In an experiment concerned with methods of measuring blood pressure, systolic bl

Question

In an experiment concerned with methods of measuring blood pressure, systolic blood pres-sure in mmHg was measured for each of 90 randomly selected people, once using a machine that automatically measures blood pressure and once using conventional equipment with a human operator. The data are available in the file systolic.scsv. The purpose of the following analysis is to determine whether machine measurements are systematically different from the human measurements.

(a) Explain briefy why the data should be treated as paired rather than two independent samples.
(b) Use MATLAB to perform a matched pairs t-test. State the value of the test statistic,the reference distribution with appropriate degrees of freedom, and the P-value. Do
you reject or retain the null hypothesis at the 5% level of signi cance? Make sure to include the code you used in MATLAB to run the test, along with the output (do
NOT include any code used to import/write out data).
(c) Determine if the assumptions of the test are met.

systolic.scsv

machine human diff 170.35 179.41 -9.06 117.38 124.28 -6.9 196.79 212.72 -15.93 88.93 88.58 0.35 180.23 169.66 10.57 109.42 185.42 -76 159.42 154.41 5.01 140.88 145.38 -4.5 139.75 146.65 -6.9 171.88 163.58 8.3 150.99 192.41 -41.42 154.41 203.56 -49.15 141.58 173.83 -32.25 135.32 158.73 -23.41 116.71 123.35 -6.64 203.45 255.12 -51.67 189.85 233.74 -43.89 190.16 218.06 -27.9 79.13 70.31 8.82 135.09 166.04 -30.95 182.52 172.13 10.39 192.21 202.09 -9.88 146.9 167.18 -20.28 121.34 122.56 -1.22 159.81 162.05 -2.24 181.06 182.69 -1.63 177.73 199.59 -21.86 164.5 164.17 0.33 143.84 130.68 13.16 191.49 220.67 -29.18 141.81 155.85 -14.04 175.25 172.89 2.36 136 152.06 -16.06 138.61 142.72 -4.11 104.74 135.25 -30.51 148.98 175.76 -26.78 141.88 154.86 -12.98 157.25 174.67 -17.42 211.88 215.12 -3.24 174.08 197.9 -23.82 105.92 77.99 27.93 123.01 150.25 -27.24 136.02 161.35 -25.33 142.54 130.89 11.65 152.98 149.33 3.65 186.57 178.29 8.28 164.21 176.49 -12.28 189.59 266.15 -76.56 189.83 197.27 -7.44 156.52 180.75 -24.23 198.22 220.45 -22.23 182.08 199.05 -16.97 143.63 136.61 7.02 167.65 189.34 -21.69 103.52 120.06 -16.54 125.49 143.88 -18.39 129.61 135.68 -6.07 121.63 136.75 -15.12 174.74 241.26 -66.52 91.89 113.07 -21.18 151.35 144.5 6.85 131.14 140.44 -9.3 166.29 155.85 10.44 146.89 149.93 -3.04 160.49 141.37 19.12 146.34 163.96 -17.62 157.77 155.32 2.45 99.34 104.23 -4.89 116.39 145.49 -29.1 125.45 116.64 8.81 111.33 117.35 -6.02 135.56 170.53 -34.97 155.1 175.09 -19.99 143.65 154.99 -11.34 179.7 172.29 7.41 188.71 198.77 -10.06 131.7 136.23 -4.53 145.95 153.64 -7.69 195.52 193.13 2.39 93.82 96.29 -2.47 139.55 172.09 -32.54 156.48 161.33 -4.85 143.69 146.15 -2.46 127.56 157.74 -30.18 128.73 110.84 17.89 190.11 199.15 -9.04 217.1 241.97 -24.87 119.92 115.01 4.91 198.68 228.83 -30.15 90.19 103.71 -13.52

Explanation / Answer

(a) Data should be treated as paired rather than independent samples because measuring blood pressure, systolic blood pres-sure in mmHg was measured for each of 90 randomly selected people using machine measurements and human measurements on same person. We have a pair of blood pressure (machine measurement and human measurement) for each person.

(b)

MATLAB code for paired T-test

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

data = csvread('Blood_Pressure_data.csv',1,0);
Machine_measurements=data(:,1)
Human_Measurement=data(:,2)

[h,p,ci,stats] = ttest(Machine_measurements,Human_Measurement,'Alpha',0.05)

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

Output

h=1 (Hypothesis test result- 0 or 1)

and p-value= 1.3084e-08 (<0.05)

ci = ( Confidence interval)

-16.8340
-8.7248

stats =

tstat — Value of the test statistic.

df — Degrees of freedom of the test.

sd — Pooled estimate of the population standard deviation (for the equal variance case) or a vector containing the unpooled estimates of the population standard deviations (for the unequal variance case).

tstat: -6.2626
df: 89
sd: 19.3587

So significant difference is there between two measurements of blood pressure as p-value,0.05.

Related Questions

Navigate

• Browse (All)
• Browse I
• Subjects

---

---

Integrity-first tutoring: explanations and feedback only — we do not complete graded work. Learn more.