In: Statistics and Probability
An insurance company is interested in estimating the population mean cost of basic dental
cleaning at dentists in Saskatoon. Suppose there are only two dentists in Saskatoon:
Dentist A and Dentist B. Suppose also that the cost of basic dental cleaning varies only
depending on how well the patient practices regular dental hygiene, so that the cost of
basic dental cleaning roughly follows a Normal distribution regardless of the dentist.
The insurance company selects 8 sample patients and sends them to both Dentist A and
Dentist B. They send the patients in random order, such that half of the patients are seen
by Dentist A first, and half are seen by Dentist B first, so as not to bias the results. The
cost of basic dental cleaning for these 8 patients seen by both Dentists A and B are
provided below. The insurance company would like to determine whether the population
mean cost of basic dental cleaning by Dentist A is different from the population mean
cost of basic dental care by Dentist B. Let the population of costs of basic dental care
from Dentist A be population 1.
Patient 1 2 3 4 5 6 7 8
Dentist A $100 $120 $125 $110 $95 $105 $120 $115
Dentist B $150 $100 $140 $100 $95 $105 $100 $120
Conduct an appropriate hypothesis test using the critical value method. [10 marks]
NOTE: You are encouraged to use Excel to calculate the sample mean(s) and sample
standard deviation(s) for this question. If you use Excel for this, provide the entire
command (e.g. if you take the average of 1,2,3, then write down =AVERAGE(1,2,3)).
Round average and standard deviation calculations to 2 decimal places.
Solution:
Here, we have to use paired t test for the difference between population means.
Null hypothesis: H0: The population mean cost of basic dental cleaning by Dentist A is not different from the population mean cost of basic dental care by Dentist B.
Alternative hypothesis: Ha: The population mean cost of basic dental cleaning by Dentist A is different from the population mean cost of basic dental care by Dentist B.
H0: µd = 0 versus Ha: µd ≠ 0
This is a two tailed test.
We assume level of significance = α = 0.05
Test statistic for paired t test is given as below:
t = (Dbar - µd)/[Sd/sqrt(n)]
From given data, we have
Dbar = -2.50
Sd = 22.6779
n = 8
df = n – 1 = 7
α = 0.05
Critical Values = -2.3646 and 2.3646
(by using t-table)
t = (Dbar - µd)/[Sd/sqrt(n)]
t = (-2.50 - 0)/[ 22.6779/sqrt(8)]
t = -0.3118
Here, we have
-2.3646 < t < 2.3646
So, we do not reject the null hypothesis
There is insufficient evidence to conclude that the population mean cost of basic dental cleaning by Dentist A is different from the population mean cost of basic dental care by Dentist B.