Q5
The dataset, BAC{HoRM}, is from a study to compare the blood alcohol concentration (BAC) of subjects using two different methods (Breathalyzer Model 5000 or labtest BAC in a laboratory) on 15 subjects. Replicate the analysis by following the sleep example of the class note.
## id test bac
## 1.1 1 1 0.160
## 2.1 2 1 0.170
## 3.1 3 1 0.180
## 4.1 4 1 0.100
## 5.1 5 1 0.170
## 6.1 6 1 0.100
## 7.1 7 1 0.060
## 8.1 8 1 0.100
## 9.1 9 1 0.170
## 10.1 10 1 0.056
## 11.1 11 1 0.111
## 12.1 12 1 0.162
## 13.1 13 1 0.143
## 14.1 14 1 0.079
## 15.1 15 1 0.006
## 1.2 1 2 0.145
## 2.2 2 2 0.156
## 3.2 3 2 0.181
## 4.2 4 2 0.108
## 5.2 5 2 0.180
## 6.2 6 2 0.112
## 7.2 7 2 0.081
## 8.2 8 2 0.104
## 9.2 9 2 0.176
## 10.2 10 2 0.048
## 11.2 11 2 0.092
## 12.2 12 2 0.144
## 13.2 13 2 0.121
## 14.2 14 2 0.065
## 15.2 15 2 0.000| test | bac |
|---|---|
| 1 | 0.1178 |
| 2 | 0.1142 |
| test | bac |
|---|---|
| 1 | 0.0524911 |
| 2 | 0.0519810 |
| 15 | 10 | 14 | 7 | 8 | 4 | 11 | 6 | 13 | 1 | 12 | 2 | 9 | 5 | 3 | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| id | 15 | 10 | 14 | 7 | 8 | 4 | 11 | 6 | 13 | 1 | 12 | 2 | 9 | 5 | 3 |
| bac | 0.006 | 0.056 | 0.079 | 0.081 | 0.104 | 0.108 | 0.111 | 0.112 | 0.143 | 0.160 | 0.162 | 0.170 | 0.176 | 0.180 | 0.181 |
##
## Paired t-test
##
## data: subset(dta5, test == "1")$bac and subset(dta5, test == "2")$bac
## t = 1.0447, df = 14, p-value = 0.3139
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
## -0.003790739 0.010990739
## sample estimates:
## mean of the differences
## 0.0036##
## Paired t-test
##
## data: bac[test == 1] and bac[test == 2]
## t = 1.0447, df = 14, p-value = 0.3139
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
## -0.003790739 0.010990739
## sample estimates:
## mean of the differences
## 0.0036##
## Paired t-test
##
## data: bac by test
## t = 1.0447, df = 14, p-value = 0.3139
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
## -0.003790739 0.010990739
## sample estimates:
## mean of the differences
## 0.0036## Analysis of Variance Table
##
## Response: bac
## Df Sum Sq Mean Sq F value Pr(>F)
## id 14 0.075156 0.0053683 60.2791 4.851e-10 ***
## test 1 0.000097 0.0000972 1.0914 0.3139
## Residuals 14 0.001247 0.0000891
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1## Analysis of Variance Table
##
## Response: bac
## Df Sum Sq Mean Sq F value Pr(>F)
## id 14 0.075156 0.0053683 60.2791 4.851e-10 ***
## test 1 0.000097 0.0000972 1.0914 0.3139
## Residuals 14 0.001247 0.0000891
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
| x |
|---|
| -0.015 |
| -0.014 |
| 0.001 |
| 0.008 |
| 0.010 |
| 0.012 |
| 0.021 |
| 0.004 |
| 0.006 |
| -0.008 |
| -0.019 |
| -0.018 |
| -0.022 |
| -0.014 |
| -0.006 |
| -0.022 | -0.019 | -0.018 | -0.015 | -0.014 | -0.014 | -0.008 | -0.006 | 0.001 | 0.004 | 0.006 | 0.008 | 0.01 | 0.012 | 0.021 |
##
## One Sample t-test
##
## data: ehs_d
## t = -1.0447, df = 14, p-value = 0.3139
## alternative hypothesis: true mean is not equal to 0
## 95 percent confidence interval:
## -0.010990739 0.003790739
## sample estimates:
## mean of x
## -0.0036檢定結果： p-value = 0.3139，表示兩種方法測得的BAC沒有顯著差異。
#Equivalence test效果量測試：差多少才算是有差?
例如：A藥比B藥使患者多睡半小時，藥效有差嗎?A生比B生數學多考10分，能力有差嗎?(使用前要先由共識決定差多少有差。)
Equivalence test (Schuirmann, 1987)
H0:The mean bac between tests is different beyond the equivalence bound of (0, 0.05)
H1:The mean bac between tests is the same.
Conclusions 血液酒精濃度差異值的效果量檢測結果不顯著。