Researchg methodlogy for the tech student
- 1. M.Sc. Data Science Research Process
- 2. Importance of Research Process • Efficient and well-planned activities always see success • Therefore, one need to efficiently plan a research activity, execute it meticulously, and publish it for outside world • Researchers should follow research process to get optimized research outcomes • The research process consists of a number of closely related activities • But such activities overlap continuously rather than following a strictly prescribed sequence
- 3. Importance of Research Process • At times, the first step determines the nature of the last step to be undertaken • If subsequent procedures have not been taken into account in the early stages, serious difficulties may arise which may even prevent the completion of the study • One should remember that the various steps involved in a research process are not mutually exclusive; nor they are separate and distinct • They do not necessarily follow each other in any specific order and the researcher has to be constantly anticipating at each step in the research process the requirements of the subsequent steps
- 5. Research Report/Thesis Writing/Publication/ Patent Phases in Research Selection of Domain Formulating a Research Problem (Tentative) & Identification of Key Words Literature Review Identification of Variables/Parameters and Research Design Preparing the Research Proposal Redefining Research Problem and Objectives (Final) and Hypotheses Data Collection and Representation Testing of Proposed Design on Collected Data/Hypothesis Testing Results and Analysis
- 7. Selection of Domain • The very first step in the process of research • Research domain can be selected from area of interest, identified gap in the literature and individual skill set • If the researcher is working in research organization or laboratory the researcher has to work on the domain of organization's interest • Every time, selection of brand new topic/domain for research is not required, in each field there are some topics that are old but still there is scope for research • Even though there are different parameters for selection of domain, one should go with an area of his/her interest and current and future market trends
- 8. Selection of Domain • Researcher’s interest is very important in basic research as outcome may not be immediately applied for product design or other social or commercial purposes • Identification of an area of research and defining research problem are most important steps in the journey of research • One needs to give sufficient time and thought for the problem formulation • Before deciding domain or selecting problem one need to understand his/her passion, strengths, weaknesses, likes and dislikes • Area of research should be of your interest and it should have potential applications after completing your research
- 9. Selection of Domain • One should avoid selection of overdone topics unless you have some unique aspect for it • Also, a topic should not be too general and old • One invention gives rise to more opportunities for research in a particular domain • First Aero plane designed by Wright brothers in 1903 was simple craft and if you look at today’s aero planes, they are totally different
- 10. Formulating a Research Problem (Tentative) and Identification of Key Words • For an in-depth literature survey researcher has to first define tentative research problem definition and identify the related keywords for literature search • Before starting with research; researcher should be equipped with all the tools and domain knowledge required for research • Research requires ability to go in depth of particular topic • Discussion of the state-of-the-art with colleague and domain expert is very important and helps a lot to finalize research problem • Attending workshops, conferences and or short term training programs is integral part of research work right from conception of problem to completion of work
- 11. Formulating a Research Problem (Tentative) and Identification of Key Words • Researcher should prepare the research tentative plan of research work • Time span and tasks may change with domain, expert level of researcher, purpose of research, and so on. • Research statement should be clearly defined and its objectives should be SMART (Specific, Measurable, Achievable, Realistic, and Time-bound) • Initially, one can start with tentative problem definition and review the state of the art literature related to topic
- 12. Literature Review • The literature survey is a comprehensive study of technical and authorized content related to research keywords • It is very important step in the initial phase of research, however, this step is revisited by researcher number of times during research journey. • Literature survey provide details of research progress of particular domain • It helps the researcher to understand the approaches, methodologies, algorithms, and datasets used by other scientists • Also, it is important to identify a research gap and contribution to an existing knowledgebase • It helps the researcher to understand the progress of domain and state of the art in the domain
- 13. Literature Review • Literature survey also helps to avoid duplication of work • Researchers should able to use previously done work as foundation and propose new idea above that may be supporting or contradicting to previous work • Good literature sources are periodicals, conference proceedings, research reports, standards, theses, dissertation reports, research articles and reports, white papers, patents, reviews, textbooks, handbooks, and encyclopedia • There are different strategies to search literature • One needs to first identify the keywords related to the problem definition and searching articles using keywords is the more precise way to search article
- 14. Literature Review • Backward and forward reference search is also popular and beneficial to understand the background of the research area • One can choose the latest article from the reputed journal written by good researcher and study all related references of that paper This is a backward reference • One can also find the important old article in the domain and find all the papers that referred the article. This is forward reference search • Researcher may add some more parameters to be studied about the research paper and can prepare a format for same to keep record of each paper • Summarization and representation of literature surveys is an important part in the theses, presentations and in journal writing
- 15. Literature Review • Literature survey can be represented in textual, tabular or graphical format using different parameters • Literature surveys can be done with respect to following points: • What • How (process schematic)/standard procedure (system block schematics) • Discussion on major steps involved • Design criteria and performance measures • Techniques currently in use • Comparative analysis (table or any suitable tool to discover and list pros and cons/ strength/weakness/future scope of existing techniques) • Scope for research/gap in research
- 16. Redefining Research Problem, Objectives (Final) and Outcomes/Formulating Hypothesis • Problem definitions should be unambiguous, clear statement that states the major objective of the research • There should be generally three to six sub-objectives defined for research work • There should be a clear indication of the research work which should not be the recurrence of the same research • There should also be an outcome which has be initialized while mentioning the research objective • Objective should be given pointwise (three to six points) • It is important to define outcomes of research work to be carried out, which should show an impact of the proposed research work
- 17. Redefining Research Problem, Objectives (Final) and Outcomes/Formulating Hypothesis • After domain selection and identifying the problem definition researchers should formulate a hypothesis • State-of-the-art literature review on related topic creates a sufficient background for formulating the hypothesis • Reporting problem with the tentative solution is a hypothesis • The hypothesis is a tentative solution based on insufficient evidence that can be true or false • “People who do not take balanced diet are obsessed than the people who take balanced diet”
- 18. Redefining Research Problem, Objectives (Final) and Outcomes/Formulating Hypothesis • In the end, researchers either approve or disapprove the null hypothesis • Generally it is easier to disapprove the null hypothesis as only one counter example is sufficient • If the outcome does not support the null hypothesis, we conclude with an alternate hypothesis • The null hypothesis is the precise statement about the parameters. Researchers either approve the hypothesis or disapprove the hypothesis • If researcher disproves the null hypothesis, all other possibilities are represented by alternative hypothesis • Null hypothesis does not provide a statistically significant relationship between variable, whereas, alternate hypothesis provides a statistically significant relationship between them
- 19. Research Proposal • Researchers should able to convince people for selected topic and objectives through the research proposal • It is very important document that is reviewed by different committees • In industry, proposal needs to be written for approval of top management, finance and marketing department • Research proposal is mandatory document to be submitted to the university or research organization during the registration for a PhD degree • It can also be written to avail the funding from different agencies • Every agency generally has its own research proposal format
- 20. Sections in Research Proposal • Research proposal formats generally include the following sections: • Introduction (Proposed Topic of Research/Rational and Significance of the Study) • Literature Survey (Background of the Proposed Research/Study of Research Work Done in the Area and Need for More Research) • Motivation • Research Statement • Objectives of the Research Proposed • Probable Methodologies/Techniques to be Used • Expected Outcome(s) (the kinds of conclusions expected and their possible value) • Plan of Research Work • References/Bibliography
- 21. Identifying Variable/Parameter and Research Design • The variable is basic quality or attribute that differs in value under different circumstances • The researcher should identify all related variables or parameters • Parameters can be identified during literature survey and it may vary depending on the proposed hypothesis • In hypothesis “Illness of children occurs mainly due to change in season,” parameters can be age, illness, duration of illness, type of illness, season during illness, weight, height, temperature, and so on • These are some of the variables considered for hypothesis
- 22. Identifying Variable/Parameter and Research Design • Variable can be either dependent or independent • Independent variables show changes in dependent variables possessing either continuous or discrete values • To prove or disprove the hypothesis one needs to decide on the design strategy of research • The research design is a systematic plan designed to obtain a solution to the research problem • Which data should be tested, which procedure/ algorithm/methodology will be applied is decided in the design phase
- 23. Identifying Variable/Parameter and Research Design • Research design is blue print of research and should be carefully planned and checked for feasibility • Good research design provides maximum outcome with minimum efforts • The researcher should think about size and type of samples and dataset to be considered, parameters to be considered • Researcher can revisit the literature review step again to check with state of the art design methodologies used by other researchers and can come up with methodology suitable for his/her own research hypothesis
- 24. Data Collection and Representation • Data can be either directly collected afresh known as primary data or already collected and used data is known as secondary data • Secondary data is already considered by researchers to test the hypothesis • Most of the times in experimental research data is collected afresh to test the methodologies, hypothesis or algorithms • If researcher decides to use afresh data, then he has to decide carefully on method of data collection • Depending on problem definition and research objectives some parameters needs to be controlled during the data collection
- 25. Data Collection and Representation • Consider the research topic “Study of effect of exercise on thyroid patient.” • Persons having hypothyroid or hyperthyroid are of interest and normal individuals are not • Thyroid level, duration of exercise, type of exercise, age, eating habits, and so on are different parameters that need to be taken into consideration during research • Data can be represented in simple text, tables, graphs, audio, video, or images • Researchers should consider sufficient amount of data with all parameters • An incomplete dataset may not give a conclusive result • Collected raw data should be first examined for errors, this is called editing of data • It should be coded and represented in user friendly format
- 26. Selection of Appropriate Method for Data Collection • Nature, Scope and objective of enquiry constitute the most important factors affecting the choice of a particular method • Funding availability decides the method for collection of data and the disposal of the research in a limited version • So selecting a cheaper method may not be efficient and effective but as to act within its limitation • Time availability is also needed to be taken into account for deciding the method of collecting the data • Another important thing that should be considered is the time precision for collection of the data
- 27. Reliability of Data Collection • The reliability can be tested by finding out the following: • Who collected the data? • What were the sources of the data? • Were they collected by using proper methods? • At what time were they selected? • What level of accuracy was desired? • Was accuracy achieved
- 28. Reliability of Data Collection • The data that are suitable for one enquiry may not necessarily be found suitable in another enquiry • In this context, the researchers must carefully scrutinize the definition of various terms and units of collection used at the time of collecting the data from the primary source originally • If the researchers find the difference in these, the data will remain unsuitable for the present enquiry and should not be used • If the level of accuracy achieved in data is found inadequate for the purpose of the present enquiry, they will be considered as inadequate and should not be used by the researcher
- 29. Selection of Appropriate Methods for Data Representation • Tabulation is a process that conserves space and reduces explanatory and descriptive statement and also provides the comparison from one state to another • Tabulation also provides various statistical computational information, which facilitates the summation of items and detection of errors and omission. • Importance of Tabulation • Each table will give a clear title that does not require an explanation • Each table will be provided with a separate number that will be easy for referring • Tables can be categorized in a chronological, geographical, and alphabetical order • Table should be made as logical, clear, accurate, and simple as possible
- 30. Selection of Appropriate Methods for Data Representation • Most of the research outcomes consist of data that are of same kind; these types of data are needed to be reduced for the better understanding of the result • Classification of the data depends on the nature of a fact that is involved • Classification can be categorized under two sections such as numerical and descriptive • Descriptive refers to qualitative phenomenon which cannot be measured quantitatively and depends on the specified attributes • Whenever the data is classified according to the attributes, researchers should give a clear picture related to the attributes
- 31. Selection of Appropriate Methods for Data Representation • The numerical classification refers to the phenomenon which will measure the data quantitatively using statistics • Data related to the salary, classes of age, and so on, comes under these classes • Such data are known as statistics of variables and are classified on the basis of class interval. • Graphical representation helps to understand the data easily • All statistical package, offers a wide range of graphs • In case of qualitative data, the most common graph is bar chart and pie chart
- 32. Testing of Proposed Design on Collected Data/Hypothesis Testing • Hypothesis testing is expressed as either a null hypothesis or alternative hypothesis • Validity of the hypothesis should be tested on the basis of the data collected by the researcher • The procedure for hypothesis testing refers to all of those steps that we undertake for making a choice between the rejection and acceptance of a null hypothesis
- 33. Limitations in Hypothesis Testing • Results cannot be expressed with full certainty; they are probabilistic • Testing is not a decision-making activity in itself, the researcher should not use it in a mechanical way • Tests don’t explain the reason why the dissimilar result has been obtained due to fluctuation • Significance of the results is been validated on the basis of the probabilistic conditions which cannot be explained fully • Inference the statistical data cannot provide the evidence for the truth of the hypothesis
- 34. Results and Analysis • The result is an important section of the research • If the topic is totally new then one cannot compare his/her results with existing • Here, the hypothesis should be tested with multiple approaches and that results can be compared • If already some scientists worked on methodologies or the topic it is important to compare your results with the existing state of the art results from the literature • While comparing results one should look at the considered parameters, dataset, and so on
- 35. Results and Analysis • Results should be represented in the visual format using tables, figures and or graphs • It should be properly labelled, clear and easy to understand • Carefully choose the type of graph taken for result representation • The relationship between the data in tables/graphs should be explained along with observations • Result section may include the problems faced during collection of data and complete analysis of results
- 36. Research Report Writing • There are different purposes for writing the research report • The research report is a medium to convey research outcomes, contributions, findings and results to the outside world • It decides the quality of your research work. Without approval of experts in domain, the research is incomplete • Publication of research in open, referred international journal is very important aspect of the research • Research report can be in the form of the theses for academic research and it is compulsory and important part of the completion of the degree • Patents, copyrights, and white papers are also possible outcomes of the research
- 37. Research Report Writing • Communicating ideas is important aspect of research • People should agree that your work is creative and novel • For good writing, lot of good reading is recommended to understand how people communicate their ideas • Different agencies provide their own formats for writing the report • Format and length of the report depend on the type agency • The researcher should first start with an introduction and can continue with literature survey, methodology, datasets, and results • Then conclusion section can be added and at last abstract of the paper, report or theses can be written • These are general guidelines suggested for novice researchers and not hard and fast rule
- 38. Sections in Research Report • Abstract • Introduction • Review of Literature • Problem Definition and Objectives • Methodology • Observations and Results • Discussion • Summary • Conclusion • Publications • Bibliography/References • Appendixes
- 39. Issues in Research Report • Researchers should be aware of plagiarism issues, copyright issues, procedures, and penalties • Plagiarism is to literally steal the work of another person, presenting it as original research without proper citation • There are different agencies including Turnitin, Unkund that verify the copied content in written document with its source • After completion of writing, a plagiarism check of the written document by an authorized agency is recommended • While reporting own work, if the researcher is referring to work by other scientists or researchers, all of this should be acknowledged and added to the references
- 40. Research Report Writing • References can be organized in alphabetical or chronological order • Acknowledgments to people, funding agencies, or other organizations that helped directly or indirectly in the completion of the research work should be added. • The writer should have good command over the language especially the grammar of the language in which the work is reported • In science and technology domain research work is generally reported in the English language
- 41. Criteria for Good Research • The purpose of the research should be clearly defined and common concepts be used • The research procedure used should be described in sufficient detail to permit another researcher to repeat the research for further advancement, keeping the continuity of what has already been attained • The procedural design of the research should be carefully planned to yield results that are as objective as possible • The researcher should report with complete frankness, flaws in procedural design and estimate their effects upon the findings
- 42. Criteria for Good Research • The analysis of data should be sufficiently adequate to reveal its significance and the methods of analysis used should be appropriate • The validity and reliability of the data should be checked carefully • Conclusions should be confined to those justified by the data of the research and limited to those for which the data provide an adequate basis. • Greater confidence in research is warranted if the researcher is experienced, has a good reputation in research and is a person of integrity
- 43. Qualities of Good Research • Good research is systematic: It means that research is structured with specified steps to be taken in a specified sequence in accordance with the well defined set of rules • Good research is logical: This implies that research is guided by the rules of logical reasoning and the logical process of induction and deduction are of great value in carrying out research • Good research is empirical: It implies that research is related basically to one or more aspects of a real situation and deals with concrete data that provides a basis for external validity to research results • Good research is replicable: This characteristic allows research results to be verified by replicating the study and thereby building a sound basis for decisions
- 44. Problems Encountered by Researchers in India • The lack of a scientific training in the methodology of research is a great impediment for researchers in our country • There is insufficient interaction between the university research departments on one side and business establishments, government departments and research institutions on the other side • Most of the business units in our country do not have the confidence that the material supplied by them to researchers will not be misused and as such they are often reluctant in supplying the needed information to researchers • Research studies overlapping one another are undertaken quite often for want of adequate information • There does not exist a code of conduct for researchers and inter- university and interdepartmental rivalries are also quite common
- 45. Problems Encountered by Researchers in India • Many researchers in our country also face the difficulty of adequate and timely secretarial assistance, including computerial assistance • Library management and functioning is not satisfactory at many places and much of the time and energy of researchers are spent in tracing out the books, journals, reports, etc., rather than in tracing out relevant material from them • There is also the problem that many of our libraries are not able to get copies of old and new Acts/Rules, reports and other government publications in time • There is also the difficulty of timely availability of published data from various government and other agencies doing this job in our country • There may, at times, take place the problem of conceptualization and also problems relating to the process of data collection and related things