Professional Documents
Culture Documents
and STATISTICAL
TREATMENT
Forms Of Data
Processing
Example:
How effective is the teaching of Professor
Snape in Mathematics to Electrical
Engineering students?
f
x
fx
20 x 4 = 80
X= fx/f
30 x 3 = 90
X=260/100
40 x 2 = 80
X= 2.6 or
3
10 x 1 = 10
(much
effective)
Total: 100
260
DATA PROCESSING
Converting information either
manually or by machine into
quantitative and qualitative forms.
Coding
Categorization
Tabulation of
Data
DATA MATRIX
Presentation of data usually in
tabular form.
Gives picture of the results of the
study.
Univariate Matrix
Involves only one variable.
Quality
Attributes
Color
7.75
Odor
8.15
Flavor
8.30
Texture
8.00
General Acceptability
8.20
Scale :
9- Like extremely
8- Like very much
7- Like moderately
6- Like slightly
5- Neither like nor dislike
4- Dislike slighty
Bivariate Matrix
Involves two variables.
JOBRELATED
PROBLEM
S
STAFF NURSES
PRIVATE HOSPITALS
Mean
Interpretation
GOVERNMENT
HOSPITALS
Mean
Interpretation
2.1
3.2
Serious problem
3.3
Serious problem
3.0
Serious problem
4.4
3.4
Serious problem
3.9
4.2
2.0
SCALE :
5- Very very serious problem
4 Very serious problem
3- Serious problem
2- less serious problem
Multivariate Matrix
Has three or more variables in the
Quality
Luncheon Meat
table.
Attribut
es
Milkfish
Offal
Mean
Goatfish
Offal
Mean
Siganid
Offal
Mean
Sardines
Offal
Mean
Color
7.7
7.9
7.5
7.1
Odor
8.0
8.0
7.3
7.2
Flavor
8.4
8.2
7.9
7.6
Texture
8.1
8.0
7.8
7.7
General
8.3
Acceptability
8.0
7.7
7.5
Scale:
9- Like extremely
8- Like very much
6- Like slightly
5- Neither like nor dislike
DUMMY TABLES
Used in planning, summarizing,
organizing and analyzing the data on
how the different variables differ with
each other.
Job
Performa
nce
Hospitals
Private
Frequen
cy
Government
Percent Frequen
cy
Percent
Total
Freque
ncy
Percen
t
Outstandin
g
Very
Satisfactor
y
Unsatisfact
ory
Total
175
100
125
100
300
100
STATISTICAL TREATMENT
Using Arithmetic mean in scaling.
4
3
2
1
INCORRECT STATISTICAL
TOOL
Percentage in scale options (4 3 2 1)
is incorrect or inappropriate
statistical tool to scale options.
7.5 % - very much effective
45.0 % - much effective
27.5% - effective
20.0% - not effective at all
UNIVARIATE
STATISTICAL
TREATMENT
MIGUEL, NORUEL JAN G.
Arithmetic Mean
The appropriate statistical tool for
Univariate problems
Example:
(9,8,7,6,5,4,3,2 and 1)
Experimental Research
Example Problem:
What is the acceptability of the
like extremely
5 votes
like very much
23 votes
like moderately
2 votes
like slightly
neither like or dislike
dislike slightly
Where:
Weighted arithmetic mean
Sum of all the products of f and x; where f is the
frequency of each weight and x is the weight
Sum of all the frequency/subjects
Given:
5
23
2
9
8
7
Solution:
Total
f
5
23
2
30
x
9
8
7
fx
45
184
14
243
Quantitative mean
Qualitative description
(like very much)
Descriptive Research
Example Problem:
Of the 200 staff nurses in private and
Given:
25
50
100
15
10
5 2
4 1
3
Solution:
(serious)
Total
f
25
50
100
15
10
200
x
5
4
3
2
1
fx
125
200
300
30
10
665
Quantitative
mean
(serious)
Qualitative description
BIVARIATE
STATISTICAL
TREATMENT
Experimental Research
Statistical tools are:
* t-test
* linear correlation
The t-test
formula:
Where:
- mean of the first variable - mean of the second variable
- variance of
- variance of
- total number of operations
- total number of operations
of first variable
of second variable
Example:
Linear Correlation
Where:
Step
Step
Step
Step
Step
2.
3.
4.
5.
6.
Example:
Descriptive Research
Statistical tools are;
* Linear Correlation
* z-test
Linear Correlation
Where:
- Spearman rho
- Sum of the squared differences between
ranks
N - Number of cases
The z-test
Where:
Z z-test
- Percentage of first group of subjects or first variable
- Percentage of second group of subjects or second variable
- Pooled percentage of and
Q =1-P
- Number of cases for the first variable
- Number of cases for the second variable
MUTIVARIATE
STATISTICAL
TREATMENT
Multivariate
Experimental Research
F-test or ANOVA (analysis of variance);
Kruskal-Wallis One-way Analysis of
Variance, and;
Friedmans Two-way Analysis of
Variance by Ranks.
F-test as Statistical
Tool in Multivariate
Experimental
Research
Supplemental
Feeds
Replication
s (kg)
Tot
al
1
2
3 (k
g)
18
10
24
15
12
36
16
13
13
42
Where:
F
=F-test
MSR
= Mean Square for
Replication
MSTrt = Mean Square for Treatment
MSE = Mean Square for Error
Step 1.
Partition of sum of
squares for replication,
treatment, error, total
by using the
appropriate formula.
Sum
of Squares for
Replication ()
SSR = - CF ; CF =
Where:
SSR = Sum of Squares for Replication
= Sum of the squared total of each
Replication
Trt = Number of Treatment
CF = Correction Factor
Sum
of Squares for Treatment Formula ()
SSTrt = - CF
Where:
SSTrt
Sum
of Squares for Total () Formula
SST =
- CF
Where:
SST
= Correction Factor
Sum
of Squares for Error Formula()
= -(+)
Where:
= Sum of Squares for Error
= Sum of Squares for Total
= Sum of Squares for Replication
= Sum of Squares for Treatment
Step 2.
Divide the sum of squares
for replication, treatment,
total and error with their
corresponding degrees of
freedom, N1, to get the
mean squares.
Formula:
MS =
Where:
MS = Mean Square
SS = Sum of Squares
df = Degrees of Freedom
REPLICATION
(M)
M=
TREATMENT (M)
M
ERROR (M)
M=
Step 3.
Step 4.
Determine if the computed F-value is
significant, if the computed F-value is
equal or greater than the tabular Fvalue; and not significant, if the
computed F-value is less than the
tabular F-value.
CV > TV = significant
where:
CV = computed value
5.14
4.76
10.92
9.78
dfE =
Step 5.
Prepare the F-test or ANOVA
Table by entering the values in
Steps 1, 2, and 3.
Source Degree Sum Mean Obser
of
s of
of
Squar ved F
Varianc Freedo Squa
es
e
m
res
Tabula
rF
1%
5%
Replicat
ion
32
16
12.03** 10.92
5.14
Treatme
nt
120
40
Error
1.33
30.075 9.78
**
4.76
Formula:
=
Where:
= Freidmans two-way ANOVA
by ranks
= Sum of the ranks
N
= Number of rows
= Number of columns
Methods of teaching
Subje
cts
FR
FR
FR
FR
1
2
3
4
5
6
7
8
9
10
90
88
85
92
87
95
80
75
83
77
3
4
3.5
3
3
3
3
3.5
4
3
85
83
80
87
82
90
72
70
80
75
1
1
1.5
1
1.5
2
1
1.5
2
1
91
87
85
93
88
96
81
75
82
79
4
2.5
3.5
4
4
4
4
3.5
3
4
87
87
80
91
82
91
75
70
79
76
2
2.5
1.5
2
1.5
1
2
1.5
1
2
Total
SA
USA
33.0
TPS
13.5
DA
36.5
17.0
df = K-1 = 3
df.01(3) = 11.34**
another
statistical tool used in multivariate
research problems both in experimental and
descriptive researches.
H = - 3(N-1)
Where:
H = Kruskal-Wallis analysis of variance by
ranks
N = Number of cases in all samples combined
n = Number of cases in each sample
= Sum of ranks in each column
1
1
2
2 3
Wt. Rank
Wt. Rank
Wt. RankWt.Wt.
Rank
Rank
43
5
4
Wt.
Wt.Rank
Rank Wt. Rank
Wt. Rank
3.3
3.0
3.1
3.2
3.4
3.5
3.6
1.1
3.9 10
38
2.8
4.1 540
0.7
3.8 637
0.8
3.7 736
0.5
4.0 439
0.9 8
0.4 3
0.2 1
0.3 2
1.0 9
3.3
32
3.0
29
3.1
30
3.2
31
3.4
33
3.5
34
3.6
35
32
1.2
29
1.4
30
2.1
31
2.4
33
2.6
34
1.3
35
2.2
1.8
2.3
11
13
20
23
25
12
21
17
22
1.23.9
11
38
1.44.1
13
40
2.13.8
20
37
2.43.7
23
36
2.64.0
25
39
1.3 12
2.2 21
1.8 17
2.3 22
2.9
1.7
2.8
1.9
1.5
1.6
2.0
2.7
2.5
1.1
28 10
16
2.8 5
27
0.7 6
18
0.8 7
14
0.5 4
15
0.9 8
19
0.4 3
26
0.2 1
24
0.3 2
1.0 9
Chi-square
() as a Statistical Tool for
Where:
= Chi-Square
O = Observed frequency
E = Expected frequency
Steps
in solving
1.
Step
O = R+C
Step 2.
E = [(R)(C)]/N
Step 3.
OE
Step 4.
Step 5.
Another
formula for 2 x 2 table
Where:
=
Chi-square
N = # of rows
L = observed frequency of cell L
P = observed frequency of cell P
M = observed frequency of cell M
C = observed frequency of cell C
P L + P
M +
M C
C
L + P +
M + C
FRIEDMAN
S TWO-WAY
ANOVA
B Y R A N K S A S S TATI S T IC A L TOO L U S E D I N
M U LTI VAR IATE D E S C R I P T IV E R E S E A RCH
BY: RONNIEL JAY MILLAN
FORMULA:
Where:
Xr2
= Number of rows
K
= Number of columns
= Sum of ranks
EXAMPLE
Three different groups of subjects exposed
to the same set of observations on the
adequacy of facilities and equipment in
fishery schools as perceived by key
officials, fishery teachers, and students.
TABLE:
Facilites and
Equipment
Key Officials
X
FR
Fishery Teachers
X
FR
Students
1.Fishing ground
2.Aquaculture
apparatus
3.Fish Capture
4.Fish Processing
5.Lighting Facilities
6.Water Facilities
7.Ventilation
8.Classrooms
9.Laboratory rooms
10.Aquaculture
equipment1
11.Fish capture
equipment
12.Fish Processing
R1
3.18
2.27
2
3
3.20
2.20
3
2
3.11
2.15
1
1
1.54
1.64
2.36
2.18
3.09
3.36
3.27
2.36
1
1
1
1
3
2
2
3
1.63
1.65
2.47
2.33
3.08
3.37
3.20
2.22
3
2
3
3
1.5
3
1
1
1.57
1.88
2.37
2.24
3.08
3.31
3.295
2.25
2
3
2
2
1.5
1
3
2
1.73
1.98
1.82
2.91
1
21.0
2.96
2
27.5
3.18
3
23.5
FR
SOLUTION:
= [212 + 27.52 + 23.52] 3(12)(3+1)
[441 + 756.25 + 552.25] 36(4)
= (1749.5) 144
= 0.08833 (1749.5) 144
Xr2 = 1.79 Not Significant
df
= K1
; df = 3 - 1
df
=2
; df0.1(2) = 9.21**
KRUSKALWALLIS
ONE-WAY
ANOVA (H)
B Y R A N K S A S S TATIS T IC A L TOO L I N
M U LTI VAR IATE D E S C R IP T IV E R E S E A RCH
(T IE D O B S E RVATIO N S )
KRUSKAL-WALLIS
ONE-WAY ANOVA (H)
-is a rank-based nonparametric test that
can be used to determine if there are
statistically significant differences between
two or more groups of an independent
variable on a continuous or ordinal
dependent variable.
FORMULA:
Where:
T = t3 t (t is the number of tied observations in a tied
group of observations)
N = Number of observations in all K samples as a
whole
T = Sum of all group of ties
EXAMPLE:
Teaching-oriented
Teachers
Administration-oriented
Teachers
Research-oriented Teachers
Performance Rank
Performance Rank
Performance Rank
8-25.5
6-9.0
10-40.5
2-1.0
8-25.5
8-25.5
6-9.0
10-40.5
8-25.5
8-25.5
6-9.0
6-9.0
6-9.0
8-25.5
8-25.5
6-9.0
10-40.5
8-25.5
6-9.0
4-2.0
391.5
10-40.5
8-25.5
6-9.0
8-25.5
6-9.0
8-25.5
10-40.5
6-9.0
6-9.0
8-25.5
8-25.5
6-9.0
10-40.5
8-25.5
6-9.0
10-40.5
8-25.5
8-25.5
8-25.5
10-40.5
8-25.5
8-25.5
10-40.5
8-25.5
10-40.5
R1
R2
328.5
R3
315.0
Legend:
10 - Outstanding
8 - Very Satisfying
6 - Satisfactory
4 - Fairly satisfactory
2 - Unsatisfactory
COMPUTATION OF TIE
SCORES:
(25.5) T = t3 t
= 203 20
T = 7980
(9.0) T = t3 t
= 133 13
T = 2184
(40.5) T = t3 t
= 103 10
T = 990
COMPUTATION OF
FORMULA:
0.87754
SUBSTITUTING OF
FORMULA OF H-TEST
Not Significant
df
=k1
=31
=2
df0.1(2) = 9.21**
F-TEST OR
TWO-WAY
ANOVA
A S S TATIS T IC A L TOO L I N M U LTIVAR IATE
E X P E R IM E N TAL R E S E A RCH
F-TEST OR TWO-WAY
ANOVA
-is the statistical used for multivariate
experimental research. It involves three or
more idependent variables as bases of
classification.
FORMULA:
Where:
F
= F-test
= Number of panelists
CF
= Correction factor
= Number of sample
= Correction factor
= Correction factor
EXAMPLE:
Fish Offal Luncheon Meat
Panelists
Total
Milkfish
Sardines
Goatfish
22
24
25
21
22
Total
39
37
38
114
SOLUTION:
Computation of Sum of squares for Samples Formula (SS S)
SSS =
=
=
=
SSS =
SST = 5.6
Source of
Variance
Degrees of
freedom
Sum of
Squares
Mean Square
Observed
F
Tabular F
1%
Samples
0.4
0.2
1.0
8.65**
Panelists
3.6
0.9
4.5
7.01**
Error
1.6
0.2
Total
14
5.8
THANK YOU
AND GOD
BLESS