UECM1224 Probability and Statistics II

Tutorial 8

Q1. A 2009 survey asked workers aged 18 – 29 to state the factor that was most
important to them in a job or career. 26% of these workers said that the most
important factor was advancement opportunity, 26% mentioned benefits, 18%
indicated money, 11% said that being their own boss was most important, 10%
preferred flexible hours, 5% indicated number of hours and 4% mentioned none or
don’t know. A recent random sample of 250 such workers were asked the same
question. Their responses are summarized in the table.

Factor Advancement Benefits Money Own Flexible Number None

boss hours of hours
Frequency 70 60 52 18 32 11 7

Test at the 5% significance level whether the current distribution of priorities of all
such workers differs from the distribution of priorities in the 2009 survey.

Q2. A factory operates four production lines. Maintenance records show that the daily
number of stoppages due to mechanical failure were as shown in the table below
(it is possible for a production line to break down more than once on the same day).
You may assume that  f =1400,  fx =1036.

Number of stoppages, x 0 1 2 3 4 5 6 or more

Number of days, f 728 447 138 48 26 13 0

(a) Use a  2 distribution and a 1% significance level to determine whether the

Poisson distribution is an adequate model for the data.

(b) The maintenance engineer claims that breakdowns occur at random and that
the mean rate has remained constant throughout the period. State, giving a
reason, whether your answer to (a) is consistent with this claim.

(c) Of the 1036 breakdowns which occurred 230 were on production line A, 303
on B, 270 on C and 233 on D. Test at the 5% significance level whether
these data are consistent with breakdowns occurring at an equal rate on
each production line.

Q3. The manager of an assembly process wants to determine whether the number of
defective articles manufactured depends on the day of the week the articles are
produced. She collected the following information.

Day of Week Mon Tues Wed Thur Fri Total

Nondefective 85 90 95 95 90 455
Defective 15 10 5 5 10 45
Total 100 100 100 100 100 500

Is there sufficient evidence to reject the hypothesis that the number of defective
articles manufactured is independent of the day of the week on which they are
produced? Use α = 0.05.

Tutorial 8 - 1
Ho :
... same as ...
2009

H ,: ... differs from 2009

...

Advancement Benefit Money Own boss Flexible hours Number of hours None

O 70 68 52 18 32 II 7

E 25010 26) .
25010 26) . 25010 18) .
25010 11) .
230(0 1) . 250(0 05) ·
25010 047 .

= 65 = 65 = 45 = 27 5
.
= 25 = 12 . 5 = 10

2
2 (70 -
65)2 160 -
65)2 152 45)
-
2
118-27 . 572132-25 & 111-12 5)2. (7-10)2
X = f t + t f = 8 1799
5
.

65 65 45 27 .
23 12 5
.
10

x =
0 . 05
,
V = K -
1 = 6 X = 12 592
.

X Xa

With 2 do not the current distribution of

0 05
rejectHo and conclude that priorities of such workers the same
=
.
we all is
,

that 2009
as in
survey.
 N

-
39

E-1036, In n

Let X be the daily number of suppliers,

Ho : X follows Poisson

U: X does not follow Poisson

-EfH =
103- = 0 74 .

0 74% .
-
0
343
.

Po = P(X =
0) =

0!
= 0 4771
.

E = 1400 (44771) = 667 94

.

0 74'ce -0 34) :

p
.

= P(X = 1) = = 0 3531 .
E = 494 34 .

1 ! ·

0 742(e-0
.
.

74(
↑ 2
= P(X = 2) =

2 !
= 0 1306
.

,
E = 182 84 .

P = p(X 3) = = 0 03222.

E = 45 108 .

8 27 .

i
Py p(X 4) 0 005961 P(X = 4) 83454 < 5
E
= = = . =

E = 8 23 + 1 372
. .

(1 ....
)
P(X5) 0 00098 =
9 772
0 0008823
= .
.

↑= = p(X =
5) = .

E
= 1 . 23522 < 3
E =
1 372 M
E 23522
.

= 1 .

P(X6) = 1 -
0 .
4771 -
0 .
3531 -
0 .
1306 -
0 .
03222 -
0 ,
005961 -
0 . 0008823 = 0 . 0001367

399 772 .

9472
M
1728 -
667 .
(447 -
494 34721138 .
-
182 . 8472(48-49 10832 (26 -
8
-
.
3454)2 (13-1 23522)
.
2

x
.

= + t f t f
667 94 494 34 182 84 45 188 #494 1. 23522
. .
. .

9 777
.

= 108 8744
.

x = 0 01
.
v = 5-1 - = 3 Xa = 11 345 .

, ,
 2
X2 > Xa

with a = 0 01 we
reject to and conclude that X does not follow Poisson distribution.
.

b) Not consistent because from (a) the Poisson model was not adequate.

C) Production O E

A 230 1036/4 =
259

B 303 &

C 270
~

D 233

1036 1036

1230-299)2 1303-25972 1270-259 22 1233-25 972

X2 = + + + = 13 799
.

259 259 259 259

2 = 0 09 .

,
V = 4 -1 =
3
,
Xx = 7 . 815

X2 > Xo

With < 0 05 Ho and conclude that the data not consistent

: we
reject are ...
= .

,
 the number of defective Independent of
Ho : ... the
day...
H, : . . .
dependent...

of week Wed Thur Fri

Day Mon Thes

defective 85(455x100 91)90(91) 95(91) 95(91) 90191)

I

Non =

Defective 15 (100 91 - = 9) 10 (9) 5(9] 5(9) 10 (9)

190-9172 115-9 ( 19-972

2
195-9172 93
9172193-9173
2 2
189-91) I 90 -
110 -
15-9)
x2 =
f
91
-
91
,
91
t t + +
91 91 9 9 9 9

110 -
9)2
t =
8 547
.

x = 0 05.
V = (5 1722-1)
- = 4 Xa = 9 488
.

, ,

X2 Xa

: With a = 0 05.
we do not rejectHo and conclude that ..
is independent.
,
 A B C D Total

correct 59( (

incorrect 21 ( ( 25( & 30( ( 40( (

Total so 100 90 110

x = 0 01 .
V = (4 1) (2
- -
1) =
3
,
UECM1224 Probability and Statistics II

Q4. A survey of randomly selected customers who visited a restaurant in K.L. showed
the following results.

Quality of Service
Gender Above Average Average Below Average Total
Female 7 24 29 60
Male 8 26 6 40 GRT
Total 15 50 35 100 N
-
Using α = 0.05, does the sample present sufficient evidence to conclude that quality
of responses is dependent of the gender of the respondents?

Q5. An accountant wants to test the hypothesis that the proportion of incorrect
transactions at four client accounts is about the same. A random sample of 80
transactions of client A reveals that 21 are incorrect; for client B, the sample
proportion is 25 out of 100; for client C, the proportion is 30 out of 90 sampled; and
for client D, 40 are incorrect out of a sample of 110. Conduct the test at α = 0.01.

Answer
Q1.  2 = 8.1799

Q2. (a)  2 = 108.8744 (c)  2 = 13 .7992

Q3.  2 = 8.547

Q4.  2 = 11.73

Q5.  2 = 4.2317

Tutorial 8 - 2