Game Analytics Resources | Unlocking the Power of Data for ...



Developing a Research Proposal From the course Research Methodology, Aalborg University Copenhagen, 2012 versionThis is a template for a research proposal (a research plan), for a professional piece of empirical research, within an area of interest to you. It is written specifically for student projects from undergraduate to master’s thesis levels. The template is designed for work in HCI, Interaction Design, Information Science etc. where humans are involved one way or another. The basic structure will work well across a number of academic fields however. General information Consult examples of research plans, you can find many examples on the Internet. Consult the literature on planning projects – again, check the Internet. In the below a structure for your project proposal is proposed, with notes on what should go into each section. You are free to deviate from this structure, but should do so with good and argued reason. When planning your empirical work for your thesis project, go back through the lectures and consider the points raised about the scientific method, pros and cons of qualitative vs. quantitative methods, measurement scales, etc. These are the elements you need to combine into a coherent framework: A research plan. Key points [text in red font only relevant for AAU students]Reports should be max. 7 standard pages for one student, 10 pages if 2 students, 12 if 3 students (officially, you can for the RM course, but NOT for the synopsis you are handing in to the study board, write 8 pages if one student, or 5 pages per student if in groups) (you will easily get there using graphs etc.). There is no maximum but remember that having superfluous content counts negatively – only write what needs to be written. Date of delivery: See Moodle. Assign one group member (if multiple people in the group) as the overall responsible for each section. This does not mean you should not work together, but it helps having someone responsible for specific components in terms of work division. You will need to prepare a presentation for the class on December 1st. This is a group presentation, not one person talking, where applicable. Time allocated to each group is max. 5 minutes. Resources: It is important that you consider that the research plan should be focused on a piece of empirical work that you can feasibly carry out. As part of your research plan should be a rough budget estimating the costs of the proposed work, and a timeline. The proposed work will most commonly be empirical/analytical/design and development-oriented, but it can also be purely theoretical. The proejct can be data-driven, and does not need to involve human participants, but if empirical it must adhere to the formal conditions for empirical work in Information Science. If you want to do a pure theory project, which is allowed by the Study Rules for the degree, please consult your instructors for additional information. We recommend using the standard template for IV reports, in terms of formatting and content structuring (with this guide superseding the template where any conflict arises). Note that the standard template describes a situation where a piece of research is reported, whereas the research proposal is focused on a situation where research has not been conducted. The basic structure is therefore a little bit different. Proposed structure of the research proposalInsert student name, mailing address and university email address. TitleSupervisor requests, degree you are enrolled in. Add any other formalia as pertinent to your degree/programme. Section 1: Executive summaryMax. 150 words providing the most essential information. Imagine you were proposing some experimental research to your manager in a company or supervisor at a university. Emphasize the importance of the proposed work and the potential gains in terms of e.g. new knowledge. Include a table of contents after the title and executive summarySection 1: Aims (introduction) and problem definition In this section you describe the fundamental aims of the empirical work, including the theoretical framework you adopt (if any) and the hypotheses you will be evaluating (if any, e.g. explorative work). You should provide a bulleted list of the specific aims of the project, as well as a list of the hypotheses/problems you want to test. You should also provide an introduction to the actual work you will be carrying out. Each hypothesis/problem should be clearly spelled out, and the relationships between the variables involved. Present graphically how you expect your independent and dependent variables interact if conducting experimental research. If working exploratively, describe the involved variables and the expected correlations or connections between factors. Your projects must be related in some way to Information Science and must adhere with the principles of ethical research with humans where humans are involved. See for example the Northeastern University IRB guidelines for research. This also means you need to consider e.g. recruitment and informed consent of experiment participants, interviewees and so forth. Section 2: Background In this section you describe the background for the proposed research, e.g. writing more about the basis for the problem you wish to address, and describe related work (you will need to find associated literature (Wikipedia is not literature!) and describe your plan in relation to the existing knowledge in the field). Provide examples of associated work in the literature and outline why your proposed research will be different or otherwise advance the state-of-the-artProvide an overview of selected theory (if any) and how the theory is applicable to the problem under scrutiny. Consider the empirical methods used by others and select those which makes sense given the problem and the constraints of the project. Explain how these are the right for the project, and how they mesh. Section 3: SignificanceThis is where you describe the importance, or motivation, for performing the research, and describe any benefits that work might lead to. You will need to write about the applied contributions of the research planned, and/or the scientific contribution. You can write about the importance of examining the area in general (e.g. why investigation diabetes is important – because many people suffer from it etc.), and why your specific proposed work is interesting/important. You will need to argue for the importance of the work you propose – i.e. why should we care and from what perspective? Section 4: Research design, methods and theoryThis will be the main section of the research plan. In this section you describe the theories you wish to employ and argue for their selection, as well as the empirical methodological framework, any physical setups, the procedure for conducting the work and any experiments, and the detailed relationships between the variables and measures involved. You will describe the population and sampling criteria, recruitment strategies, etc. as relevant. You will need to argue for why the proposed research design is the right one to examine the problem you are interested in. If you use grounded theory and similar approaches you may not be able to identify a theoretical framework this early in the project, but you can still provide examples of such theories that might be relevant, and argue for why they might be relevant. Important elements that can be present include (depending on your specific angle/project some or all of these will be more or less relevant): 4.1 ModelsThe first thing you need to do is build a research model. The model describes each of the variables involved, and the measures you will employ. The model also describes how the relationships (proposed associations) between the variables work. It also contains information about potential confounding variables and where in the process they might affect the results. The research model should also describe when measures are applied. There are many ways it can be visualized, e.g. using flow charts or similar models. The sample research plan also includes an example. The second, optional, element is a hypothesis model or factor model. This is basically a relationship model that describes the hypothesized relationship between the independent and dependent variables in the experiment, or the correlations between variables for explorative/observational work (see figures below – many more examples on the Internet). In connection to the hypothesis model, you will need to describe each of the independent and dependent variables, and how you expect to measure them. E.g. via a survey, behavior logs, etc. Measures have to be realistic – do not go into the domain of science fiction. The research models provides the basic structure for the research. It provides the reader of the proposal of a quick overview of the proposed work, and generating these models forces you to consider how the different elements of the work are connected (i.e. improves planning). 4.2 Variables and measuresNext you need to describe in detail each variable involved, and how you intent to measure it. If you use established methods, e.g. a survey developed by other authors for evaluating happiness, be sure to critically evaluate the method. Similarly, if you design your own measure, consider the pros and cons of doing so. Remember the difference between e.g. measuring how many dinosaurs that die from a meteor strike as a total number vs. as a number per square mile – these two measures gives us very different levels of information. Also take into account the four types of measurement options we have discussed in the course (ratio scale, interval scale etc.). 4.3 Population and settingYou need to describe where the empirical work is to be carried out. It is not enough to write “lab”. Which lab? Why that one? Or if fieldwork, why there? You need to argue for why the location chosen makes sense (e.g. lab vs. fieldwork), and the pros and cons of the proposed location. For example describe the conditions present, and the potential problems that could occur and how to solve them (this last bit goes into Section 5). Next comes the population: which population are you targeting, and can it be described? Do we know any features of it? (age, gender composition, etc.). Can you find any data on the population? Define the population in as much detail as possible. 4.4 Sampling Next you need to describe the sample you are aiming for, and how you anticipate carrying out the sampling process. If you cannot feasibly obtain a random sample, what biases do you expect might affect the results, and how do you anticipate accounting for them? You do not need large sample sizes. Much experimental work in IS and HCI is carried out on small sample sizes, e.g. 30 people. What is important is considering the pros and cons of small vs. large sample sizes, and taking the appropriate precautions. For example, if using small samples, carefully note the limitations possible in terms of inference to the parent population. You also need to describe recruitment (if necessary) – how do you anticipate finding the participants needed? If e.g. using a hospital as a basis for the work, participants could be patients. If you are performing an online experiment, you may need to use online advertising, and account for potential bias, etc. You also need to describe if you will be using a vetting procedure (i.e. check participants have the right properties before admitting them into the experiment). Same process if obtaining data from external sources. 4.5 Timeline and protocolYou need to describe a timeline for the project. The timeline should include both the overall plan for the entire project, but also an overview of the process of running each component of the project (initial investigation, experiments, design, user testing …). For example, if you are planning on performing a usability test on 50 participants, with 5 participants being in the lab at any one time, the overall project timeline needs to show the dates you will be calling in each group, and the experiment timeline will outline the testing process itself (from when participants arrive at the lab until they leave). The latter is referred to as en experiment protocol, and is usually very detailed. 4.6 BudgetA rough overview of the expected costs of the work, including salaries, equipment etc. Do not spend too much time on the details – we do not need to find the lowest price for a PC of sufficient strength to run the application you want to work with – overall cost estimates are fine, but should cover the main expenses involved (salaries, equipment, facilities, participants etc.). Consider whether the costs are feasible. Section 5: AnalysisIn this section you describe the analysis plan, i.e. how you expect to analyze the data that the project will generate, whether you use quantitative or qualitative methods or a mixture.This section is important because your analysis plan must relate to the way you are collecting data. For example, if you are running an experiment where you will be collecting data from participants at specific temporal intervals, you may want to perform on-going analysis to check whether data are being collected properly etc.In this section you also need to consider the potential problems and risks that may occur, both in terms of the project itself, in terms of potential damage to participants, but also factors that affect your ability to successfully run the work (i.e. describe risks to your participants and how to mitigate them and risks to your ability to complete the research proposed). For example, participant dropout if running a longitudinal study. The same goes if you are working with a client, who might or might not be available when needed. If using external data, controlling the quality of these. Etc. You need to describe how you will handle confidentiality and protection of user rights, address ethical concerns etc. You need to outline a strategy for data safety and monitoring. Section 6: ReferencesUse full references in APA style (or the style your university requires). Section 7: AppendicesThis is where you attach any appendices. For example, if you are planning on using human participants, you know you will need a letter of consent. This should be attached to the research plan. Other examples include e.g. preliminary analysis of pilot test data. ? 2012 Anders Drachen ................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download