h0 평균시간이같다
h1 평균시간이다르다 (일한는사람과 안하는사람의 수면시간)
EC1_1 (경제)
BP16_1 (수면시간)
age
sex
> d = read_sas("hn22_all.sas7bdat")
> df = d[,c("EC1_1","BP16_1","sex","age")]
> df2 = df[df$sleep<=24 & df$job<=2,]
> df3 = na.omit(df2)
> qqnorm(df3$sleep)
> var.test(sleep~job, data=df3) //분산분석 (두 그룹의 수면시간은 다르다는 결과)
F test to compare two variances
data: sleep by job
F = 0.7063, num df = 2999, denom df = 2023, p-value < 2.2e-16
alternative hypothesis: true ratio of variances is not equal to 1
95 percent confidence interval:
0.6519469 0.7647096
sample estimates:
ratio of variances
0.7063021
> t.test(sleep~job, data=df3, var.equal=F) #경제활동하는 그룹이 더 조금잔다라는결과.
Welch Two Sample t-test
data: sleep by job
t = -2.8051, df = 3824.5, p-value = 0.005055
alternative hypothesis: true difference in means between group 1 and group 2 is not equal to 0
95 percent confidence interval:
-0.18358953 -0.03253432
sample estimates:
mean in group 1 mean in group 2
6.692333 6.800395
> df4 = df3[df3$age>29 & df3$age<61,] //나이 범위 설정(30~60세)
> var.test(sleep~job, data=df4)
F test to compare two variances
data: sleep by job
F = 0.75265, num df = 1754, denom df = 586, p-value = 1.578e-05
alternative hypothesis: true ratio of variances is not equal to 1
95 percent confidence interval:
0.6578813 0.8572541
sample estimates:
ratio of variances
0.7526457
> t.test(sleep~job, data=df4,var.equal=F)
Welch Two Sample t-test
data: sleep by job
t = -3.5302, df = 899.14, p-value = 0.0004364
alternative hypothesis: true difference in means between group 1 and group 2 is not equal to 0
95 percent confidence interval:
-0.32168595 -0.09180483
sample estimates:
mean in group 1 mean in group 2
6.653561 6.860307
---------------------------------------------
> df_man = df4[df4$sex==1,] #남녀구분
> df_woman = df4[df4$sex==2,]
> var.test(sleep~job, data=df_man)
F test to compare two variances
data: sleep by job
F = 0.70199, num df = 885, denom df = 84, p-value = 0.01909
alternative hypothesis: true ratio of variances is not equal to 1
95 percent confidence interval:
0.4999865 0.9448459
sample estimates:
ratio of variances
0.7019855
---------------------------------------------
> var.test(sleep~job, data=df_woman)
F test to compare two variances
data: sleep by job
F = 0.77898, num df = 868, denom df = 501, p-value = 0.001447
alternative hypothesis: true ratio of variances is not equal to 1
95 percent confidence interval:
0.6655903 0.9086936
sample estimates:
ratio of variances
0.7789812
-----------------------------------------
> t.test(sleep~job, data=df_man,var.equal=F)
Welch Two Sample t-test
data: sleep by job
t = -3.0455, df = 95.654, p-value = 0.003002
alternative hypothesis: true difference in means between group 1 and group 2 is not equal to 0
95 percent confidence interval:
-0.7318348 -0.1542627
sample estimates:
mean in group 1 mean in group 2
6.627540 7.070588
---------------------------------------------
> t.test(sleep~job, data=df_woman,var.equal=F)
Welch Two Sample t-test
data: sleep by job
t = -2.1339, df = 943.12, p-value = 0.03311
alternative hypothesis: true difference in means between group 1 and group 2 is not equal to 0
95 percent confidence interval:
-0.27760306 -0.01161521
sample estimates:
mean in group 1 mean in group 2
6.680092 6.824701
---------------------------------------------
> df_means = aggregate(BP16_1~EC1_1, data = df_economic, FUN=mean)
> print(df_means)
EC1_1 BP16_1
1 1 6.784549
2 2 7.163386
> df_means = aggregate(BP16_1~EC1_1, data = df_economic, FUN=sd)
> print(df_means)
EC1_1 BP16_1
1 1 3.153079
2 2 5.947984
'R' 카테고리의 다른 글
음주와콜레스테롤 aov (0) | 2024.10.21 |
---|---|
운동하는 사람들 중에서 체중의 평균이 그룹간 차이가 있나(anova) (0) | 2024.10.20 |
t-test, anova-test (0) | 2024.10.19 |
검증의 종류 (0) | 2024.10.19 |
귀무가설의 검증(비교군집이 2개일 때, 3개일 때 구분) (0) | 2024.10.14 |