DATA PROCESSING & STATISTICAL

TREATMENT

Group 5
Data Processing

It is a means of converting Involves input, throughput, and output

information either manually or by mechanisms:
machine into quantitative or
qualitative forms for use in research Input involves the responses to the
analysis.
measuring instrument of the
subject of the study.
Once the measuring instruments
have been retrieved, the next activity Throughput includes the statistical
of the researcher is to process the procedures and techniques.
raw data into quantitative and
qualitative forms. They are both
involved in data processing to arrive Output, the result of the study
at precise analysis and interpretation
of the results. which are presented in the data
matrix form.
3 Basic Steps in Data Processing

1 Categorization of Data
refers to the grouping of subjects under study according to the
objectives or purpose of the study.

EXAMPLE:

The main purpose of the study is to determine the problems met by

science and mathematics instructors and professors in a certain
state college in relation to:

a. Administration and supervision The subject may be categorized:

b. Instructional materials 1. as a whole

c. Teacher factor 2. as Science and Mathematics mentors

d. Student Factor 3. as qualified and non-qualified

4. according to performance

5. in terms of experience
3 Basic Steps in Data Processing

2 Coding of Data
is the process of driving codes from the observed data. In qualitative
research the data is either obtained from observations, interviews or
from questionnaires. The purpose of data coding is to bring out the
essence and meaning of the data that respondents have provided.

EXAMPLE:

If the number of categories in variable is 30, then the numerical codes

should have two digits, such as 01,02 - 30
3 Basic Steps in Data Processing

3 Tabulation of Data
• tallying and counting the raw data to arrive at a frequency distribution
and to facilitate in organizing them in a systematic order in a table or
several tables.

• Tabulation can be done either manually or by machine.

Manual Tabulation
Machine
Data are manually
tallied by hand and for Total number of cases
small number of cases is more than 100
only.
Data Matrix

1 Univariate
Presentation of data is usually in
tabular form, and this is called Univariate - involves only one variable
Data Matrix and data processing A variable in univariate analysis is just
output is either quantitative or a condition or subset that your data
falls into. You can think of it as a
qualitative. “category.” For example, the analysis
might look at a variable of “age” or it
might look at “height” or “weight”.
3 types of Data Matrix
Mentor Perceived Adequacy of Instructional Materials
at four Levels of SUC
1. Univariate Matrix Level of SUC Mean
Descriptive
interpretation
2. Bivariate Matrix
1 3.6 Very adequate
3. Multivariate Matrix
2 3.4 Adequate

3 2.9 Adequate

4 2.8 adequate

Scale:
4 – very adequate 2 – fairly adequate
3 – adequate 1 - inadequate
Data Matrix

2 Bivariate Matrix 3 Multivariate Matrix

The Bivariate involves two variables. The Multivariate matrix has three or
more variables in the table.

No. of
Quiz Grade Recitations
Absences
1 100 95

1 95 95

2 88 85

3 88 83

4 80 80

7 65 75

8 63 70

9 60 65

10 50 60
Dummy Tables

Dummy Tables are helpful in preparing for the

data matrix because they are used in planning,
summarizing, organizing, and analyzing the data
on how the different variables differ with each
other.

Dummy tables are almost similar with the real

tables except that only the total number of
variables, total number of cases and percentages
are presented.

Unlike real tables, the frequency, mean,

percentage, descriptive interpretations and
totals are shown.
Statistical Treatment

A statistical treatment is a summary of the

procedure, including statistical methods used. 𝑓𝑥1 + 𝑓2𝑥2+ . . . 𝑓𝑘𝑥𝑘 𝑎 σ 𝑓𝑥
𝑋= +⋯
𝑓1 + 𝑓2 . . . 𝑓𝑘 σ𝑓

where:
The authors read various theses and dissertations in
X = Weighted arithmetic mean
which the researcher used percentage in scales used σ 𝑓𝑥 = Sum of all the products of f and x
similar to that in the foregoing problem. This is a where f is the frequency of each option

common error of researchers. For comparison purposes and x is the weight of each option
σ𝑓 = Sum of all the subjects
on the of correct and incorrect statistical tools used in
organizing the data.
The percentage formula is as follows
𝑓
% = x 100
𝑁

where:
% = percentage

f = frequency

N = Number of cases
Statistical Treatment

SCHEMATIC PRESENTATION

It is not enough that a researcher formulates only one

research problem in his study, but preferably two or more to
be able to test the significant difference between the
responses of the subjects. In scientific research, there is a
statistical tool for every research problem. Independent Variable Dependent Variable

Qualified Mentors
Level of SUCs
Instructional
Non- Qualified Materials
Independent Variable Dependent Variable
Mentors
Science & Level of SUCs
Schematic presentation of Bivariate Matrix Variables
Mathematics Instructional on the Adequacy of Instructional Materials as
Mentors Materials Perceived by Science and Mathematics Mentors in
the different levels of SUCs.

Schematic presentation of Univariate Matrix Variable

On the Adequacy of the Instructional Materials as
Perceived by Science and Mathematics Mentors in
the different levels of SUCs
Statistical Treatment

Of three examples of matrices, it is only in the

Independent Variable
univariate matrix problem that testing of the
significant difference of the response of
subjects cannot be applied, because only one
Salmon Style
variable is involved.
Dependent Variable

Spanish Style Statistical Tools to Test the Significant Difference

Acceptability of of the Variables
Canned Milkfish
Testing the significant difference of variables is applicable
French Style
when 2 or more variables are involved in the study as
Bivariate multivariate matrices.

Japanese Style 2
12
𝑥𝑟 = σR𝑖2 − 3𝑁(𝐾 + 1)
𝑁𝐾(𝐾 + 1)
Diagram of Multivariate Matrix Variables on the
𝑥r2 = Friedman two way analysis of variance
Acceptability of Canned Milkfish in Salmon, Spanish,
N = number of rows
French and Japanese Styles.
K = number of columns
Ri = ranks in the ith columns
Statistical Treatment

To apply the foregoing formula, consider the following steps.

Step 1:

Statistical Tools to Test the Significant Difference Find the squares of samples (SSS) by summing all the data

of the Variables columns to get EYi2 and dividing the sum by the number of
panelists. Subtract this value from the correction factor (CF) to
σ𝑌12
obtain SSS= - CF.
To test the significant difference of the multivariate matrix. 𝑝

Step 2:

Determine the Sum of squares of panelists (SSp) by summing all

𝑀𝑆
𝐹= the rows total (σ𝑌𝑖2) and divide by the number of samples (
σ𝑌12
)
𝑀𝑆𝑒 𝑆
σ𝑌12
and subtract it from the CF to arrive at SSS= 𝑆
- CF .

MS = Mean Square Step 3:

Mse = Mean Square error Find the sum of squares’ total of each score to get SSt and
subtract it from the CF or SSt= σ𝑌12 - CF .
Statistical Treatment

Step 4:

Get the sum of squares of error (SSe) by summing SSs + SSp and
find the difference of this value of SSt or SSe= SSt – (SSS + SSp).

Step 5:
𝑀𝑆
Compute F by using 𝐹 =
𝑀𝑆𝑒
Step 6:

Find the degrees of freedom (df), N – 1, as baseline to determine

the F value is significant or not.

Step 7:

If df has been obtained choose the level of probability either

0.01 or 0.05 and find the tabular value in the F-table
Statistical Tool for Research Design

“ Tostudy,
determine the correct statistical tools to be used in the
the researcher should diagnose carefully the specific Chi-square
problems and the measuring instruments.
” Is used to compare the differences of two or more
variables “with data that can be readily transformed

into frequencies”. (Guilford & Fruchter, 1973)

Some suggested Statistical Tools for Descriptive Design:

Chi – square
Friedman’s two way analysis of variance Formula: 2
2 σ 0 −𝐸
Kruskal – Wallis one way analysis of variance by ranks 𝑥 =
𝐸
Correlation

Pooled percentages where:

𝑥 2 = 𝑐ℎ𝑖 𝑠𝑞𝑢𝑎𝑟𝑒
Z - score 𝐸 = Expected frequency

0 = Observed frequency
Statistical Tool for Research Design

“ The suggested statistical tools for experimental design are

the T-test, Analysis of variance (ANOVA), and correlation.
”
T-test is used to determine two variable
means which differ significantly.

𝑥1 − 𝑥2
𝑡 =
𝑆𝐷12 𝑆𝐷22
+
𝑁1 𝑁2
where:
𝑡 = 𝑇 − 𝑣𝑎𝑙𝑢𝑒
𝐸 = Expected frequency

0 = Observed frequency
DATA ANALYSIS & INTERPRETATION
Data Analysis

Data analysis may be defined as an examination of There are 10 types of Data Analysis
data or facts in terms of quantity, quality, attribute,
1. Univariate Analysis
trait, pattern, trend, relationship among others so
2. Bivariate Analysis
as to answer research questions which involves
3. Multivariate Analysis
statistical techniques and procedure.
4. Normative Analysis
5. Status Analysis
6. Descriptive Analysis
The basis in analyzing research data are
7. Classification Analysis
specific problems/objectives, hypothesis,
measuring instruments and statistical tools. 8. Evaluative Analysis
9. Comparative Analysis
10. Cost Effective Analysis
10 types of Data Analysis

3 Multivariate Analysis
This type of analysis tests three or more independent
1 Univariate Analysis variables at a time on the degree of relationship with
the dependent variable.
This type of analysis tests a single variable whether
the variable is similar to the population from which Statistical tools which may be used in this type are:
it has been drawn.
Analysis of Variance (ANOVA) for experimental design
Friedman two-way ANOVA including Friedman and
Kurskal-Wallis ANOVA and Chi-square for descriptive
design.

2 Bivariate Analysis Bear in mind that Friedman and Kruskal-Wallis

Bivariate Analysis tests two variables on how ANOVA are applicable both for experimental designs.
they differ with each other.

Common statistical tool to be used in Bivariate Analysis are:

4 Normative Analysis
• Correlation Coefficient (used in descriptive and experimental design)
• z-test The result of the study is compared with the norm.
(used in experimental design) the statistical tool which may be used for this type
• t-test
are arithmetic mean and standard deviation.
10 types of Data Analysis

7 Classification Analysis
5 Status Analysis
This type of analysis is usually employed in natural
Stresses real facts relating to current conditions in a
science subject such as Botany, Zoology, Biology,
group of subjects chosen for study. The common
Phycology, Ichthyology, Conchology and the like. The
statistical tools which may be used in this type are,
statistical tools may be used in this type are mean and
arithmetic mean, standard deviation and z-test, for
t-test are used to test the significant difference of the
the difference between means.
means.

6 Descriptive Analysis 8 Evaluative Analysis

Is the type of analysis appraises carefully the worthiness
Merely describes the characteristics, composition,
of the current study. The statistical tools that commonly
structures, substructure, that occur as units within
used in this type are, the weighted arithmetic mean,
the larger structure. The statistical tools commonly
percentages, Friedman’s two-way ANOVA and z-test.
used in this type are the arithmetic mean, chi-square
and Friedman two-way ANOVA.
10 types of Data Analysis

9 Comparative Analysis 10 Cost Effective Analysis

In this type of analysis, the researcher consider at In this type of analysis is applicable in comparing
least two entities (not manipulated) and stablishes a the cost between two or more variables and to
formal procedure for obtaining criterion data on the determine which of the variable is most effective.
basis of which he can compare and conclude which
of the two is better. In other words cost-effective analysis (CEA), the
researcher should determine which of the variable is
The common statistical tool used in this type are, more economical and effective.
mean, variances and t-test.
 𝑀𝑆
𝐹=
𝑀𝑆𝑒

THANK YOU!