Chapter Five: Data Collection and Analysis Techniques

5.1 Primary and Secondary Data

Information sources are usually classified into two broad categories, Primary and secondary.
The primary data are those which are collected a fresh and for the first time and thus happen to
be original in character. Original data related to the past events could also be considered as
primary data. The major sources of primary data include documents such as diaries of
eyewitness, court records, statistical reports (census results), tape-records, films, letters and
autobiographies.

On the other hand, secondary data are those which have already been gathered by someone
else and which have already been passed through the statistical process. These include history
textbooks or historical studies based on the actual data, statistical research based on census
data, news paper reports of an event not written by an aye witness, biographies and other
second-hand descriptions.

The methods of gathering primary and secondary data differ since primary data are to be
originally collected, while in case of secondary data the nature of data collection work is merely
that of compilation. Therefore, a considerable care should be taken by the researcher while
choosing data gathering tools for the two types of data sources. The following section is
devoted to the description of various data gathering instruments..

5.1.1 Data Gathering Tools for primary Data.

There is a wide variety of primary data gathering tools available to researchers; deciding which
to use depend on:

- the particular methodology or type of research that has been selected;

- What is feasible in a given research situation;
- What is likely to yield the most appropriate information?

All these items are decided upon in the context of the particular research problem.

Some of the more important data collecting methods for survey or descriptive research studies
are questionnaire, schedules, interview, observation, focus group discussions and case study
methods.

Each of them is briefly discussed below.

1) Questionnaire is the most popular type of data gathering instrument. It is practiced by

individual researchers, private and public organizations, and even by governments. As
stated by kothari (1990: 124), a questionnaire consists of many questions ( items)
printed or typed in a definite order on a form or set of forms. The questionnaire is

1
 usually mailed ( or administered personally) to the respondents who are expected to be
literate and able to understand the items and provide answers properly. The
respondents are expected to answer the questions by themselves.

Types of Questionnaires

As far as the format (nature) of the items in a questionnaire is concerned, the questionnaire is
classified as either open-ended or closed-ended type

Questionnaires that call for short or check responses are known as closed form or restricted
type. They include a set of questions to which respondents can reply in a limited number of
ways.

On the other hand, the open-form or unrestricted type of questionnaire calls for a free
response in the respondent's own words.

Often researchers use one of them exclusively or both of them in combination. In view of
Bailey (1982), each type of questionnaire has its advantages and disadvantages.

Advantages of closed- ended ( fixed-alternative) questions:

1. The answers are standards, and can be compared from person to person.
2. The answers are much easier to code and analyze, and often can be coded directly from the
questionnaire, saving time and money,
3. The respondent is often clearer about the meaning of the question ( that is a respondent
who is unsure about the meaning of the question can often tell from the answer categories
what is expected). Thus there are fewer frustrated respondents who answer " don't know'
or fail to answer at all. This helps the return rate, since frustration over a single question
can lead the respondent to discard the entire questionnaire.
4. The answers are relatively complete ( if all appropriate answer categories are provided), and
a minimum of irrelevant responses are received. For instance, asking rural respondents the
open-ended question, " How often do you go to town?" may provide such relevant and
unusable answer as " whenever I wish " or " when I can get a ride," when what is required
is an estimate of the frequency. A close ended question with the response categories "once
a week or less, two to five times a week, every day," will be more likely to elicit usable
information.
5. Closed- ended questions is often easier for a respondent to answer as he or she merely has
to choose a category, while formulating an original answer for an open- ended question can
be much more difficult.

2
Disadvantages of closed-ended questions:

1. It is easy for a respondent who does not know the answer or has no opinion try to guess the
appropriate answer or even to answer randomly.
2. The respondent may feel frustrated because the appropriate category for his or her answer
is not provided at all or is not provided in sufficient detail, and there is no opportunity for
the respondent to clarify or qualify his or her answer.
3. There may be too many answer categories to print on the questionnaire, or there may be so
many categories that the respondent could not remember them all if they were read aloud
by the interviewer. This causes rereading and repetition, can lead to errors, and results in a
much lengthier and more expensive interview.
4. Differences in interpretation of what was meant by the question may go undetected,
whereas in an open-ended question one might be ale to tell form the written answer hat
the respondent misinterpreted the question.
5. Variations in answers among the different respondents may be eliminated artificially by
forced- choice responses.
6. There is more likelihood of a clerical error as the respondent may circle a three when he or
she meant to circle a two. Such mistakes do not occur with open-ended responses.

The advantages of open-ended questions:

1. They can be used when all of the possible answer categories are not known, or when the
investigator wishes to see what the respondent views as appropriate answer categories.
For example, the open-end question, what are the major problems confronting Addis Ababa
at present time? May reveal some findings that the researcher did not anticipate, in
addition to the expected ones (theft, smog, juvenile delinquency, smog, taxes, crime).
2. The allow the respondent to answer adequately, in all detail he or she likes, and to clarify
and qualify his or her answer.
3. The can be sued when there are too many potential answer categories to list on the
questionnaire. For example, one might receive 101 answers to the question of problems
facing Addis Ababa, many of which have a frequency of only a few each. It would be
awkward to list all 101 problems on a questionnaire.
4. They are preferable for complex issues that cannot be condensed into a few small
categories.
5. They allow more opportunity for creativity or self-expression by the respondent. He /she
feels the answers are uniquely his or hers instead of being forced upon him or her by the
researcher.

Disadvantages of open-ended questions:

1. They may lead to collection of worthless and irrelevant information beyond the researcher's
expectation,
2. Data are often not standardized from person to person, making comparison or statistical
analysis such as computation of percentages difficult.

3
3. Coding is often very difficult and subjective, leading to low intercoder reliability.
4. Open-ended questions require superior writing skills, better ability to express one's feelings
verbally, and generally a higher educational level than do closed-ended questions.
5. Open-ended questions designed to be general and to explore all dimensions of the subject
may be too general for the respondent to understand what is meant, requiring the use of
robes or of more specific follow-up questions administered by the data collector. This
feature makes open-ended questions generally unacceptable for mailed-out or self-
administered questionnaires, which generally rely upon less complex closed-ended
questions.
6. Open-ended questions can require much more of the respondent's time and effort, and
may engender a high refusal rate.
7. Open-ended questions require more paper and make the questionnaire look longer,
possibly discouraging some respondents who do not wish to answer a lengthy
questionnaire.

Comparison of open and closed Questions

In view of their usage, the two groups of items are compared in the following table (Bailey,
1982:126- 127).

Table 5.1 The Comparison of Closed and Open Questions

Closed - ended questions Open- ended questions

 They should be used where the answer
categories are discrete, distinct, and
relatively few in number.
 Most of them measure variables that are
nominal (e.g. gender or skin color), ordinal
( e.g. educational level), or intervally
measured questions that have been
collapsed questions that have been
collapsed into relatively few ordinal
categories (e.g. income questions
collapsed in the form $101-$300,$301-
500,etc,
 They are generally self-contained can be
answered quickly, and require fewer
instructions than open-ended ones. Thus,
closed-ended questions can be used with a
sample that has a lower educational level
and are generally considered more
appropriate for mailed-out and self-
administered questionnaires.
 They are used for complex questions that
cannot be answered in a few simple
categories but require more detail and
discussion.
 They are used to elicit the respondent's
unique views, philosophy or goals.
 They are especially helpful in preliminary
investigations in which the researcher has
not yet decided which characteristics of
the phenomenon are relevant to his/her
study and needs to describe all potentially
relevant characteristics in detail.
 They are preferred wherever accuracy,
detail, and exhaustiveness are more
important than time or simplification of
coding and data processing.
 They are used to measure Variables that
are ratio and intervally scaled, as it is
impractical to list all possible answers.

4
A great deal of questionnaires contain a mixture of both. Most contain a number of questions
on simple nominal background variables (e.g., gender), which are fixed alternative, and a
number of other closed -ended questions (agree/disagree or yes/no). They may also contain
many open-ended thought questions. A questionnaire containing primarily fixed-alternative
questions should contain at least one open-ended question (at the end of the questionnaire) to
determine whether anything of importance to the respondent has been omitted.

Response -category format

A) Open-ended questions. Response categories for open-ended categories generally consist

of only a blank space in which the respondent can write his or her answer. The
researcher can regulate the amount the respondent writes by how much space is allowed
if he is careful in enough in estimating sufficient space for complete answer. Some open-
ended questions, particularly those requiring a single number (age at last birthday) for an
answer, may require only a blank (-----) for a response category.
B) Closed-ended questions. Response categories for closed-ended or fixed-alternative
questions are somewhat more complicated. Most answers are variables having two or
more discrete categories. Category style depends at least partially upon whether the
variable is (1) nominal (discrete nonnumerical categories such as male or female), (2)
ordinal (rank-ordered categories such as most favorable, neutral, least favorable), (3)
interval ( ordered categories for which intervals between all ranks are equal, such as
the Fahrenheit temperature scale), and (4) ratio ( which is merely interval measurement
with a fixed zero point, such as age).

In view Bailey (1982:128 ), the basic rule for writing answer categories is to provided all possible
answers in as clear and uncluttered a fashion as possible. The means of answering the question
(check a box , circle a number) should be explicit. Nominal scales. For nominal questions with
factual answers ( as opposed to opinion), the general practice is to list each alternative,
providing a blank, a box to be checked, or a number to be circled.

For example:

Check the appropriate blank

1. Sex: male X ; Female ____

2. What type of car do you drive? Ford ___ Toyota ____ Mitsubishi ____

Check the appropriate box

1. Sex Male X ; Female 

2. Similarly made

5
 Circle the appropriate number

1. Sex: male 1 ; Female 2

Inventory method.

Another response category called inventory method is also used. An inventory is list of items to
be checked or marked by the respondent. For example:

Check all the ways in which you have learned some bit of news or information in the last week.

____ Television
____ Radio
____ Newspaper
____ magazine
____ word of mouth
____ Other (please specify)
Grid format

The box format, often called a grid, is frequently used with inventories. According to Bailey
( 1982: 131), a grid is a boxed format for an inventory that provides an evaluation for each item,
or may even be a two-way inventory. As illustration of this point, see Figure 5.1 below.

Figure 5.1. A Grid Gathering Information on Both the type of Illness and Type of Treatment.

Nose Chest Gargle Inhalants Throat syrup Throat Throat Cough

Drops Rub sprays candy Tablets medicine

Chest cough
Dry cough
Smoker's cough
Severe cough
Sinus pains
Common cold
Sore throat
Infected tonsils
Dry throat
Catarrh
Indigestion
Asthma
Bronchitis

6
Ordinal/ Likert scales. A response mode for such questions could be illustrated as follows.
Mathematics is essential for all business studies students: Strongly agree  agree  uncertain
 Disagree  strongly disagree 

It is often desirable , where possible, to construct the same answer format for a whole series of
question. If this can be done, one set of response category labels can serve for a whole set of
categories if the labels are placed above the categories like titles.

For example

Agree Disagree
 If you start trying to change things
very much, you usually make them ____ ____
worse.
 All groups can live in harmony in this
country without changing the ___ ____
system in any way

The obvious advantage of such a design is that the instructions for completing all questions in
the series can be stated only once, at the beginning, and need not be repeated. Generally not
all questions in the questionnaire will have the same response series. But often there will be
several different sets of fixed-alternative questions, each set having different set of categories.
levels depending on the nature of the study and the researcher's interest. Some of the
commonly used response categories include the following:

1. Strongly agree / agree/ neutral/ disagree/strongly disagree/unable to answer.

2. Excellent/adequate/barely adequate/inadequate/terrible/unable to answer.
3. Excellent /good /all right /poor /bad /unable to answer
4. Often/ sometimes/ almost never
5. Radical left /very liberal /somewhat liberal /middle/somewhat conservative / very
conservative/radical right.
6. Certainly true /quite often true/ seldom true/never true/ don't know
7. Very important/ important/ Somewhat important/ not important/ don't know.

Ranked responses.

Another response category format for designing questions is to present a list of topics and have
the respondent rank-order them.

Examples.

7
 1. Please rank each category below to indicate how important it is to you as a source of
financing. Rank the most important source "1", the second most important source "2", etc.
___ Full-and/or part-time work
___ Savings
___ Parental aid
___ Loans
___ Scholarships
___ Grants
___ Veterans' benefits

2. The following are some of the problems faced by citizens of Addis Ababa. Please order
them in terms of importance, from ( most important) to 5 (least important).
___Smog
___Traffic
___Taxes
___Crime
___Drug Addiction
The key to writing a good ranking item is to make sure your directions are complete and that the
criteria for he ranking are clearly specified. It is also quite important to keep the list of items to be
ranked no longer than ten items, to keep them all on the one page, and to make clearer whether
you want all items rank or just, say, the top five.

Interval scales. These include the questions of the form:

1. What is your age? 10-14  15-19  20- 24 

2. What is your daily income? Below 5 Birr  6-10 Birr  11-15 Birr 
Over 15 Birr 

Question Order

When combining questions to form the final questionnaire the researcher must decide how
many questions to include and in which order to present them. In this regard, a few general
rules are forwarded by Bailey (1982: 138 -142). These include the following:

1. Put sensitive questions an open-ended questions late in the questionnaire. If sensitive

questions, such as those about sexual behavior, income, or religion, are placed first, the
respondent may react to them adversely and refuse to continue.
Open -ended questions should also be placed late, even if they deal with nonsensitive
issues, because they generally require more thought and writing from the respondent and
therefore take longer to answer than fixed- alternative questions. Thus the respondent may
refuse to continue or to complete answering all questions.

2. Ask easy-to- answer questions first. The first questions on the questionnaire should be easy
to answer but not a threat to the respondent. The first question should generally e about a

8
 fact rather than an opinion or a belief and should require little thought. A question about
gender or age is preferable to one about philosophy of life Opening questions usually deal
with background information such facts as age, gender, marital status, residence,
occupation, and education (but not income, religion, or race). If possible, the opening
questions should be interesting ones that will stimulate the respondent to complete the
questionnaire.
3. Ask information needed for subsequent questions first. It is often helpful, particularly for
questions that may tax the respondent's memory or that deal with other members of the
family, to first elicit information, such as family members' names, to be used in asking
subsequent questions.
4. Place questions in logical order. The average person probably arranges his or her responses
to question in some general order, such as time sequence and sub topics of study. Thus the
researcher should follow this practice when constructing a questionnaire. For instance, if
the study is targeted on fertility survey of a given family, there could be a number of
subtopics (frame of references) that could be used to design questions. These include job
history, contraceptive practices, birth rates, etc. If the respondent is asked about the job
history, he/she finds it easy to respond when the time sequence is met, either from first job
to the present or vice versa. Similarly, it s advisable to ask questions about all the
respondent's jobs before moving to questions about contraceptive practices, i.e., first job,
etc.; instead of first job, contraceptive practices, second job, contraceptive practices, etc.,
order.

To sum up, constructing a questionnaire should establish a sequence such as a time and
care must be taken against mixing topics or frames of reference as well as against jumping
abruptly from one frame of reference to another. These help to maintain the respondent's
train of thought.

5. Avoid establishing a response set. A response set is a tendency to reply to items in

particular way, regardless of the question's content or the correct answer. In a
questionnaire design, a response set could be created due to either of the following:
(a) "Social desirability" the tendency to agree with statements that are socially desirable or
supported by norms;
(b) "acquiescence", the tendency to answer yes rather than no or to agree rather than
disagree,
(c) a preference for strong statements versus moderate or indecisive ones;
(d) "question sequence", sequences of questions asked in a very similar format produce
stereotyped responses, such as a tendency to endorse the right-hand or the left-hand
response, or to alternate in some simple fashion.
In such situations, the researcher has to change the order of the questions by looking at
the pilot-test results. he/she may randomize question order or vary the question and/or
answer format from question to question.

9
6. Separate reliability-check question pairs.

Pairs of questions, one stated positively and one negatively, are often used to check
reliability. For example, we might ask " Abortion should be legalized ? agree / disagree)" at
one point in the questionnaire and "Abortion should not be legalized ( agree/disagree) "
later.

7. Place scale items according to response required. The researcher must decide whether
present the scale items in a single group or to divide them.
8. Vary question by length and type. The researcher has to vary the placement of questions as
to question format, response format, length, and whether open-or closed-ended ones.
This may help keep the respondent interested.
9. Determine whether funnel technique is applicable. In funnel technique; broad, general or
even easy to answer open-ended questions are asked first, and then the funnel is narrowed
by asking more specific questions. This technique puts the respondent at ease first
nonthreatening (even irrelevant ) questions. By using the funnel technique in conjunction
with "filter" questions, the researcher can determine whether specific questions apply to
the respondent, and thus avoid asking questions that do not. A filter or screening questions
is one that determines whether future questions are applicable to the respondent. For
example, instead of asking the respondent how many cigarettes he or she smokes or what
brand, the researcher first asks whether the respondent smokes. If the answer is no, then
the specific smoking questions are skipped.

Level of wording of Questions

The fundamental characteristic of a good questionnaire item is that it is clearly understood.

When respondents find your question difficult to understand or answer, he or she is likely to
mentally change it in some way that makes it easier to answer. In other words, when your
questionnaire items are difficult, various respondents may understand the items differently and
provide different answers. This may put your study at risk of failure.

However, by the use of the following methods (suskie, 1995::44-51), you can ensure your
question will be clearer to your respondents, so they will interpret it in the same way:

1 Keep it short. Short, straightforward items are easier to understand than compound or
complex statements and therefore yield more accurate answers. They also mean a shorter
questionnaire and therefore an increased response rate. Lengthy questions may cause the
respondent to lose focus and may also cause fatigue and impatience.
Here is an example of a question that could be made much shorter by rewriting it:

Which of the following statements best describes your enrollment pattern at Addis Ababa

Science and Technology University?

10
 ___ I was enrolled primarily as a full-time student.
___ I was enrolled primarily as a part-time student
2 Make sure it asks only one question . Rather than ask , " should 18-year old have the right
of an adult? ", specify which rights your mean: voting, drinking, Marriage, or whatever. A
"double-barreled" item like the following asks several questions that might each be
answered differently: "Is this job a step toward career goals or related to your major? To
avoid this problem, check your use of the words " and" and " or" and limit your questions to
one adjective or adverb each.
3 Keep it readable. Keep the vocabulary level low, go over each word in your item and make
sure it is as simple as possible. Avoid the jargon of your field and the use of double
negatives. If you use negative items, underline then and/or put them in all capital letters or
boldface to draw attention to them: Not, EXCEPT, ETC. Also avoid words which mean
different things to different people.
4 Make all definitions, assumptions, and qualifiers clearly understood.
Make all definitions, assumptions, and qualifiers clearly understood. Clarify terms that could
be misunderstood. A question like "Do you use the IBM mainframe computer?" can be
interpreted many ways. One person could say that anyone who's ever looked at a printout
is technically a user; another could say that only those with "hands on" experience are
users. Similarly, rather than asking for age ask for year of birth or age as of January 1.
Terms like "adequate," "Minority " "value" " how much", "usually," "access." “convenient,"
"most," " quality," "procedures, " " diversity," "not too often," "acceptable," " very often,"
and "now" are full of ambiguities. Even seemingly straightforward words like " you" and
"never" can be misinterpreted ( is "You" singular or plural? Does "never" mean absolutely
never" or "virtually never"?).

When asking for an estimate of something spell out the period of time you are interested in:
last week? Per day? When you ask for salary, similarly specify whether you want gross or
net salary and whether you want salary rate per year, per month, biweekly, weekly, or per
hour. In fact, specify the unit of measurement for all numerical responses. Do you want the
student's course load expressed as number of courses or of number of credits? Label the
response blank with the appropriate unit of measurement to help clarify your question.

When asking for a rating or comparison, make sure you give a clear point of reference. Spell
out the criteria to be used. Do you want the rated on courtesy, cheerfulness, helpfulness,
or knowledge? In terms of employees here or employees in general?

If your question includes a qualifying phrase (e.g., "except for housing, " if you commute"),
place it at the beginning of your stem. Your sentence structure may not be the best, but the
qualifier won't be overlooked.

Make sure your questions have no hidden assumptions written into them. How many
assumptions can you find in this classic "Have you stopped beating your wife?

11
 When it comes to definitions, assumptions, and qualifiers, consider the feelings of your
respondent. Don't phrase definitions in such a way that he or she feels stupid. Make sure
the respondent has all the information necessary to answer the question. Be careful of
presuming, say, that the respondent has graduated from high school.

5 Avoid making significant memory demands. Don't ask alumni, for example how they felt
about your institution as freshmen. You can't possibly get accurate answers; too much as
happened to your respondents between then and now that will color their memories.
Asking "How many times did you visit the library last week?" Will give you more valid data
than asking, "How many times did you visit the library last year?"
Consider this question:

How much do you usually spend on textbooks each semester?

This item requires the respondent to recall and make a mental average of spending over
several semesters. The question would be greatly improved if it were rewritten as:

About how much did you spend on textbooks last fall?

It's more likely that your respondents could answer this question more accurately because
its memory and cognitive demands are not as great.

6 Avoid asking for very precis responses. Questions asking for a very precise response (such
as annual salary to the exact dollar, grade point average to two decimal place, how many
times the respondent visited the library or called home in the past year) make severe
memory cognitive demands on respondents and will not be answered accurately. Rather
than asking for an exact amount, use a multiple-choice or rating scale format.
Respondents will also respond less accurately to a question with many response categories ( for
instance, a question asking the respondent to identify which one of ten listed factors was most
important in deciding to attend your school), than a question with just a couple of response
categories.

7 Avoid asking for broad generalizations about attitudes or opinions. It is harder for people
to make broad generalizations, especially about their feelings or opinions, than to describe
concrete actions. Opinion questions are also often harder to analyze and interpret than
behavioral "Vignette" questions, since respondents, views come from different frames of
reference. Student opinions of our advisement program, for example, may be based in
part on the quality of their high school advisement experiences, the experiences of their
friends, or on stories they've heard advisement at other colleges. Your questionnaire will
therefore be less ambiguous and more reliable if you make as many items as possible concrete
or "behavioral". Asking for a behavior rather than a feeling or opinion. For example, instead of
asking, " Do you consider yourself a hardworking student?" ask, How many hours per day

12
 do you usually study?" or "How many hours did you study yesterday? Try asking several
questions and inferring an opinion rather than simply asking for the opinion. Don't try to make
your questions too specific, however. A large number of " trivial" questions many cause
confusion or a loss of interest and will lengthen your questionnaire considerably.

8 Make the item easy fast to answer. The respondent should easily understand how to
decide on an answer and how to record it. A simple answer format requiring only a few
checks is better than a complicated scoring key.
9 Keep it interesting. Don't let your items get pedestrian or monotonous. The six items
below would quickly frustrate almost any respondent:
10 Avoid biased, loaded, leading, or sensitive questions. While no ethical researcher writes
these kinds of questions on purpose, it's easy to write them inadvertently. Here are some
examples (all from real surveys!):

Do you agree that computer literacy should be an objective or requirement of the college?

Are you concerned that higher education spending is rising faster than health-care spending?

Do you favor requiring full-time professor at institutions which receive state funds to spend a minimum
number of hours teaching?

Should harsher penalties be enacted for crimes committed against the elderly?

Do you feel there is too much power concentrated in the hands of labor union officials?

Are you in favor of forcing state, country, and municipal employees to pay union dues to hold their
government jobs?

These questions are biased because they are all presented as statements that respondents
must agree or disagree with, and respondents generally find it easier and less threatening to
agree than disagree. They are also "loaded" with non- neutral terms, such as "harsher",
"power" and "forcing." Furthermore, the questions present arguments on only one side of the
issue; as many of the questions now stand, few would disagree with them.

To avoid biased or loaded questions, try the following:

a) Image you're writing questions from the opposite of view. For example, if you're trying to
collect data to support the need for expanded counseling services, imagine you're trying to
cut back on them. Would you still ask the same questions and phrase then the same way?

b) Ask about the pros and cons of an issue. Don't ask the respondent to make criticisms
without giving him of her a chance to praise as well. And don't ask the respondent if he or

13
 she is in favor of a new program or extending the hours that an office is open (who wouldn't
be ?) without noting the costs and liabilities.
c) Avoid questions that people are uncomfortable answering honestly.
People tend to be comfortable with the status quo, to deceive themselves reluctant to
admit any inferiority or wrongdoing. Many will not answer honestly if you ask them
whether they smoke, watch TV, go to art museums or church, or vote. Many will also not
report accurately their salary, their age, their job, their grade point average, whether they
failed a particular course, and whether they were rejected from other schools to which they
applied. Many people will indicate they agree with the statement, "I believe all people
should be respected as equals," even if they demonstrate prejudice or intolerance toward
certain groups.

Be very careful phrasing any questions whose responses have any potential level of social
desirability. In some instances, such as asking about age or salary, people may respond
more truthfully to broad categories than a request for a specific figure.

Questions worded in the first person ( "Do you cheat?") will be answered differently from
those in the third person (Do your friends cheat?"). Some respondents may find third-
person questions easier to answer honestly.

Try not to phrase questions in a psychologically threatening way. Rather than ask if a
respondent does something good or bad (e.g., vote, cheat on tests), ask how often he or she
does it, it also helps to ask about very specific behaviors. If you ask, "Did you glance at your
neighbor's paper during the last math test?", some respondents may not even realize you
are asking about cheating behavior. Others may recognize your intent but still feel can
answer your question honestly without labeling themselves "cheaters."

Bottom line: Few researchers have the skill to conduct the level of research needed to elicit
accurate answers to sensitive questions. If you're not confident you'll get an accurate,
honest answer, don't ask the question.

d) If appropriate, let people admit they don't know or can't remember. Forcing an opinion
from people who have one or a memory from people who can't remember one is
tantamount to forcing them to lie. Consider, for example, the following statement: Unit
college's administrators talk about the importance of diversity, but never see them
attending diversity programs." Students who do not attend diversity programs themselves,
or who do know many administrators, cannot respond validly to this statement.

e) Use two different versions of the question in two different parts of the questionnaire. Do
this in your pilot test and see how the responses differ then decide which one(s) to use in
the final version.

14
f) Use multiple assessments with a variety of methods, formats, and topics. The only way
you can be sure your questions aren't biased or leading is to cross-check your responses
against other data. This concept is called "triangulation"

Cover Letter or Introductory statement of a Questionnaire.

After the questions have been written with a reasonable level of wording and their order
decided upon, the remaining tasks in questionnaire construction consists of writing an
introductory statement or cover letter. This is very important part of a questionnaire
construction because it justifies the study to the respondent and often determines whether
she/he cooperates or not. As such, the introductory statement should be concise, courteous
and businesslike.

The cover letter is usually written on the letterhead of the organization conducting or
sponsoring the survey in order to help legitimize the survey to be conducted to the respondent.
In addition, the introductory statement should:

(1) state the purpose of the study and its significance;

(2) explain who the data collector is, the basis of its authority, and why it is conducting the
study;
(3) tell how and why the respondents were selected;
(4) explain why their answers are important;
(5) tell how to complete the forma and list the person to call if help is needed to complete the
form;
(6) provide assurance of confidentiality and anonymity when appropriate;
(7) explain how the data will be used;
(8) explain who will have access to the information;
(9) Provide mail-back instructions (if necessary); and
(10) present the response efforts as a favour and thank the respondents for their cooperation.

Below is an example of introductory statement, which has been extracted from "Effective
Management" in school's project" in UK:

Dear Respondent,

The aim of this study is to identify examples of management structures and processes in
individual schools which staffs of these schools have recognized as effective practice. The
investigation has who main phases:

15
- the present questionnaire survey to obtain heads' and teachers' views;
- case studies in a selection of schools using interviews with heads and staff to explore the
issue in more depth.

The purpose of this questionnaire is to obtain your perceptions and views on various aspects of
management in the school. It forms a major part of the project and the information you give
will enable us to establish a clear picture of teachers' views of school management. Your
cooperation to responds is very important to this survey because it represents hundreds of
others who are not included in our sample.

Please answer every question. The first section of the questionnaire seeks basic factual
information where answer should be writ in the space provided or circled to indicate the
appropriate response.

The rest of the questionnaire has been designed as a series of statements where your
agreement/disagreement can be indicated by circling the appropriate number. Please answer each
question by circling the number which most closely represent your opinion using the following 5 points
scale:
5= Strongly Agree, 4= agree, 3= Don’t know,

2= Disagree, 1= strongly disagree.

Space is provided at the end of each section for you to add further comments or explanation.

We would stress that all in information you provide will be treated in the strictest confidence.

We thank you very much for your cooperation

Typographical and other considerations of a questionnaire.

The size, style, density of type and duplication are signposts to guide the respondent's eyes and
to signal the kind of information being presented. In general, a carefully laid-out, grammatically
flawless looks shorter and more professional, is much easier to complete, and therefore yields a
higher response rate and better quality data. In this regard, suskie (1995: 64-65) has suggested
important remarks as follows:

1. Cleanly format your questionnaire. Putting the text into two columns rather than running
it across makes the survey easier to read and shorter in appearance.
2. Use plenty of "white space." Use generous margins and spacing. If your questions fill only
three quarters of a page, spread them out.

16
3. Use a smaller (10 - or 11 - point) font. It makes your questionnaire look not only shorter
but also more professional.
4. Stack the responses to multiple-choice items vertically. A respondent can find and mark
the appropriate response more easily when the choices are listed in a column rather than
across one or two lines:
Confusing: ___ yes ___ No

Better: ___ yes

___ No

Responses in one column are also easier for you and your helpers to find and code, and
columns of responses create white space that adds eye appeal.

5. Make spaces for check mark with professionally drawn boxes. Such as , brackets such as
[ ], parentheses such as ( ), or simply a short line such as ___ .
6. Proofread your questionnaire carefully for grammar and spelling. A single error will
detract from the professional appearance of your questionnaire. Ask some lay people
outside your field to read your questionnaire and let you know of any directions, items, or
words they find hard to understand.
7. Reproduce your questionnaire by the best means possible. Offset printing or top-quality
photocopying look best and are not too expensive. Dittos, mimeographs, and fuzzy
photocopies look antiquated and unprofessional and can be hard to read.
8. Use quality paper. Subtly colored paper - nothing garish or too bright to be readable - may
be a nice touch with some groups.

Limitations of a Questionnaire.

1. The questionnaire cannot be used with children and illiterates.

2. The return rate of the mailed questionnaire is often as low as 40 percent to 50 percent. As a
result of such a poor response rate, the data obtained are often of limited validity.
3. Sometimes respondents may not like to respond in writing to the question of intimate and
confidential nature or to the questions involving certain controversial issues.
4. It is sometimes difficult to formulate and phrase questions on certain complex, delicate and
intricate problems.
5. There is no check on a respondent who misinterprets a question or gives incomplete or
indefinite responses.
6. Sometimes the respondent may modify his earlier original responses to the questions when
he finds that his responses to latter questions are contradicting the previous ones.

17
 2) Interview Method
The interview is a process of communication or interaction in which the subject or interviewee
gives the needed information verbally in a face-to-face situation (through personal contact ) or
through telephone calls, which are called personal interview and telephone interview
respectively. In personal interview method, the interviewer asks questions in a face-to face
contact to the other person(s), called interviewee(s). [ At times the interviewee may also ask
certain questions and the interviewer responds to those, but usually the interviewer initiates
the interview and collects the information.]

On the other hand, in case of telephone interview, the interviewer contacts the respondents
(interviewees) and gathers relevant information through telephone calls. Such type of data
collection is helpful in industrial surveys and commonly used in developed nations.

In developing nations, however, telephone interview is rarely practiced. Most of research

studies which are conducted in those countries make use of personal interview method.

Advantages and Disadvantages of Interview studies.

The use of interview technique for data collection does have certain advantages and
disadvantages

a) Advantages:
1. More information and that too in greater depth can be obtained.
2. Interviewer by his own skill can overcome the resistance, if any, of the respondents; the
interview method can be made to yield an almost perfect sample of the general
population.
3. There is greater flexibility under this method as the opportunity to restructure questions
is always there, specially in case of unstructured interviews.
4. Observed method as well can be applied to recording verbal answers to various
questions.
5. Personal information can as well be obtained easily under this method.
6. Samples can be controlled more effectively as there arise no difficulty of the missing
returns; non-response generally remains very low.
7. The interviewer can usually control which person(s) will answer the questions. This is
not possible in mailed questionnaire approach. If so desired, group discussions may also
be held.
8. The interviewer may catch the informant off-guard and this may secure the most
spontaneous reactions than would be the case if mailed questionnaire is used.
9. The language of the interview can be adopted to the ability or educational level of the
person interviewed and as such misinterpretations concerning questions can be
avoided.

18
 10. The interviewer can collect supplementary information about the respondent's personal
characteristics and environment which is often of great value in interpreting results.

b) Disadvantages. There are also certain limitations of the interview studies, which include
the following.
1. It is a very expensive method, specially when large and widely spread geographical
sample is taken.
2. There remains the possibility of the bias of interviewer as well as that of the
respondent; there also remains the headache of supervision and control of interviewers.
3. Certain types of respondents such as important officials or executives or people in high
income groups may not be easily approachable under this method and to that extent
the data may prove inadequate.
4. This method is relatively more-time-consuming, specially when the sample is large and
recalls upon the respondents are necessary.
5. The presence of the interviewer on the spot may over-stimulate the respondent,
sometimes even to the extent that he may give imaginary information just to make the
interview interesting.
6. Under the interview method the organization required for selecting, training and
supervising the field-staff is more complex with formidable problems.
7. Interviewing at times may also introduce systematic errors.
8. Effective interview presupposes proper rapport with respondents that would facilitate
free and frank responses. This often a very difficult requirement.

Although the interview method is confronted with a number of difficulties, every effort has to
be made to make it effective and efficient data gathering tool. To this end, interviewers should
be carefully selected, trained and briefed. They should be honest, sincere, hardworking,
impartial and must possess the technical competence as well as necessary practical experience.
Occasional field checks should be made to ensure that interviewers are neither cheating, nor
deviating from instructions given to then for performing their job efficiently.

3) Observation Method
Observation is the process in which one or more person observe what is occurring in some real
-life situation, and they classify and record pertinent happenings according to some planned
scheme (Koul, 1996:168). Observation method appears to be very valuable instrument in a wide
range of research studies.

As a good research tool, observation needs proper planning, expert execution, and adequate
recording and interpretation ( Koul, 1996,p. 168).

19
A) Planning for observation

The planning for observation includes definition of specific activities or units of behavior to be
observed, the nature of the groups of the subjects to be observed, the scope of observation -
individual or group, determination of the length of each observation period, deciding about the
tools to be observed in making the observation and recording , etc.

According to Good (1966: 244-245), planning for observation includes the following factors:

1. An appropriate group of subjects to observe.

2. Selection and arrangement of any special condition for the group.
3. Length of each observation period, interval between periods, and number of periods.
4. Physical position of the observer and possible effect on the subject or subjects.
5. Definition of specific activities or units of behavior to be observed.
6. Entry of frequencies or tallies in the record, as a total for the entire observation period or by
sub-division of time within the observation period.
7. Scope of observation, whether for an individual child or for a group.
8. Form or recording , including consideration of mechanical techniques and such quantitative
factors as number, time, distance, and spatial relationships.
9. Training of the observer in terms of expertness.
10. Interpretation of observations.

B) Execution of Observation
An expert execution of observation includes:

1. Proper arrangement of specific conditions for the subject(s) to be observed.

2. Assuming the proper role or physical positions for observing.
3. Focussing attention on the specific activities, or units of behavior under observation.
4. Handling well the recording instruments to be used.
5. Utilizing the training and experience fairly well in terms of making the observation and
recording the facts.
Classification of observation methods.

Observation methods are classified in various ways depending on the relationship between the
subject and the observer or the nature of the categories of the data to be observed.

In view of the subject-observer relationship, observation is classified into three: direct or

indirect, disguised or undisguised, and participant or non-participant or non-participant (cooper
& Emory, 1995,pp 334-35). When it is seen from the point of view of the data of interest,
observation is classified as either structured or unstructured ( Kaoul, 1996, p.1699). Each of
these types of observation will be briefly discussed below.

20
1. Direct Vs indirect observation
Direct observation describes to situation in which the observer is physically present and
personally monitors what takes place. This approach is very flexible because it allows the
observer to react to and report subtle aspects of events and behavior as they occur. He/she is
also free to shift places, change the focus of observation, or concentrate on unexpected events
if they should occur. A weakness of this approach is that the observers' perception circuits may
become overloaded as events move quickly; they must later try to reconstruct what they ere
not be able to record. Also, observer fatigue, boredom, and distracting events can reduce the
accuracy and completeness of such observation.

Indirect observation is the term used to describe studies in which the recording is done by
mechanical. Photographic, or electronic means, For example, a special camera that takes one
frame every second is mounted in a department of a large store to study customer and
employee movement. Such methods are less flexible than direct observation, but they are
much less biasing and may be less erratic in accuracy. Another advantage of such indirect
systems is the permanent record can be analyzed to include many different aspects of an event.
Electronic recording devices, which have improved in quality and declined in cost, are being
used more frequently in observation research.

2. Disguised Vs undisguised observation

A second factors affecting the observer-subject relationship concerns whether the presence of
the observer should be known to the subject. When the observer is known, there is a risk of
atypical activity by the subjects. The role of the observers should be disguised in situations
where people behave differently if they know they were being observed.

Observers use concealment to shield them-selves from the object of their observation. Often
technical means are used such as one-way mirrors, hidden cameras, or microphones. These
methods reduce the risk f observer bias but bring up a question of ethics. Hidden observation is
a form of spying, and the propriety of this action must be reviewed carefully.

A modified approach involves partial concealment. The presence of the observer is not
concealed, but the objectives and subject of interest are. A study of selling methods may be
conducted by sending an observer with a salesperson who is making calls on customers.
However, the observer's real purpose may be hidden from both the sales person and the
customer (e.g. she may pretend she is analyzing the display and layout characteristics of the
stores they are visiting)

21
3. Participant Vs non-participant observation

In the participant observation, the observer becomes more or less one of the group under
observation. In such situations, the observer will be in the sight of the person being observed
and may actually take part in some activity with the observed individual or group. The observer
may play any one of several roles in observation, with varying degrees of participation as a
visiting stranger, an attentive listener, an eager learner, or a more complete role as participant
observer.

In the non-participant observation, observer takes position where his presence is not
disturbing to the group. He may follow in detail the behavior of one individual or may describe
one or two behavior characteristics of a dozen r more individuals.

4 Structured Vs unstructured observation

Structured observation start with relatively specific formulations. The observer sets up in
advance categories of behavior in terms of which he wishes to analyze he problem, and keeps
in mind the time limit under which he has to make the observation. Structured observations
are executed in controlled situations like classroom or laboratory settings. They are much too
formal and are designed to provide systematic description to test casual hypotheses. They are
more appropriate for conclusive research studies. In the Unstructured observations, it may not
be possible to categorize behavior in advance of observation. Instead of using predetermined
categories, the observer considers aspects f behavior in terms of their context or the situations
of which they are part. Unstructured observations are more appropriate for exploratory
research studies.

C) Recording and Interpreting the observation

The recording of the observation data may either be simultaneous or soon after the
observation. In the former case, the observer goes on recording his observation data
simultaneously with the occurrence of the phenomena observed. In the latter case, the
observer undertakes to record his observation not simultaneously with his actual observation
process, but immediately after he has observed for a unit of time while the details are still fresh
in his mind.

In viewing, classifying and recording behavior, the observer must take utmost care to minimize
the influence of his biases, attitudes and values on the observation report. Most observers are
also subject to fatigue, halo effects, and observer drift a decay in reliability or validity over time
that affects the coding of categories.

22
The observers should be thoroughly versed in the requirements of the specific study. Each
observer should be informed of the outcomes sought and the precise content elements to be
studied. Observer trials with the instrument (if applied) and sample videotapes should be used
until a high degree of reliability is apparent in their observations. Where there are imperative
differences between observers, they should be reconciled. It is also advisable to develop an
observation form while making observations. The specific behaviors to be observed and
recorded should be listed on this form. The observation form should be simple and the
behaviors listed on it may be clearly specified and examples be given Where necessary.

Advantages and disadvantages of Observation.

a) Advantages:
 It can be useful in locating data about non-readers, young children, people with
mental disorders, and laboratory animals.
 Interaction between the observer and a subject can be initiated and carried on.
 The procedure can identify level of skill development.
 It provides a record for future comparisons.
 It provides rationalization and artificial presentation.
 It does not require a special laboratory setting.
b) Disadvantages:
 It is time consuming
 It is costly to collect data.
 Procedures of data collection may alter natural ongoing.
 The data may reflect observer’s bias.

4. Focus Group discussion (FGD).

With origins in sociology, focus group discussion has become widely applicable in business
research.

A focus group discussion is a discussion made by a panel of 8 to 12 respondents led by a trained

moderator. The moderator uses group dynamics principles to focus or guide the group in an
exchange of ideas, feelings, and experiences on a clearly understood topic. The output of the
session is a list of ideas and behavioral observations with recommendations of the moderator.

The qualitative data obtained from group interview might be used for quantitative testing.
Particularly, in exploratory research, the qualitative data that focus groups produce may be
used for enriching all levels of research questions and hypotheses and comparing the
effectiveness of design options.

23
Qualities of a moderator:

The moderator should be equipped with sufficient skill so that he/she can maintain a high
degree of interaction among group members. Unskilled moderators typically find themselves
conducting individual interviews with each of the participants rather than stimulating
interaction within the group. It is only with interaction the focus group discussion can:

- Provide the desired spontaneity of response by participants,

- Produce the degree of emotional involvement essential to produce "depth" level
responses, and
- Produce the kind and degree of rapport which facilitates a "give and take " exchange of
attitudinal and behavioral information.
The key qualifications for moderator include the following

1. Kindness with firmness. In order to elicit necessary interaction, the moderator must
combine a disciplined detachment with understanding empathy.
2. Permissiveness. While an atmosphere of permissiveness is desirable, the moderator must
be all at all times alert to indications that the group atmosphere of cordiality disintegrating.
3. Involvement. Since the principal reason for the group interview is to expose feelings and to
obtain reactions indicative of deeper feelings, the moderator must encourage and stimulate
intensive personal involvement.
4. Incomplete Understanding. The most useful skill of the group moderator is his or her ability
to convey lack of complete understanding of the information being presented.
5. Encouragement. Although the dynamics of the group situation facilitate the participation
of all members in the interaction, there may be individuals who resist contributing.
6. Flexibility. The moderator should be equipped prior to the session with a topic outline of
the subject matter to be covered. By committing the topics to memory before the
interview, the moderator may use the outline only as a reminder of content areas omitted
or covered incompletely.
7. Sensitivity. The moderator must be able to identify, as the group interview progresses, the
informational level on which it is being conducted, and determine if it is appropriate for the
subject under discussion. Sensitive areas will frequently produce superficial rather than
deep responses. Depth is achieved when there is a substantial amount of emotional
response, as apposed to intellectual information indications of depth are provided when
participants begin to indicate how they feel about the subject, rather than what they think
about it

24
Benefits of Focus Group Discussion.

If properly used, group interview does have the following contributions to a given research
study.

1. Synergism. The combined effect of the group will produce a wider range of information,
insight, and ideas than will the accumulation of the responses of a number of individuals
when these replies are secured privately.
2. Snowballing. A bandwagon effect often operates in a group discussion situation in that a
comment by one individual often triggers a chain of responses from the other participants.
3. Stimulation. Usually after a brief introductory period the responders get “turned on “in
that they want to express their ideas and expose their feelings as the general level
excitement over the topic increases in the group.
4. Security. The participants can usually find comfort in the group in that their feelings are
not greatly different from other participants and they are more willing to express their ideas
and feelings.
5. Spontaneity. Since individuals aren't required to answer any question in a group interview,
their responses can be more spontaneous and less conventional, and should provide a more
accurate picture of their position on some issues.
6. Serendipity. It is more often the case in a group rather than individual interview that some
idea will drop out the clue.
7. Specialization. The group interview allows the use of a more highly trained, but more
expensive, interviewer since a number of individuals are being “interviewed
“simultaneously”.
8. Scientific Scrutiny. The group interview allows closer scrutiny of the data collection process
in that several observers can witness the session and it can be recorded for later playback
and analysis.
9. Structure. The group interview affords more flexibility than the individual interview with
regard to the topics covered and the depth with which they are treated.
10. Speed. Since a number of individuals are being interviewed at the same time, the group
interview speeds up data collection and analysis process.

5. Case study method.

According to kothari (1990:140), the case study method is a method of study that involves a
careful and complete observation of an institution, a cultural group or even the entire
community. It is a method of study in depth rather than breadth. The case study places more
emphasis on the full analysis of a limited number of events or conditions and their
interrelations. It deals with processes that take place and their interrelationship. Thus, case
study is essentially an intensive investigation of the particular unit under consideration. The
object of the case study is to locate the factors that account for the behavior-patterns of the
given unit as an integrated whole. In case study method, efforts are made study each and every
aspect of the concerning unit greater details and then from case data generalizations and
inferences are drawn.

25
A) Characteristics of case study method.
The important characteristics of the case study method are as follows:

1. Under this method the researcher can take one single social unit or more of such units for
his study purpose; he may even take a situation to study the same comprehensively.
2. Here the selected unit is studied intensively i.e.; it is studied in minute details. Generally,
the study extends over a long period of time to ascertain the natural history of the unit so
as to obtain enough information for drawing correct inferences.
3. In the context of this method we make compute study of the social unit covering all facets.
Through this method we try to understand the complex of factors that are operative within
a social unit as an integrated totality.
4. Under this method the approach happens to be qualitative but not quantitative. For
instance, under this method we not only study how many crimes a man has committed but
shall peep into the factors that forced him to commit crimes when we are making a case
study of a man as criminal. The objective of the study may be to suggest ways to reform the
criminal.
5. In respect of the case study method an effort is made to know the mutual inter-relationship
of causal factors.
6. Under case study method, the behavior pattern of the concerning unit is studied directly
and not by an indirect and abstract approach.
B) Phases of case study Method
Major phases involved in case study are as follows:

1. Recognition and determination of the status of the phenomenon to be investigated or the

unit of attention.
2. Collection of data, examination and history of the given phenomenon.
3. Diagnosis and identification of causal factors as a basis for remedial or developmental
treatment.
4. Application of remedial measures i.e., treatment and therapy this phase is often
characterized as case work.
5. Follow-up program to determine effectiveness of the treatment applied.

C) Advantages and disadvantages of case study method.

Advantages: Proper application of the case study method, can bring about the following
benefits:

(i) It enables the researcher to understand fully the behavior pattern of the concerned unit.

26
(ii) It helps the researcher to obtain a real and enlightened record of personal experiences
which would reveal man's inner strivings, tensions and motivations that drive him to
action along with the forces that direct him to adopt a certain pattern of behavior.
(iii) This method enables the researcher to trace out the natural history of the social unit
and its relationship with the social factors and the forces involved in its surrounding
environment.
(iv) It helps in formulating relevant hypotheses along with the data which may be helpful in
testing them. Case studies, thus, enable the generalized knowledge to get richer and
richer.
(v) The method facilitates intensive study of social units which is generally not possible if
we use either observation method or the method of collecting information through
schedules. This is the reason why case study method is being frequently used,
particularly in social researches.
(vi) Information collected under the case study method helps a lot to the researcher in the
task of constructing the appropriate questionnaire or schedule for the said task requires
through knowledge of the concerning universe.
(vii) The researcher can use one or more of the several research methods under the case
study method depending upon the prevalent circumstances. In other words, the use of
different methods such as depth interviews, questionnaires, documents, study reports
of individuals, and the like is possible under case study method.
(viii) This method is a means to well understand the past of a social unit because of its
emphasis of historical analysis. Besides, it is also a technique to suggest measures for
improvement in the context of the present environment of the concerned social units.
(ix) Case studies constitute the perfect type of sociological material as they represent a real
record of personal experiences which very often escape the attention of most of the
skilled researchers using other techniques.
(x) Case study method enhances the experience of the researcher and this in turn increases
his analyzing ability and skill.
(xi) This method makes possible the study of social change. On account of the minute study
of the different facets of a social unit, the researcher can well understand the social
change then now. This also facilitates the drawing of inferences and helps in
maintaining the continuity of the research process. In fact, it may be considered the
gateway to and the same time the final destination of abstract knowledge.
(xii) Case stud techniques are indispensable for therapeutic and administrative purposes.
They also of immense value in taking decisions regarding several management
problems. Case study data are quite useful for diagnosis, therapy and other practical
case problems.
Disadvantages: Important drawbacks of the case study method may as well be highlighted.

(i) Case situations are seldom comparable and as such the information gathered in case
studies is often not comparable. Since the subject under case study tells history in his

27
 own words, logical concepts and units of scientific classification have to be read it into it
or out of it by the investigator.
(ii) Real information is often difficult to be collected because the subjectivity of the
researcher does enter in the collection of information in a case study.
(iii) The danger of false generalization is always there in view of the fact no set rules are
followed in collection of the information and only few units are studied.
(iv) It consumes more time and requires lot of expenditure. More time needed under case
study method since one studies the natural history cycles of social units and that too
minutely.
(v) The case data are often vitiated because the subject may write what he thinks the
investigator wants; and the greater the rapport, the more subjective the whole process
is.
(vi) Case study method is bases on several assumptions which may not be very realistic at
times, and as such the usefulness of case data is always subject to doubt.
(vii) Case study method can be used only in a limited sphere; it is not possible to use it in
case of a big society.
(viii) Response of the investigator is an important limitation of the case study method. He
often thinks that he has full knowledge of the unit and can himself answer about it. In
case the same is not true, then consequences follow. In fact, this is more the fault of
the researcher rather than that of the case method.

5.1.2 Collection of Secondary Data.

Many researchers use data from the available data archives (either in the form of documents
or survey results and codebooks) for secondary analysis. Secondary data are gathered or
authored by another person. They refer to the data which have already been collected and
analysed by someone else. Secondary data may either be published data or unpublished
data. Usually published data are available in:

a) Various publications of the central, regional and local governments;

b) Various publications of foreign governments or of international bodies and their
subsidiary organizations;
c) Technical and trade journals;
d) Books, magazines, and newspapers;
e) Reports and publications of various associations connected with business and
industry, banks, stock exchanges, etc.;
f) Reports prepared by research scholars, universities, economists, etc., in different
fields; and
g) Public records and statistics, historical documents, and other sources of published
information. The sources of unpublished data are many; they may be found in
diaries, letters, unpublished biographies and autobiographies and also may be

28
 available with scholars and research workers, trade associations, labour bureaus
and other public /private individuals and organizations.

5.1.3 Selection of Appropriate Method for Data collection

There are various methods of data collection. As such the researcher must judiciously
select the method(s) for his own study, keeping in view the following factors:

1. Nature, scope and object of enquiry . This constitutes the most important factor affecting
the choice of a particular method. The method selected should be such that it suits the
type of esquire that is to be conducted by the researcher. This factor is also important in
deciding whether the data already available (secondary data ) are to be used or the data not
yet available (primary data ) are to be collected.
2. Availability of funds. When funds at the disposal of the researcher are very limited, he will
have to select a comparatively cheaper method which may not be as efficient and effective
as some other costly method. Finance, in fact, is a big constraint in practice and the
researcher has to act within this limitation.
3. Time factor. Availability of time has also to be taken into account in deciding a particular
method of data collection. Some methods take relatively more time, whereas with others
the data can be collected in a comparatively shorter duration. The time at the disposal of
the researcher, thus, affects the selection of the method by which the data are to be
collected.
4. Precision required. Precision sought is another important factor to be considered at the
time of selecting the method of data collection. But one must always remember that each
method of data collection has its uses and none is superior in all situations. For instance,
telephone interview method may be considered appropriate (assuming telephone
population) if funds are restricted, time is also restricted and the data is to be collected in
respect of few items with or without a certain degree of precision. In case funds permit and
more information is desired, personal interview method may be said to be relatively better.
In case time is ample, funds are limited and much information s to be gathered with no
precision, then mail-questionnaire method can be regarded more reasonable. When funds
are ample, time is also ample and much information with no precision is to be collected,
then either personal interview or the mail-questionnaire or the joint use of these two
methods may be taken as appropriate method of collecting data. Where a wide geographic
area is to be covered, the use of mail-questionnaires supplemented by personal interviews
will yield more reliable results than either method alone. The secondary data may be used
in case the researcher finds the reliable, adequate and appropriate for his research.

5.2 Processing, Analysis and Interpretation of Data

5.2.1 Data Processing

The data, after collection, have to be processed and analysed in accordance with the outline laid
down for the purpose at the time of developing the research plan. Processing implies editing,

29
coding, classification and analysis refers to the computation of certain measures along with
searching for pattern of relationship that exists among data-group. Interpretation follows analysis
and it simply means explaining the patterns and implication’s of the analysed data.

Processing Operations: The processing operations are as follows:

1. Editing- Editing of data is a process of examining the collected raw data (especially in
surveys) to detect errors and omissions and to correct these. It involves a careful scrutiny of
the completed questionnaires and /or schedules. It is done to assure that information received
are complete as much as possible and have been well arranged to facilitate coding and
tabulation.
a) Field editing – It consists in the review of the questionnaire by the investigator. This type of
editing in necessary in view of the fact that respondents writing styles often can be difficult
for others to decipher. This sort of editing should be done as soon as possible after the
interviews.
b) Central editing –It should take place when all forms or schedules have been completed and
returned to the office. This type of editing implies that all forms should get a thorough
editing by single editor in a small study and by a team of editors in case of a large inquiry.
2. Coding – It refers to the process of assigning numerals or other symbols to answers so that
responses can be put into a limited number of categories or classes. It is necessary for
efficient analysis and through it, the several replies may be reduced to a small number of
classes, which contain the critical information required for analysis. Coding decision should
usually be taken at the designing stage of the questionnaire.
3. Classification- Most research studies result in a large volume of raw data, which must be
reduced into homogeneous groups if we are to get meaningful relationships. This fact
necessitates classification of data, which is the process of arranging data in groups or classes
on the basis of common characteristics, such as:
a) According to attributes- Data are classified on the basis of common characteristics, which
can either be descriptive or numerical.
b) According to class intervals-The numerical characteristics refer to quantitative
phenomenon, which can be measured through some statistical units. Data relating to income,
production, age, weight, etc. come under this category. Such data are known as statistics or
variables and are classified on the basis of class intervals.
4. Tabulation- It becomes necessary for the researcher to arrange the same data in some kind
of concise and logical order, when a mass of data has been assembled. This procedure is
referred to as tabulation. It is the process of summarising raw data and displaying the same in
compact form for further analysis. It is an order arrangement of data in columns and rows.

Reasons for Tabulation

a) It conserves space and reduces explanatory and descriptive statements to a minimum.
b) It facilitates the process of comparison.
c) It facilitates the summation of items and the detection of errors and omissions.
d) It provides a basis for various statistical computations.

30
Problems in Processing

The following two problems of processing affect the data used for analysis purposes:

a) The problem concerning “do not know” responses - The researcher often comes across
some responses that are difficult to handle, while processing the data. One category of such
responses may be “Do not know”. It is of little significance, when its response groups in
small. It becomes a matter of major concern, in which case the question arises, when it is
relatively big.

b) Use of Percentages- Percentages are often used in data presentation for they simplify
numbers, reducing all of then to a 0 to 100 range. Through the use of percentage, the data are
reduced in the standard form with base equal to 100, which in fact, facilitates relative
comparisons. While using percentages, the following rules should be kept in view by
researchers

1. Two or more percentages must not be averaged unless each is weighted by the group size
from which it has been derived.
2. Use of too large percentages should be avoided, since a large percentage is difficult to
understand and tends to confuse, thereby defeating the very purpose for which percentages
are used.
3. Percentages hide the bases from which they have been computed. If this is not kept in view,
the real differences may not be correctly read.

5.2.2 Analysis of Data

It means the computation of certain indices or measures along with searching for pattern of
relationship that exist among the data groups. Analysis, particularly in case of survey or
experimental data, involves estimating the values of unknown parameters of the population and
testing of hypotheses for drawing inferences. Analysis may, therefore, be categorised as
descriptive analysis and inferential analysis.

a) Descriptive Analysis – It is the study of the distributions of one variable. This study provides
us with profiles of companies, work groups, persons and other subjects on any of a multitude of
characteristics such as size, composition, efficiency, preferences, etc. This sort of analysis may
be in respect of one variable or in respect of two variables or in respect of more than two
variables. In this context, we work out various measures that show the size and shape of a
distribution(s) along with the study of measuring relationships between two or more variable.
b) Inferential analysis- It is concerned with the various tests of significance for testing
hypotheses in order to determine with what validity data can be said to indicate some conclusion
or conclusions. It is concerned with the estimation of population values. It is mainly on the basis
of inferential analysis that the task of interpretation is performed.
Statistics in Research
The role of statistics in research is to function as tool in designing researches, analysing its data
and drawing conclusions there from. Most research studies result in a large volume of raw data,
which must be suitably reduced so that the same can be read easily and can be used for further
analysis. In fact, there are two major areas of statistics viz., descriptive statistics and inferential

31
statistics. Descriptive statistics concern the development of certain indices from the raw data.
Inferential statistics concerned with the process of generalisation and is also known as sampling
statistics.

The important statistical measures that are used to summarise the survey research data are:
1. Measures of central tendency or statistical averages;
2. Measures of dispersion
3. Measures of asymmetry (skewness)
4. Measures of relationship

Measures of central tendency or statistical average - It is also known statistical averages.

Mean, median and mode are the most popular averages.
Measures of dispersion - An average can represent a series only as best as a single figure can,
but it certainly cannot reveal the entire story of any phenomenon under study. In order to
measure the degree of scatter, the statistical device called measures of dispersion are calculated.
Important measures of dispersion are :
(a) range
(b) mean deviation, and
(c) standard deviation.

Type of Tests
There are two general classes of tests of significance: parametric and non-parametric. Parametric
tests are more powerful because their data are derived from interval and ratio measurements.
Non-parametric tests are used to test hypotheses with nominal and ordinal data. Parametric
techniques are the tests of choice if their assumptions are met.
Some assumptions for parametric tests include:
1. The observations must be independent. That is, the selection of any one case should not
affect the chances of any other case to be included in the sample.
2. The observations should be drawn from normally distributed populations.
3. These populations should have equal variances

Non-Parametric Tests
Most phenomena or variables in the behavioural science can hardly be described quantitatively
and the mode of gathering information on them requires the use of nominal and ordinal scales
which do not meet the standard requirements of parametric statistics. We need inferential
statistical tests that can be valid under general statistical assumptions. These tests belong to non-
parametric statistics.
Non-parametric procedures are not generally concerned with population parameters. Non-
parametric statistics are distribution-free statistics. The distribution of the variables is immaterial
since the tests are not concerned with population parameters.

Advantages of non parametric statistics

i. Most non-parametric statistics are used with minimum assumptions. Their use is not
restricted and their chance of being used improperly is minimal.

32
ii. Some measurement scales (particularly, nominal and ordinal scales) are weaker than
others. But non-parametric statistics can be effectively used even when the data are
measured on weak measurement scales.
iii. They are easy to compute and interpret. No sophistication in mathematics and or statistics
is required of any researcher or student using them. Therefore, non-parametric statistics
have become the comfort zone of researchers and students, particularly those whose
understanding of mathematics and statistics is limited.

Disadvantages of non-parametric statistics

i. Since non-parametric statistics are simple to calculate and apply, they are often abused by
researchers and students who do not bother to understand when each non- parametric statistics
can be used.
ii. Researchers and students often prefer non-parametric statistics when it is obvious that
parametric statistics are more appropriate. Thus, some important information is wasted.
When to use non-parametric statistics:
i. The hypothesis which is to be tested does not involve population parameter.
ii There is no assumption of normality about the distribution of the variables.
iii. When data are generated from weaker measurement scales, e.g. ranking, frequency counts
and some subjective measuring scales.
iv. Result are needed fast and no statistical elegance( beauty) is required.

Chi-square Test

The chi- square test is an important test amongst the several tests of significance. It is
symbolically written X2 (Pronounced as ki-square), statistical measure used in the context of
sampling analysis for comparing a variance to theoretical variance. As a Non- parametric test, it
can be used to determine if categorical data shows dependence or the two classifications are
independent. It can also be used to make comparison between theoretical populations and actual
data when categories are used. Thus, this technique through the use of which it is possible for all
researchers to

(a) Test of goodness of fit

(b) Test the significance of association between two attributes

Conditions for the application of Chi- square Test

The following conditions should be satisfied before chi-square test can be applied:
a) Observations recorded and used are collected on a random basis.
b) All the items in the sample must be independent.
c) No group should contain very few items, say less than 10. In case where the frequencies are
less than 10, re-grouping is done by combining the frequencies of adjoining groups so that
the new frequencies become greater than 10.
d) The overall number of items in all the groups must also be reasonably large. It should
normally be at least 50, however small the number of groups may be.

33