Module -III

Data Collection and Analysis: Introduction, Need for Data Collection, Methods of
Data Collection, Principles for Accessing Research Data, Data Processing, Data
Analysis, Presentation of Data, Error Analysis, Scientific Models. Scientific
Methodology - Introduction Rules and Principles of Scientific Method, Hypothesis,
Testing of Hypothesis, Basic concepts, Procedure, Important parametric tests: z-test, t-
test, χ2-square test, F test.

Data Collection and Analysis

In research methodology, "data collection and analysis" refers to the systematic

process of gathering and interpreting information to address research questions or
objectives. Data collection involves the gathering of relevant data from various
sources, such as surveys, interviews, observations, experiments, or existing datasets.
Researchers employ appropriate methods and tools to ensure the accuracy, reliability,
and validity of the collected data.

Once the data is collected, it undergoes analysis, which involves organizing,

interpreting, and summarizing the information to derive meaningful insights. Data
analysis techniques vary depending on the nature of the data and the research
objectives. Common methods include descriptive statistics, inferential statistics,
qualitative analysis, content analysis, thematic analysis, and coding.

The primary goal of data collection and analysis is to extract valuable insights,
patterns, and relationships from the data to address research questions or hypotheses
effectively. It allows researchers to draw conclusions, make inferences, and support
their findings with empirical evidence. Moreover, data collection and analysis are
iterative processes, often requiring multiple iterations to refine research questions,
gather additional data, or adjust analysis techniques based on emerging insights.
Overall, effective data collection and analysis are critical components of the research
process, guiding researchers in generating reliable and valid conclusions.
Need for Data Collection

Data collection is essential because it provides the raw material needed for research,
analysis, decision-making, and problem-solving. Without data, we would lack the
information necessary to understand trends, make informed decisions, or develop
effective strategies. Whether it's in scientific research, business operations, or social
studies, data collection allows us to gather evidence, identify patterns, and draw
meaningful conclusions. It helps us address questions, test hypotheses, and validate
theories. In essence, data collection is the foundation upon which much of our
knowledge and understanding is built.

Methods of Data Collection

The task of data collection begins after a research problem has been defined and
research design/ plan chalked out. While deciding about the method of data collection
to be used for the study, the researcher should keep in mind two types of data viz.,
primary and secondary. The primary data are those which are collected afresh and for
the first time, and thus happen to be original in character. The secondary data, on the
other hand, are those which have already been collected by someone else and which
have already been passed through the statistical process. The researcher would have to
decide which sort of data he would be using (thus collecting) for his study and
accordingly he will have to select one or the other method of data collection. The
methods of collecting 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.

1. COLLECTION OF PRIMARY DATA

We collect primary data during the course of doing experiments in experimental
research but in case we do research of the descriptive type and perform surveys,
whether sample surveys or census surveys, then we can obtain primary data
either through observation or through direct communication with respondents in
one form or another or through personal interviews. This, in other words, means
 that there are several methods of collecting primary data, particularly in surveys
and descriptive researches. Important ones are:
(i) observation method
(ii) interview method
(iii) through questionnaires
(iv) through schedules
(v) other methods which include
(a) warranty cards;
(b) distributor audits;
(c) pantry audits;
(d) consumer panels;
(e) using mechanical devices;
(f) through projective techniques;
(g) depth interviews
(h) content analysis.

OBSERVATION METHOD

Observation is a helpful way for researchers to watch and record behavior without
relying on people's answers. They plan carefully and make sure their observations are
accurate. There are two types: structured, where everything is planned, and
unstructured, which is more flexible. Sometimes, researchers join the group they're
studying (participant observation), while other times, they watch from a distance (non-
participant observation). Both ways have their pros and cons. Controlled observation
follows strict plans, while uncontrolled observation is more casual. Overall,
observation can give valuable insights, but it can also be expensive and limited.
Interview Method

The interview method of collecting data involves presentation of oral-verbal stimuli

and reply in terms of oral-verbal responses. This method can be used through personal
interviews and, if possible, through telephone interviews.

(a) Personal interviews: The personal interview method involves direct interaction
between an interviewer and respondents, either in person or indirectly through
cross-examination. Structured interviews follow a set format, while unstructured
interviews allow for flexibility in questioning. Clinical interviews delve into
underlying feelings or motivations, while non-directive interviews encourage
respondents to freely discuss a topic with minimal questioning. This method
offers advantages such as obtaining in-depth information, flexibility in
questioning, and the ability to control who answers. However, it is costly,
susceptible to bias, time-consuming, and may introduce errors. Successful
interviews require careful selection, training, and briefing of interviewers, along
with creating a friendly atmosphere, asking intelligent questions, and recording
responses accurately while maintaining impartiality and focus. Regular checks
ensure adherence to guidelines and address any issues promptly.

(b) Telephone interviews: Telephone surveys are a quick and cost-effective way to
gather information by calling people on the phone. They're good because they're
flexible, fast, and have higher response rates. You can easily reach people who have
phones and record their answers without making them uncomfortable. But, there are
limits. People only have a short time to answer, and not everyone has a phone, so
your survey might not cover everyone. Plus, it's hard to ask detailed questions, and
there's a chance the person asking the questions might influence the answers.

COLLECTION OF DATA THROUGH QUESTIONNAIRES

The questionnaire method is commonly used for big surveys by many groups, like
individuals, organizations, and governments. It works by sending a set of questions to
people by mail, and they answer and send it back. This method is liked because it's
cheap and doesn't have the interviewer's bias. People can take their time to answer, and
it's good for reaching people who are hard to find. But there are problems too.
Sometimes people don't send back the questionnaire, and it's slow. Also, it only works
if people are educated and willing to cooperate. Once the questionnaire is sent, it's
hard to change, and some questions might be hard to understand or get skipped. Before
doing a big survey, it's smart to do a small test first (pilot survey) to check if the
questions are good. If the questionnaire isn't done right, the survey might not work. So,
it's important to make sure the questions are clear and in the right order before sending
them out.

Researcher should note the following with regard to these three main aspects of a
questionnaire:

1. General form: A questionnaire can be structured or unstructured. Structured

questionnaires contain predetermined questions presented consistently to all
respondents, ensuring standardization in responses. Questions may be closed
(requiring 'yes' or 'no' answers) or open-ended (allowing free responses). Structured
questionnaires are straightforward to administer and analyse but may limit the depth of
information obtained. Conversely, unstructured questionnaires provide a general guide,
allowing interviewers to formulate questions based on respondents' responses. This
method captures a wider range of data and is effective for exploring new topics or
probing attitudes and reasons. Pretesting unstructured questionnaires can inform the
creation of structured ones for the main study, balancing standardization with
flexibility and depth.

2. Question sequence: The order of questions in a questionnaire is really important. It

helps make sure people understand the questions and give good answers. We should
start with easy and interesting questions to keep people interested. It's best to avoid
asking tough or personal questions at the beginning. After that, we can ask the
important questions related to what we're studying. The order should make sense to the
person answering. If we're testing the questions first, we can change the order if
needed. Tricky questions should come later so we can still get useful information even
if they're not answered. Overall, the order of questions affects how people answer, so
we need to plan it carefully.

3. Question formulation and wording: When making a questionnaire, it's super

important to make sure the questions are easy to understand and don't bias the answers.
Each question should focus on one idea and fit into the overall plan. There are two
main types of questions: multiple choice and open-ended. Multiple choice questions
are easy to answer and analyse, but they might limit people's opinions. Open-ended
questions let people answer in their own words, which can give more insight, but
they're harder to analyse. Usually, a mix of both types is best. It's also crucial to use
simple and clear language that everyone can understand, and to avoid words with
multiple meanings or emotional impact. The questionnaire should be short, organized
logically, and leave personal or tough questions for later. Providing space for answers,
clear instructions, and using attractive paper can encourage people to respond. Control
questions help ensure the answers are reliable, and including options like "don't know"
or "no preference" is helpful too. Overall, making a good questionnaire takes practice
and attention to detail.

COLLECTION OF DATA THROUGH SCHEDULES

This method is a lot like using questionnaires, but with a twist. Instead of people
filling out forms themselves, specially trained folks called enumerators go to them and
ask the questions in person. They record the answers right there on a form called a
schedule. Sometimes, the people being asked the questions fill out the form with the
help of the enumerators. The enumerators explain the purpose of the questions and
help out if anyone has trouble understanding them. Choosing the right enumerators is
super important—they need to be smart, honest, and patient. This method is great for
big surveys, like population censuses, but it can be pricey. Usually, it's used by
governments or large organizations.
DIFFERENCE BETWEEN QUESTIONNAIRES AND SCHEDULES

Questionnaires and schedules are both used to collect data in research, but there are
some key differences between them:

1. Method of Collection: Questionnaires are usually sent through mail and filled out by
respondents on their own, while schedules are filled out by research workers or
enumerators who can help interpret questions.

2. Cost: Questionnaires are cheaper because they only require preparation and mailing,
but schedules are more expensive as they involve hiring and training enumerators.

3. Response Rate: Questionnaires often have higher non-response rates, while

schedules, filled out by enumerators, usually have lower non-response rates.

4. Identity of Respondent: In questionnaires, the identity of the respondent may not be

clear, but in schedules, it is known.

5. Speed: Schedules are faster because enumerators collect data directly, while
questionnaires may take longer as respondents may delay returning them.

6. Personal Contact: Questionnaires lack personal contact as they're sent and returned
by post, whereas schedules involve direct personal contact with respondents.

7. Literacy Requirement: Questionnaires require literacy and cooperation, but

schedules can gather information from illiterate respondents.

8. Sample Distribution: Questionnaires allow for wider and more representative

sample distribution, while schedules may face challenges in reaching a wider area.

9. Accuracy of Information: Schedules generally yield more accurate information

because enumerators can clarify questions, reducing the risk of incomplete or wrong
responses.

10. Success Factors: Questionnaire success relies on the quality of the questionnaire
itself, while schedule success depends on the honesty and competence of enumerators.
11. Physical Appearance: Questionnaires need to be visually appealing to attract
respondent attention, while schedules filled by enumerators don't require this.

12. Observation Method: Schedules can be combined with observation method, but
this isn't possible with questionnaires.

SOME OTHER METHODS OF DATA COLLECTION

Let us consider some other methods of data collection, particularly used by big
business houses in modern times.

1. Warranty cards: Warranty cards are small cards that come with products like
appliances or electronics. They ask you questions about the product, like your name
and address, and they're usually inside the product's packaging. The dealer wants you
to fill out the card and mail it back to them.

2. Distributor or store audits: Distributor or store audits are checks done by

distributors or manufacturers using their salespeople. They visit retail stores regularly
to gather information. Instead of asking questions, they observe things like inventory
levels. These audits help estimate market size, share, and buying patterns. They're
usually done on a group of stores and focus on sales trends and promotions' impact on
sales.

3. Pantry audits: The pantry audit method estimates what people consume by looking
at what's in their pantry. Investigators check what items are there, how much, and their
prices. This helps understand what consumers buy and their preferences. Sometimes,
direct questions are asked about why certain products were bought. This method may
not need repeated visits. But, it may not always show preferences, especially if
promotions influence sales a lot.

4. Consumer panels: A consumer panel is like a group of consumers who agree to

track their consumption habits regularly and share this information with researchers
when needed. There are two main types: transitory and continuing. Transitory panels
measure the impact of specific events or changes, like before and after an advertising
campaign. Continuing panels collect data over a longer period, often on various
consumer behaviours. These panels are typically operated through mail surveys.
However, ensuring the panel represents the broader population and avoiding biases are
key challenges with this method.

5. Use of mechanical devices: Modern businesses often employ mechanical devices

like eye cameras, pupillometric cameras, psychogalvanometers, motion picture
cameras, and audiometers to gather information indirectly. These tools help in
understanding consumer behaviour and preferences. Eye cameras track where people
look on images or text, while pupillometric cameras measure pupil dilation to gauge
interest. Psychogalvanometers assess body excitement levels in response to visual
stimuli. Motion picture cameras record buyer movements in stores, revealing reactions
to packaging or labels. Audiometers track TV program preferences by fitting devices
to television sets. Additionally, psychologists use projective techniques to uncover
underlying motives or attitudes through indirect means like pictures or inkblots,
requiring specialized training to interpret responses accurately.

We may now briefly deal with the important projective techniques.

(i) Word association tests: In these tests, people are asked to say the first word
that comes to mind when they hear another word. For example, if they hear
"cold," they might say "hot." Researchers use lists of words to see which ones
people associate most strongly. This helps in marketing to understand what
qualities people link to different brands. It's a quick and easy way to gather
information, especially for advertising research, but it works best with widely
known words that have clear meanings.
(ii) Sentence completion tests: These tests build on word association techniques.
Instead of just saying the first word that comes to mind, people complete
sentences like "persons who wear Khadi are..." to show what they associate
with Khadi clothes. Analysing these responses helps understand people's
attitudes. It's a quick way to gather information and develop ideas for
questionnaires, but it can be tricky when responses have multiple meanings.
(iii) Story completion tests: In these tests, instead of completing sentences, people
finish stories created by the researcher. They're given part of a story and
asked to come up with an ending.
(iv) Verbal projection tests: These are the tests wherein the respondent is asked to
comment on or to explain what other people do. For example, why do people
smoke? Answers may reveal the respondent’s own motivations.
(v) Pictorial techniques: There are several pictorial techniques. The important
ones are as follows:
(a) Thematic apperception test (T.A.T.): The TAT shows people pictures,
some normal and some strange. People then talk about what they think is
happening in the pictures. The investigator uses these answers to understand
the person's personality and attitudes.
(b) Rosenzweig test: In this test, people are shown cartoons with speech
balloons. They are asked to fill in an empty balloon with their own words. By
looking at what people write, researchers can learn about their attitudes.
(c) Rorschach test: In this test, there are ten cards with inkblot prints. The
designs are symmetrical but don't have any specific meaning. People are
asked to describe what they see in these inkblots, and their responses are
interpreted based on a predetermined psychological framework. Although this
test is commonly used, there are concerns about its validity.
(d) Holtzman Inkblot Test (HIT): The Holtzman Inkblot Test (H.I.T.) is
similar to the Rorschach Test but uses 45 inkblot cards instead of 10. It
assesses perception based on colour, movement, shading, and other factors.
Each respondent provides one response per card, which is interpreted at three
levels: form appropriateness, shading and colour, and movement. The test
provides more information than the Rorschach Test because of the variety of
stimuli. However, its usefulness for personal selection and vocational
guidance is still uncertain. Some modifications involve presenting multiple-
choice options for responses to improve interpretation. Administering the test
 individually and conducting post-test inquiries are essential for understanding
respondents' perceptions and personalities accurately.
(e) Tomkins-Horn picture arrangement test: The Thematic Apperception Test
(TAT) involves showing respondents a series of pictures and asking them to
tell a story about each one. The pictures often depict ambiguous situations,
allowing the respondent to project their thoughts and feelings onto the
images. The stories they create provide insights into their personality,
attitudes, and emotions.

vi) Play techniques: Play techniques involve asking subjects to act out or improvise
situations where they take on different roles. This allows researchers to observe
traits like hostility, dominance, sympathy, or prejudice. For example, children
might be given dolls representing different racial groups to play with, and their
interactions with these dolls can reveal their attitudes towards those groups. This
method, known as the doll-play test, is commonly used in sociology studies to
understand deep-seated feelings and attitudes through play.

(vi) Quizzes, tests and examinations: This technique involves indirectly assessing
candidates' specific abilities. It uses a combination of long and short questions
to test their memorization and analytical skills.
(vii) Sociometry: Sociometry is a method to understand social relationships within
a group. It looks at who individuals prefer or dislike in different situations.
Through sociometry, researchers study the underlying motives of people.
Sociograms are used to visualize these social connections, showing leaders
and followers within the group. This technique has been used to study how
ideas spread among medical practitioners regarding drugs.

7. Depth interviews: Depth interviews are intensive explorations of people's inner

thoughts and motivations, commonly used in motivational research. They delve deep
into unconscious desires and feelings, needing skilled interviewers and time. These
interviews can be either projective or non-projective. Projective interviews use indirect
questions to uncover information, while non-projective ones aim to understand
attitudes and behaviours directly. Both types offer valuable insights into people's
psychosocial dynamics.

8. Content-analysis: Content analysis involves examining the content of written or

spoken materials like books, newspapers, or speeches. In the past, it was about
counting specific features, but now it focuses on understanding the overall message or
meaning. Bernard Berelson is associated with this approach, which is like comparing a
quick chat to a detailed interview. Content analysis helps understand the essence of
verbal materials. For instance, when reviewing research articles, it includes examining
their content. This can range from simple tasks like counting specific elements to more
complex analyses, such as studying the media's attitude towards education.