Module IV

Collection, Analysis and Interpretation of Data

Scope of the Module

This module consists of the following lessons:

Lesson 1. Tools of Research
Lesson 2. Descriptive Statistics
Lesson 3. Inferential Statistics

Overview of the Module

The module actually consists of two major parts, data collection a nd

interpretation of data. Data collection would refer to the different instruments that
would generate data needed for the study, their construction and methods of
validation as well as evaluation and scoring.

The second portion of the module consists of statistical techniques and

methods used for data analyses, both descriptive and inferential.

Objectives of the Module

After studying this module on collection analysis, and interpretation of

data, you will be able to:

1. distinguish between direct or empirical data and documentary or

secondary data;
2. identify methods in generating data;
3. construct instruments needed in the research, determine their
methods of validation and evaluation and scoring.

1
Research 1A
Module IV
Lesson 1. TOOLS OF RESEARCH

Lesson Objectives:

After studying this lesson on tools of research, you will be able to:

1. determine the different tools of research

2. identify the types/forms under each instrument;
3. state the advantages and disadvantages of each instrument, if any;
4. construct the various instruments; and
5. recognize the parts of a questionnaire, a common tool of research.

Introduction

The collection of meaningful data depends largely on the quality of the

instruments employed in the process. The significance of the findings of a
research becomes meaningless if the instrumentation is questionable. A
researcher, therefore, must exercise utmost care in choosing the research tools
particularly if he plans to collect the data himself. He would need to choose an
appropriate data-collection instrument that would ensure the relevance, reliability
and validity of the data collected. Several devices are employed for collecting
primary data, but three of them – questionnaire, interview, and observation – are
used most extensively.

Direct or Empirical Documentary Data or Secondary Data

Before constructing instruments to generate data, the writer determines

the kind of data that are needed for the research report. The research report may
call for documentary or secondary data and empirical or direct data, depending
on the statement of the problem. It must be emphasized that the statement of the
problem is very important because it will be the focus of data collection,
instrumentation, and interpretation.

Documentary Data. These are data which can be found in government

offices, in schools, private organizations, public organizations like the library or
business organizations, or other agencies which may have the data. In other
words, the writer does not gather the data. Because of this, the data generated
are referred to as documentary data or secondary data. In basic research, the
use of purely documentary data is generally discouraged, except for some
studies that will utilize only documentary data. But for in depth treatment of data,
such data are oftentimes reverted to some form which is the creation of the
writer.

2
 For example, if the writer seeks information on the relationship of
academic performance of high school students in the academic subjects and with
their corresponding scores in the NCEE, then this is a correlational study which is
based on documentary data. Academic grades and NCEE grades are found in
the office of the registrar and are categorized as documentary data. Or if the
study is about the prices of real estate and the variations of prices according to
cities, then the study likewise is anchored on documentary data. Real estate
prices are found in the Assessor’s office of the cities.

Empirical Data. Empirical or direct data are data which are generated
from people through the use of research techniques such as the questionnaire or
through interviews. There are studies that need empirical data. Below are some
topics of research which need the use of questionnaire.

1. Major Sources of Stress and Coping Strategies of Middle Managers

of a Manufacturing Firm in Metro Manila
2. Decision-Making Structures and Loci of Decision Making of
Managers of a Bottling Company in Metro Manila
3. The Relationship between Level of Job Satisfaction of Personnel
and Leadership Style in a Certain Organization
4. The Use of Marketing Mix Factors by Salesmen of a
Pharmaceutical Company

Questionnaire and Schedule Techniques

Definitions.

1. The questionnaire is a device for securing answers to questions by

using a form which a respondent fills in himself. If the set of questions
asked are filled in by the interviewer himself in a face-to-face situation
with the interviewee, it is referred to as schedule.
2. The questionnaire has been defined as “a written or printed form used in
gathering information on some subjects consisting of a set of questions
distributed to and filled out by the respondent. The schedule has been
regarded as “a form filled out by the researcher or completed in his/her
presence.” (Sanchez:1998)

Forms of question. Determined by the objective of the particular question, the

forms are classified as follows:

1. Open-ended – questions in which the respondent is asked to

provide his own answers to the questions. Example. What do you
think as the most crucial problem confronting the Philippines today?
2. Close-ended – questions in which the respondent is asked to select
his answer from among a list provided by the researcher. Example:

3
 in the question above, the researcher structures the responses,
such as:
• Peace and order situation
• Rice shortage
• Graft and corruption
• Environmental issue
• Unemployment
• Poverty
Close ended questions provide a greater infirmity of responses and are
more easily processed; however, in the researcher’s structuring of responses
some important issues might be overlooked. On the other hand, open ended
questions may lead to answers essentially irrelevant to the researcher’s intent.

In the constructions of close – ended questions, two guidelines should

always be followed.

1. Response categories should be exhaustive; that is, they

should include all the possible responses that might be
expected. Often, a researcher adds a category labeled
“Others (please specify).”
2. response categories must be mutually exclusive; that is, the
respondent should not feel compelled to select more than
one response.

Advantages and Disadvantages of the Questionnaire

Advantages: According to Sanchez (1998), the questionnaire, which is

commonly used as the main data-gathering instrument, has the following
advantages:

The respondents may have greater confidence in their anonymity and thus
feel free to express views they fear might be disapproved or might get
them into trouble.
It allows respondents to respond in their time frame.
By its very nature, the questionnaire is likely to be a less expensive
procedure than an interview.
It requires much less skill to administer a questionnaire than an interview.
It can be administered to large number of people across a wide
geographic area

Disadvantages:

The respondents fail to dig deeply enough to provide a true picture of

opinions and feelings.
Representatives sample of data cannot be obtained by a mailed
questionnaire if the population includes nonreaders.

4
 A lower response rate.

Guides to Questionnaire Construction

1. make items clear and unambiguous. Be precise so that the respondent

knows exactly what you want him to answer to.
2. Avoid double-barreled questions which ask respondents for a single
answer to a combination of questions. Example. Do you agree or
disagree with the statement “Congress should reduce our foreign debt
payments and increase the budget for education”.

This question should be split into two: Do you agree or disagree with the
statement. “Congress should reduce our foreign debt payments? Do you agree
or disagree the Congress the budget for education?

3. Questions should be relevant to most respondents.

4. Avoid long and complicated items; short items are best.
5. Avoid negative items. Example: “The Philippines should not recognize
Taiwan.”
6. Avoid biased items and leading questions. Example: “ Do you agree or
disagree with your teacher’s view on abortion?”
7. Avoid questions that may in some way be psychologically threatening
to the person answering.
8. Sequence or arrange your questions properly
9. Give clear instructions and introductory comments where appropriate
10. Pretest your questionnaire before using it in your study.
11. Prepare a letter of transmittal to accompany your questionnaire
12. Choose appropriate format for presenting a series of response
categories such as using boxes, brackets or parentheses. Avoid using
slashes and underscores as they require considerably more effort and
the result is not very neat, especially if response categories are single
spaced.

Interview

Definition. A research method that involves the collection of data through direct
verbal interaction between the interviewer and the interviewee or respondent.

Sanchez (1998), noted that personal interview may be formal or informal.

In formal interview, the respondents are asked questions from carefully prepared
questions or schedules and their answers are recorded in a standardized form;
while in informal interview, the interviewer may not have a prepared set of
questionnaires but just a number of important points around which the interview
centers.
In many surveys, the formal personal interview is most usual.

5
Types. There are two types of interview.

1. Structured interview – the interviewer uses a set of standardized

questions and questioning procedures: questions are asked in a fixed
and sequential manner. The interviewer may not repeat any questions
and may not modify, alter or follow – up any questions asked unless
the interviewee does not hear the interviewer.
2. Unstructured interview – a more flexible and open interview. The
interviewer can modify, repeat, elucidate the questions asked and may
follow up the responses of the interviewee without deviating from the
purposes of the interview.

Advantages:

1. attains higher response rate than the questionnaire;

2. decreases number of “don’t knows” and “no answers”;
3. guards against confusing items;
4. interviewer can observe respondent’s behavior, appearance, etc.;
5. questions not readily grasped by the interviewees can be rephrased
with proper emphasis and explanations when necessary;
6. the interviewer has greater opportunity to appraise the accuracy and
validity of re[plies;
7. interviewees often feel free to express personal and confidential
information which they would not ordinarily do in writing.

Disadvantages:

1. the very adaptability gained by the interpersonal situation leads to

subjectivity and possible bias;
2. considerable training is required before the individual can successfully
carry out an interview study;
3. the interview is time consuming and therefore limits the number of
subjects from whom data can be obtained.

Principles in Conducting an Interview

The following are the principles to be followed in conducting an interview as

cited by Sanchez (1998):
1. Preparatory Stage
To find out how and what the people who are to serve as
respondents of the survey think and feel about the areas of interest in the
survey is the main purpose of the preparatory stage. This is followed by
the formulation of a set of questions to be tried out with small diverse
samples of respondents in order to discover how well these questions
actually get the information and how the respondents react emotionally to

6
 them. The process may have to be repeated in order to determine the
interviewing procedure that appears to work out satisfactorily.
2. Setting
The place where the interview will be conducted should be free
from disturbance, noise, or intrusion.
3. Establish Rapport at the Outset
Efforts should be taken to establish at the very beginning a friendly
atmosphere between the interviewer and the person to be interviewed.
During the interview process, a genuine liking and respect for each other
must be developed. A state of “rapport” exists between the interviewer
and the respondent when the latter has accepted the research goals of the
interviewer and actively seeks to help him or her in obtaining the
necessary information.
4. Carrying the Interview Forward
The process of carrying the interview forward is greatly aided by
writing the questions in a fashion which most closely approximates
conversation while probing for those items for which the research is being
conducted.
5. Recording the Interview
The interview must be recorded to ensure that all information
gathered is complete and accurate. This will make the interviewer feel at
ease while analyzing the information gathered and relating them to the
other interviewers.
Note-taking or tape recording are the usual methods for preserving
the information collected in the interview. The chief advantage of the note-
taking method is that it facilitates data analysis since the information is
readily accessible. Its advantages is that it may disrupt the effectiveness
of the communication between the interviewer and respondent. On the
other hand, the use of the tape recorder has these advantages: 1) tape
recorded data can be played back more than once and can be studied
more thoroughly than the limited notes taken during the interview; 2) tape
recorded data can be evaluated by persons other than the interviewer;
and 3) the interview process can be speeded up because there is no
necessity for extensive note-taking.
On the other hand, its disadvantage is the presence of the tape
recorder changes the interview situation to some degree: respondent may
be reluctant to express his feelings freely if he knows that his responses
are being recorded.

6. Closing the Interview

In closing the interview, gratitude and appreciation should be
shown and expressed with a friendly smile and sincerity for the
respondent’s generosity.
The following are measured in an interview:
a. opinions e. weaknesses
b. attitudes f. experiences

7
 c. information g. difficulties
d. interests

The researcher in his/her write-up must be objective, impartial, and

accurate. To check the accuracy of the information, the same question
must be asked in various ways not only once but twice or even more, and
further checked by observation to see if the actual behavior of the
respondent tallies with his/her verbalized replies.

Observation

Definition – process of ascertaining what people think and do by watching

them as they express themselves in various situations, selected to typify the
conditions of informal living or to represent some special set of factors. It is
recognized as the most direct means of studying people when one is interested
in their overt behavior.
Types:
1. Simple, Uncontrolled Observation
1.1 Participant – used when investigator is accepted as a
member of the group either known or unknown to group
being observed.
1.2 Non- participant – used when it is impossible to be genuine
participant in all ways, whether known or unknown to a
group or controlled.

2. Systematic or Controlled Observation

2.1 Controls on the observer in the form of mechanical devices,

team observation, films and recordings, schedules and
inventories, development of elaborate categories for locating
and coding observed behavior quickly.
2.2 Controls on the observed – situation is contrived or
manipulated in which definite stimuli are introduced.

Characteristics of Observation for Research

1. Observation is specific; that is, observer looks for definite things and
not just for general impressions.
2. Observation is systematic – the length of the observation periods, the
interval between them and the number are carefully planned.
3. Observation is quantitative – the number of instances of a certain type
of behavior are usually recorded.
4. Observation must be recorded immediately
5. Observation implies expertise – observer must be trained to do such
work.

8
 6. Observations claims to be scientific because the results can be
checked and substantiated, either by comparing the results of different
observers or by repeating the study.

Tools of Observation

1. checklists or schedules
2. time sampling – random sampling of time permits control over the
observable forms of occurrence during a number of definite time
intervals that are systematically spaced;
3. behavioral diaries and anecdotal records;
4. mechanical instruments – motion pictures, videotape recorders and
sound recordings preserve details in a reproducible from which can be
studied repeatedly and intensively by the investigator and checked by
other research workers.

An example of an observation guide is presented in the following page which

was used in observing a class demonstration in the teaching mathematics.
Observation Guide
Event Sample Illustrations/Events Tallies
1. Motivation
a. recall of previous lesson
b. devises used
2. Concept Development
a. inductive method used
b. use of pictures and devices
c. clarity of explanation
d. active participation of students
3. Generalization
4. Evaluation

To determine the reliability of the observations, the formula of Mussen (1960)

may be used:

No. of agreements among observers

r = -----------------------------------------------------
total numbers of dimensions observations

where r = the reliability coefficient

Interpretations of the reliability coefficient or r:

.90 – 1.00 very high
.80 - .89 high

Below the .80 would mean that there is a need for proper orientation
among the observers.

9
 Personal Information Sheet: A Sample

Direction: Please cross (x) the blanks which best describes you. (Please
check).

1. Code Number ______(Do not fill this one.)

2. Sex: 2.1 Male _________ 2.2 Female _________
3. Civil Status:
3.1 Married ____________
3.2 Single ____________
3.3 Widower ____________

4. Educational Attainment
4.1 Elementary ____________
4.2 Secondary ____________
4.3 College ____________
4.4 Post College ___________

5. Age:
5.1 29 – below ____________
5.2 30 - 39 ____________
5.3 40 – 49 ____________
5.4 50 – above ____________

6. Length of Service with the Company

6.1 5 – below ____________
6.2 6 – 10 ____________
6.3 11 – 15 ____________
6.4 16 – above ____________

7. Level of Management
7.1 Top ____________
7.2 Middle ____________
7.3 Rank and File ____________

The Personal Information Sheet is used to elicit information on the

personal characteristics of the respondents so that a profile maybe drawn of
them. Such profile is placed together with the topic: respondents of the Study.
The reader may be interested to know what type of respondents were used in the
study.

As you will observe the questionnaire above, the items there concern
about the respondents himself. With the questionnaire above, you will gain

10
insights on how to prepare the questionnaire. You can add items that may add
more information about the respondent.
The questionnaire is so constructed and numbered so that the data can be
decoded and computerized. You will further observe that the age and length of
experience have open intervals on both ends.

Here is a questionnaire constructed by a researcher Dr. Dionisio C. Caro

in his masteral’s thesis, entitled “Perceptions on the Cubao – Makati – Escolta
Routes: An Assessment.”

A. The cover letter

January 15, 1989

Pasay City
Metro Manila

Dear Respondent:

I am a graduate student of the University of the East working for a

Master in Business Administration (M.B.A.) degree and for a partial
requirement for the degree, I am working on a research that would
reflect your assessment of the Love bus Service on the Cubao-Makati-
Escolta route.

As a passenger, you are competent to make judgments on the Love

bus Service in terms of some criteria like:
1. the effectiveness of the love Bus as a transportation service;
2. collection of fares;
3. time scheduling;
4. convenience of the trip; and
5. the facilities inside the bus.

The assessment are described in terms of the extent to which the Love
Bus has made the ride comfortable and pleasant, and responses
therefore are geared towards the measurements of that extent. The
options to the questionnaire are: to a large extent, to a certain extent,
not sure, to a limited extent, and not at all. The options are assigned
points as follows.

Options Assigned Points

To a large extent 5
To a certain extent 4
Not sure 3
To a limited extent 2
Not at all 1

11
 It is requested that you give your assessments frankly and objectively
so that the service of the Love Bus may be improved.

Information given by you will be held highly confidential.

Thank you for making this a part of your day.

Thank you again.

Dionisio C. Caro
Researcher
B. The Assessment Questionnaire

Directions: Please encircle the number which best describes your answer.
The options are:

To a large extent -5
To a certain extent -4
Not sure -3
To a limited extent -2
Not at all -1

1. Effectiveness of the Love bus

a. Reduction of private cars

1) Travel has been made easy on route by
instituting the Love Bus service. …..5 4 3 2 1
2) People seem to use the Love Bus more
than the other buses of means of transport. …..5 4 3 2 1
3) People would rather use the Love Bus rather
their private cars. …..5 4 3 2 1
4) The use of Love Buses had reduced the number
of private cars traveling the same route …..5 4 3 2 1
5) The Love Bus service has reduced the use of
Gasoline …..5 4 3 2 1
6) The Love Bus has reduced traffic in the
route that they ply. …..5 4 3 2 1
b. Service
1) The Love Bus travels on schedule …..5 4 3 2 1
2) The Love Bus does not deviate from its route …..5 4 3 2 1
3) The driver and the conductor are courteous …..5 4 3 2 1
4) The driver observes strictly traffic rules …..5 4 3 2 1
5) The conductor is honest and returns the correct
change …..5 4 3 2 1
6) The driver stops at the passenger’s place. …..5 4 3 2 1
7) I enjoy taking the Love Bus on the Cubao

12
 Makati- Escolta route …..5 4 3 2 1
8) The driver announces clearly on intercom
places where it would stop. …..5 4 3 2 1
c. Equipment
1) The Love Buses are generally in good condition …..5 4 3 2 1
2) The air condition equipment functions well …..5 4 3 2 1
3) The seats are clean …..5 4 3 2 1
4) The bus is clean …..5 4 3 2 1
5) The passengers are well taken care of
by the conductors …..5 4 3 2 1

d. Collection of Fares
1) The fares of the Love Bus service is reasonable
and fair …..5 4 3 2 1
2) Ticketing is immediate and changes are
Immediately taken …..5 4 3 2 1
3) Considering the service the Love Bus gives,
the fare collected is just and fair. …..5 4 3 2 1

C. Problems Encountered by the Passengers

Options: Strongly agree -5

Agree -4
Not Sure -3
Disagree -2
Strongly Disagree -1
1. Smoking should be allowed …..5 4 3 2 1
2. Drivers and Conductors of the Love Buses
Are not courteous and polite. …..5 4 3 2 1
3. Sometimes the conductors does not give
The correct change …..5 4 3 2 1
4. Sometimes the driver deviates from the
Usual route …..5 4 3 2 1
5. The driver is reckless and drives very fast …..5 4 3 2 1
6. The buses are not in good running condition …..5 4 3 2 1
7. The air condition is out of order …..5 4 3 2 1
8. Only a few units are operating …..5 4 3 2 1

D. Suggestions to Improve the Love Bus Service

1. Reduce flat rate for everybody. …..5 4 3 2 1

2. Introduce student concession fares …..5 4 3 2 1
3. Introduce half fare for children …..5 4 3 2 1
4. Introduce prepaid fares at a discount …..5 4 3 2 1
5. Introduce transfer tickets. …..5 4 3 2 1
6. Increase number of operating love buses

13
 In different routes …..5 4 3 2 1
7. Love Buses service should be extended
till late in the evening. …..5 4 3 2 1
8. There should be reduced delay in the
collection of fares. …..5 4 3 2 1
9. There should be automatic fare collection. …..5 4 3 2 1
10. Bus should leave when full, regardless
of time schedule …..5 4 3 2 1

The above questionnaire has not been printed in the full text, but
substantial enough for the student to see how a questionnaire should be
constructed. The student should note the less number of items in a certain topic,
and the similarity of responses for ease of statistical analysis and interpretation.

Note also that with the assignment of points, scoring is very evident. For
verbal interpretations of the results, arbitrary intervals are set, an example of
which is shown below:

Verbal Interpretations Intervals

To a large extent-strongly agree 4.5 – above

To a certain extent – agree 3.5 – 4.4
Not sure 2.5 – 3.4
To a limited extent – disagree 1.5 – 2.4
Not at all – strongly disagree 1.4 – below

The responses are arranged in a 5 continuum in a Likert Scale. The first

two responses are positive, while the third (not sure or uncertain) serves as the
fulcrum of the responses, while the last two responses are negative, which
respond to the characteristics of a probability distribution.

Now go over the manuscript and when you feel that you are ready, you
may take the Self-Progress Check test.

(SEE NO. 3 OF “INSTRUCTIONS TO THE STUDENT”)

14
Research 1A
Module IV
Lesson 1
SELF – PROGRESS CHECK TEST

General instructions:

• Read and understand each statement carefully.

• Write your answer on the space provided.

A. Identification: Identify the following. Write your answer on the blank

before each number. (10 points)

___________1. Type of observation where the researcher becomes a part of

the respondents of the study.
___________2. Kind of interview where a questionnaire may be used as an
interview guide.
___________3. A kind of questionnaire where all the answers are found in
the option.
___________4. A face-to-face encounter between the researcher and the
respondent.
___________5. The characteristics of a good tests wherein it seeks to
measure what it intends to measure.
___________6. Means the test items are consistent, reliable and stable.
___________7. The ability of an instrument to make the discriminations
required for the research problem.
___________8. The degree to which the measure is independent of personal
opinions and subjective judgment.
___________9. Data that are generated from questionnaire replied to by
people.
___________10. Data gathered from offices, both public and private.

B. Essay. Direction: In not more than 10 sentences, answer the following

question briefly. (10 points)

1. Distinguish a formal interview from an informal interview.

(SEE NO. 4 OF “INSTRUCTIONS TO THE STUDENT”)

15
Research 1A
Module IV
Lesson 2. DESCRIPTIVE STATISTICS

Lesson Objectives:

After studying this lesson on descriptive statistics, you will be able to:

1. present data through tables and figures;

2. analyze data using statistical methods; and
3. interpret data

Statistical Methods Used in Data Interpretation

Once the questionnaires are in, data are culled, tallied. Sometimes at this
moment, the researcher stops to work as he does not know what to do. One can
always ask the help of statisticians, to guide him to work on the data.

However, here are some of the statistical techniques and methods whioch
can help you.

Descriptive Statistics

The main function of statistics is to tell stories about the data or describe
data. To describe data, three types of descriptive statistics may be used: 1)
statistics of location, 2) statistics of distance, and 3) statistics of association.

The Statistics of Location. The first thing that a researcher should compute are
the statistics that will describe what are common or what the representatives of
the group are. These are called the measures a of central tendency. The
measures of central tendency are the average and the most commonly used
averages are the mean, median, and the mode.

Mean. The arithmetic mean is described as:

a. a computational average
b. it is the sum of the variables divided by the number of variabales
or in formula form:

n
x = Sx/n
i =1i

where x = the arithmetic mean

n = the number of variables
i = the interval

16
c. it is used when the values of the variables are not extremes.
Example 1. When one is asked to compute the average household income in
Forbes park, the mean is used as a measure of central tendency,
since their earnings are almost the same.

Example 2. If one is asked the average of the following scores:

34, 37, 30, 29, 32. 30, 33. 39. 39, 38
The arithmetic mean is the best measure of central tendency
because the values of the variables are not extremes.

There is also another mean which is very much used in the interpretation
of data especially from questionnaire which are assigned points. And this is the
weighted mean which is solved by the formula:

n
Wm = S fi Xi /n
i = 1
where: Wm = the weighted mean
f = frequency
x = the variable
n = the number of variables

Example 1. Find the weighted mean of the following variables:

5, 5, 5, 4, 4, 3, 3, 3, 2, 2

3(5) + 2(4) + 3(3) + 2(2)

Wm = -------------------------------------
10

15 + 8 + 9 + 4 36
= --------------------- = --------
10 10

= 3.6

This is very useful when options to the items of a questionnaire are

assigned points.

Median. Another measure of central tendency is the median. It has the following
characteristics.
1. It is a positional average.
2. it is the position in the distribution where half of the variates is
greater and half is lesser. This implies ranking.
3. It is used when the values of the variables are extremes.

17
Example 1. When one is asked on the average household income of the people
living in Quiapo, the best measure of central tendency is the
median, because in this place, there are very many poor and very
many rich people, so that the household income is heterogeneous.
Hence the use of a positional average.

Example 2. Find the average of:

100, 80, 60.40. 0
Note that the values in this distribution are extremes.
We rank the variables and we have
100
80
60 – this is the midpoint, hence this is the median
40
0

Note that if you change 100 to 1,000, the median is not affected, since it is
a positional average.

Example 3. Find the median of the following:

100, 60, 40, 20, 0, 80
Rearranging the variables, we have:
100
80
60
The midpoint lies between 60 and 40
40 60 + 40
20 the median is ------------
0 20
= 50

The median is the midpoint between 60 and 40, which is 50. This is the
procedure when the number of variables are even.

Mode. The mode is another commonly used measure of central tendency. It has
the following characteristics.
1. It is a nominal average
2. It is the variable with the greatest frequency
3. it is used fro qualitative data

Example 1. A researcher wants to know the average qualification of teachers in

a certain school. This is the information.

30 Ph.D.s
58 MAs
100 BSE

18
 The mean and median have no meaning in this problem, but rather the
most representative group is the BSE, and such is the variable with the greatest
frequency or the mode.

Example 2. What is the mode of the following variables?

40, 23, 45, 45, 45,46, 45

In this case, the mode is 45, which occurs more time than the others.

It is also possible for a distribution to be bimodal or trimodal, or there are

two or three variables with the greatest frequencies.
The three measures of central tendency are called the statistics of location
or points, since as points, they are easily located in the normal distribution as
shown below.

[m,f(m)]

50% 50 %

Of the three measures of central tendency the arithmetic mean is the most
stable.

Statistics of Scale or Distance. However, although we are able to describe

what is representative of the group using the measures of central tendency, their
descriptions, however, are not enough. This is because the measures of central
tendency are not unique, and if given cannot describe a definite distribution.

Let us take this example. If

_ _ _____
A = B, then XA = XB, if the two distributions and B are equal distributions,

then the corresponding measures of central tendency are the same.

_ ___
A B A=B XA = XB
100 80
80 100
60 60
40 20

19
 20 40
0 0

_ _
But if XA = XB , does it also mean that A = B?
_ _
Let us see. If XA = XB, will A = B?
_
XA = 60 could also be the mean of the distribution 100, 80, 60,40, 20.

But it could also be the mean of the distribution of the 60, 60, 60, 60, 60
and 60.

Therefore from this explanation and example, means are not unique in this
sense that it can represent infinite number of distribution. Hence we need more
refined methods to describe data, or a description of variations or scatter of the
variables.

This gives the rational for the use of the measures of variation or the
statistics of scale. Let us just take the most common measures of dispersion or
variation.

e. Range. The range is the distance between the smallest

and the largest variable or:
R = LV – SV
However, the range is not a good measure of variation or description,
since it is involved only with two variables, and the rest of the variables lose their
identity. It is used only to solve other statistical values.

f. Standard Deviation. The standard deviation has the

following description.
1. it is sometimes called the root-mean-square.
2. it is the square root of the squared deviations from the mean
divided by the degrees of freedom or by n-1. A degree of
freedom is a number which determines the number of variables
which are free to vary under certain conditions or restrains.

Example 1. Give 5 numbers whose sum is 100.

The student can give any four numbers, like

25

30
These four numbers are free to vary

10

20
 25
10 But 10 is fixed otherwise the sum would not be equal
to 100. So the degrees of freedom is 5 – 4 = 1, or the
----------- general rule which is n - 1
100

2. In formula form, it is solved by:

_
S=
S (x – x)2
-----------------------

n–1
Where s = the standard deviation
n = the number of elements
x = the mean
x = the variable

The standard deviation describes the homogeneity and heterogeneity of

the variables. The greater the value of the standard deviation, the more scattered
the variables.

There are other measures of scale such as the quartile deviation and
mean deviation, but the most frequently used not only in description of data, but
also in testing hypotheses is the mean.

Of the different measures of scale, the standard deviation is the most

commonly used, and the variance(s) is the square of the standard
deviation

When the mean is used as the measure of central tendency, the best
measure of scale or deviation is the standard deviation, while if the median is
used, the best measure of central tendency is the quartile deviation. But
inferential statistics utilize very much the mean and the standard deviation in the
tests of hypothesis.

The figure below shows that there are six standard deviations at the base
of the normal curve. The area in the probability curve, 1 standard deviation above
and below the mean 48.26 percent, while are of 2 standard deviations above and
below the mean is 95.95 percent, while the area 3 standard deviations above and
below the mean is 99.99 percent.

21
 -3 -2 -1 M 1 2 3

Once the necessary values are computed, they are arranged in tables.
Below is an example of a frequency and percentage distribution of the age
groupings of respondents of the study. Note that in the preparation of the table,
the title is found on the top. The different categories are arranged and expressed
in percentage in two decimal points. Note that at the bottom of the table, 6he
mean age and standard deviation are given.

Table 1

Frequency and Percentage Distribution of Respondents According to Age

Groupings

Age Groupings Number of Percent Ranking

Respondents
1. 29 - below 5 10.00 3.5
2. 30 - 39 20 40.00 1.5
3. 40 – 49 20 40.00 1.5
4. 50 - above 5 10.00 3.5
Total 50 100.00

___ --
x = 39.50 years
s.d. = 8.34 years

Interpretation of Data

The table shows the age groupings of the respondents of the study. Two
large groups are noted, namely, those whose ages fall on the 30-39 years and on
the 40 – 49 age groups. There are 20 respondents in each group showing each a
proportion of 40.00 percent. There are only very few of the respondents whose
ages fall on the youngest and the oldest group. However, a mean age of 45.345
years, showing that the respondents are in their middle forties, strong and prime
in their lives.

Note: The interpretation does not use all the numbers in the table, but
only those that stand out prominently. Leave the rest to the imagination of the
readers.

The above data could be graphed to make data more vivid and clear.
Below is the vertical bar graph for it.

22
 50
40
30
20
10
29- 30-39 40-49 50 - Age Groups
below above

Figure 1. Age Groupings of the Respondents of the Study

Below is an example of how data are interpreted. The data are

hypothetical based on the items of a questionnaire for the Skill Mix Theory. Note
that the table summarizes data of the human relations skills of managers in terms
of the weighted mean, the standard deviation verbal interpretation and ranking.
Not all the values are also discussed in the interpretation but only those that are
important ones. Other values are left to the reader for them to interpret.

Table 2
Impressions of Manager Respondents on their Human relations Skills

Items Weighted Standard Verbal Rank

1. He makes his subordinates 4.231 0.987 Very Good 4
work like a team
2. he lets his subordinates 4.523 0.887 outstanding 1
know where they stand
3.He understands the 4.331 0.678 Very Good 3
problems of his subordinates
4. He lets them know of the 3.875 1.126 Good 5
problems of management
5. He shows his loyalty to his 4.391 0.684 Very Good 2
subordinates
Overall Mean 4.207 0.852 Very Good 2

Note the interpretation of the table.

Table 2 shows that the assessment of the managers of their own human
relations skills were varied and ranged from “good to outstanding”. They
assessed themselves as outstanding in their ability in being firm and letting their
subordinates know where they stand (4.523), or when trouble comes, the
managers stood by their personnel. They were rated low however in their ability
to share with personnel the problems of the company. This might be due to the
fact that in general, the problems are confidential. Overall results with an overall

23
weighted of 4.207 managers nevertheless described themselves as “very good”
in human relations skills.

Now you have seen how data are presented, in tabular, or in figure form,
or in textual from. You should gather insights into how tabular data are being
presented, analyzed and interpreted.

(SEE NO. 3 OF “INSTRUCTIONS TO THE STUDENT”)

24
Research 1A
Module IV
Lesson 2

SELF – PROGRESS CHECK TEST

General instructions:

• Read and understand each statement carefully.

• Write your answer on the space provided.

A. Identification: Identify the word referred to by the sentence or the group

of words. (10 points)

_______________1. An inventory or list of information desired by the

researcher.
_______________2. The most commonly used technique to generate data.
_______________3. A research technique where the researcher comes face-
to-face with the respondent.
_______________4. An interview where an interview guide is used.
_______________5. A type of observation where the researcher is a part of the
respondents of the study.
_______________6. It is used as a measure of central tendency when the
values of the variables are not extremes.
_______________7. It is used as a measure of central tendency when the
variables are extremes.
_______________8. It is a measure of central tendency which implies ranking.
_______________9. It is a positional average.
_______________10. It is a computational average.
_______________11. Refers to the nominal average.
_______________12. It is a variable with the greatest frequency.

_______________13. Refers to the root mean square.

_______________14. Another name for measures of central tendency.
_______________15. It is best used when the options are assigned points.
_______________16. They are not unique.
_______________17. It determines the homogeneity or heterogeneity of the
variables.
_______________18. A method of data presentation where the title is found on
the top.
_______________19. It helps make data clear and vivid.
_______________20. Title is found at the bottom of the page.

25
B. Essay. Direction: In not more than 10 sentences, answer the following
question briefly. (10 points)

1. Discuss the Characteristics of Median as a measure of central

tendency.

(SEE NO. 4 OF “INSTRUCTIONS TO THE STUDENT”)

26
Research 1A
Module IV
Lesson 3. INFERENTIAL STATISTICS

Lesson Objectives:

After studying this lesson on inferential statistics, you will be able to:
1. select statistical techniques in interpreting data;
2. use relationships or correlational analyses techniques; and
3. apply comparative analyses techniques.

Inferential Statistics

We have seen that descriptive statistics are necessary in interpreting

data. But it should not end there. The researcher should use more sophisticated
means of interpreting and analyzing data.

Research should not be merely description of data, but should indulge in

finding relationships between the variables in the data or comparing situations. At
this point, it is imperative that when one does a piece of research work, one
should seek the guidance and advice of a statistical consultant.

This module will just contain some of the most common statistical
techniques and methods for 1) correlational or relationship, and 2) comparative
analyses.

1. Relationship Techniques
a. Finding relations between variables
1. Pearson r correlation coefficient

One of the most popular techniques of finding relationships between

variables is the Pearson r correlation coefficient. Technically defined as:

covariance
r = ------------------------------------

var x var y

it results into a formula called the product moment formula which is:

n Σxy - Σx Σy
r = ------------------------------------------

[nΣx2 – (Σx)2][nΣy2 – (Σy)2]

27
 where x = the independent variable
y = the dependent variable

This correlation coefficient describes how the variations of one variable

go with the other, whether direct or inverse, whether positive or negative. So that
the Pearson r does not only indicate the direction of the relationship but also the
amount of the relationship.

However, there are basic assumptions for the use of the Pearson r and
among these are:
1. that the distribution are normally distributed; and
2. that the relationship is linear.

Some statisticians say that the first assumptions is near to impossibility,

but that the second one must be a must.

Since the Pearson r is descriptive, a t-test is used to determine the

significance of a Pearson r with the formula:
_____
r Ön–2
t = ----------------------------------

____2_
Ö 1 -r

where r = the correlation coefficient

n = the number of pairs

The t – test uses a null hypothesis in the form:

H0: p = 0

Against an alternative hypothesis in the form:

Ha: p1 0

A linear regression analysis follows with a regression line to be used in

the form:

Y = a + bx + e

Where e = 0, but is normally distributed, and where:

A = y – bx

Or the ordinate of the regression equation and

n Σ xy – Σx Σ y

28
 b = ------------------------------------------
n Σ x2 – (Σx )2

is the regression coefficient or the slope of the line. The regression equation
provides the method of predicting the behavior of y given a value of x.

The r2 (%) determines the amount of overlap which is explained by the

regression equation solved.

2. Spearman rho Rank Correlation Method

When one is not sure about the normality of the distributions, the
researcher may utilize a ranking method known as the Spearman rho Rank
Correlation method. It is one of the easiest correlational techniques.

It is solved by the formula:

1 - 6 Σ D2
P = -------------------------
N (n2 – 1)

Where : D2 = the difference of the ranks

n = the number of pairs

The Spearman rho Rank Correlation coefficient can either be tested by a

t-test or given values in a certain table.

When the variables are qualitative, the researcher may use the chi-
square test.

The chi-square formula is:

X2 = Σ (0 – Σ)2/Σ
where : 0 = the expected frequencies and
Σ = the expected or the theoretical frequencies.

The chi-square test uses a null hypothesis in the form:

H0 : 0 = S

2. Comparative Analyses
1. Parametric tests.

There are a few of the parametric tests which are commonly used for
comparative analysis.

29
 When comparing the means of two independent samples, a t – or a z –
test is used depending upon the sizes of the sample and knowledge of the
population or the sample variance.

On the other hand, determining the significant differences of three or

more means of independent samples, a one-way analysis of variance may be
used, however, with the basic assumptions present, like: randomness and
homogeneity of variances. If the t-test meets the required level of significance, a
Scheffe’s test is used to determine the significant difference of any pair of means.

2. When normality of the variables are not determined, a Friedman’s

analysis of variance may be used, especially so when the sizes of the sample are
extremely different. Otherwise, a qualitative method, or a chi- square may also be
used.

(SEE NO. 3 OF “INSTRUCTIONS TO THE STUIDENT”)

30
Research 1A
Module IV
Lesson 3
SELF – PROGRESS CHECK TEST

Identification: Identify the word that is being referred to in the sentence or

group of words. (20 points)

_______________1. Inferring the characteristics of the population based on the

sample.
_______________2. It determines how the variations of one go with the other.
_______________3. The kind of relationship when the Pearson r is found
positive.
_______________4. The kind of relationship when the Pearson r is negative.
_______________5. It is known as the regression equation.
_______________6. The a in the regression equation.
_______________7. It is known as the slope of the regression equation.
_______________8. One of the basic assumptions of the use of the Pearson r
which is difficult to attain.
_______________9. The more important requirement for the use of a Pearson r.
_______________10. Descriptive hence must be tested.
_______________11. The test for the significance and reliability of a Pearson r.
_______________12. It is known as the proportion overlap explained by the
regression equation.
_______________13. It is used as a correlational analysis when one is not sure
of the normality of the variables.
_______________14. A test used for comparative analysis when there are two
independent samples and the sample size is small.
_______________15. A test used in comparing two sets of independent
samples, when the samples are large and when standard deviation for the
population is given.
_______________16. It is used when comparing three or more independent
samples.
_______________17. It is one of the requirements for the use of analysis of
variance, which described the nature of the variables.
_______________18. Of variances is also required for the use of the t-tests.
_______________19. It is used when the t-test meets the required level of
significance.
_______________20. It can also be used for finding relationship between
variables, especially if they are qualitative.

(SEE NO. 4 OF “INSTRUCTIONS TO THE STUDENT”)

31
Research 1A
Module IV

ANSWER KEYS TO THE SELF-PROGRESS CHECK TESTS

Lesson 1

A. Identification: (10 points)

1. participant 6. reliability
2. structured 7. sensitivity
3. closed-ended questions 8. objectivity
4. interview 9. empirical
5. validity 10. secondary

B. Essay: (10 points)

1. Distinguish a formal interview from an informal interview. The

discussions should reflect the following ideas:

In formal interview, the respondents are asked questions from

carefully prepared questions or schedules and their answers are recorded
in a standardized form; while in informal interview, the interviewer may not
have a prepared set of questionnaires but just a number of important
points around which the interview centers. In many surveys, the formal
personal interview is most usual.

Lesson 2

A. Identification: (20 points)

1. assessment questionnaire 11. mode
2. questionnaires 12. mode
3. interview 13. standard deviation
4. structured 14. averages or statistics or
location
5. participant 15. weighted mean
6. mean 16. measures of central tendency
7. median 17. standard deviation
8. median 18. table
9. median 19. table
10. arithmetic median 20. figure

32
B. Essay: (10 points)

1. Discuss the Characteristics of Median as a measure of central tendency.

The discussions should reflect the following ideas:

The characteristics of median as a measure of central tendency are

as follows: it is a positional average; it is the position in the distribution
where half of the variates is greater and half is lesser. This implies
ranking; and it is used when the values of the variables are extremes.

Lesson 3 (20 points)

1. Inferential statistics 11. t-test
2. correlational coefficient 12. r2 (%)
3. direct 13. Friedman’s Analysis of
variance
4. inverse 14. t-test
5. y=a+bx+c 15. z-test
__ __ 16. One-way analysis of
6. a = y - bx variance

n Sxy –Sx Sy
7. b = ------------------------ 17. homogeneity

n Sx2 - (Sx)2
8. distributions are normally distributed 18. ratio
9. that the relationship is linear 19. Scheffe’s test
10. Pearson r 20. chi-square test

33
34