Assignment guidance and marking scheme
This is your summative assessment project based on the two elements of the course: environmental analysis (JPS) and environmental modelling (NK). Please submit one project in two parts (format is up to you: Word, a PDF or a R notebook file).
NOTE: if you select the latter please work through the documents to make sure the code generates the output. Each code clip must be self-contained. Parameters, data etc are not passed down the document like a R script.
PART A: Environmental analysis (2000 words equivalent) [Marking scheme for Part A is appended below].
Use one of three datasets listed in the project datasets on the CANVAS pages [more on this later in the module]. You can also use a dataset that is relevant to you. We can discuss what this might be. There may be an opportunity to use other REM datasets but please make sure it is not the same data that you are analysing as another module. You cannot gain credit twice for the same piece of work!
Your task is to develop, analyse and validate an appropriate statistical model for the data you have selected. Your project will include evidence of extensive data exploration, hypothesis testing, model development, interpretation and validation.
PART B: Environmental Modelling (2000 words equivalent) [Marking scheme – below; scroll down)
Develop and investigate a mechanistic model within a specific area relevant to the broad scope of the REM masters programme. You may develop a new model or modify any mechanistic model presented within the practicals. Subsequently, you should present either: (1) a calibration and evaluation of the model performance, or (2) a sensitivity analysis and/or Monte Carlo analysis of the developed model.
This is a methods module so we are not expecting a project write up on an ecological or modelling problem. Instead, present the R code from your statistical and mechanistic models and an analysis/interpretation of the results. The R code must be fully annotated so as to be clearly and easily understandable to a reader who is moderately proficient in R but who has no knowledge of the work you are presenting. Integrate appropriate figures and tables and code snippets into each part of the project to show a narrative of how and why you did what you did, what the models mean (interpret them) and how and why you validated them.
PLEASE upload your R script and associated datafiles as appendices to the project. We will run the script to ensure it works!
We will evaluate your projects on the basis of:
(i) Your justification of decisions you took on what analytical/modelling routes you used and why you selected them, (ii) the interpretation of the model results, (iii) model validation, and (iv) quality of your R code. Use supporting reading where available. Remember 10% of the grade will be allocated on the basis of the quality and annotation of your code.
File format: Use your university ID as the first element of the filenames for the Word, script and data files.
EasyDue™ 支持PayPal, AliPay, WechatPay, Taobao等各种付款方式!
E-mail: firstname.lastname@example.org 微信:easydue