Question

A digital blood pressure gauge is under fire for its inaccuracy, so the medical association has selected a group of 15 patients. They were asked to have their (systolic) blood pressure taken first by an accredited physician (with accurate readings of systolic pressure), and then to go home and use the digital blood pressure gauge themselves. The data for the 15 patients tested:

Patient	Physician	Gauge
A	112	126
B	109	108
C	139	116
D	141	123
E	120	138
F	99	123
G	128	119
H	118	122
I	116	116
J	120	118
K	111	114
L	123	108
M	114	130
N	121	123
O	132	127

Question: Do average systolic readings made by patient users of the gauge differ significantly from those taken by physicians? Use a 95% confidence interval (CI).

Answer 1

Physician	Gauge	Difference
112	126	-14
109	108	1
139	116	23
141	123	18
120	138	-18
99	123	-24
128	119	9
118	122	-4
116	116	0
120	118	2
111	114	-3
123	108	15
114	130	-16
121	123	-2
132	127	5

Sample mean of the difference using excel function AVERAGE(), x̅d = -0.5333

Sample standard deviation of the difference using excel function STDEV.S(), sd = 13.5376

Sample size, n = 15

95% Confidence interval :

At α = 0.05 and df = n-1 = 14, two tailed critical value, t-crit = T.INV.2T(0.05, 14) = 2.145

Lower Bound = x̅d - t-crit*sd/√n = -0.5333 - 2.145 * 13.5376/√15 = -8.0302

Upper Bound = x̅d + t-crit*sd/√n = -0.5333 + 2.145 * 13.5376/√15 = 6.9635

-8.0302 < µd < 6.9635