Sei sulla pagina 1di 12

Laboratory work nr.

2
1. Compute the simple means: arithmetic, geometric, harmonic and quadratic. Use the Excel functions:
AVERAGE, HARMEAN, and GEOMEAN for the first 3 means. Compute median, mode for ungrouped data.
Use the QUARTILE function to find out the quartiles. Find the relationship between the means.
155
156
158
158
159
160
160
163
164
165

169
169
172
172
173
173
173
173
173
173

173
173
174
174
174
174
175
175
175
175

175
175
175
176
176
176
176
176
177
178

178
180
180
180
180
183
183
183
184
187

I copied the values of the height (in cm) of 50 pupils in a school in Excel and used these values to compute the
following steps.
Using the AVERAGE function I found the Arithmetic mean: Arithmetic mean= 172,76
Using the GEOMEAN function, I computed the Geometric mean: Geometric mean= 172,59
Using the HARMEAN function, I computed the Harmonic mean: Harmonic mean= 172,42
To compute the quadratic mean I took the following steps:
1. Using the SUMSQ function, I found the sum of all the values squared=1495130.
2. Then, using the function COUNT, I found the total number of values, which is 50.
3. Then I took the average of the square values SUMSQ/COUNT=29902,6.
4. Finally, I took the square root of the average values, using the SQRT function. SQRT(29902,6)=
172,9237. Quadratic mean= 172,92
5.
Between the types of means computed for the same data set we always have the relation:
Quadratic mean Arithmetic mean Geometric mean Harmonic mean.
-

In our case:
172,92>172,76>172,59>172,42
Using the MEDIAN function, I computed the median. Median=174 (the middle value of the data set).
Using the MODE function, I computed the mode. Mode=173 (the value that occurs more frequently).
Using the QUARTILE function in Excel I computed the quartiles:
Q0=Qmin= 155
Q1=
172
Q2=Me=
174
Q3=
176
Q4=Qmax= 187

2.) Compute mean absolute deviation, the standard deviation for ungrouped data.
To compute the MAD and standard deviation for ungrouped data, I created the following table in Excel:

Xi
di=Xi-Mean |Xi-Mean| di^2
155
-17,76
17,76 315,4176
156
-16,76
16,76 280,8976
158
-14,76
14,76 217,8576
158
-14,76
14,76 217,8576
159
-13,76
13,76 189,3376
160
-12,76
12,76 162,8176
160
-12,76
12,76 162,8176
163
-9,76
9,76
95,2576
164
-8,76
8,76
76,7376
165
-7,76
7,76
60,2176
169
-3,76
3,76
14,1376
169
-3,76
3,76
14,1376
172
-0,76
0,76
0,5776
172
-0,76
0,76
0,5776
173
0,24
0,24
0,0576
173
0,24
0,24
0,0576
173
0,24
0,24
0,0576
173
0,24
0,24
0,0576
173
0,24
0,24
0,0576
173
0,24
0,24
0,0576
173
0,24
0,24
0,0576
173
0,24
0,24
0,0576
174
1,24
1,24
1,5376
174
1,24
1,24
1,5376
174
1,24
1,24
1,5376
174
1,24
1,24
1,5376
175
2,24
2,24
5,0176
175
2,24
2,24
5,0176
175
2,24
2,24
5,0176
175
2,24
2,24
5,0176
175
2,24
2,24
5,0176
175
2,24
2,24
5,0176
175
2,24
2,24
5,0176
176
3,24
3,24
10,4976
176
3,24
3,24
10,4976
176
3,24
3,24
10,4976
176
3,24
3,24
10,4976
176
3,24
3,24
10,4976
177
4,24
4,24
17,9776
178
5,24
5,24
27,4576
178
5,24
5,24
27,4576
180
7,24
7,24
52,4176
180
7,24
7,24
52,4176
180
7,24
7,24
52,4176
180
7,24
7,24
52,4176
183
10,24
10,24 104,8576

183
183
184
187

10,24
10,24
11,24
14,24

10,24
10,24
11,24
14,24
277,28

104,8576
104,8576
126,3376
202,7776
2829,12

3.) Make the discrete variable (frequency distribution). Compute the means: arithmetic, geometric,
harmonic and quadratic. Compute median, mode, quartiles, mean absolute deviation, the standard
deviation, coefficient of variation.
Discrete variable frequency distribution:
Xi
155
156
158
159
160
163
164
165
169
172
173
174
175
176
177
178
180
183
184
187
Total:

Fi
1
1
2
1
2
1
1
1
2
2
8
4
7
5
1
2
4
3
1
1
50

I created the following table in Excel to help me in the computing of the arithmetic, geometric, harmonic and
quadratic mean, the mode and median.
Used to
compute
the
arithmetic
mean:
Xi
155
156
158
159
160
163
164
165
169
172
173
174
175
176
177
178
180
183
184
187
Total:

Fi

Cfi
1
1
2
1
2
1
1
1
2
2
8
4
7
5
1
2
4
3
1
1
50

1
2
4
5
7
8
9
10
12
14
22
26
33
38
39
41
45
48
49
50

Xi*Fi
155
156
316
159
320
163
164
165
338
344
1384
696
1225
880
177
356
720
549
184
187
8638

Used to
compute the
harmonic
mean:

Used to
compute the
geometric
mean:

Used to
compute the
quadratic
mean:

Fi/Xi
Xi^Fi
xi^2*Fi
0,006451613
155
24025
0,006410256
156
24336
0,012658228
24964
49928
0,006289308
159
25281
0,0125
25600
51200
0,006134969
163
26569
0,006097561
164
26896
0,006060606
165
27225
0,01183432
28561
57122
0,011627907
29584
59168
0,046242775
8,02359E+17
239432 mode group
0,022988506
916636176
121104 median group
0,04
5,02651E+15
214375
0,028409091
1,68874E+11
154880
0,005649718
177
31329
0,011235955
31684
63368
0,022222222
1049760000
129600
0,016393443
6128487
100467
0,005434783
184
33856
0,005347594
187
34969
0,289988853
1495130

(the product of each xi to the power fi ).

For the Discrete frequency distribution, the Mode will correspond to the variable most occurring, the category with
the highest frequency. In our case, the variable with the highest frequency is 173 (frequency=8). Mode=173
To find the median, first we must find the median location (and compute the cumulative frequency).

The cumulative frequency must be greater or equal than 25,5. Looking at the previous table we can say that the CFi
that is greater or equal than 25,5 is 26, so the Median= 174.

Quartiles:

In order to approximately find out the Q1, Q2 and Q3 I computed the Ogive(a cumulative line graph).

Ogive

60
50

45
38 39

40

48 49

50

41

33
30

26
22

20
10
1

9 10

12

14

0
155156158159160163164165169172173174175176177178180183184187

From the Ogive, we can see the positions where the quartiles lie and thus can approximate them as follows:

Mean absolute deviation, the standard deviation, coefficient of variation.


I computed the following table in Excel to help me in the computing of the Mean absolute deviation, the standard
deviation and the coefficient of variation for the discrete frequency distribution.
Xi
155
156
158
159
160
163
164
165
169
172
173
174
175
176
177
178
180
183
184
187
Total:

Fi
1
1
2
1
2
1
1
1
2
2
8
4
7
5
1
2
4
3
1
1
50

|Xi-Mean|
Xi(mean=172,76) |Xi-mean|*Fi (Xi-Mean)^2
Mean
(Xi-Mean)^2 *Fi
-17,76
17,76
17,76
315,4176
315,4176
-16,76
16,76
16,76
280,8976
280,8976
-14,76
14,76
29,52
217,8576
435,7152
-13,76
13,76
13,76
189,3376
189,3376
-12,76
12,76
25,52
162,8176
325,6352
-9,76
9,76
9,76
95,2576
95,2576
-8,76
8,76
8,76
76,7376
76,7376
-7,76
7,76
7,76
60,2176
60,2176
-3,76
3,76
7,52
14,1376
28,2752
-0,76
0,76
1,52
0,5776
1,1552
0,24
0,24
1,92
0,0576
0,4608
1,24
1,24
4,96
1,5376
6,1504
2,24
2,24
15,68
5,0176
35,1232
3,24
3,24
16,2
10,4976
52,488
4,24
4,24
4,24
17,9776
17,9776
5,24
5,24
10,48
27,4576
54,9152
7,24
7,24
28,96
52,4176
209,6704
10,24
10,24
30,72
104,8576
314,5728
11,24
11,24
11,24
126,3376
126,3376
14,24
14,24
14,24
202,7776
202,7776
166
277,28
1962,192
2829,12

4.) Make the polygon by frequencies.


Polygon of frequencies
9
8

7
6
5

5
4

4
4
3

1
1

0
155 156 158 159 160 163 164 165 169 172 173 174 175 176 177 178 180 183 184 187

5.) Compute the coefficients of skewness and kurtosis.


Using the KURT function in Excel I computed the kurtosis. Coefficient of kurtosis= 0,16188
Using the SKEW function I computed the skewness. Coefficient of skewness= -0,7758
6.) Make a conclusion about the characteristic of the distribution
The mode of the data set is equal to 173, so we can say that the height that is most common among the pupils is 173
cm. The mode of the distribution is clearly seen in the polygon of frequencies too. In order to judge the shape of the
distribution we need to compare the mean the median and the mode.The arithmetic mean is equal to 172,76, the
median is equal to 174 and the mode is 173. The Median(174)>Mean(172,76), so the distribution of the data is
skewed to the left. The mean is equal to 172, 76 cm so we can say that in average the pupils have a height of 172, 76
cm.
The Kurtosis characterizes the relative flatness of a distribution compared with the normal distribution. Positive
kurtosis indicates a relatively peaked distribution. Negative kurtosis indicates a relatively flat distribution. In our
case, we have a relatively peaked distribution because the coefficient of kurtosis is 0,16188.
Analysing the quartiles we can say that:
or 25% of the pupils have a height less than or equal to 172 cm.
or 50% of the pupils have a height that is less than or equal to 174 cm.
or 75% of the pupils have a height that is less than or equal to 176 cm.
or 50& of the pupils have a height between 172 cm (Q1) and 176 cm (Q3).

7.) Group the pupils into 8 equal intervals by their heights. Then work only with this distribution
(continuous variable).
I computed the range of variation and the width size of the group (the nr of groups is given=8). These values will help
in the creating of the intervals and the table.

I created this table in Excel:


Gr.
1
2
3
4
5
6
7
8

Interval
155-159
159-163
163-167
167-171
171-175
175-179
179-183
183-187

Total:

Fi

Cfi
4
3
3
2
14
15
4
5
50

4
7
10
12
26
41
45
50

Ximid Ximid*Fi
157
628
161
483
165
495
169
338
173
2422
177
2655
181
724
185
925
8670

8.) Compute the average, median, mode and quartiles. Are there any differences comparing with the
results from the first point, third point?

To compute and find the location of the mode we choose the interval with the greatest frequency (15) : 175-179

Location for Median:

Quartiles:

Mean=
Mode=
Median=
Q0=
Q1=
Q2=
Q3=
Q4=

Mean, Mo,
Me,
quartiles for
ungrouped
and discrete
fr
distribution:
172,76
173
174
155
172
174
176
187

Mean, Mo, Me,


quartiles for
continuous
frequency
distribution:
Mean=
173,4
Mode=
175,2
Median=
174,71
Q0=
155
Q1=
171,14
Q2=
174,71
Q3=
178,06
Q4=
187

If we compare the average, median, mode and quartiles of the continuous frequency distribution with the results
from the first point, third point we can say that there are differences between them. The mean, mode and median
are totally different, only the minimum and maximum quartiles are obviously equal.

10

9.) Make a histogram.

Histogram
16

14

15

14
12
10
8
6

2
0
155-159 159-163 163-167 167-171 171-175 175-179 179-183 183-187

10.) Compute the mean absolute deviation, the standard deviation and the coefficient of variation.
Mean=
Gr.
1
2
3
4
5
6
7
8
Total:

Interval
155159
159163
163167
167171
171175
175179
179183
183187

Fi

Ximid
4

157

161

165

169

14

173

15

177

181

185

50

173,4
Ximid-mean

|Ximid-mean|

|Ximid-mean|*Fi

(Ximid-mean)^2

(Ximid-mean)^2 * Fi

-16,4

16,4

65,6

268,96

1075,84

-12,4

12,4

37,2

153,76

461,28

-8,4

8,4

25,2

70,56

211,68

-4,4

4,4

8,8

19,36

38,72

-0,4

0,4

5,6

0,16

2,24

3,6

3,6

54

12,96

194,4

7,6

7,6

30,4

57,76

231,04

11,6

11,6

58

134,56

672,8

64,8

284,8

718,08

2888

11

11.) Compare your height with the height of these pupils. If you were a member of this group, what
percentage of them would be smaller than you?
My height is 160 cm. Looking at the data set we can see that there are 5 pupils that have a height lower than 160cm.

10% of the pupils have a height that is smaller than mine (160cm).
12.) Make a conclusion regarding the accumulated not more than 10 sentences. You may compute any
other statistical indicator you may consider necessary for the conclusion.
Analyzing the Continuous frequency distribution we can say that:
Mean (173,4) <Median (174,71) <Mode (175,2) so we have a negatively skewed distribution. This is also seen by
the polygon of frequencies and the histogram which displays a non-symmetrical distribution of the data. . The
distributions has a symmetry modified by extending the tale toward . The coefficient of skewness, which is
-0,7758, also tells us that the distribution has a tale longer toward , also called left hand distribution, or skewed
to the left, because it is close to -1.
The main relative measure of variation is the coefficient of variation. This coefficient provides a comparison
between the magnitude of the deviation and the magnitude of the mean. The coefficient of variation is 33,31%. It is
between 17% and 35% so the mean has a medium level of representativeness and the continuous frequency
distribution has a medium level of homogeneity.
Analyzing the chart below we can say that the majority of the pupils (30%) have a height that is between 175cm and
179 cm and only 4% of them has a height between 167cm and 171cm.

179-183,
8%

155-159,1598% 163,
183187,
10%

6%
163-167,
6%
167-171,
4%

175-179,
30%

171-175,
28%

Nr.
1
2
3
4
5
6
7
8

Interval
155-159
159-163
163-167
167-171
171-175
175-179
179-183
183-187
Total:

Fi
4
3
3
2
14
15
4
5
50

Fi %
8
6
6
4
28
30
8
10
100

12

Potrebbero piacerti anche