Part I
(This part is worth 18 marks. All questions carry equal weight in this part. Please answer by circling one
correct answer. Please do not try to insure yourself or reduce risk by marking more than one correct answer)
1. Which of the following statements is false?
i) The mean of a random variable will always be positive while its standard deviation can never be negative
ii) The standard deviation of a random variable can never be zero
iii) The mean of a random variable can be smaller than its standard deviation
a. i) only
c. None of them
d.
All of them
Ans: a
2. Chebyshevs rule tells me that approximately at most 16 % of data values are outside of z standard
deviations of the sample mean. If so, the value of z is
a. 1.091
b. 2.35
c. 2.5
Ans: c
3. In a Confidence Interval determination with sample size 8, if the critical value is 2.998, we state
a)
b)
c)
d)
With 98% confidence that the margin of error is 2.266 if the population standard deviation is 2
With 99% confidence that the margin of error is 1.5899 if the sample standard deviation is 1.5
With 99 % confidence that the margin of error is 0.795 if the population standard deviation is 0.75
With 98% confidence that the margin of error is 1.06 if the sample standard deviation is 1
Ans: d)
4. The Expected Value computed using the formula xf(x) is a numerical measure of the mean of
a. The probability distribution of a population
b. The random variable for a population
Ans: c
5. Which of the following statements are false?
i) A normal distribution can have a probability mass function
ii) In some cases, such as for sampling distributions, the standard normal distribution may have a standard
deviation other than 1 since we need to divide by n
iii) The z-score for a positive value in a sampling distribution always increases when sample sizes are
increased
b) i) only
Ans: a)
(b) 0.035
(c) 0.050
(d) 0.127
(e) 0.965
1. In a symmetrical bell-shaped distribution for a continuous random variable X, the empirical rule tells us
that approximately 81.5% of the values lie between a and b, with the mean situated at 10 and the variance
being equal to 16. Compute a and b. (There will be two sets of possible values). Also, compute P(X < a)
and P(X > b) for one set.
Ans.: The empirical rule allows us to infer that 81.5% = 47.5% (2 standard deviations) + 34% of values
(one standard deviation) lie between a and b. No other values are provided by the empirical rule. (If the
students have used the Normal Distribution tables to get the answers, please award NO marks). If the mean
is 10 and the standard deviation is 4, then the two sets of possible values are:
[10 8, 10+ 4] or [2, 14]
[10 4, 10 + 8] or [6, 18]
PART III
(This section is worth 20 marks. Please be very careful with your mathematical notation.)
1. A large hospital wholesaler, as part of an assessment of workplace safety, gave a random sample of 40 of
its warehouse employees a test (measured on a 0 to 100 point scale) on safety procedures. For that sample
of employees, the mean test score was 72 points, with a standard deviation of 6 points. Determine and
interpret a 95% confidence interval for the mean test score of all the companys warehouse employees.
Ans: 95% C.I = 72 + 1.96*6/sqrt (40) = [70.14, 73.86].
Interpretation: The above means that there is 95% chance (or probability or confidence) that the
Population Mean will lie between 70.14 and 73.86 (or in that interval).
2. In the past decade there have been extensive antismoking campaigns to try and reduce the
proportion of smokers in the population. In 1982, a survey of 350 adult females revealed that 148
smoked. In 1989, 488 adult females were surveyed and 163 smoked. Let p represent the
proportion of adult female smokers.
i.
Ans.
H0: p1989 = 0.422857
Ha: p1989 < 0.422857
ii.
Ans. There is some, but not overwhelming evidence, that the proportion of smokers has
decreased.