Lecture 01: Research Methods PDF

Lecture 01: [Music] foreign [Music] I would like to welcome you to an eight lecture series on research methods from Allama Iqbal open University in this course we will see what is the research process how to critique research and in this critiquing part criticism is not the only thing...

Lecture 01: [Music] foreign [Music] I would like to welcome you to an eight lecture series on research methods from Allama Iqbal open University in this course we will see what is the research process how to critique research and in this critiquing part criticism is not the only thing that we focus on we also focus on other things that we will be discussing in this course we will see how to read search through and summarize others research and how to conduct your own research we will discover the difference between natural and social science and their research methods and we will also find out that computer science and information technology both have component that belongs to natural science and social science so when we are applying research methods on computer science we have to take care of of both aspects of Sciences we will reveal mysteries of basic statistics and demonstrate their importance as an important researchers tool we will discover how to measure behaviors traits and attributes that interest you as a researcher we will see how to develop systems and collect the kind of data that allows you to become a publishable researcher in today's lecture we will see what research is and what it isn't we'll also look at attributes of high quality research we'll briefly look at the history of research and model of scientific inquiry here I have tried to Define research research is studious inquiry or examination investigation aimed at the discovery and interpretation of facts revision of accepted theories or laws in the light of new facts or practical application of such new or revised theories or laws as you can see this definition is coming from Webster's new College year dictionary and um it sounds to me a pretty good run on sentence so I have given brief version of my own definition of research and I say that research is a process through which new knowledge is discovered we apply research methods in a situation where we want to discover something new that did not exist before here's a an important question why do we conduct research the first answer is that we want to increase the reliability of our reported findings or we want to test a theory or we want to determine relationship between two or more variables or we want to extend the range of study by including more variables and one of the most important things why we conduct research because we want to publish it and remember if you don't publish no one will recognize you as a good researcher and probably this is a good place to to explain the difference between a conference paper and a journal paper there is a concept of peer revision peer revision is that once I conduct some research someone else in the field is probably more capable of making a judgment whether my research method whether my research findings whether my research project as a whole we will try to answer this question that why should we conduct research we should conduct research because it increases reliability of the reported findings or because we want to test the theory with an experimental design or otherwise or we want to determine relationship between two or more variables or we might still be interested in conducting research because we want to extend the range of the study by including more variables and an important thing why do we conduct research because we want to publish it if we do not publish no one will recognize you as a good researcher so what's the difference between conference paper in a journal paper and what is the concept of peer revision we do peer revision because once I conduct some research probably someone else in the field is more capable of making a judgment whether my research is significant or not they can guide me whether my research method was according to the research objectives that I had for that research project so what is the concept of peer revision in peer revision once I do some research I would like to hand it over to someone else in the field who can probably make a better judgment whether my research is significant or not conference papers well it depends the type of conference and the level of conference but usually it is easy to get published and get accepted for research presentation in a conference then normally a journal paper one of my teachers told me that one Journal paper is equivalent to three conference papers while there is no hard and fast rule that how many conference papers are equal to how many Journal papers but you should remember that usually Journal papers require more input more hard work and contribution to Theory so if someone asks us a question why study research method in one sentence I can probably answer it as follows because research methods develops understanding of common research methods promotes transfer of knowledge reliability and replicability of research results it also allows for determining the extent of research significance I would like to take a moment to explain the concept of research significance research is we are developing algorithms that enable us to communicate internet in low bandwidth situations so there is a difference between research significance so research method study of research method allows us to find out whether the research is significant or less significant another important thing to remember that application of research methods vary slightly for social science research and natural science research social science courses include sociology political science literature and natural science Fields include such as biology physics chemistry mathematics now computer science has a component that comes from social science and that comes from Natural Science for example when we are developing new algorithms we are dealing with the number crunching we are probably somehow dealing with some aspect of Natural Science but when we are trying to study people's behavior when they adopt technology when they do not use technology or use technology that type of research comes from social science so what are the attributes of high quality research first of all it is based on the work of others Newton once said after he was recognized for his laws of motion and he was regarded as a big researcher and big scientist he said I am standing on the shoulders of giants before me by saying that he meant that whatever he did not come only from him his work was based upon the work of Copernicus and even before that Muslim scientists and it goes even before that even before Muslim scientists ionian era researchers Greek researchers so the second important attribute it is replicable if someone else wants to do this research they can replicate the research study another important attribute it is generalizable to other settings we will see this concept of generalizability uh in a moment but this is an important attribute of high quality research it is based on some logical rationale and tied to a theory so in this course actually we will be looking at some of the theories that come from computer science background and we will see how do we tie those theories to our research method or to our research objectives that have some logical rational another important thing it is accomplishable it is a doable you take on a research study that you cannot do or you cannot accomplish or you cannot complete that type of research is not considered an important one because you did not complete it okay now there could be an argument that since research is a continuous process a cyclical process you cannot complete research all by itself so once you define objectives of your research you try to make sure that you achieve those objectives and if you could not achieve those objectives you have to document those uh hurdles those difficulties that you faced in the process of completing that research another attribute of high quality research is that it generates new questions and that research is cyclical in nature you answer some question some new questions are raised so you try to answer those questions that leads to further question and this process never ends research is incremental research is a political which means that this is an activity that should be undertaken for the good of society not for the good of politics and the most important of all it is publishable if you do not publish your research no one will know what you have done so you share your research by publishing it and once you publish your research others will know what you have accomplished and they wouldn't have to spend time in redoing the same thing that you have done they will build upon your research as we saw that was an attribute of high quality research it is incremental brief history of research let's review history of research briefly irrigation civilization about 5 000 BC at that time Some Humans they started thinking about nature and meaning of life ionian philosophers around 600 to 200 BC includes names such as Plato's Socrates Aristotle whose writings are still considered inspiration for a lot of researchers even though most of their research most of their philosophies have been proven wrong but it still gives us a point to start for example Aristotle considered founder of deductive reasoning deductive reasoning is drawing conclusions from General things to specific things for example in his book he gave an example all men are mortal Socrates is a man therefore conclusion is that Socrates is Mortal in the field of logic all all these sentences have their own value they are considered very important they're called major premise minor premise and then conclusion so these researchers these scientists deserve this credit next era is Lam first contribution even even before going into the details of what Muslim scientists did the most important contribution was that today we know about Plato we'd know about Socrates we know about Aristotle because their books were translated in Arabic then Muslim scientists they had contribution in mathematics medicine social science and this was probably the first time when experimental research method was developed and used new experimental equipment and utensils and methodological details were developed surgical equipments and instruments were developed another important thing a circumference of Earth different types of chemicals without going into details that why Muslims were so much involved into developing these chemicals we do understand that Muslims had a lot of contribution to science and research in this era after that Francis Bacon is usually credited with this concept even though there are controversies that surround this definition of Francis Bacon that the this definition is coming from text earlier than Francis Bacon but without going into the details Francis Bacon said that according reasoning means that methodological experimentation and careful observation would lead to new scientific knowledge and valuable discoveries this means specific to General in deductive reasoning we were drawing conclusions from General observations to specific things inductive reasoning is the opposite we draw conclusion from specific things to General things in 1662 Royal Society of London was founded and perhaps it was the first modern Research Institute in Europe Franklin Institute 1824 perhaps the first U.S Research Institute and it was the first Institute that received government grant Charles Darwin 1859 led to Modern experimentational design even though his theories and his findings were based upon observation and that led to the development of his theory but his contribution is that his theories led to a lot of experimentation Frederick trailer he's considered father of Scientific Management when he did his time and motion studies to coordinate employees and their work National Science Foundation in U.S 1949 in Henry Truman's era and there were a lot of Behavioral Science movements in research for example antonym Hawthorne studies that state that when employees are being monitored their productivity differs similarly herzberg's industrial hygiene Theory Maslow's hierarchy of needs theory X Theory Y all these Behavioral Science movements belong to this era so what is scientific method how do we conduct research according to general definition scientific method is a shared approach to understanding the world it is a standard sequence of steps in formulating and answering questions so scientific method is basically sequence of steps that we take whenever we conduct any research study scientific method includes the following steps identification and definition of a problem this identification could be based upon researchers observation this leads to formulation of hypothesis or formal research questions what is a hypothesis hypothesis is basically a tentative answer to the research questions but how do we find out whether that answer was scientifically true or not for that we need to collect data we need to organize it and we need to analyze it based upon that analysis we formulate the conclusion based upon that conclusion we either verify the hypothesis or reject them or we might simply modify those hypothesis so in today's lecture we covered what is research and what it isn't we saw the definition then we discussed attributes of high quality research we also briefly reviewed history of research and at the end of the lecture we saw a model of scientific inquiry the steps that we take to conduct a research study lecture um [Music] thank you Lecture 02: [Music] [Applause] thank you Rahim I would like to welcome you to the second lecture of an eight lecture series on research methods from Allama Iqbal open University in today's lecture we will be looking at applied versus basic research we will also look at types of research variety of research methods and a criteria to see which methodology to choose from applied versus basic research applied research has no immediate application for example theoretical development of algorithms mathematical formulas comes under basic research applied research has immediate applications for example development of better application communication systems new technologies development for commercial purposes so these are two important types of Applied research and basic research there are two major types of research methods based upon the level of control non-experimental research and experimental research non-experimental research further is characterized in these three categories descriptive historical and correlational experimental research is characterized in two categories through experimental and quasi-experimental non-experimental study describes the relationships between variables but this type of research does not test cause and effect relationship let's look at the details of experimental research descriptive research describes characteristics of existing phenomena it provides broad picture of a situation and usually it serves as a basis for other types of research an example could be a survey of software programmers to determine computer aided software engineering tools so this type of research is called descriptive research in which you do not conduct any experiment and this type of research basically describes characteristics of an existing phenomena and the variables that play an important role in that phenomena the next type of research is called historical research historical research describes past events in the context of other past or current events it also relates to the concept of primary and second resources of data an example would be historical influence of Pakistan software export board on software export so this type of research is called historical research next type of research is called correlational research usually the questions in this type of research aim at finding out what several events have in common asks whether knowing one event can allow prediction of another event if we knew one of the things we should be able to predict the other one but this type of research does not imply causation that one event causes the second thing it does not focus on that type of research an example would be relationship between years of software development experience and programmers productivity so this type of research aims at correlational studies between one two or more variables next type of research is called experimental research and again we said that there are two types of experimental studies true experimental research and quasi-experimental research and the basic purpose behind experimental research is to conduct and discover causal relationship one thing causing another thing true experimental research researcher assigns participants to groups treatment variable is controlled by researcher and control of potential causes of behavior is done so we need two groups in this type of research one group that goes through that treatment that research is aiming to conduct in experimental research for example we have two groups of patients or in Dono patients foreign but one of the medicine is the actual medicine one of the group is getting the actual medicine and the other group is not getting the actual medicine now we have to look for changes in the level of their treatment so this type of research is called true experimental research in computer science perspective for example we want to find out technology adoption we are developing a new software for which we want to find out how people behave when they use it in that situation we will develop two groups one group we will give them the new software the other group we will probably not give them the new software they'll do the way they have always done their work and then we compare the productivity if productivity was to be measured so in this type of research researcher has complete control of the environment and control of potential causes of behavior is very important second experimental type research is called quasi experimental research in this type of research participants are pre-assigned into groups if we do not have any control over assigning people to one group or the other this type of research method is very useful when researcher cannot manipulate all the variables for example we want to conduct research on employees productivity in the workplace by using software agents since our objective is to find out productivity in the work environment we do not have control on assigning people to different roles they are already assigned we cannot make up one person a CEO and the other person act as an accountant or the other person as software developer since we are doing it in a real-time environment those people are are pre-assigned to those groups so we do not have complete control over the environment that's why this type of research is called quasi experimental research based upon data collection method there are two types of research the first one is called quantitative research the other one is called qualitative research quantitative research involves large number of data crunching usually we need a sticks to do this type of research for example we conducted a survey of programmers on some issue and we got results from let's say thousands of programmers or hundreds of programmers we need power of statistics to analyze that data similarly we are developing a new algorithm to increase a particular software's efficiency so we conduct different types of tests and we get a number of readings so results from a network latency efficiency test for a particular software let's say it could be agents in a grid environment so that type of research is called quantitative research the other type of research the qualitative research method examines behavior in natural social cultural and political context usually results in non-quantitative data for example an in-depth interview with a CEO of a software house to find out management issues so in this situation we might only focus on probably limited number of software houses so that we can have an in-depth analysis so this type of research method or data collection method is called qualitative research this diagram shows information systems research framework and in this situation you see there is an environment is research factor and knowledge base but here relevance and rigor of research are too important aspects in any research method for example the research method that the research study that you are conducting is it relevant to computer science or is it relevant to the situation where you actually wanted to conduct that study and then the second thing is rigor how much rigor you have put into that research foreign now you can do a lot of hard work you can spend hours and hours in collecting data but it's possible that the data that you're collecting or the analysis that you are doing is irrelevant it does not have relevance to your field or your research study so we have to balance between relevance and rigor we have to make sure that both aspects of research are there there are variety of research methods to choose from for example there could be research studies that are speculative in nature and they are more like commentary type research one of the examples is Nicholas cars 2003 article that was titled I.T does not matter and in that article he made a claim that I.T is an infrastructural technology just like Railway system or electrical production system and since and and he compared it that since those two types of infrastructural Technologies failed at the end Investments did not pay off as much as people claimed information technology is the same it provides a mechanism for people to conduct business so people should not invest heavily in information technology and usually this this type of research leads to huge controversies and debate for example one of the persons who responded to This research said that yes everyone now has access to same type of systems same type of computers same type of network so one should not have superiority on the other one or in business terms is called competitiveness but the person who responded to this type of research he said that if this was the case then let's take an example we have three comparators doing the same type of business and we give them free money 100 million dollars are you really sure that all three competitors are going to invest in the same way and are going to get the same type of results in business terms is called return on investment the answer is no if the answer is no does it mean that money does not matter or take another example you give 100 best graduates to accompany to two competitors are both those companies going to use their talent in the same way and are going to get the same return on investment the answer is most probably no does that mean Talent does not matter so therefore comparison between I.T and Railway line or electrical generation system is not fair so I gave you this example to understand that speculative or commentary type research leads to sometimes leads to a lot of controversy the second type of research is called Frameworks or conceptual models in this type of research we try to develop Frameworks and conceptual models for something Library research we usually go and study a lot of research papers and try to do some type of literature analysis and this type of literature analysis focuses on what others have done so far and it also analyzes that what future research problems might be there another type of research is called case study research and in case study we usually focus on one or few important cases we could do a survey and usually survey is done with multiple people or organizations we could have a field study or a field experiment field study or field experiment are done in the real life environment for example you want to increase productivity of software programmers you need to conduct an experiment in the field in the real office environment but how do you do that before doing an experiment you need to study that environment that's where field study comes in that leads to field experiment or there could be another research method called laboratory experiment in laboratory experiment usually true experimental research type things are considered again in that type of research we will have two group of people and two group of people will be treated differently so that we can have difference in the outcome if there is a difference in the outcome of Group 1 and group two we say that the treatment was effective or treatment caused that change if there was no difference between the two groups at the end of the experiment we say that the treatment did not had any effect another type of research is called design science research in design science we develop new artifacts of a software system for example we might write a set of code we might have executables we might have different types of interfaces so development of actual software systems falls under design science and when is it called research when you prove scientifically that your new artifact new software system is better than others or at least is different from others we have just described qualitative type research we can also have research on secondary data someone else has already collected a lot of data and we just need to analyze it and that analysis could be done in a scientific manner for example census data is collected from all over the country and we need someone to actually do that secondary analysis analysis research study on that secondary data another type of research method is called content analysis example [Music] for example what do we do in that we have the original content and we have different people trying to interpret it so we are doing an analysis what does a certain verse mean so it's not only confined to religious Affairs it could be done in other environments for example from computer science context you might try to analyze people's communication on email for example how do people communicate when they use email and how do people communicate when they use telephone or when they use video conferencing so that's where content analysis type research method is used we can also have combination of research methods more than one research methods for example we do qualitative study to find out what type of data do we need to collect or for example to conduct a survey so we are using qualitative research to develop a survey instrument or we can use any combination of these things for example Fields field study and field experiment or it could be more than two research methods as well this will be called triangulation if you receive the same type of results by using different types of research methods it adds power to your research and Analysis so once we have so many research methods which method to choose for your particular research study here is the answer that it depends which method to use in which type of environment it depends on research question the research model degree of maturity of Teresa's topic if you are pursuing a research topic that has not been explored yet no one knows about it you need to probably conduct a qualitative type research so that you can do a in an in-depth analysis of that field however if the research field has been mature and a lot of researchers have already focused on that you probably need to confirm the theories based upon previous research so in that situation you probably need to collect a lot of data and probably you'll choose quantitative type research similarly it also depends upon external and internal validity it also depends on prior research by others and two last important things investigators own interest and strengths for example if a researcher is really good statistician choosing qualitative type research method might not make good sense on the other hand if a researcher is strong in writing in observing and writing probably qualitative type research would be a better option but there are different criteria for each of the research method and the last one publication Outlet and the audience where are you going to publish that research so different publication Outlets have different types of criterias and you need to keep those things in mind before going into submitting your research we understand that without publication research is not considered significant so in today's lecture we focused on applied and basic research we discussed different types of research and you also looked at a variety of research methods we also discussed how to choose an appropriate research methodology for your own research collections [Music] thank you [Music] Lecture 03: [Music] thank you [Applause] thank you smell like Raman Ibrahim I would like to welcome you to the third lecture on a series on research methods today we will be talking about an extremely important aspect of research methods which is how to select an outlet for your research publication we have talked about this concept in the first lecture that it is extremely important to publish your research if you do not publish your research no one will know what have you done so far and others will not know where to begin their research so today's lecture is titled where to publish your research and in this lecture we will be talking about different publishing Outlets we'll be talking about rankings journals books book chapters and conferences major Outlets include books book chapters journals and conferences before we choose an outlet for publication of the research material we have to think about this criteria this criteria includes ranking of the journal or the conference different organizational preferences future requirements personal preferences field specific requirements and impact factor of the research publication Outlet I will start with journals journals are considered foremost research publication outlets and one of the major reason is their extensive peer review process we should remember that this peer review process does not happen overnight it takes a lot of time in the peer review process and even when the paper has been accepted you might not be able to get your paper published in the next issue so you might need to wait so therefore journals usually have a lead time of several months and several years in addition to that journals have very low acceptance rate and acceptance rate means for every 100 articles submitted they might be accepting maybe 10 20 or or maybe even three or four articles so acceptance rate is also important before you decide on sending your publication into a journal the next point is timeliness of research publication in journals that timeliness is not assured and one of the major reasons is that peer review that extensive peer review process we should remember that whenever we select a journal for our research publication it might take a lot of time before we can actually get it published another important concept is ranking of the journals all journals are not considered equal some journals are considered more prestigious than others there are different mechanisms for ranking a journal and I have given detail of one of those methods for example this inter College year Studies Institute isi performs ranking of different journals in different types of fields and in that ranking they have their own criteria that criteria includes a lot of factors about a journal some of those factors are citation count citation count means number of times articles from that publication have been cited this current year the next criteria is article count total number of Articles published this year and this is needed because some journals are published quarterly some some journals are published annually and some journals are published fortnightly probably every two weeks so a journal being published every two weeks will have a lot more articles and there is a lot more chance that the Articles being published in those in that type of Journal will be cited more so if you need to compensate for the number of publications of a journal in a year relative to number of citations from that journal similarly impact factor impact factor is called frequency of an article's cited life cited half-life age of majority of article references how long an article has been referenced and citation means when someone else conducts research and if that person uses or builds upon your research that person cites that person gives reference to your research so how many times an article is being cited is called citation so cited Half-Life means age of majority of Articles being referenced some articles are referenced over and over again maybe even 10 years after they have been published they still have a lot of value in them they'll keep publishing they'll keep citing on the other hand some some articles may not be uh cited after a year of their publication so Half-Life will vary similarly last important factor that I have included here is immediac index which means how quickly an article is referenced so for example if a journal has issued its new issue last month and another Journal is accepting articles this month and a researcher is using an article being published last month in submission this month it's immediac Factor will be higher so these are a few different criterias on which journals are ranked so whenever you decide on sending your publication to a journal you should consider whether your publication needs to go in a top tier Journal or a second tier journal or a third year Journal here are some examples of these rankings this is sorted by isi impact factor for example in these journals the top Journal becomes ACM Computer Service ACM computer survey has been cited 17 731 times this year but but its impact factor is 7.4 and usually Half-Life cited half-life of the Articles being published in this journal is 7.3 years similarly bioinformatics Mis quarterly Mis quarterly has been cited 2395 times this year but its impact factor is 4.978 but the cited Half-Life is greater than 10 years an article being published in this journal gets cited more than 10 years after its publication so this type of Journal will be considered an extremely important journal and this constitute top tier journals let me give you another example IEEE pattern analysis this has been cited 13 053 times but its impact factor is 3.8 even though it's being cited more its impact factor is still low because of the other factors Maybe This Journal has more frequency of publication compared to Mis quarterly or ACM Computer Service similarly but this is not the only ranking mechanism there are other mechanisms as well and depending upon the field these rankings will vary for example this list was for computer science closely related field as information systems or mis in Mis journals are ranked differently for example in this ranking Mi is quarterly becomes first and if you look at this this is one mechanism of ranking a journal this is the second one third one so here are a lot of ranking mechanisms and overall in this ranking and average has been taken so you can see that according to this mechanism Mis quarterly was second in the list but according to all these type of mechanisms it was one so on average Mis quarterly becomes the top tier Journal in Mis we move on to the next publishing Outlets books you can publish your research in the form of a book and there are a lot of benefits attached to it first of all publishing research in books has different meaning for different organizations some organizations value it higher than others so for example if you are working in one University or one research organization they might consider publishing a book a very prestigious thing compared to another organization or another University where publishing a book is not a big deal so depending upon the organization where you work or where you will work you have to decide whether to publish a book or to send your publication in a journal look at this one usually it takes a lot of time to get published and therefore knowledge published in books is relatively old similarly timeliness of research publication is not assured an important aspect peer review process is not mandatory you might get your book peer-reviewed but publishing a book does not require peer review process and we know that one of the important aspects of research is peer review so this is one of the reasons that books are not usually considered pardoned or a preferred way of publishing your research nonetheless books are important another mechanism would be book chapters usually book editors solicit researchers for book chapters and they maintain some level of peer review process however this peer review process is not as extensive as journals in journals usually there is a concept of blind peer review a researcher will not even know who is actually reviewing the research so there is a level of transparency however in this case when we are publishing book chapters that transparency is difficult to maintain because you know who the editor is importance of book chapters vary from organization to organization just like it varied for books and since it takes a lot of time timeliness of research publication is not assured so we have seen that starting from journals which is considered the most prestigious uh outlet for research publication to books and book chapters timeliness is difficult to assure so what do we do if we want to report our research immediately and want to get feedback from other researchers the best mechanism in that case would be a conference conferences are very important Avenue of presenting publishing and peer reviewing research this is a kind of double peer review process because first of all in most conferences you will be reviewed by someone before presenting your research if you are accepted when you are presenting your research in a conference everyone attending your presentation is a reviewer their questions their guidance will be an extremely important resource for further research in the field and I have seen a lot of researchers who go to conferences to get that feedback but again importance of presenting in a conference vary from organization to organization some organizations consider it very prestigious to attend a conference and present in a conference similarly there is again ranking of conferences all conferences are not the same some conferences have acceptance rate of less than four percent even stricter than some of the journals but those type of conferences are considered very prestigious there are international conferences and local conferences similarly due to direct peer review while presenting conferences become an extremely important Avenue for publishing and presenting your research importance of presenting in a conference varies from organization to organization as I discussed earlier this was the discussion on how to select an Avenue or an outlet for publishing your research now we will be talking about how you will select literature for your research to build a case for your research method in this literature review you have to consider these things selecting a problem reviewing literature and then writing the literature review usually research starts with an idea it leads to research questions and that leads to research hypothesis we had discussed earlier that a hypothesis is a tentative answer to the research question before you do anything else you have to do literature review so that you understand you have confidence on what others have done in that field here's another diagrammatic representation of this process so if we have to make it a flowchart we'll describe it that we need to Define an idea in as general terms as possible by using Journal sources we'll we'll be talking about what these General sources are in a minute then we will search through second resources then we'll search through primary sources will organize our notes and then write a proposal so we have to follow these steps before we can even start conducting our research or start applying research method literature review provides greater confidence in selected topics because it lets you know what others have done and it provides researcher with new ideas and approaches not previously considered if you do not review literature how will you know what others have done what approaches others have used so this is an important aspect where we have to make sure that we understand what has been done in the past and what has not been done in the past so that your research is important and significant it identifies individuals conducting research in a related field and these are those individuals whom you may want to contact for review or for their guidance because they have already published in that area it also gives you ideas concerning research method and their designs in similar studies and it may reveal sources of data unfamiliar to you because a lot of data might have already existed so you might not need to collect data you might just need to analyze the existing data it may introduce you to measurement tools that work effectively and it may also reveal problems and methods of dealing with them similarly it may help you interpret and make sense of your results and help you relate your results to work of others now once we know what is the importance of literature review let us talk about what are the elements of literature review these elements include reports of studies of closely related problems that have been investigated in the past design of study including procedures employed and data Gathering instruments used populations that were sampled if it is a quantitative research method and sampling method employed you will also try to find out extraneous variables that could have affected the findings of previous research variables that were defined in the literature faults that could have been avoided and recommendation for further research this is where we start our own research there are some pitfalls to avoid remember you may need to change your first idea similarly don't pick a trivial project don't try to do more than what is possible try to do something new and don't fall in love with your idea so if we avoid these things we can assure that our research is flexible and that it's doable if you remember research being doable was one of the important aspect of research method similarly we have to Define our interests personal experiences or first-hand knowledge might give us some idea of where to start for example if we want to understand how programmers program or how new tools of programming impact productivity of researcher of programmers we need to start right there we can go to the professor and ask them if they have some idea and think about what has not yet been asked that will be an extremely important mechanism where you will be able to find out the research idea here is an example idea is open classroom and academic success problem could be what is the effect of open versus traditional classrooms on reading level so this was the question a research hypothesis could be the children who are taught reading in open classroom setting will read at a higher level grade than children who are taught reading in a traditional setting so what was the process we came up with an idea we raised a question and we tried to answer that question tentatively this answer is hypothesis and we understand that hypothesis is an educated guess this hypothesis doesn't come from sky this has to be grounded in theory in literature what others have done based upon that now you will follow the next process you might want to design an experiment in which you have two groups of students one type of students going through an open classroom setting and the other other group of students going through traditional classroom environment and then you compare the results of the same teaching material at the end of the course at the end of that experiment or you might already have these type of classes traditional classrooms and open classrooms in different schools or universities or colleges you might simply go there and collect data with a questionnaire so it will depend on your interest and depth of the knowledge or depth of the research process that you want to employ different sources of information there were three major sources of information the first one was General sources what it does is it provides an overview of a topic and provides leads to where more information can be found and examples include Daily Newspaper news weeklies popular periodicals magazines trade books these type of things are called General sources they should only be used to develop an idea once you have developed an idea that that's the area that you want to research you go on to secondary sources they provide level of information once removed from the original work for example books on specific subjects and reviews of research after refining your idea then you move on to primary sources and what are these primary sources journals abstracts and scholarly books and what this does is it contains original reports of original work or experiment the next one is how to use General sources General sources provide a general introduction to topic area and some clues about where to find more information internet is an important General resource but be cautious about it because internet is currently an unmoderated and unregulated media so internet is good for fun and having ideas but be cautious in the use of those resources that you get from internet you should make sure that resource is reliable enough secondary sources second resources provides scholarly summaries of research and they have references to more sources on a particular topic how do we use primary sources primary sources provide reports of original research and we understand the journals are the foremost resource for this type of original research reports now the question is why are journals the best sources we have already discussed this in the previous section of this lecture that we select journals for our publication Outlet because they are considered more prestigious because of their peer review process so same goes here journals are important because of their peer review process experts review the article and make recommendations and many submitted articles never get published only those articles get published who have done something really innovative something new thank you very much we'll see you in the fourth lecture in the series next week thank you Lecture 04: [Music] [Applause] billah I would like to welcome you to the fourth lecture in this series on research methodology in this lecture we will Define what ethics are we'll also look at different legal issues with unethical behavior in research guidelines for ethical standards will talk about plagiarism and we'll talk about ways how to use others research in building a case for our own research study here is dictionary meaning of Ethics ethics is a system of moral principles or values now the question is what is a principle a principle is a basic truth law or assumption what is value a value is a principle standard or quality considered worthwhile here are some problems with research unethical behavior in research in a survey of 1100 researchers at American universities with 245 respondents 88% believed that fraud was uncommon among researchers however 32% reported that they had a colleague that they suspected of falsifying data and when asked for major motives of that intellectual fraud or unethical Behavior the reported desire for fame and recognition job security firm belief or wish to promote a theory laziness similarly misrepresentation failure to credit or fully credit deserving contributors or crediting or over crediting someone who did not deserve that much credit so these are all unethical behaviors you'll be surprised to find out the history of unethical behavior in research I'm giving you some examples only from American perspective but this does not uh mean that these are the only unethical behaviors in research European history is also full of unethical behaviors some of those examples include tusky syphilis study in this research researchers injected healthy people with syphilis virus without their know and what they wanted to do was they wanted to find out how syphilis virus develops in a body over a long period of time so basically those researchers played with the lives of human beings similarly radiation experiment in jails the uh the researchers bombarded uh prisoners with radiation without their knowledge and the purpose was to find out effects of radiation on human body similarly cancer cell experiment that researchers knowingly and willfully grow grew cancer cells in their uh patients biological agent studies researchers sprayed the whole city of San Francisco with a chemical agent and LSD experiments uh CIA uh Central investigation uh agency of finan of America they conducted these type of experiments this led to 1974 US national research Act in which they mandated that all federally funded research proposals must go through an Institutional review board history of European research is also full of these types of examples so without going into further details of these incidents the lesson that we can learn from these things is that we should be highly ethical in our research problems in our research methods let's see the foremost thing whenever you involve someone in your research you should do is you should provide an informed consent letter that informed consent letter should include the following purpose of the research identity and credentials of the researcher the research process how long the participants will be involved in offer to withdraw from research at any time for any reason potential benefits to participants as well as to society potential harm or discomfort to individual an assurance that results will be kept in strictest confidence how to get a copy of the results how you can be reached should anyone one have any questions and a place for for prospective subjects to sign indicating that they agree to participate or if you are involving children in your research their parents should be able to sign that consent letter similarly what procedure will you use to ensure confidentiality of the data you should also answer a question how do you plan to debrief the participants for example you're your research study uh might focus on something where you do not really want to tell your participants what you are doing at the beginning some Behavioral Studies and some types of Behavioral Studies you might want to do that but you should have a mechanism to debrief those uh participants what have you done but before we do that there should be um an Institutional Review Board in different organizations or a third party who should be able to tell us whether the research that we are conducting poses harm to the individuals who are participating in that type of research and an important thing to consider is that why do we do research anyway we do research to increase our understanding of how and why we behave the way we do if this is behavioral type of research now you might think that in uh computer science we might not be dealing with these types of issues we might not be dealing with human beings for example if you're doing research on grid Computing or we developing um agent systems but think about it computer science is a broad field computer science includes those things that I just mentioned but it also also includes behavioral aspect how people use technology how people adopt technology so we should consider all those things uh when we are conducting these type of research obviously it's not as dangerous as um injecting syphilis virus into subject's body but still there could be some psychological impacts of your behavioral type research in computer science so you should be aware of those things and you should make sure that the participants are also aware of those things and there is no risk or harm to the participants here are a few basic principles of ethical research you must protect participants from physical or psychological harm you must ensure that information about individual does not become public you must not CSE people to participate you must ensure that participants know and understand what is involved when using children you must gain permission from parents and use extra caution information about participants should be held in confidence at the end of the experiment you must explain the study to each participant similarly we should use computer simulation if possible if treatment is harmful try to find population that has already already been naturally treated for example if this is um medical research there might already be people who have developed that type of disease that you want to uh conduct your research on always secure informed consent if possible report summary data rather than individual results and an important aspect always get your design reviewed now we will be talking about another type of ethical issue and the interesting aspect is that this type of ethical issue does not only relate to your research it also relate to your everyday assignments and that issue is plagiarism what is plagiarism plagiarism is the act of presenting words ideas images sounds or the creative expression of other as your own you might have seen people copying their assignments well researchers do that too and there are some serious consequences when they are caught a study of almost 4500 students at 25 schools in United States suggested cheating is a significant problem in high school 74% of the respondents admitted to one or more instances of serious test cheating 72 2% admitted to Serious cheating on a written assignment just imagine so if children in schools are cheating what type of researchers would they make so we have to make sure that we avoid this type of ethical issue and how do you know whether you have plagiarized or not if you have included the words and ideas of others in your work that you neglected to site you did not explain that this idea is coming from someone else that means you have plagiarized similarly you have had help you wouldn't want your teacher to know about means you have cheated on that assignment similarly if you do not want your reviewers to know about something you have quoted or you have included in your research or your readers for that matter you have probably plagiarized so you always want to avoid this type of unethical Behavior there are two types of plagiarisms intentional plagurism and unintentional plagurism intentional plagurism includes copying of someone else's work buying or borrowing papers cutting and pasting blocks of text from electronic source without documenting media borrowing without documentation web publishing without permission of the creators and when I said documentation I meant proper citation an unintentional type of plagiarism careless paraphrasing what is paraphrasing paraphrasing is describing someone else's ideas in your own words so a careless paraphrasing in which you did not site you thought since you have rephrased it you have explained it in your own words now it's your idea no it's still not your idea the idea came from someone else you need to cite it you need to give reference you need to give credit to the original author of that idea similarly poor documentation or quoting excessively that your paper is full of quotations quotation itself is not plagiarism but if you over quote if every second sentence of your document or your paper or research study is quoted from someone else it is plagiarism even though it's unintentional type of plagiarism similarly failure to use your own voice a research study must include your own voice you use someone else's research to build a case for your own study okay now here are some of the excuses that over the period of time while teaching this class I have got from different students why should we be ethical why should we not copy why should we not pleasu eyesee and the reasons behind that questions are as follows when you copy you cheat yourself you limit your own learning the consequences are not worth the risk it's only right to give credit to authors who whose ideas you use citing gives authority to the information you present citing makes it possible for your readers to locate your Source research is not a me versus them game it's about learning to learn and cheating is unethical Behavior anyway so what are the possible consequences if you search this topic on internet you'll find out a lot of stories about a lot of researchers being caught and what happens to them when they are caught their reputation as a researcher is at stake they sometime lose their jobs and it is very difficult difficult to build a credibility again once you have been caught plagiarizing just imagine if your architect cheated on their math exam what type of engineer would they be if your doctor had cheated on their exams what type of diagnosis would they do for you so think about this in any field and you'll find out that cheating and plagiarizing is not only unethical it is highly dangerous for others so we always want to avoid plagarism we do not want to plagiarize so does this mean we need to site everything the answer is no we don't need to site everything especially the ones that are common knowledge and what is common knowledge facts that are widely known or information or judgments considered common knowledge we do not need to document them we do not need to site them for example if you say word is round it's a common knowledge you don't need to cite this similarly if someone says computer networking is transmission of ones and zeros it's a common knowledge everyone knows about that we don't need to cite this so how we how do we find out whether something is common knowledge and something is not common knowledge you do it by this mechanism when you are doing your literature review if you find the same thing with three authors that means probably that fact is common knowledge and we do not need to site that but remember whenever you are in doubt site give reference we do not need to document we do not need to give reference when we are discussing our own experiences observations or reactions we do not need to site that because there are our own ideas our own feelings our own findings our own observations and reactions and experiences similarly when uh we are compiling the results of original research from scientific experiments or surveys or or whatever type of research method you have used we don't need to site that others will cite that when they will use our research for their own research method for their own research studies similarly if you're using come common knowledge we have already talked about that okay now here's another question if I change a few words am I okay the answer is no paraphrasing changing a few words or explaining the same idea in different words in your own words means paraphrasing paraphrasing original ideas without documenting your source is plagiarism we cannot do that but we can borrow from the word work of others in our own we have already talked about that that uh good research always builds upon others work what others have done so how do we do that so how do we use others research we use others research in the following three manners we quote we paraphrase we summarize have to make sure to use a blend of different Source materials making sure that your own voice is heard your voice should not get lost in those uh different types of citations you should use citations that's an important aspect of research but you should make sure that your voice remains important so how do we quote quotations are are exact words of an author copied directly from a source word for word you sometime even use quotation marks to make sure that the readers understand that this is your this is not your idea and you are borrowing that idea to build your own case make sure you must site every quotation when do we use quotations we use quotations when we want to add power of an author's words to support our own argument or if we want to disagree with an author then we site when then we quote similarly if you want to highlight particularly uh eloquent or powerful phrases or passages then you quote when you are comparing and contrasting specific points of view or if you want to note the important research that precedes your own and you can see I have cited this text these are the people who actually wrote this one similarly how do we paraphrase paraphrasing means rephrasing the word words of an author putting his or her own thoughts in your own words when you paraphrase you rework The Source idea words phrases and sentences structure with your own however as we have discussed earlier it's still not yours it's someone else's you need to cite this type of paraphrasing you need to cite that you need to give reference when do we paraphrase when we plan to use information in our own note cards and wish to avoid plagurism we want to avoid overusing quotations want to use our own voice to present information this is where we paraphrase how do we summarize summarizing involves putting the main idea of one or several writers in your own words including only the main points summaries are significantly shorter than the original and take a broad overview of the source material once again it is extremely important to site those sum summaries when do we summarize when we want to establish background or offer an overview of a topic want to describe knowledge about a topic want to determine the main ideas of a single Source this is where we summarize so how do we know how do we decide which strategy strategy to use where and when when you are reading your literature during that review process you include any direct quotes or unique phrases in quotation marks or mark them with a big Q this will help you understand that that this is uh the code that you want to include in your in your own research and make sure that the writer's name is identified similarly when you want to paraphrase when you reading write a big p similarly include page numbers and Source reference so that you can go back and check for accuracy as you write now we move on to the next important aspect how do we site how do we document that we are citing something from someone else this is called intext documentation when you you are using someone else's work in the middle of your own paragraph So how do we do that the purpose is to give immediate source information without interrupting the flow of paper or the project the research study intext documentation is taken very seriously inaccurate documentation is as serious as having no documentation at all we need to be brief uh provide brief information in intext documentation and it should match Source information in the works cited or the references section at the end of your research paper you must use an original idea from one of your sources whether you quote or paraphrase it you summarize original ideas from one of your sources you use factual information that is not common knowledge you quote directly from a source you use a date or fact that might be disputed that's that's when you document and how do we site there are different styles just like there were different mechanisms to rank journals and conferences similarly there are different mechanisms to site your in text documentation some of those styles are APA Styles American Psychological Association they have their own style similarly MLA or the Howard style usually we cite author's last name and year of publication in intext documentation in the absence of an author cite the title and the year of publication if you're using more than one book or p paper by the same author list the last name year title in a sequence if you identify the author and the title in the text just list the year of publication you should remember that every journal and every conference has a different style for this type of documentation so whenever you are submitting your paper to a journal or a conference you should make sure to follow their style some would say that we will follow Howard style so all the specifications of Howard style will go into that and that journal or the conference does not provide any further information on that so you need to go back to the source of Howard style and find out how to site but sometimes uh rather most of the time conferences and uh journals will provide you with a s sample paper and you'll use that sample paper to build your own case this is the end of uh our fourth lecture in the series I hope to see you next week Allah n [Music] Lecture 05: [Music] n [Applause] bman I would like to welcome you to the fifth lecture on the series of lectures on research methods today we will be talking about uh some quantitative research methods in today's lecture we'll focus on the following things we'll talk about different variables we'll find out what do we mean when we say relationship between independent and dependent variables we'll talk about some important types of variables we'll also revisit hypothesis samples and populations and the concept of significance variables is a class of outcomes that can take one or more value the more precisely a variable is measured the more useful the measurement is so what is a variable a variable uh is an outcome of a treatment a variable is something that can take different values for example you must have U studied this concept in your basic mathematics that X is a variable and X can assume any value so those values could be uh discrete numbers or those values could be for example uh male female or those values could be level of Education High School higher Secondary School College University or undergraduate graduate doctoral uh these values will go in a variable now what is dependent variable the outcome of a research study is called a dependent variable or the effect it depends on the experimental treatment now remember we talking about quantitative research methods so outcome of your experiment the effect of your experiment is called dependent variable now uh what is independent variable then independent variable includes the treatments or the conditions under control of the researcher the causes those are called independent variables because researcher can manipulate those variables and there should be at least two different values of independent variable if there are no two values available for that variable how will you find out the difference whether the treatment had an effect or treatment did not had an effect so there should be at least two levels of independent variable or the treatment or the cause now what happens to the effect if the effect is same with both the treatments or both levels of the independent variable it means there was no change there was no uh impact of independ change in the independent variable on the dependent variable relationship between independent variable and dependent variable independent variables should not be confounded what this means is that levels do not vary systematically with other variables if there is a change in other variable in some other variable that is not under study for This research level of this independent variable should not change if it does that is a confounded variable and that means you do not have as a researcher you do not have complete control over independent variable and if you do not have a complete control on uh independent variable you cannot say anything for sure for the dependent variable dependent variable should be sensitive to changes in the independent variable so we should have a pair of dependent and independent variable in such a way that the dependent variable should be sensitive to changes in independent variable if dependent variable doesn't change to any change in the independent variable it means that probably there is no relationship between the two variables well that could be a finding in itself but uh whenever we are selecting dependent and independent variable we should uh keep these things in mind there are other important types of variables one type of variabl is called control variable and it has a potential influence on the dependent variable so what do we do with this variable we control this we do not let uh control variable impact the values of dependent variable that's why it's called control variable the other important type of variable is called extraneous variable or intervening variable intervening variables have unpredictable impact on dependent variable and they cannot be controlled that's why they are called intervening variable or extraneous variable so what do we do if we cannot control them we need to infer their effect on the dependent variable for example if you are using working Professionals in their actual professional environment for your research study um maybe for uh uh judging impact of using software agents in in the work environment there are a lot of variables that you cannot control for example someone might be sad or extraordinarily happy so that will impact the uh results of your study so that type of variable that you cannot control because you are doing your research in the actual work environment you call them extraneous variable now you can U make an attempt to control for it by using University students and make them behave like professionals now you might be able to control those variables but you compromise on another thing you will not get the actual results that you would in an actual work environment similarly the next U important type of variable is called moderator variable and these variables are related to either independent variables or dependent variables and they hide the true true relationship between independent and dependent variables so we need to take uh the effect of moderat variable into consideration before we finalize our study here's a summary of uh different variables their impact and their definitions dependent variable a variable that is measured to see whether the treatment or the manipulation of independent variable had an effect this is called outcome variable results variable or Criterion variable independent variable is the one that is manipulated to examine its impact on dependent variable this is called treatment variable or factor variable or even predictor variable control variable is a variable that is related to dependent variable the influence of which needs to be removed it is called restricting variables extraneous or intervening variables are the ones that are related to either dependent or independent variables and are not part of the experiment these are called threatening variables moderator variables are the ones that are related to either dependent or independent variables and have an impact on dependent variable these are also called interacting variables hypothesis we have already talked about this concept in one of our previous lectures here we will see how to scientifically write down your hypothesis a hypothesis reflects General problem under study restates the general problem in a form that is precise enough to allow for testing if you remember we called the hypothesis a tentative answer an educated guess and we said that hypothesis should always be written in a declarative form form which is testable a hypothesis is a tentative assumption made in order to draw out or test its logical or empirical consequences there's another definition a belief based on an observation that has not been tested yet or an educated guess now we will talk about different types of hypothesis the first important type of hypothesis is called null hypothesis and null hypothesis always assumes that there is no relationship between independent and dependent variable under the study so what do we mean by that that the mean theoretical average of population 1 is equal to the population uh theoretical average of population 2 and this is how we donate it this is H KN which means null hypothesis colon mu1 is equal to mu2 what does this mean this means that there is no difference between the theoretical averages of population one and population two and that's how we donate it and we always start with the concept with the notion that there is no difference and there is no relationship between dependent and independent variable that's what we assume that's the starting point for all research purpose of the null hypothesis is to provide a starting point for analysis and it is accepted as true unless proved otherwise and it assumes that chance may have caused any observed differences when do you develop hypothesis because we have discussed this thing in our literature review section that it might be based upon an observation so you observed that if something happens it has a different effect and now when you want to actually test it you assume that the observed difference was because of a chance because something else had happened so you want to remove that bias of your previous observation before you conduct this study and it provides a benchmark for comparison the second type of hypothesis is called research hypothesis research hypothesis is a statement of inequality as compared to null hypothesis null hypothesis was statement of equality that there is no difference due to treatment now we assume that there is difference and we uh describe that relationship exists between independent and dependent variable and how do we donate it scientifically we call it a H1 and we put a colon and now you see we we are not using mu rather we using X1 bar X1 is the theoretical average of sample one and we say that it is not equal to theoretical average of sample two we use an inequality symbol here and what does this mean this means that there is some relationship between independent and dependent with variable that's why when we were using those two levels of treatment one theoretical average of one sample is different from the theoretical average of the other one because there was some difference okay now there is another interesting question that you might ask that we used population in the previous hypothesis and we are using sample in here that's because null hypothesis talks about the whole population whatever we do in our research we want to generalize that research if you remember from our first lecture when we were talking about uh characteristics of good research we said that good research is generalizable and for example if you want to find out behavior of Pakistani people certain thing you might not be able to actually go and ask every single Pakistani we are about 160 million people it is practically impossible to go to ask the same question to every single person so what do we do we ask that question to a sample to few people and based upon the responses from that sample we generalize that result so when we are doing our um experiment we always consider population that's why we are using X1 bar and X2 bar instead of mu1 and mu2 mu always uh stand for population xar stand for sample now there are different types of relationships between research hypothesis one type of relationship is called non-directional research hypothesis it means that two groups that you had are different but there is no Direction specified in the hypothesis and how do you say that that H1 colon X1 is not equal to X2 we do not know which one is greater or which one is lesser this leads to the second type of uh research hypothesis which is called directional research hypothesis the groups are different and the direction is specified for example X1 bar is greater than X2 bar or the other way around so these are called directional and nondirectional research hypothesis what is the purpose of research hypothesis because they can be directly tested during a research process and they provide us a place where we can compare them against null hypothesis here are some of the differences between null hypothesis and research hypoth hthis null hypothesis is a statement of equality between variables research hypothesis is inequality between variables null hypothesis always refers to population research hypothesis refers to sample null hypothesis is indirectly tested research hypothesis directly tested null hypothesis is uh stated using Greek symbol me whereas research hypothesis is stated using Roman symbol xar null hypothesis is implied research hypothesis is explicit what makes a good hypothesis we have talked about these things that it should be stated in a declarative form that it asks a question it answers a question and it uh posts relationship between variables it reflects Theory or literature it should be brief and to the point and it should be testable Now we move on to uh the important concept of the difference between samples and population a sample is a representative portion of a population whereas population is the entire group of interest and we have already discussed that results from samples should be generalizable to population significance means that what is the probability that your observed differences are from the treatment and not from chance we need to find out the significance of your results because there could be influences of more than one treatment you might be thinking that you are controlling your independent variable but there might be impact of intervening variable or moderat or variable that you cannot control so in that type of situation you need to find out you need to Tell the readers what is the significance of your research findings and once again what significance significance is the probability that observed differences resulted from treatment and not from chance significance level is usually associated with risk that you are not 100% certain that the null hypothesis is incorrect so that risk assessment is very important okay now what types of statistical treatments you can apply on the data that you collect as a result of this experiment or a survey or any other quantitative research method that you employ for your own research there are two important types of analysis that you can perform first type of analysis is called univariate analysis in which you are interested in only one type of variable and you might be interested in the impact of many variables on a dependent variable now we will talk about different types of statistical treatments that you can apply on the data that you have collected as a result of your research study this research study could be experimental based or it could be survey or it could be any other type of quantitative research method in which you have got a lot of data you might be interested in finding out the difference between an independent variable and dependent variable what if you have more than one inde dependent variables you cannot use univariate analysis in that case you need to use multivariate analysis what are the different things that we use in statistics we calculate mean we calculate mod we calculate median we do T tests we find out Z scores we do regression regression is um uh the projection that in future if the trend in the past follows this will be the future result we want to find out correlations between different variables we might even go on to um structured equation modeling and there are other different types of um statistical treatments that can be applied to the data that We Gather in this lecture we are not going into the details of those statistical treatments because they they become part of Statistics course but we need to identify we need to understand that these types of mechanisms exist that we can apply on our data that we gather as a result of the experiment or any other quantitative research method so once again why do we do all these things we do all these things because we want to increase our understanding of how and why we behave and do the things we do this is what's called research method I look forward to seeing you in another episode of this lecture thank you very much Allah [Music] n Lecture 06: so i would like to welcome you to the sixth lecture in the series on research methods today we will be talking about different important research methodologies in information systems and computer science research today's lecture is mainly based upon an article research methodologies in mis an update which was published in communications of the ais in november 2004 we have uh talked about these things briefly in our second lecture today we will be talking about these things in more detail first of all let's review definition of what a research methodology is we have discussed that an overall process guiding the research project is called research method or a primary evidence generation mechanism or that we are not looking at specific methods or techniques during various phases of the project for example we are not talking about statistical methods at this time so this is what we call research methodology and this is a list of various important research methods that we employ in various fields whenever we conduct research this list has already been shared with you in the second lecture and how do we decide on which of these methodologies to choose from we have described this criteria that it depends on the research question research model degree of maturity of research topic external or internal validity prior research by others investigators in trust and strengths and publication outlet and audience so now let's talk about these methodologies in detail the first methodology was speculation or commentary type research articles this type of research is driven by thinly supported arguments or opinions with little or no empirical evidence what do we mean by empirical evidence we mean that we do not have a lot of data to support these type of arguments or opinions but people strongly believe in those type of things usually experts and leading authorities in the field write these type of research papers the other research model or research method that we use is development of frameworks and conceptual models this type of research intends to develop framework or a conceptual model an organizing framework model of important variables in research domain is proposed possibly with relationships evaluation and verification may await future work so what happens in this type of research is that we identify different variables and we put them in some order and we try to build some relationship that what is a dependent variable what is an independent variable and how do they interact what is the impact of intervening variables what are the impacts of uh confounding variables so all these things all those variables and their relationships in a diagram is the development of research framework another important type of research is library research in library research we basically review existing literature we summarize or we may make some extensions to the existing body of knowledge when we want to go beyond just summarizing existing literature we call it literature analysis and literature analysis includes criticizing analyzing and extending existing literature and attempt to build new groundwork some of the examples include meta-analysis now what is meta-analysis the word meta comes from computer science in which we say meta is data about data for example we have a journal or we have an article that article has some content in it but there is some other information that comes before the journal before that article for example name of the author the journal number of pages page numbers year of publication volume number issue number so all that type of data is called metadata so when we apply the same concept in research what we do is we gather number of articles on a certain topic and we try to derive unifying themes from those articles for example what type of methodology has been used in most of the research for example if we are reviewing 300 articles on a certain topic that review will give us a very good overview of what type of methodology to use if we want to conduct research in that particular field we will also know about what questions have have already been asked and what questions have already been answered what questions are open questions what questions are a trivial question in that particular field so literature analysis becomes an important research method now this could include qualitative and quantitative uh data analysis we will be talking about quantitative and qualitative uh data data analysis uh in in greater detail later in this lecture but for the time being in meta analysis we do have both quantitative and qualitative data data to analyze and this sometimes even defines future direction of a certain certain field so typically each previous study is treated as a data point in literature analysis case study case study is a study of single phenomena for example an application a technology a decision in an organization or logical time frame and one of the examples that i can quote here is uh experience of a company in implementing an erp system erp systems are called enterprise resource planning systems enterprise resource planning systems have different types of consequences different organizations have different types of experiences with that some organizations implement this type of system very easily others do not some organizer spends millions of dollars and still do not get the value out of that uh information system that they are trying to implement so why did that happen why was it successful in the companies where it was successful and why did it fail in the companies where it did fail so that type of research analyzing the causes and uh doing an in-depth analysis constitutes case study and usually unit of analysis is one organization one company one person and another example of case study could be in allama iqbal open university we have uh phds in iqbal yat one person so studying about iqbal his literature that he wrote all those things they become case study mostly this is part of qualitative data analysis or qualitative research method survey is another important research method it's a quantitative in nature and this research uses pre-defined and structured questionnaire to capture data from individuals in service usually we provide some space for people who are responding to the questionnaire for some open form entry however mainly survey is meant to get data on particular questions in either yes no type of answers or rating on a scale or select the best option so multiple choice questions type questionnaire is mailed out usually now today with the use of technology we might email this type of survey or we may fax the service and get responses but usually response rate is pretty low sometimes if if a researcher gets 10 response rate it is considered extremely good so surveys are an important research method the next important research method is called field study in field study what we do is we go and study single or multiple and related processes or phenomena in a single or multiple organizations an important thing to note is that investigator or the researcher spends time in the actual organization and obtains data in the actual setting sometimes this type of research is also called ethnography and in ethnography researcher actually goes in a certain ethnic environment and tries to do things and tries to think about things the way actual people do so as a result the findings of the of the research are from the real people's perspective or that organization's perspective so they try to understand how people behave in a certain culture in a certain organization in a certain environment how do they act how do they react so for example if a researcher goes and spends time in for example afghanistan or in some latin american culture and tries to find out how people live their lives in that type of environment that type of study will be called ethnography next research method is called field experiment research in organizational settings that manipulate and control various experimental variables and subjects this type of research is usually expensive because it involves involvement of real people real organizations and this type of study will have serious consequences in the output of that organization for example a company wants to uh to implement certain software or as i gave an example they want to implement an erp system but in the previous attempts people were not willing to adopt that system so now they want to do another attempt to implement that system so they might do an experiment in a small portion of their organization instead of launching it in a full-fledged organizational environment so that type of research will be called field experiment next important research method is called laboratory experiment in lab experiments usually research is done in a simulated lab environment that manipulates and controls various experimental variables and subjects typically it is conducted in a classroom environment with students next research method is called design science this involves development of artifacts creation and evaluation of i.t artifacts what are it artifacts the programs we write the coding the executable files software engineering documents requirement analysis design architecture all those documents all those valuables or deliverables are called artifacts if your research involves the development of those type of documents or writing code or doing programming that type of research will most probably be classified as design science because you are designing new architectures new systems new softwares in design science uh problem solving paradigm is followed and in this problem-solving paradigm it creates and evaluates ideas practices and technical capabilities and products qualitative research in qualitative research what we do is that we try to understand people and social and cultural context within which they live this method may include ethnography action research case research uh interpretive studies examination of documents and texts and we will have our next lecture on this topic where we will have a demonstration of a qualitative research method next research method is called secondary data research method in this type of research method we utilize existing data from organizational resources for example financial and accounting reports archival data and published statistics next research method is called content analysis in this type of research method we uh analyze text and we examine that text in a systematic way by identifying and grouping different themes and code them we classify them and we develop categories relevant example would be from religion tafsir what researchers do is they try to understand the meaning of quran in the context of hadees and interpretations done by other researchers so that type of biblical analysis is called content analysis now this doesn't have to be quran uh all the time this could be an email this could be a letter this could be analysis of any textual uh correspondence between two organizations between two people so this is again an example of qualitative research method next research method is called combination research method and as the name suggests it is a combination of different research methods there is a concept of triangulation triangulation means that one research method gives you certain results you employ another research method and if you get the same results your findings are more reliable if you use another research method it will become even more reliable so triangulation of methods is very important well triangulation of method is not the only thing that makes your research reliable triangulation of data sources could also make your research more reliable for example you collect data from experts you collect data from existing documents from published reports and you collect data from surveys you conduct simulations so you are employing different types of research methods to generate different types of data it includes quantitative data it includes qualitative data and you combine them and as a result your findings are considered more reliable in combination we also cross validate the results and it provides more confidence in the findings next research method is called survey research in survey research we have again two types of surveys uh one is exploratory or descriptive survey or explanatory survey in exploratory survey researcher tries to become familiar with the topic and finding the value of various various variables for example opinion research what people think about certain phenomena in explanatory research we try to find causal relationship between variables for example in technology adoption the basic model is technology attributes those attributes affect attitude attitude effects intention and intention effects use we have already spent some time on experiments uh when we were talking about quantitative research method we'll talk about experiments in a little bit more detail here experiments an experiment is a test under controlled conditions that is made to demonstrate the validity of a proposed theory or knowledge principles or examine the validity of a certain hypothesis in other words why do we need to conduct experiments we look at the the scientific research process we make some observation and we get curious we ask questions we study literature after studying that literature we develop tentative answers to that question now whether those tentative answers to those research questions are valid or not we need to conduct some study we need to do some experiment now there are some basic principles to experimental designs the foremost principle is that a planned interface in the natural order of events by the researcher should be selected the goal is to observe the effect of single or multiple independent variables on a dependent variable each combination of independent variable is called a treatment we have already talked about it in our last lecture a selected condition or a change is introduced there may be several treatments subjects are assigned to different treatments treatments are administered and their impact on subjects is observed next is simulation simulation is a design science methodology and it is a term applied to the process of modeling the essential features of a situation and then predicting what is likely to happen by operating with the model on a case-by-case basis we are developing an algorithm for efficient communication we do not have access to real networks and even if we have access to real networks we do not want to overload that network because that network will be handling real-time data that could be very important data and and if we overload the system things might get delayed so we do not want to uh mess around with the real life system so what do we do we do simulation in that that type of situation today we have discussed different research methods and we know that there are many methodologies that exist beyond the ones that i have described today and no research method is better uh from one another all research methods have their own benefits their own power and they have their own disadvantages so which method do we choose we go back to the criteria that i had explained at the beginning of this lecture according to that criteria we select which methodology is the most appropriate in a given situation for your particular research question but you should be very careful in the choice of a methodology because it takes a lot of time energy and effort to do a research project and you do not want to get into into a situation where after doing that research project you find out that you had chosen inappropriate research method for for that type of research however you should be very careful about the choice of a particular research method for your research project a research project involves a lot of time commitment a lot of energy and you do not want to find out at the end of the study that the research method that you have used for that type of research was inappropriate so you should spend some time in selecting an appropriate research method for your research project thank you very much we'll see you in the next lecture allah you Lecture 07: thank you thank you assalamualaikum I would like to welcome you to the seventh lecture in the series of lectures on research methods before we go into the discussion of qualitative methods we need to spend a little bit more time one more time to find out the difference between quantitative and qualitative research methods in quantitative research methods we focus on collecting objective set of data we employ different types of research methods for example surveys experiments simulations and once again we want objective data for example if we ask someone for gender it will yield an objective answer similarly in qualitative research methods we focus on collecting subjective information we might not have a lot of numbers to Crunch in this type of research method and data is mostly in the form of spoken or written words images Impressions or even gestures some of the examples of research methods that belong to qualitative research include interviews case studies content analysis observations natural experiment photography ethnography histography social media for example if we ask a question what role information communication Technologies are playing in the socio-economic development of Pakistan to different people we might get different answers so we asked that question to different people with a varying degree of expertise and backgrounds these are experts these are end users Regulators stakeholders or or even people who are having independent opinion and we do an important thing we carefully record everything this recording could be in the form of audio or video but we have to make sure that we record everything in natural sequence what socio-economic benefits is the progress in ICT sector yielding variety of socioeconomic benefits of ICT utilization in Pakistan are appearing I'll give you a prime example of the ready availability at fairly reasonable cost of the mobile phones mobile phone is transforming the way people interact with each other readily available mobile phones in schools to Headmasters headmistresses teachers and students are making it possible for the Young Learners to interact with the teachers and that to me is a Quantum Leap in terms of the educational change that one sees in Pakistan what socio-economic benefits is the progress in ICT sector yielding interestingly in Pakistan ICT meaning information and communication Technologies is the component relating to to telecommunication has grown tremendously over the last 18 months after the market was deregulated and with the proliferation of mobile phones throughout Pakistan there are almost 45 million of them and the fact that the cost of calling has gone down to almost less than uh what used to be the cost of a local call you can call New York a