Basics of Research Methodology

Introduction:

    Research plays a vital role in human progress. It's like a tool we use to find out the truth about things. It's not just searching for answers, but a deep and focused search. Research helps us find solutions by using the scientific method, which is a key part of every field of study. It's how we investigate and learn new things.

    Ranganathan elaborates and elucidates the meaning of research as a “critical and exhaustive investigation to discover new facts, to interpret them in the light of the known ideas – laws and theories – to revise the current laws and theories in the light of the newly discovered facts and to apply the conclusions to some practical purpose”.

    "Research" is a purposeful intellectual activity aimed at exploring a phenomenon to expand existing knowledge. It involves systematically collecting, organizing, analyzing, and interpreting data to solve societal problems. This becomes scientific when specific methods are used to gather, record, measure, and analyze data in a research cycle – from problem selection to final report writing. In essence, it's a systematic thinking process and scientific method for studying problems and finding solutions. However, research design, application, and outcome may differ across subjects.

    Authors categorize research approaches into two main types: exploratory research and conclusive research. Another classification focuses on research based on approach, purpose, and nature: pure, fundamental, and theoretical research fall under one category, while applied research falls under another.

    Research can be approached in two main ways: exploratory and conclusive. It can also be classified based on its purpose and nature: pure (fundamental and theoretical) and applied research. The research process involves stages like defining the research, designing it, selecting data collection methods, and creating a report. Formulating clear hypotheses is a key step, serving as potential answers to research questions. The empirical part of research uses concrete evidence from collected data to draw conclusions about a specific population or observations. This unit provides a concise overview of research conduct.

STEPS IN THE RESEARCH PROCESS

The process of research invariably consists of three stages –

the primary stage, secondary stage and final stage.

Each of these stages consist of several steps and thus the any research process comprises the following steps:

i) Identifying and Formulating the Research Problem

ii) Defining the objectives

iii) Formulation of Hypotheses

iv) Literature Review

v) Research Design

vi) Sample Design

vii) Data Collection

viii) Analysis and Interpretation of Data

ix) Writing Research Report

x) Conclusions



i) Research Problem:

Selecting the right research problem is a significant step that contributes to solving half of the research challenge. Identifying a research problem is a crucial task that initiates the research process. It requires a good understanding of the subject area in which the study will take place. Defining the research problem is a process that involves a reasonable level of knowledge within the broader subject domain. This step is essential for effectively analysing and exploring the subject area. Additionally, this process includes:

Identification of the broad field or subject area of interest;

• Narrowing down the broad area into sub areas;

• Selecting a topic of interest as well as the currently most important topic;

• Raise research questions;

• Formulate objectives; and

• Assess the objectives.



ii) Formulating Objectives in Research:




The primary purpose of research is to evaluate the necessity for investigating a chosen subject or issue. Objectives represent the aims we intend to achieve through our study. Clarity and specificity in wording objectives are crucial. Objectives typically include main and sub-objectives. The main objective presents a comprehensive statement about the study's core focus. Sub-objectives are specific elements of the topic investigated within the study's primary framework.
iii) Identifying Variables and Formulating Hypotheses:

This step involves clearly identifying the independent and dependent variables used in the research. The hypotheses need to be specific enough to facilitate statistical testing. Formulating and verifying hypotheses are central to the research process. A hypothesis is a speculative idea about a phenomenon. Scientific research involves creating hypotheses and testing them using observed data. If the data contradicts the hypothesis, it is rejected; if there's alignment, it's accepted. For instance, consider the example: 'There is no connection between book circulation and in-library usage.'
iv) Literature Review in Research:

After choosing a research problem, the researcher examines available literature to understand how the problem has been addressed and to identify recent information. This involves consulting relevant sources, including the World Wide Web and the Internet. Literature review is crucial for clarifying and focusing the research problem, enhancing the research methodology, and identifying the exact problem. It entails searching for existing literature in the study area and reviewing the selected literature.
v) Research Design Preparation:

Creating a research design is a vital step in the research process. It's the overall strategy for conducting the research. Young defines research design as the organized and systematic planning of a research endeavor. It serves as the conceptual framework guiding the research. The research design outlines how data will be collected, analyzed, and measured. Essentially, a research design is a structured plan for guiding a research study. Its purpose is to efficiently gather pertinent information while minimizing the use of effort, time, and resources.

Research design is comparable to an architect's blueprint for a construction project. Just as an architect plans a blueprint before building, a researcher creates a plan for their study. This plan, often referred to as a Research Design or Research Strategy, outlines the structure and approach of the study. The design may be a specific presentation of the various steps in the research process. They are:

• The selection of research problem;

• The presentation of the problem;

• Literature Review;

• The formulation of hypotheses;

• Research Methodology;

• Data collection;

• Hypotheses Testing;

• Interpretation; and

• Report Writing.




vi) Types of Research Design

Research design varies based on the type of research. The research problem determines the type of research, which then guides the methods used. The hypothesis shapes the required data, leading to specific data collection and analysis techniques.

Boyd categorized research designs into Exploratory and Conclusive, each with subdivisions. Three main research designs are in use: Survey, Historical, and Experimental.

vii) Sampling: A Brief Overview

Sampling involves selecting a smaller group from a larger population for study. It's like taking a snapshot that accurately represents the whole population. This process uses a probability-based approach to ensure accurate reflections of population patterns and subgroups.


In technical terms, sampling is about choosing a subset of entities from a population, which could be individuals, books, trees, patients, etc., depending on the context. This chosen subset, known as a sample, should reflect the characteristics of the entire population. Think of it as a "Small Universe" – a microcosm that mirrors the bigger picture. The goal is to make estimates about population parameters like mean, standard deviation, and correlations.



Designing a sample involves three key factors:

1. Sample Units: What will be included in the sample? For instance, when studying book usage patterns, should we sample books themselves or transaction records?

2. Sample Size: How many units should be in the sample? Larger samples might seem more reliable, but they don't always have to be huge; they just need to accurately represent the population.

3. Sampling Method: How should the sample be chosen? Various techniques, such as random sampling or stratified sampling, help us make informed selections.




Ultimately, sampling enables us to make meaningful inferences about a population without having to examine the entire group. By understanding different sampling methods, we can choose the best approach for our specific research context.

VIII) Sampling Methods:

Sampling methods are divided into two categories:

a) Probability Sampling

b) Non-probability Sampling

Probability Sampling includes:

i) Simple Random Sample: Every population unit is known and has an equal chance of selection.

ii) Stratified Random Sample: Population is divided into distinct groups (e.g., age or location) and random samples are taken from each group.




iii) Systematic Random Sample: Sampling in a systematic way, often at regular intervals.

iv) Multi-stage Sampling: Sampling conducted in multiple stages.

v) Cluster Sample: The population is divided into groups, and a sample is drawn from these groups.




Non-probability Sampling includes:

i) Convenience Sample: Easiest-to-reach population units are selected.

ii) Purposive / Judgment Sample: Units are chosen based on researcher judgment for accurate data collection.

iii) Quota Sampling: Special form of stratified sampling where a predetermined number from each group is selected.



ix) Common Pitfalls of Sampling in Research:

Research often aims to create effective public policies and solutions through surveys, which are built upon a sample population. This involves designing questionnaires and formulating questions that align with the chosen sample. However, there are six significant pitfalls associated with sample-based data collection:




1. Inadequate Representation: The sample might not accurately represent the entire population, leading to skewed results.

2. Selection Bias: If the sample is not chosen randomly, it may introduce bias and affect the validity of the findings.

3. Small Sample Size: A small sample might not provide reliable insights and may not be statistically significant.

4. Non-Response Bias: When a significant portion of the selected sample does not respond, it can impact the accuracy of the results.

5. Self-Selection Bias: Participants who choose to be part of the sample might differ from those who don't, leading to distorted conclusions.

6. Misleading Questions: Poorly framed or confusing questions can result in inaccurate responses and misleading data.

These pitfalls highlight the importance of careful sampling techniques and question formulation to ensure the reliability and relevance of research outcomes.

DATA COLLECTION AND MEASUREMENT

i) Data Collection After the selection of a proper research problem, and design of research the final step is to make a framework on plan of action for the conduct of research and this involves the data collection. After all these steps, comes the stage involving the collection of the data; this data is required during the various phases of study. So now we will study about the details of the sources of data collection, importance of data collection and also about the various methods that can be used for data collection. The construction of a research instrument or tool for data collection is an important part of a research. The research findings are totally dependent on valid and reliable data, one has to be careful in selecting the tool for data collection. The sources of data can be divided into;

a) Documentary source and;

b) Field source;

but the more popular and accepted sources of data are classified as;

• Primary Source and

• Secondary Source

c) Primary data are information generated to meet the specific requirements of the investigation to be made.

A method refers to the way of gathering data and some of the methods are;

• Observation

• Questionnaire and Schedules

• Experimentation

• Simulation

• Interview

• Projective Technique

d) The method of collecting secondary data is briefly classified into two main factors:

• Internal – available within organisations and institutions

• External – consisting of personal sources and the public sources

ii) Data Measurement Levels:

In statistics, data is measured using four main levels: Nominal, Ordinal, Interval, and Ratio. These levels vary in how useful they are for research.


1. Nominal: This level has no meaningful ranking among values. It categorizes items or subjects based on qualitative classifications. Examples include gender, nationality, and language.

2. Ordinal: Data at this level can be ranked, but the differences between values are not consistent or meaningful. For instance, a survey rating of "good," "better," or "best."

3. Interval: Here, the differences between values are consistent, but there is no true zero point. Temperature measured in Celsius or Fahrenheit is an example.

4. Ratio: This level has consistent differences between values and a true zero point. Examples include height, weight, and income.


Nominal data, for example, distinguishes between items based on categories like gender or ethnicity.

Ordinal Data: Allows ranking data (1st, 2nd, 3rd, etc.), but doesn't indicate relative degree of difference. Examples: "sick" vs. "healthy," "guilty" vs. "innocent," "false" vs. "true," or ranked opinions like "completely agree" to "completely disagree."




Interval Data: Shows differences between items, but not their ratios. Can express ratios of differences (e.g., one difference is twice another). Also referred to as "scaled variables."




Ratio Data: Has meaningful zero point and defined measurement distances, allowing for flexible statistical analysis. Ratios are meaningful. Interval data lacks ratios (e.g., 20°C isn't "twice as hot" as 10°C).
Data Collection Methods and Tools/Instruments:

Once the research problem is defined and a study design is created, a research instrument is built, and a sample is chosen. Data collection involves methods like interviews, mail questionnaires, experiments, and observations. Inferences and conclusions are drawn from this data for the study.

Some of the methods are briefly described below:

Historical Method:

Historical Research systematically evaluates and combines evidence to establish facts and draw conclusions about past events. It involves interpreting source documents for authority, content, and meaning, as well as understanding records in relation to general laws, trends, and hypotheses. It's not just uncovering facts; interpretation is a vital aspect of legitimate historical research.

ii) Survey Method:

Surveys come in two types:

Structured
Unstructured.

In structured surveys, all respondents are asked the same formal questions. In unstructured surveys, we adapt the interview based on respondents' answers. Survey research can be direct or indirect. In the direct approach, we ask straightforward questions about behaviors and thoughts. For example, "Why don’t you buy Nokia cell phones?" In the indirect approach, we inquire about preferences, like, "What kind of cell phone do you like?" This can help reveal consumer preferences. Surveys can gather diverse information, but respondents might hesitate due to privacy concerns, lack of time, or inability to recall details.

Experimental Method:

The experimental method, also known as empirical research, relies on data. This type of research involves the researcher having control over the variables being studied and deliberately changing one of them to observe its impact. In this approach, the researcher directly obtains information from the source and takes specific actions to generate the needed data. Experiments can be replicated, and they tend to yield consistent results.

iv) Contact Method: Data collection through mail, telephone, personal interviews, etc. Collects large data at low cost per respondent, but has a low response rate and limited control over respondents.

v) Case Study Method: A qualitative research approach involving in-depth analysis of a specific social unit, often referred to as an insight-simulation study.

vi) Questionnaire Method: A tool for structured surveys/interviews, consisting of questions presented to respondents. Can be flexible, but requires careful development and testing before large-scale use.

vii) Observation Method: Human or mechanical observation of actual behavior, structured or unstructured. Accurate for certain data, but challenging for capturing feelings, beliefs, and attitudes.


DATA PRESENTATION

After the data has been collected, the next question is ‘What is to be done with the data that is collected’? The next step is the data has to be processed and presented for facilitating its analysis as per the plan of research process. The processing of data comprises, editing, coding, classification and then presentation of the data in Tabular, Graphical forms. It is desirable that appropriate tables and graphical methods are employed for the logical presentation of the data.

Broadly, the types of tables and graphical representation of data are listed below:

i) Tabulation of Data After the classification of data, they are presented in the Tabular form, to make the implicit qualities and quantities of data explicit and meaningful. The tabulation of data can be only means and useful for research analysis. There are different types of statistical tables. They are:

a) Simple and complex tables;

b) One way tables;

c) Two way tables;

d) Three way tables; and

e) Manifold Tables. Further the Tables may be

f) Frequency table or

g) Response table.




ii) Graphical Representation of Data Several types of graphs and/or charts are used to present the collected/ tabulated data. The common graphical methods used are; bar charts, two dimensional diagrams, pictographs, pie-charts and arithmetic or line graphs/charts. 

Four popular methods used are:

• Bar Diagram;

• Histogram;

• Frequency Polygon; and

• Ogive or Cumulative Frequency Curve.

DATA ANALYSIS AND INTERPRETATION: Analysis and interpretation of data are the most crucial aspects of research. Processing and analysing data involves a number of closely related operations; these are performed with the purpose of summarizing the collected data; this also involves organization of database in a manner that they answer the research objectives. It is through a systematic analysis that the underlying features of data are revealed and valid generalisations are arrived at.

Analysis of data is to be made with reference to the object of the study and its possible effect on scientific discovery, and with reference to the research problem at hand or hypothesis. Interpretation means drawing inferences from the collected facts after the analytical study. Interpretation has two major aspects, namely establishing continuity in research through linking the results of a given study with those of another and the establishment of some relationship with the collected data. In order to make the collected data to be more meaningful, after its tabulation and representation is to lay down the procedure for its analysis. 

To be effective the procedures employed are:

• An intensive review of the data, with reference to research objectives;

• Analysis with suitable techniques and results; and

• Selection of the results to study hypothesis and objectives of research.




The types of analysis of data include;

 Descriptive analysis – consisting of Bivariate, 

Sequential and Multivariate analysis and 

Casual analysis – consisting of Correlative analysis and Inferential analysis.

If you want to analyse data using computer, you should be familiar with the appropriate software. In this area, knowledge of computer and statistics play an important role. The most common software is SPSS for windows. However, data input can be a long and laborious process, and if data is entered incorrectly, it will influence the final results. In present day contest, the software EXCEL has many features for statistical computations.

The interpretation of e data is the final step, after the tabulation and or representation and analysis of data. The interpretation in simple terms is the analysis of the results which has to be done very carefully and logically taking into consideration the object of research. The interpretation should avoid wrong and erroneous expression that may lead to misinterpretation of the data.

REPORT WRITING AND CONCLUSIONS 

Report writing is the culmination of any research investigation, as the research worker is obliged to report his/her study on its completion, it is a social obligation.

The contents of research report can be targeted to various users. Target users for any research report may be among the,

a) Academic community

b) Sponsors of research or

c) Layman. T

he components and contents of report may include among others;

a) Materials and methods

b) Tabular and Pictorial Representations

c) Implications and suggestions for further research.

CONCLUSIONS In this Unit a brief and simple description of Research Methodology is presented as a prelude to the conduct of research. More descriptive account on each one of the topics presented in this Unit can be obtained by referring to any standard textbook on Research Methods, as broad and a macro structure of the topics to be studied is given here.

However, at the end of the research a reporting has to be done of the entire research work. In this process, the end result would to answer the following questions:

• Was our initial hypothesis correct?

• What if my findings are negative?

• What are the implications of the findings for the theory, for the background assumptions, or relevant literature?

• What recommendations can we make for public policies or programs in this area?

• What suggestions can be made for further research on this topic?

In the entire process of research, mostly several statistical methods are applied. This involves summarization and presentation of data in textual, tabular and graphical form, which are in short described in the preceding sections. Some of the common quantitative and statistical computational methods employed to aid interpretation of the data involve are enumerated here below. Some of the simple arithmetic procedures and few are complex statistical techniques. They are:

• Measurements of central tendency :mean, median, mode;

• Measures of dispersion: range, mean deviation, variance and standard deviation);

• Probabilities and probability distributions;

• Correlation coefficient, regression coefficient, estimates of dependent variables ; fitting various linear and nonlinear models;

• Analysis of variance; and

• Testing of Hypothesis: z-statistic, t-statistic, chi-square, F-statistic. In the Unit – 19 the Testing of Hypothesis is discussed.