SOCS0100 Computational Tools for Reproducible Social Science

Second Summative Assignment

Guidelines for Completing and Submitting SOCS0100 Assignment:

• This assessment is due on 13 January 2025, 1pm and shall be submitted on Moodle.

• Late submission results in penalties. There is no exception to late submission penalties, unless an extenuating circumstances application has been successfully made. Please see the details here.

• You are expected to submit a compressed (zipped) folder: in the folder, you should include the coversheet, the main body of your report (html) (i.e., answers to the questions), any tables, figures, dataset, and integrated chunks of code you may use in your report, your main code files

(e.g., R scripts, app.R, Quarto docs), and README.md file.

• Word limit of this assignment is  1500. This word count excludes tables, chunks of code, figures, and table and figure legends, references, but includes any footnote or endnote you may use. Exceeding this limit will result in penalties.

• This is an assessed piece of coursework for the SOCS0100 module; collaboration and/or discussion with anyone is strictly prohibited. The rules for plagiarism apply and any cases of suspected plagiarism of published work or the work of classmates will betaken seriously.

• The coursework will be assessed against the criteria set in the UCL UG-ESSAY GRADING SCHEME, a pdf of which could be seen in the assessment submission area of the course on Moodle. In addition to those general guidelines, further specific factors will affect the marks: Correctness of your code, clarity of arguments, rigour in processing, analysing, and presenting the tasks, creativity and novelty in your answers, and the ability to demonstrate that key concepts treated in the module are understood well.

• Please read the below guidelines and AI-usage policy carefully to avoid losing unnecessary    marks.

Assessment PartI

•   The second assessment aims to evaluate your proficiency in applying fundamental computational techniques in automated  data collection and building an interactive dashboard to explore the real-world data you formed.

•   Additionally, you will be required to add an exhaustive overview of the dataset, the significance of the data source in social sciences, and some insights from the interactive dashboard into a structured report while critically engaging with ChatGPT. This assignment is designed to foster both technical skills and critical thinking in using computational tools.

PartI-A Automated Data Collection (30 points)

•   Employ one of the following automated data collection techniques  covered in the module (static web-scraping; dynamic web-scraping; or APIs) in R to comprehensively gather data from a chosen source. In-class exercises can guide you on what sort of web source you can scrape for this part.

•   Methodically document each step of your data collection process, elucidating  the rationale behind every decision.

•   Provide code snippets complemented by explanatory comments, ensuring that your data collection procedures are both transparent and reproducible.

PartI-B Data Exploration and Contextualisation (10 points)

•   Clean and tidy your data for the next steps. Please show your data wrangling process in detail.

•   Provide an exhaustive overview of the dataset you formed, encompassing characteristics such as dimensions, data types, and a comprehensive description of each variable.

•   Clarify the rationale behind your selection of the data source, emphasising its potential significance in social sciences.

Assessment Part II

Part II-A Building an Interactive Dashboard with R Shiny (30 points)

•  Create  an  interactive  dashboard  using  R  Shiny  that  utilises  the  refined  dataset.  Your interactive dashboard should be reproducible through your app.R file. Note that you are not expected to embed the Shiny app into your Quarto document.

• Design and implement a minimum of three interactive visualisations within the R  Shiny dashboard, derived from your dataset.

• Offer clear and concise interpretations of each visualisation, elucidating any emerging trends, patterns, or noteworthy observations in your report (you don’t need to add visualisations into your report).

• Elaborate on the specific visualisations chosen, justifying their selection, and elucidating their contributions to a deeper comprehension of the data source.

Part II-B Reproducibility (10 points)

•   Uphold the principles of reproducibility by  sharing your replication materials in the compressed (zipped) folder, encompassing all pertinent code and a meticulously crafted README.md file. Prioritise the meticulous documentation of your   codebase, rendering it accessible and comprehensible for potential replication.

Part II-C Critical Engagement with AI: ChatGPT (20 points)

You are expected to provide your reflections based on the points below in the report.

• Embrace ChatGPT as a collaborative tool in your computational process, soliciting its code refinement.

• Engage in a critical evaluation of ChatGPT's contributions to your project within a dedicated section of your report

• Analyse the value-added by ChatGPT, highlighting instances where it offered insightful perspectives, refined code, or innovative problem-solving approaches.

• Dilate upon any constraints or challenges that surfaced during interactions with ChatGPT, contributing to a well-rounded assessment.

• Reflect upon how ChatGPT impacted your assessment's trajectory, shaping its outcome and affecting your overall learning experience.

AI-Usage Policy in This Assessment:

In this module/assignment, students are permitted to use only ChatGPT for specific defined processes within the assessment.

This can be utilised to enhance and support the development of specific skills in specific ways, as specified by the module leader and required by the assessment. As per the requirements, for instance, students are asked to use ChatGPT for critically evaluating their code in automated data collection, data processing, and creating interactive dashboards in this assessment. In doing  so,  students  are  expected  to  highlight  instances  where  ChatGPT  offers  insightful solutions, refined code, or worsens student’s solutions and code.

Except critical engagement with ChatGPT through code refinement, this module prohibits all other use of artificial intelligence (AI), including large language models, to author or co-author  formative or  summative work. This prohibition includes the  following practices  and  any practices similar to them:

• Writing parts or all of an assessment;

• Generating outlines, structures and high-level arguments for essays;

• Rewriting or paraphrasing text from other sources for use in written work.

Language and writing review are not prohibited, defined as having a third-party or software check areas of academic writing such as structure, fluency, presentation, grammar, spelling, punctuation,  and  language  translation.  However,  language  review  may  be  considered Academic Misconduct if substantive changes to content have been made by the reviewer or software or at their recommendation, which would suggest that the reviewer or software had either produced or determined the substantive content of the work.

Including content generated by AI tools will not be considered academic misconduct only if it is clearly signposted (by, for example, quotation marks) and attributed (by including a reference to the tool and date of use). However, similarly to quoting Wikipedia, quoting an AI system is unlikely to be a valuable addition to your work and unless clearly relevant to an argument may negatively impact the perceived quality of your work.

Suspected use of AI technologies other than specified one in the assessment may lead students to be subject to an Investigatory Viva.