Data Processing

Data Processing

Data processing involves input, throughput, and output

mechanisms.
Input involves the responses from the research
instrument by the subjects of the study.
Throughput includes statistical procedures and techniques
Output is the results of the study which are presented in
data matrix form.
Data Processing

In data processing both quantitative and qualitative forms are

involved to arrive at exact analysis and interpretetation of the
results.

A numerical value (quantitative) is useless without descriptive

interpretations (qualitative) of the former.
Data Processing

Data processing consist of the three basic steps,

namely: (1) Categorization, (2) coding, and (3)
tabulation of data.
Data Processing

Categorization of data refers to the

grouping of the subjects under study
according to the objectives of the purposes
of the study.
Data Processing

Coding of data. After the subjects have been

categorized into groups and subgroups, data are next
coded by the investigator. Informations frome the
questionaires, tests, interview schedules, rating scalem
and many others must be transferred into coded items
to facilitate the tabulation of data. The codes must be
numerical or alphabetical. The former is commonly used
due to sufficient number coverage and is fit for the
computer processing.
Data Processing

Tabulation of data . Once the raw have been given

numerical codes, they are ready for tabulation. This is
done by talllying and counting the raw data to arrive
at a frequency distribution and to faciliate in
organizing the in a systematic order in a table or
several tables. Tabulation can be done either manually
or by machine like electric computer or a mechanical
counter-sorter.
Data Processing

In manual tabulation, data are manuakky tallied by

hand. This is applicable if the total number of cases
or subjects is small. But if the total number of
subjects is more than 100, devices like mechanical
counter-sorter or an electric computer is
recommended to speed up the process.
Data Matrix

Presentations of data is usually in tabular form and

data processing output is either quantitative or
qualitative.

Data matrix are useful in analysis and interpretation

because they gice a clear picture of the results of
the study.
Data Matrix

Three types of Data Matrices :

Univariate matrix
Bivariate matrix
Multivariate matrix
Data Matrix

Univariate matrix. A univariate matrix involves only

one variable. For instance, the research problem is to
determine the acceptability of the quality attributes
of milkfish luncheon meat from offal of boneless
milkfish as evaulated organoleptically by 20 trained
panelists using the 9-point Hedonic Scale(9, like
extremely to 1, dislike extremely).
Data Matrix
Data Matrix

Bivariate matrix . The bivariate matrix involves two

variables. For instance, the problem is “CORRELATION
BETWEEN JOB-RELATED PROBLEMS AND JOB
PERFORMANCE OF STAFF NURSES IN PRIVATE AND
GOVERNMENT HOSPITALS IN ILOILO CITY”. In this kind
of problem, staff nurses in IloIo City may be
categorized into private and government hospitals.
Data Matrix
Data Matrix

Multivariate matrix. The multivariate matrix has

three or more variables in the table.
Data Matrix
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 real tables except that only the total number
of variables, total number of cases , and precentages are
presented. Unlike real tables, the frequency ,mean,
percentage, desriptive interpretation and total are shown.
Dummy Tables
Statistical Treatment

It is a must that researchers diagnose the problem

by using the appropriate statistical tool to arrive at
accurate and definite interpretation of results.
Incorrect Statistical Tool
Statistical Treatment

Percentage is incorrect or inappropriate statistical tool to

scale options due to vague interpretation of results.
Univariate Statistical Treatment

The appropriate statistical tool for univariate problem is

the weighted arithmetic mean and the like.
Statistical Treatment

Example:
“How effective is the teaching of Miss Y in Biology
to teacher in education students?” Of the 200
teacher eduction students, 15 said very much
effective or 4;90, much effective or 3;55, effective
or 2; and 40, not effective at all or 1.
Incorrect Statistical Tool

The author read several theses and dissertations in which

the researchers used percentage in scale options (4,3,2,1)
and the like. This is common error. Percentage is
incorrect or inappropriate statistical tool to scale options
due to vague interpretation of the results.
Incorrect Statistical Tool
Univariate Statistical Treatment

The appropriate statistical tool for univariate

problem, both experimental and descriptive
designs, is weighted arithmetic mean for
scales options(i.e., 9,8,7,6,5,4,3,2 and 1)
and the like.
Univariate Statistical Treatment

Experimental research .For instance, the

experimental research problem is “What is the
acceptability of the flavor of fish burger from
offal of boneless milkfish?” Of the 30 panelist who
evaluated the product using the 9-point Hedonic
Scale, % rated like extremely or 9;23 rated like
very much or 8; and 2, likely moderately or7.
Univariate Statistical Treatment
Univariate Statistical Treatment

Descriptive research. In descriptive research,

weighted arithmetic mean is appropriate for scale
options (i.e, 5,4,3,2 and 1) and the like for
univariate research problem. For instance, “ How
serious are the job-related problems met by the
staff nurses in private and government hospitals in
Iloilo City in relation to administration of top
management when classifed as a whole?”
Univariate Statistical Treatment

Of the 200 staff nurses in private and government

hospitals in Iloilo City, 25 staff nurses said very,
very serious or5;50 said very serious or 4;100,
serious or 3;15, less serious or 2;10, not serious at
all or 1.
Univariate Statistical Treatment
Bivariate Statistical Treatment in Experimental
Research

Experimental research. The statistical tool for bivariate

(two variables) problem in experimental research are t-
test and linear correlation. For instance, the
experimental research problem is ”Is there a significant
difference on the man catch of squid using horizontal and
vertical fishing for 21 days operation?”. The appropriate
statistical tool is t-test.
Bivariate Statistical Treatment in Experimental
Research
Bivariate Statistical Treatment in
Experimental Research
Bivariate Statistical Treatment in
Experimental Research
Bivariate Statistical Treatment in
Experimental Research
Bivariate Statistical Treatment in Descriptive
Research

The statistical tools used for bivariate

research problems are z-test and linear
correlation. For instance, the research
problem is “What is the relationship between
capital (X) and profit (Y) of milkfist burger
from offal of boneless milkfish?”
z-test as Bivariate Statistical Tool in Descriptive
Research

z-test between precentages. The z-test is used to

determine the significant difference between two
percentages of related in individuals in which the
data are collected through survey. The z-test
formula as formulated by Ferguson and Takane
(1989)
z-test as Bivariate Statistical Tool in Descriptive
Research

Suppoese the researcher wishes to conduct a study on the job-related

problems met by research-oriented and non-research-oriented Biology
faculty at the Norhthern Iloilo Polytechnic State College. The specific
problem is “Is there a significant difference between the job-related
problems met by research-oriented and non-research oriented Biology
faculty at the Northern Iloilo Polytechnic State College in relation to
instructional materials?’
There were 27 research-oriented Biology faculty members and 18 non-
research-oriented faculty members. Of the 27 research-oriented Biology
faculty members, 27 or 100% considered instructional materials as a
problem and only 13 or 72.22% of the non-research-oriented Biology
faculty members said that instructional material is a problem.
Multivariate Statistical Treatment

Experimental research. The statistical tools

used in multivariate experimental research
problems with three or more variable are F-
test or ANOVA (analysis of variance),
Kruskal-Wallis One-way Analysis of Variance,
and Friendman’s Two-way Analysis of
Variance by Rank.
F-test as Statistical Tool in Multivariate
Experimental Research

F-test or two-way analysis of variance (ANOVA)

involves three or more independent variables as bases
of classification. For Instance, the researchers wishes
to study the effect of fish meal, trash fish, bread
meal,and fish silage as supplemental feeds upon the
growth of grouper on the effect of fist meal, trash
fish, bread meal, and fish silage as supplemental feeds
upon the growth of grouper cultured in fish cages for
three replications.
Friedman’s Two-Way ANOVA as Statistical
Tool of Multivariate Experimental Research

Friedman’s two-way analysis of variance

(ANOVA) is also a statistical tool used
both in experimental and descriptive
multivariate research problems.
Kruskal-Wallis One-Way ANOVA is Statistical Tool
for Multivariate Expermental Research

Kruskal-Wallis one-way analysis of variance

(ANOVA) by ranks is another statistical tool
used in multivariate research problems both
in experimental and descriptive researchers.
Kruskal-Wallis One-Way ANOVA is Statistical Tool
for Multivariate Expermental Research

Descriptive research. The statistical tools

used in multivariate descriptive research
problems are chi-square(X^2), Friedman’s
two-way ANOVA by ranks (X^2), and Kruskal-
wallis one-way ANOVA by ranks(H).
Chi-square (X^2) as Statistical Tool for
Multivariate Descriptive Research.

Chi-square (X^2) represented by dancing X symbol is used

as a statistical tool in multivariate descriptive research
problems. Chi-square test (X^2) are of many types, for
instance, 2x2 table, 2x3 table, 3x2 table, 3x3 table, 3x4
table, 4x4 table, and many others.
Chi-square (X^2) 2x2 table. In chi-square (X^2) test 2x2
table or fourfold table, two discrete variables are involved
to test if these variables are independent from each other.
The variables are usually nominal. The values are arranged
in the form a 2x2 table with two rows and two columns.
Friedman’s Two-Way ANOVA by Ranks as
Statistical Tool Used in Multiveriate Descriptive
Research

• Friedman’s two-way analysis of variance by ranks

is used when the data from K realted samples
consist of at least an ordinal scale and have been
drawn from the same set of observation to
different population. The data is a set of
observation from sample individuals where
observations are made on each group of
individuals under more than two condotions.
Friedman’s Two-Way ANOVA by Ranks as
Statistical Tool Used in Multiveriate Descriptive
Research

The researcher wishes to determine the significant

difference of the adequacy of facilities and
equipment in fishery schools as perceived by key
officials, fishery teachers, and Fishery students.
The research problem is “Is there a significant
difference on the adequacy of facilities and
equipment in fishery schools as perceived by key
officials, fishery teachers, and Fishery students?”
Kruskal-Wallis’ One-Way ANOVA (H) by Ranks
Statistical Tool in Multivariate Descriptive Research
(Tied Observations)

When ties occur between two or more

observations, each observation is given the mean
of the ranks for which is tied. The formula of tied
observations of Kruskal-Wallis one-way analysis of
variance (H) by ranks is the same with
formula(7.13) but there is addition of divisor.
F-test or Two-way ANOVA as Statistical Tool
in Multivariate Experimental Research

F-test or two-way ANOVA is the Statistical tool used

for multivariate experimental research. It involves
three or more independent variables as bases of
classification.
Suppose a researched wishes to determine the
flavor acceptability of fishh offal luncheion meat
from milkfish, sardines, and goatfish. These fish
offal luncheon meat were evaluated by five
trained panelists.