Data Collection and

Analysis Procedures
Writing the Data Gathering Procedure

The data gathering procedure shows the

exact details on how you would be gathering
your data from the first step of soliciting permission
to conduct the study from different institutions up to
retrieval of your instrument.

The data gathering stage should start by writing

a letter addressed to the person in command in the
institution where you would be conducting the study.
Also mention whether you wish the study to be
endorsed to other institutions or individual who
might help you out in the administration. The letter
should indicate the purpose by which you are
conducting the study and an assurance that all
information gathered from them shall be dealt
with strict confidentiality according to the
 You should also indicate in this part how
you would be gathering the data, that is, if you
are using the questionnaire, and whether
you would be administering the instrument
yourself or you would be hiring enumerators,
or whether you would be floating the
questionnaires during a meeting so that
everybody would all be gathered, etc.
You should also mention if you would be
getting an informed consent from the
Respondents of the study.
 For the experimental research, you
should present a very elaborate data
gathering procedure, where you would be
surer that you would be gathering the needed
data for your study, from the first step to the
last . This data gathering procedure from
Encisco (2014 ) shows you a detailed
presentation:
QUALITATIVE
DATA
ANALYSIS
 An albino rat was used as the subject. An
incision was made on the body of the rat.
Observation was made on the body of the rat. For 5
to 7 days, there was a formation of abscess around
the incised wound of the rat . Then on the seventh
day, using sterile cotton swab, the abscess was
scrapper and which was then placed on a
thioglycollate broth culture media. Thioglycollate
broth was made by mixing distilled water which is
45 ml added with the thioglycollate broth reagent
and is then placed in a graduated cylinder which
was autoclaved for 1 to 3 hours for sterilization.
Each culture tubes contain 3 to 5 ml of
thioglycollate broth preparations. In this proposal,
the researcher used 4 ml of thioglycollate broth
preparations on 2 separate tubes.
 The cotton swab scrapings were placed on
the thioglycollate broth preparations on 2
separate tubes ( 2 culture tubes and 2 pieces
of different cotton swabs) and then the
researcher incubated it for 24 hours.

On the next day, the thioglycollate broth

preparations were observed as turbid because
of the growth of microbes in the culture media
compared on the past day which was clear
because of no signs of growth. Turbidity is
caused by the dipped swabs containing the
samples.
 And then from the thioglycollate broth
preparations, the researcher acquired some
samples using a disposable wire loop on 1
blood agar plates and on 1 mcConkey agar
plate. Using 1 loop, the researcher dips it unto
the thioglycollate broth and planted it on the
blood agar plate. On a slanting position, the
researchers press the end of the loop on a
circular motion and then slide it downwards.
Same technique was done on the McConkey
agar which was also press on a circular motion
for the primary streaking culture using the
same end of the wire loop.
 Primary streaking is done on both blood agar
plates and macConkey agar plates. And after that,
the researcher disposed the loop on containers
which had a biohazard sign for infection control.

Using another loop, the researcher streaked

from the primary culture from the marks of the
circle, the researcher stroked the loop from left to
right going downwards not too far away from each
other. Same technique was done on macConkey agar.
The samples were planted on the blood agar and
macConkey agar plates to view the comparison of
growth of the specimen on each agar plates. Then
the researcher did not yet discard the loop used on
the primary culture because its other half will going
to be used on secondary culture.
 Using other half of the loop, the
researcher streaked it unto the next side of
the plate beside the primary culture of the
blood agar plate same as the primary
culture , the secondary culture starts from
left to right motion. The researcher made
sure that the secondary culture does not
touch the primary culture to avoid
contaminants and to acquire the poor culture
being isolated from the experiment. The
movement of streaking is from left to right
motion was not too close from each other for
the reason of acquiring the pure culture of
bacteria being isolated from the experiment.
And then the researcher discarded the used
 And then the researcher used another
loop for the tertiary culture. From left to
right motion, the researcher isolated some
colonies from the secondary culture of the
blood agar plate, the researcher press the
wire loop and pressed it moving from left to
right. The tertiary culture was not close to
close to each other to observe the pure
bacterial colonies. Some technique was
done the macConkey agar plates. Each of
the primary , secondary and tertiary
cultures are made sure that they will not
touch each other to avoid contaminants and
to obtain the pure cultures being tested on
this proposal.
 After the streaking procedure, the
researcher incubated the agar plates in the
incubator in room temperature for 24 hours.
On the next day, there is an observable
growth on the blood agar plates specifically
on its primary, secondary and tertiary
cultures. There is also noted growth on the
macConkey agar plates primary, secondary
and tertiary cultures. On the blood agar
plates, there is a grayish to yellowish color of
colonies observable on its primary, secondary
cultures. There is also a noted alpha plates.
Alpha hemolytic is the greenish background
around the colonies.
 And then the researcher isolated
from the blood agar plates and isolated
it using a disposable loop and planted it
on the Mueller Hinton Agar for the
experimentations of the different
preparation of oregano.
 From the blood agar plates, the
researcher isolated some pure colonies. Pure
colonies of microorganisms are acquired from
the secondary and tertiary pure cultures.
Using a disposable loop , the colony is
acquired from the secondary and tertiary
cultures and is streaked on Mueller Hinton
Agar. The researcher drawn the loop from
upwards to downwards and then on the center
of the plate from left to right is also drawn
then in the upper part of the plate, the
researcher streaked the loop from left to right
going downwards on a vertical position. And
then streaked it again on the on a diagonal
position going downwards.
 And on its end part was streaked
on the side of the plate on a horizontal
position going downwards. After streaking
on different positions, the researcher
streaked it on the side of the plate in a
circular motion. The researcher made sure
that the streaking on different angles of the
plates were too close to each other for the
better observation of zones of inhibitions
once the antimicrobial disc were placed from
different preparations of oregano. And then
the researcher incubated the plates for 24
hours.
Lastly, you should mention the
procedure you would be undertaking
after you have retrieved a substantial
number of the questionnaire according
to your Sampling design, like tallying,
and subjecting the data using
statistical tools.
Below is a sample data gathering procedure for Descriptive Research
from Dominado (2011):

Upon the approval the questionnaire, the

researcher sought permission from the Regional Director of the
Commission on Higher Education (CHEd) to conduct the study.
The study was then endorsed by the regional director to the
presidents of the private colleges in the region. The assistance of
the Presidents of the colleges was sought to facilitate the floating
of the questionnaires and to conduct the interview. To personally
answer which arose during the administration of the
questionnaire, the proponent administered the instruments
personally. Questionnaire were retrieved after two weeks.
Although there were cases of missing and unreturned
questionnaire particularly from the faculty members, the
researcher manage to retrieve a number enough to be
considered valid according to standards. As for the students
administration of the questionnaires was done during their
classes so that 100% retrieval was achieved. The data gathered
from the administrators, faculty members and students were
checked and tallied and was made ready for presentation,
Data encoding

After the administration of the questionnaire,

the next thing to be done is the categorization of
data, which will be used later in data coding. This
will facilitate tallying, if done manually or encoding,
if done electronically.

To categorize data, responses in each variable in

the questionnaires must be analyzed. If for example
you asked for the civil status of the Respondents ,
possible answers would be single, married, single
parent, annulled, divorced, separated, abandoned
etc. while the highest educational attainment would
be no formal schooling, elementary undergraduate,
elements graduate, etc. Remember to include all
responses under a category.
 Some variables would automatically use the
responses as categories. However, there might be
cases where you should put several responses under
one category. For example, in listing down the
occupation of 300 respondents, you might arrive at
more or less 100 categories. Thus, there is a need
for you to categorize the occupations according to
standard grouping as used by organizations or
institutions. For example, the United Nations group
occupations into managers, professionals,
technicians, and associate professionals, clerical
support workers, service and sales workers, skilled
agricultural, forestry and fishery workers, craft and
related trade workers plant and machine operators
and assemblers, elementary occupations and armed
forces occupations (United Nations, 2008).
 You might need to do some readings to determine
categories of variables that you used in your study.
After categorization, the next step is for you to
prepare a coding sheet, where nominal data would be
assigned numbers ready for tallying or coding. You
have to make sure that each variable in the specific
questions and presented in the questionnaire be
categorized and assigned code. Here is a sample
coding sheet:

For profile of the respondents

Age: as is
Gender: Male:
Female:

Socio-economic Status: as is
QUANTITATIV
E DATA
ANALYSIS
Data Analysis using Statistics and Hypothesis Tests

Guidelines in Deciding the Most Appropriate Statistical Tool

An analysis of the kind of the data you are dealing with

and the type of questions which you are expected to answer
should guide you in deciding the appropriate statistical tool for
your study.

Data may be nominal, ordinal or interval. A nominal data

is also called categorical data since respondents are grouped in
categories with or without numerical or quantitative equivalent
except when counted in totality and put in a frequency table.
Example is when you group respondents into category of gender,
and count the total numbers say, male= 23 and female= 37.

An ordinal data has a numerical equivalent which assigns a

particular attribute of respondents into orders or ranks.
Academic performance is an ordinal data as a grade of 75 may
be assigned in the lowest rank and 98 to be in the highest rank.
Another example is age as the respondent with an age of 21
 An interval data is considered as both nominal and
ordinal data. Respondents are categorized or grouped
and assigned ranks or orders afterwards which may
be higher or lower, smaller or bigger, etc. Example of
the interval data is grouped monthly income of
respondents as follows:

Monthly Income of the Respondents

PhP5000-10000
PhP10001-15000
PhP15002-20000
PhP20001-25000
PhP25001-3000
Tools for Profile of the Respondents

There are two most common tools used in descriptive part

of the quantitative research ( the specific questions represented
in the boxes of the paradigm), the frequency and percentage
distribution and the measures of central tendency.

The frequency distribution is simply the total count for

each group or category of the variables while percentage
distribution is the fractional part of the category in relation to
the totality of the respondents.

The measures of central tendency are used for the general

characteristics of the group, that is if you would like to describe
your data with one generalization. Although there are three
measures of central tendency, namely, the mean, median and
the mode, in actual practice in research report writing , the
mean and the mode are most oftentimes used.
 The mean is usually referred to as the arithmetic
average, computed by means of dividing the sum of
individual scores by totality of the scores.

The mode is the most frequently occurring score

and usually used by mere inspection of the data rather
than computation . This is where the thesis writer uses
the words “majority” or “most” as the case may be to
describe the data.

In ordinal and interval data, there is a need to use

both the frequency and percentage distribution and the
measures of central tendency, like in age, where you also
need to compute for the average to arrive at a general
description of your data.
Tools for the Correlational Part of the Research

A common mistake of research report writers and

statisticians themselves is applying a correlation tool without
analyzing the kind of data that they are dealing with. You
might be correlating or associating two nominal data, or
two ordinal data or you might be correlating a nominal
data to an ordinal data or vise-versa, and this difference in
the computation and the resulting conclusion if the wrong tool
was used.

If you are correlating two nominal data, you may

use the Chi-square Test of Goodness Fit, the Chi-square
Test of Independence or the Lambda Coefficient .

If you would like to determine significant relationship

or association between two ordinal variables, the Pearson-
Product Moment Correlation, the Chi-square Test of
Goodness of Fit, the Spearman Rank Order Correlation
and the Gamma Coefficient would be the best measures.
 To determine the significant correlation or
association between an ordinal data and a nominal
data, the Chi-square Test of Goodness Of Fit, the
Regression Analysis and the Point-Biserial Coefficient
of Correlation would be the best tools to use.

Given the several options in solving for the

correlation based on the different kind of data, as in
other mathematics problems, the rule is always to use
what ever is the most reliable and convenient to
the user without sacrificing the quality of the output .
There may be a short and long methods of
computations, but at the end of the day, it is answer to
the specific questions and resulting conclusion and
recommendations that matter.
 It is also important to note that rules on the use and
choice of the correlation tools apply in all modes of computation,
whether manual or electronic computation using computer
softwares. This is emphasized because when you assign numeric
codes to nominal data for electronic computation and solve or
correlation coefficient, the computer will give you an answer
which end with a fallacious interpretation.

Example , let us correlate gender to the number of hours

spent in social networking. With a code of 1 for male and 2 for
female, and using the correlation-r in the SPSS, you might arrive
at a correlation ratio of -.45. This is interpreted as inverse
correlation which means that as the number of hours spent in
social networking increases, gender decreases, or as the number
of hours spent in social networking decreases, gender increases,
which is a fallacy. The computer gave you an answer based on
your input but you arrived at a ridiculous interpretation, because
you used a wrong correlation tool.
Tools to Determine Significant Differences ( Comparison
Data)

Groups may be compared to determine significant differences.

Statistically, this may be done using the t-test or the z-test, the
one-way analysis of variance or the two way analysis of
variance.

The t-test is used to determine if there is a significant

difference in the averages of two dependents or independent
groups.

An independent-samples t-test is used when you want to

compare the differences in the mean scores on some continuous
variable for two different groups of subjects.

Would there be a significant difference in the self-esteem of males

and females?

Do you want to compare the academic performance of children

with normal weight and obese?
 You may also compare individual persons’ responses
using two different measures or questions that have the
same scale of measurement like the Likert Scale of 1 to 5.

Do you like to compare the job satisfaction levels and the job
performance of your respondents?

Is teaching performance comparable to attitude towards the

teaching profession?

The one way analysis of variance (ANOVA) is used to

determine if there is a significant difference in the averages or
means of two or more groups within a variable. The role of the
analysis of variance simplifies what would have been a
repetitious job for the T-test/Z-test. If you were to determine
the significant difference of three groups, you would need three t-
test computations, one of the first two groups, another for the 2nd
and 3rd group and the last for the 1st and the 3rd group. So, three
computations which is done once in the analysis of variance.
 Example would be: Is there a significant
difference in the mathematics scores of the first year
college students coming from local, private and state
colleges?

The Two-way Analysis of Variance determines

the significant differences between groups of a
particular variable in relation to another variable.
Thus, two variables are actually involved. An example
would be: Is there a significant difference in parenting
style when grouped according to age?
Tools for Experimental Research

As per definition, the experimental research compares

attributes of a controlled and experimental group.
Therefore, you would also be comparing two dependent or
independent groups as you would when using the t-test or the
z-test and the one way analysis of variance.

With the dependent sample t-test, you only have one

group and you collect data from each member of the group on
two different occasions or conditions. Examples are in
experimental research where you would like to assess the
respondents at the beginning (pre test) and at the end of an
intervention activity (post test).

If for example you would like to determine if there has been

significant improvement in the test scores of the experimental
group after undergoing a 40-hour remedial intervention classes,
the t-test would be most helpful to arrive at a conclusion
 You may also use the paired sample t-test in
experimental designs with a controlled and
experimental group where one group is exposed to an
intervention, or the groups are exposed to two different
intervention methods. Is there a significant difference
in the weights of obese children after group 1 was
exposed to a regular physical activity program and
group 2 after undergoing a strictly vegetarian diet?
Who among the two groups of students performed
better, group 1 after undergoing the outcomes-based
teaching learning design, or group 2 who were
subjected to the traditional teacher centered learning
method.

To sum this section up, a matrix showing the

specific questions and the corresponding statistical
tools is presented below.
A Table Showing the Corresponding Statistical Tool for each Specific
Question in the Statement of the Problem.

Specific Question Statistical Tool

1. What is the profile of the
respondents in terms of:
1.1 age; Frequency and Percentage
distribution and mean
1.2 gender; Frequency and Percentage
distribution
1.3 highest educational attainment;
Frequency and Percentage
1.4 socio economic status; distribution

1.5 occupation of parents; Frequency and Percentage

distribution and mean
1.6 number of siblings Frequency and Percentage
distribution
1.7 ordinal birth order?
Frequency and Percentage
distribution and mean
Frequency and Percentage
distribution
2. How may the job performance Frequency and Percentage
of the respondents be described distribution and Weighted
by their peers, by themselves, Arithmetic mean
and their immediate superiors?

3. What is the job satisfaction Frequency and Percentage

levels of the respondents as distribution and Weighted
described by themselves? Arithmetic mean
4. Is there a significant
correlation between job
satisfaction levels and the profile
of the respondents?
Pearson product moment correlation, Chi-
4.1 job satisfaction and age; square, point-biserial or regression analysis
Chi-square, point biserial or regression
4.2 job satisfaction and gender; analysis

4.3 socio economic status; Chi-square, point biserial or regression

analysis
Pearson product moment correlation, Chi-
4.4 monthly salary: square, point-biserial or regression analysis
Pearson product moment correlation, Chi-
4.5 length of service; and square, point-biserial or regression analysis
Chi-square, point biserial or regression
4.6 position designation (or is there a analysis
significant difference job satisfaction of
the respondents when grouped T-test or analysis of variance depending on
according to their profile? the number of groups

5. Is there a significant (Any one of these)

correlation between job Chi-square test of Goodness of Fit,
satisfaction and job Point Biserial Coefficient of
performance? correlation, Regression analysis,
Pearson product moment correlation
 For quantitative research, you will need a scoring
system which should form part of the statistical
design. The most commonly used is the Likert (1932)
Scale where respondents are asked to respond to a
series of statements about a topic, in terms of the
extent to which they agree with them or the
frequency by which these happen to them.

An understanding of the scoring system and the

Descriptive interpretation that comes with it will
also help the thesis writer in the statistical
treatment. Look at the example below:
The format of a typical five level Likert item could be:

Scale Range Qualitative Description

5 4.21-5.00 Strongly Agree
4 3.41-4.20 Moderately Agree
3 2.61-3.40 Neither agree or disagree
2 1.81-2.60 Moderately disagree
1 1.00-1.80 Strongly disagree

Some would get rid or the middle part “neither agree nor disagree” or
“sometimes” making it a scale of 1 to 4, so that final form becomes

Scale Range Qualitative Description

4 3.25-4.00 Strongly Agree
3 2.50-3.24 Neither agree or disagree
2 1.75-2.49 Moderately disagree
1 1.00-1.74 Strongly disagree
Another Example :

Scale Range Qualitative Description

4 3.25-4.0 with very high chance or possibility
3 2.5-3.24 with moderate chance or possibility
2 1.75-2.49 with very little chance or possibility
1 1.0-1.74 no chance at all/ not possible

Below is a sample Statistical Treatment of Data taken from the proposal

of Mallari at al (2014)

Data to be gathered will be organized, tallied and processed using the

Statistical Package for the Social Sciences ( SPSS) with the following
Statistical tool:

1. For the perceptions and practices on energy and water conservation:

Frequency and Percentage Distribution and Weighted Arithmetic
Mean.
2. For the significant effect of the perceptions on the practices on energy
and water conservation: Chi-Square
3. For the significant effect of the personal related variables to the
perceptions and practices of the respondents: Chi-Square
 Some schools require that the formula for
each of the Statistical tool should be included in the
write up. Other schools would just require the tool
without the formula especially in cases where an
electronic computations like the SPSS is used. It is
again advised that you seek the assistance of your
research adviser regarding this matter.
What to do.

A. Write the statistical design of your proposal here including the

specific problems by which you are applying the statistical tools.
Don't forget to show this to your research teacher for comments,
suggestions and / or approval.

1. For specific question #1 :

2. For specific question #2:

3. For specific question #3:

4. For specific question #4 :

5. For specific question #5 :