代做Summative Assignment on SOST30022 – Network Analysis调试SPSS
- 首页 >> Matlab编程Summative Assignment on SOST30022 - Network Analysis
The essay comprises 100% of the mark for this course.
For this assessment, you are expected to write an up-to-2000-words essay. Failure to follow this requirement will result in immediate deduction of the mark.
The essay is to present your analysis of one of the provided network datasets (including the network and node attributes) provided on Blackboard in the Assessment section. You are to choose one of the datasets.
The essay must have a title, be split into logical sections featuring proper Introduction and Conclusion. The essay should be a coherent and continuous narrative describing your analysis and interpretation of the provided data. Sections should build on each other consecutively. For instance, selection of model effects should base on prior interpretation of network statistics and visualization.
Results of calculations and modelling should be properly formatted as tables according to the conventions of academic publications. Software outputs must not be copy-pasted into the essay without proper formatting andre-structuring. Most interesting results should then be presented in the textual narrative, discussed, and reflected upon with regard to the specific research context.
To get a better idea of what scientific text looks like, it is advisable to look at the papers in the journalSocial Networksand follow along the approaches to justification and reflection on the results presented there.
Importantly, you are not expected to describe the methods of analysis (e.g., what is social network analysis, how network measures are calculated and what they in general mean, or what ERGMs are). This will not add to your mark but will count towards the word limit. Rather, you are to concentrate on justification/backup of your analytical choices (including the effects to in include in the ERGM), description of the results, and thoughtful interpretation of the results.
While working on your analysis and essay, you are advised to use lecture slides, video tutorial materials, software manuals, and readings – all provided on the ‘Course content’ Blackboard page of the course.
You are welcome to consult the examples of essays that received high and low marks, provided on Blackboard in the Assessment section.
The essay must (complete absence of any of these elements or poor performance on any of them will significantly lower the mark) include:
• A brief presentation of the network at hand: what are the nodes and links, what is the research context, why it is important to apply network analysis in this particular context.
• A table with key descriptive statistics on the network calculated in UCINET, including density, triadic clustering, degree centralization, average distance, as well as average, highest and lowest node centralities (by degree, closeness, betweenness, and eigenvector). You should not provide centrality calculations on each of the nodes separately.
• A network graph visualization (counted as 300 words; can precede network statistics in the essay structure) done in UCINET, including the attributes you find relevant. Other figures (not tables added as images) are allowed and are not included in the word count. Graph visualisation functions (layout type, nodes and lines sizes/colours/shapes) should be used extensively. You are to clearly justify these choices with regard to importance of the features of the particular dataset in the specific research context (from your point of view). E.g., why node size was chosen to display age of the nodes while node colour - for their group belonging.
• An interpretation of the visualization and descriptive statistics, summarising what they tellus about properties of the network (e.g., sparse/dense, centralized/decentralized, clustered/not clustered, and soon) in the particular research context (e.g., what are the consequences for a network of street gangs being sparse?
Decentralised? Containing a lot of closed triplets? What does it tellus about capacities and limitations of these groups, their leadership, information transfer in them, and so on?). This interpretation should prepare further selection of network configurations to model, i.e., spotting the patterns in network structure that might signal the network mechanisms that can be examined further using ERG modelling.
• A table with statistics (parameter, SD,t-ratio, SACF) of a converged Exponential
Random Graph model estimated in MPNet, along with goodness of fittest to it in the appendix (GoF table does not count towards the word limit). You should present one, most reliable, model from among those containing a greater number of significant effects that make sense regarding previous analyses, conceptual considerations, research context and peculiarities of the data. The model must include parameters with attributes. Significant parameters must be indicated. You can check significance at p<0.05 or distinguish between p<0.1, <0.05, <0.01, as described in the video tutorial and during classes. Selection of model effects must be justified conceptually and empirically - with regard to data, research setting, literature, and previous descriptive analyses. The model must include effects featuring node attributes.
• An interpretation of the presented model, i.e., presentation of significant and
insignificant parameters, taking into account parameter signs and explaining these results, each on their own and in combinations. Interpretation should speak about the mechanisms underlying the formation of this network based on the model parameters; like before, interpretation must take into account the research context. For instance, which mechanism of network formation does positive and significant ATA (alternating triangles) parameter indicate when we model a network of friendships among monks? If this mechanism is operative, what are the consequences for the structuring of this network?
• Contextualization of analyses and interpretation with regard to the substantial literature on the research context (ideally, network analyses in this/similar research setting) is optional and its inclusion will increase the mark.
Evaluation criteria:
• Structure and logic of narration, coherence of the steps of analysis and interpretation.
• Understanding of key network-analytical notions, procedures, measures, and network configurations.
• Quality of justification of visualization choices with regard to representation of important features of the data.
• Depth and correctness of interpretation of visualization given the specific research context.
• Correctness and completeness of network descriptive statistics calculation.
• Depth and correctness of interpretation of network statistics given the specific research context.
• Quality of justification of effects inclusion in the model with regard to key features of the data and taking into account previous analyses.
• Quality of modelling (including Goodness ofFit).
• Depth and correctness of interpretation of modelling results given the specific research context.
• Comprehensiveness and formatting quality of figures and tables.
• Links to the literature on the research context (add to mark but do not subtract from it).
• Extra analytical/interpretation steps, well-conducted and reasonable – may add to the mark, but only in case mandatory requirements are met.
• Additional strengths/weaknesses of the essay not listed above.
Administrative arrangements:
• Essay should have a title and an accurate word count included on the frontpage. Failure to do so will lead to an automatic 2-mark deduction. Your word count should include all text in the essay (including any footnotes, tables and soon) but not the bibliography and the appendices.
• Coursework must be typed, double-spaced in a reasonable font (e.g., 12 pt, Times New Roman or Arial).
• You must submit your essay (in pdf format) by the deadline date and time via TurnItIn.
• Full details of how to submit online are available in the ‘Submission of Coursework’ folder in the relevant section on the course Blackboard website.
• Use the submission title of ‘student ID + title’ and otherwise anonymise the assignment.
• If you have technical or organizing questions on submission, appeals, resits or similar issues, you are expected to contact student administration at ugsoss@manchester.ac.uk. Note that teachers cannot alter marks after announcement, but you can appeal to the corresponding committee, asking student administration about it. Inquiries about submission technical issues, appeals, resits, or mark alteration will not be responded to by the teachers.
• You must keep a copy of your submission receipt until all work on this course is complete and you have received your final grades.
• Note that our online submission system includes ‘TurnItIn’ plagiarism detection software. Be sure that you fully understand what plagiarism is and what are its consequences; further details are included in Section 9 of the course description on the Blackboard. If, after reading the guidance, you are at all unsure about what counts as plagiarism, you should contact your Academic Advisor to discuss it.
• Contentwise questions on the assignment must be asked during drop-in sessions, during classes, during office hours, and on theDiscussion boardof the course on Blackboard.