Sie sind auf Seite 1von 7

BES Tutorial Sample Solutions, S1/13

WEEK 4 TUTORIAL EXERCISES (To be discussed in the week starting


March 25)
1. UNSW wants to measure its attractiveness in absorbing bright minds from
across Australia. They perform an experiment by inviting 100 high school
students from different public schools in New South Wales and ask them to
browse a few websites related to different universities and choose one that
they would prefer most.
(a) Is this a random sample? Can you think of any potential source of bias?
The sample is not perfectly random. First of all, only students in NSW
are sampled, therefore the attitudes of students from other states of
Australia and overseas students are missed. Also the students are all
coming from public schools, and public school graduates might have
different aspirations from private school graduates.
(b) Now assume that the sample of students is chosen to be perfectly
random. Can you think of any confounding factors?
Even if the sample is perfectly random, it should be noted that
universities have different qualities in different fields. For instance,
UNSW engineering and science might be leading faculties, but the
medical faculty might not be the top. A students choice of a university
does not only depend on the attractiveness of the University as a whole,
but also on the fact that they are leading in the particular field that the
student is interested in.
Also Australian students have historically been reluctant to travel, so
proximity has also been an important factor in determining university
choice. Therefore, to measure total attractiveness, we should control
for strengths and weaknesses across schools and differences in travel
times.

2. SIA: Managerial Roles in Asia


Pearson and Chatterjee (2003)*, conducted a survey amongst indigenous
managers working in 4 Asian countries in order to investigate the relative
importance and frequency of employment of different managerial roles. In
addition to providing responses to questions about their managerial work,
respondents provided demographic information. Part of the demographic
information from Table 1 of Pearson and Chatterjee (2003) is given below:

Sample size
Gender
Men
Women
Age (Years)
20-29
30-39
40-49
Above 49

Brunei
115

Respondents
Malaysia
Japan
143
195

Thailand
156

73.9
26.1

67.1
32.9

94.4
5.6

28.2
71.8

11.3
25.2
51.3
12.2

21.0
45.5
25.8
7.7

7.7
32.3
27.7
32.3

44.2
27.6
22.4
5.8

(a) Compare and contrast the differences in the gender and age
distributions of the respondents from the four different countries. Do
these differences necessarily imply that these are non-random samples
from the populations of managers in each of these countries?
Some discussion of the lower part of the table is required. It is
noteworthy that the young predominate in Thailand, whilst the
uppermost age group predominates in Japan. All four age distributions
have quite large differences from one another. Malaysia and Thailand
are, perhaps, the most similar to one another.
No, it does not necessarily imply that these are non-random samples.
They may reflect actual differences in the population of indigenous
mangers across countries. (Although some of the distributions do look
extreme and hence unlikely to be representative.)
(c)Pearson and Chatterjee (2003) do state that these were convenience
samples often administered by managers who were involved in post
graduate educational programs conducted by the authors. Does this
necessarily introduce any biases into the analysis of managerial roles?

Not necessarily. It depends. Suppose these samples have gender


distributions that are not representative then this is only an issue if
managerial roles differ by gender. If not, there is no problem. Even if
they do vary by gender we could control for this in our analysis. A
very simple method would be to do cross country comparisons
separately for males and females.
(c) Pearson and Chatterjee (2003) state that Clearly, the Japanese
sample was the oldest, while the Thailand sample was the youngest.
Consider the median age of respondents in Thailand and Japan and
explain whether Pearson and Chatterjees conclusion is justified
according to this measure of location. What about the median age of
Brunei respondents compared with that of the Japanese respondents?
Review how you would seek an answer to these questions if they had
been posed in terms of mean ages, rather than median ages.
Given the distributions we know that the median age for the Japan
sample is between 40 and 49 while for Thailand it is 30 to 39.
Therefore the answer is yes in terms of medians, the median age for
the Thailand sample is less than for the Japan sample. For Japan
compared to Brunei we cant be sure because in both cases the
median must be between 40 and 49 but we cant tell exactly where.
A comparison of means is even more problematic. Without the
individual data we cant calculate means without making assumptions
about the distribution within a class. For example, assuming class
midpoints to be representative might be a good starting point but then
what do we do about the open ended Above 49 interval? So while
it might seem reasonable that the means for Thailand and Brunei are
less than for Japan it is difficult to say unequivocally.
*Pearson, C.A.L. & Chatterjee, S.R. (2003), Managerial work roles in Asia: An empirical study of
Minzbergs role formulation in four Asian countries, Journal of Management Development 22, 694706.

3. (a) Explain what it means to say that two probabilistic events in a sample
space are mutually exclusive of one another.
If two events, lets call them A and B are mutually exclusive, then it
means that they do not have any simple events in common; i.e. that the
simple events that combine to make up A have no elements in common
with those that go to make up B.
(b)

Explain what it means to say that two probabilistic events in a sample


space are independent of one another.
When two events are independent of one another, it means that the
effect of conditioning on the occurrence of one of them has no effect
of the marginal probability of the other, i.e. Pr(A/B) = Pr(A).

(c)

Why can two events not at the same time be both mutually exclusive
and independent of one another?
Because if A and B are mutually exclusive, then Pr(A and B) = 0,
whereas if they are independent Pr(A and B) = Pr(A).Pr(B) 0.

4. A department store wants to study the relationship between the way


customers pay for an item and the price of the item. 250 transactions are
recorded and the following table is formed. Convert the table to a joint
distribution and find the following:
Price
category
Under $20
$20-$100
Over $100

Cash
15
11
6

Joint distribution:
Price
category
Cash
Under $20
0.06
$20-$100
0.044
Over $100
0.024
Marginal
0.128

Payment
Credit card
9
53
38

Debit card
18
52
48

Payment
Credit card Debit card
0.036
0.072
0.212
0.208
0.152
0.192
0.4
0.471

Marginal
0.168
0.464
0.368
1

(a) What is the probability that an item is under $20?


P(Under $20) = 0.168
(b) What is the probability that an item with a price tag of $43 is paid in
cash?
P($20-$100 and cash) = 0.044
(c) What is the probability that people pay for an item that is at least $20 by
credit?
P($20 and credit) = 0.212 + 0.152 = 0.364
(d) If somebody used a debit card to pay for an item, what is the probability
that the item was less than $100?
P(<$100|debit) = (0.072+0.208)/0.471 = 0.594
(e) Are price and way of payment independent?
One way to check is to compare the marginal distribution of price with

the conditional distribution of price given payment type (say cash)


P($20-$100|cash) = 0.344
implying dependence

P(($20-$100) = 0.464

5. In a small batch of 20 manufactured widgets, there are, in fact, 3 defective


ones. You, as quality control officer for the company making the widgets
decide to examine a sample of 3 widgets, selected without replacement, to
see how many defective ones are selected.
(a)Use a probability tree to evaluate the probability distribution of the
number of defectives sampled.
The tree is of the obvious kind with the first branch from a branch where
the probability of defective is 0.15 and not defective is 0.85. From the
upper of these branches at the next node the probability of defective
being selected is 2/19 and non-defective is 17/19. From the lower first
branch, the probability of a defective is 3/19 and of a non-defective is
16/19. From the nodes at the end of the 4 second branches, the 8
probabilities of defective and non-defective are, respectively, 1/18,
17/18, 2/18, 16/18, 2/18, 16/18, 3/18 and 15/18.
Since draws are made independently each time, the relevant probability
distribution of X, the number of defectives drawn in a sample of 3
without replacement is
x

P(X = x)

680/1140

408/1140

51/1140

1/1140

P(X=x) = 1

(b) How would your answer change if the sampling were done with
replacement?
The resultant probability distribution is now
x

P(X = x)

4913/8000

2601/8000

459/8000

27/8000

P(X=x) = 1

(c)Now consider a more realistic real-world problem where the


manufacturer makes widgets in large batches and does not, in fact, know
the proportion of defective widgets manufactured. Managers ask you to
set up a sampling experiment to help determine the proportion of
defective widgets in a (large) batch. How would you go about doing this
and how confident would you be in the answer that you present to senior
management?
Now there is almost no difference between sampling with and without
replacement. The population is now large and this is a realistic inference
problem for the unknown population proportion of defectives. If we call this
proportion p and select a random sample of size n and count the number of
defectives, x, say, then a feasible was of estimating the unknown
population parameter p is to choose the sample analogue = x/n.
You would probably be more confident in your estimate the larger the
sample size you had used, but you would not expect to assess p exactly.

Das könnte Ihnen auch gefallen