Description of Assessment Task and Purpose:

This assignment is concerned with research data gathering, analysis and interpretation - you will obtain a dataset, investigate quantitative methods for analysing this data using inferential statistics, design an appropriate methodology, conduct your own data analysis using appropriate software, reflect upon the results, and present your findings. The assessment involves writing a 4-page report in the format of a scientific article, and a presentation on the same subject.

Your first task is to acquire a non-trivial research dataset. This task might involve the design of a methodology for collecting data as a primary part of the research - for example, by constructing a survey to elicit data from a chosen user group. Alternatively you may take a pre-existing dataset from another source, which should be acknowledged in your report, where you then focus more on the analysis of this data. There are a wide range of sources that could be used. A few indicative examples of possible datasets could include (but are not limited to):

• ."fantasy football" player performance statistics,

• .multi-user virtual reality data,

• .performance statistics versus ground-truth data for a computer-vision recognition system, etc.

The choice of dataset can be oriented towards your own personal interests, such as data relating to your Research Project, etc., but you must submit only original work which has not been submitted for any other assignment, or which is intended to be used for another assignment (e.g. for your Research Project). If in doubt, please discuss the choice of dataset with a member of the module delivery team at the earliest available opportunity.

The next task is to investigate the existing research methods and background literature for acquiring and analysing similar datasets. Please note that the focus of this task should be more than the domain-specific methods used for solving a particular research problem (e.g. if a research paper describes the development of an artificial neural network system for stock market prediction, we are most interested in the methods used to evaluate the system performance in predicting stock market trends, not the details of the neural network model itself). There is no need to develop any artefact to generate the dataset that you subsequently analyse. Please note also that the literature review should focus on literature relating to your own research question, not "textbook" literature on research methods in general.

Based on the background research, design and implement an appropriate methodology for analysing your chosen dataset. You will need to identify an appropriate research question (or questions) and/or hypothesis to test. For example (and not limited to these examples):

• .evaluating the effect of the full moon on behaviour,

• .deciding whether hospital carpeting results in more infections,

• .selecting the best means to stop smoking,

• .testing whether acetaminophen or ibuprofen helps faster with headaches

• .etc.

Please note: you must apply inferential statistics to draw conclusions about the population under consideration from the sample data, for example, by applying techniques introduced in the module such as confidence intervals or statistical tests (t-Test, ANOVA, Spearman Correlation, etc). Your analysis must be non-trivial - for example, just applying descriptive statistics such as mean and standard deviation would not be sufficient. Descriptive statistics should however be used to justify the choice of techniques. If you are in doubt, please discuss the choice of methodology with a member of the module delivery team at the earliest available opportunity.

Take care to justify the choice of research techniques in your report, including discussion of the alternatives considered. You should make sure to explain and describe the techniques applied with sufficient clarity to allow other researchers to replicate your findings if desired. Please make sure to describe the approach and software that you used to carry out statistical testing. Consider how to best present you results, e.g. using tables, graphs, etc. Discuss the results obtained and the conclusions inferred with respect to the original research hypothesis or question. Finally, reflect on the value of quantitative approaches in respect to your study. How does this differ from qualitative approaches, and how could the latter be valuable?