Professional Documents
Culture Documents
Giulia Marcon
1) Topics: Chapters 7 and 9
Measures of variability: range, variance, standard deviation, coefficient of variation
Relations between two variables: two-entry table, conditional frequencies, side-byside chart.
2) Exercises:
Exercise 1.
A student examined all bars and coffee shops near the University and the following variables
were collected:
SEAT
is it easy to find a seat? (0=NO 1=YES)
WINDOWS number of (shop) windows
PRICE
price of the preferred sandwich
DIST
distance from the University (in meters)
SEAT WINDOWS PRICE DIST
Total
DIST 2
2.8
650
422500
3.5
500
250000
700
490000
2.4
850
722500
3.2
200
40000
400
160000
4.5
350
122500
4.2
150
22500
3.6
200
40000
18
31.2
4000 2270000
a) Compute the range and the standard deviation of the variable Distance (DIST).
b) What is the most adequate measure of central tendency (central position) to describe the
variable PRICE?
c) Use an adequate analysis (on the basis of graphical representations and/or tables and/or
statistical measures) to show if finding a seat easily depends on the number of windows of
the bars. Briefly explain your results
a) X = DIST
Range = Max Min = 850 150 = 700
N
X =
x
i =1
4000
= 444.4444
9
X2 =
x
i =1
2
i
2270000
444.4444 2 = 54691.3580
9
X2 =
X = 54691.3580 = 233.8618
b) No mode can be observed from the data. We could prefer the median with respect to the
mean in case of anomalous values or in the case of asymmetric distribution.
There are no reasons why we should prefer the median to the mean (equals to 3.4667) as
measure of central tendency. Indeed, the underneath context does not allow us to conclude
that the mean is preferable to the median. The two values are very close to each other and
both of them can be correct measures of synthesis for the Price variable.
c)
Contingency table
(absolute frequencies)
Seat
Contingency table
(marginal row frequencies)
Seat
Total
Number of
windows
No
Yes
0.3333
0.6667
1.0
0.3333
0.6667
1.0
0.3333
0.6667
1.0
Total
Total
0.3333
0.6667
1.0
Number of
windows
No
Yes
Total
side-by-side chart
0,8
Rel. Freq.
0,7
0,6
0,5
0,4
Easy to find a seat
0,3
0,2
0,1
0
N. of windows
2
The contingency table is sufficient to show us that finding a seat in the bar has the same
relative frequency (33,33%) for bars with 1, 2 or 3 windows and it is therefore independent
from that variable.
Exercise 2.
In a population of 200 married couples, information on the number of children (X) and the yearly
income of the couple in thousands of Euros (Y) was collected. The resulting data are summarized
in the following two-way table:
Y\X
[0, 30)
10
50
60
[30,60)
20
36
[60,90)
a)
Frequency
20
78
102
200
x[ 0 ,30 ) =
b) x[ 30 ,60 )
x[ 60 ,90 )
c)
x = 1.41
y = 30
C.V . X =
sX =
2
sY =
2
0.6648
= 0.4715
1.41
C.V .Y =
20.1246
= 0.6708
30
s X = 0.6648
sY = 20.1246