Overview
The overall goal of the research paper is to test the following:
- Your ability to pose appropriate research questions and hypotheses
- Your ability to design and carry out appropriate statistical analyses for a set of data
- Your ability to report the results obtained using standard methods of expression and presentation
- Your ability to discuss and interpret the results
You may want to start with feature extraction using factor analysis, and cluster analysis to get a set of features that will allow you to ask interesting questions of the data. You should use t-tests, analyses of variance (and covariance) and regression analysis as the main statistical procedures in the research paper, although other techniques such as multivariate analysis of variance (MANOVA) and discriminant analysis can be used if you want, along with non-parametric methods. You are free to segment the data in various ways prior to carrying out analyses. Feel free to test assumptions, transform the data, carry out follow-on analyses or do any other types of analysis that assist in the interpretation of the data. Note here that the focus will be on asking and answering interesting questions of the data rather than on doing a small and un-interesting analysis that is rock solid in terms of tests of assumptions.
You must use the data that is posted on the blackboard site. This data set reports the results from a US survey relating to attitudes towards social media and privacy that was carried out in 2013.
General Requirements
Any evidence of collusion in writing the answers, or plagiarism of work from other sources will result in loss of marks or other academic sanctions, although you are
welcome to cite the analyses that are in the three reports that have already been published on the survey (provided on blackboard), as long as references and citations to that work are explicit and clear. It is ok for you to cite analyses and results from any materials provided to you. However, you should also carry out your own analyses and you should always indicate in your writeup which results are yours, and which results you are referring to as the work of others.
DO NOT include statistical printouts in the body of your report, but instead summarize the statistical results within the text. When reporting the results of F-tests and other statistical analyses in the body of your report, please use the wording and style recommended by Andy Field (APA style).
Where possible, summarize your findings in terms of readable and informative tables and figures. Generally, you will run many analyses, only some of which may end up being useful or part of your report. Put samples of key statistical printouts in an appendix, but limit these statistical printouts to no more than 10 pages (re-formatted or edited, as appropriate). At the start of your analysis it is recommended that you explore and visualize the data as much as possible so that you understand how it is structured and so that you have some intuitions about the data that will help to guide your analyses.
Please create the entire report document (including appendices) as a pdf file. You will be given further instructions later on how to submit your report.
Report Structure
The structure of the report is largely implied in the grading scheme above (with the exception of the issue of persuasiveness, which should permeate the entire report). In order to simplify the grading task for the instructor, the following outline of four main sections is recommended.
1. Research Questions and hypotheses
2. Data preparation and Preliminary analysis (this is where you will test assumptions, examine distributions, etc.)
3. Analysis and Results
4. Interpretation and Discussion
Note that this is an incomplete research paper that focuses on statistical analysis. If you are working in a one person group (by yourself) you should not do a literature review nor should you refer to the literature in your discussion. There is no need for a bibliography and no other papers need to be cited for this research paper. You will only need to do a literature review if you are working in a two person group.
Research reports -
1. The privacy attitudes questionnaire (paq): Initial development and validation by Mark H. Chignell, Anabel Quan-Haase, and Jacek Gwizdka
2. "Social Media and the Cost of Caring." by Hampton, K.N., Rainie, L., Lu, W., Shin, I., & Purcell, K.
3. "Social Media and the ‘Spiral of Silence.' By Hampton, K.N., Rainie, L., Lu, W., Dwyer, M., Shin, I., & Purcell, K.
4. "Social Media Update 2013" - Maeve Duggan and Aaron Smith
5. Derivation and Evaluation of the Short-Form Communication Preferences Inventory by Lottridge, D., Baran, M., Rodrigues, S., Chignell, M.
Attachment:- Pervasive Connectivity - Questionnaire.rar