edu)
result1

# SNK多重比较
result_SNK = SNK.test(result1,"d_Education",group=F)
cat("\n")
result_SNK$comparison

cat("\n综上, 由SNK多重比较可见, 教育程度组中1-2, 1-3, 2-5, 3-5存在明显差异")

cat("\n\n###################################################################\n\n")
cat("1.2 男性和女性的教育程度构成是否相同？\n")

d_edu = data.frame(edu = data$Education, sex = data$Gender) # 组合

# 卡方检验
result2 = chisq.test(table(d_edu),correct=F)
result2$expected
result2

cat("综上, 男性和女性的教育程度构成存在显著差异(p = 0.002304,p < 0.05)")

cat("\n\n###################################################################\n\n")
cat("1.3 男性和女性的种族(Race)分布是否相同？\n\n")

d_race = data.frame(race = data$Race, sex = data$Gender) # 组合

# 卡方检验
result3 = chisq.test(table(d_race),correct=F)
result3$expected
result3

cat("综上, 男性和女性的种族分布之间未观察到显著差异(p = 0.6875,p > 0.05)")

cat("\n\n###################################################################\n\n")
cat("2. 如下两种白喉杆菌培养基培养结果是否相同？\n\n")

# 创建2x2的矩阵
d_2methods = matrix(
  c(36,10,24,135), nrow=2,
  dimnames = list("Method 2" = c("+","-"),
                  "Method 1" = c("+","-")
                  )
  )

d_2methods

# 进行McNemar检验
mcnemar.test(d_2methods,correct = F)

cat("综上, 两种不同白喉杆菌培养基培养结果之间存在显著差异(p = 0.01635,p < 0.05)")
cat("\n")

###################################################################

1.1 试评价不同教育程度(Education)的BMI是否有差别？如果有,哪两组有差别？

Call:
   aov(formula = d_BMI ~ d_Education, data = d_BMI_edu)

Terms:
                d_Education Residuals
Sum of Squares         7.84  40535.99
Deg. of Freedom           1      1239

Residual standard error: 5.719851
Estimated effects may be unbalanced
36 observations deleted due to missingness

      difference pvalue signif.         LCL         UCL
1 - 2 -1.7445839 0.0045      ** -3.08044758 -0.40872016
1 - 3 -1.3112591 0.0314       * -2.52976757 -0.09275068
1 - 4 -0.9836367 0.0584       . -2.00240571  0.03513236
1 - 5  0.1579075 0.7611         -0.86086150  1.17667658
2 - 3  0.4333247 0.4042         -0.58544430  1.45209378
2 - 4  0.7609472 0.3080         -0.45756125  1.97945563
2 - 5  1.9024914 0.0024      **  0.48391827  3.32106455
3 - 4  0.3276224 0.5282         -0.69114659  1.34639149
3 - 5  1.4691667 0.0245       *  0.13330296  2.80503038
4 - 5  1.1415442 0.0718       . -0.07696422  2.36005266

综上, 由SNK多重比较可见, 教育程度组中1-2, 1-3, 2-5, 3-5存在明显差异

###################################################################

1.2 男性和女性的教育程度构成是否相同？
   sex
edu        1         2
  1 146.8312 136.16876
  2 102.7300  95.27002
  3 159.2834 147.71664
  4 131.2661 121.73391
  5 120.8893 112.11068

        Pearson's Chi-squared test

data:  table(d_edu)
X-squared = 16.607, df = 4, p-value = 0.002304

综上, 男性和女性的教育程度构成存在显著差异(p = 0.002304,p < 0.05)

###################################################################

1.3 男性和女性的种族(Race)分布是否相同？

    sex
race         1         2
   1 131.67424 122.32576
   2  31.62255  29.37745
   3 376.36022 349.63978
   4 107.82772 100.17228
   5  14.51527  13.48473

        Pearson's Chi-squared test

data:  table(d_race)
X-squared = 2.2628, df = 4, p-value = 0.6875

综上, 男性和女性的种族分布之间未观察到显著差异(p = 0.6875,p > 0.05)

###################################################################

2. 如下两种白喉杆菌培养基培养结果是否相同？

        Method 1
Method 2  +   -
       + 36  24
       - 10 135

        McNemar's Chi-squared test

data:  d_2methods
McNemar's chi-squared = 5.7647, df = 1, p-value = 0.01635

综上, 两种不同白喉杆菌培养基培养结果之间存在显著差异(p = 0.01635,p < 0.05)