7 Data skills

7.1 General information

• Data Skills Assignment 1 deadline: November 3rd 2023.

• Data Skills Assignment 2 deadline: November 24th 2023.

• In total, this assessment is worth 15% of your final course grade. Each portfolio submission is worth 7.5% and you will be awarded a mark out of 22.

• There will be separate submission links for each Assignment where you will submit a single .Rmd file to each. You should only submit the .Rmd file.

• The assignment and submission link will be made available at least one week before the deadline.

7.2 Type of assessment/structure

• Over two data skills assignments, we will ask you to complete a series of tasks including general statistical knowledge questions, data wrangling and visualisation tasks, descriptive and inferential statistics tasks, and interpretation questions. Note that this list is indicative and non-exhaustive.

• We will not ask you anything that has not been covered yet in either the lectures or the data skills book. For example, data skills 1 would not expect you to learn material covered in week 7 or later.

• There will be approximately 11 questions in each assignment, with each question being worth 1, 2 or 3 marks to sum to the University of Glasgow 22 point scale of Schedule A.

7.3 Assessment support

• All the coding tasks you will be asked to do will have been covered in the data skills book, and the activities within the book. Theoretical tasks will have been covered in the lectures, labs, and data skills book.

• For coding tasks, it is important to try the code in the book as you go along and ask questions about what you are doing. This will really help you in the assignments.

• You are free to ask questions regarding the assignment on the available forums, in-class, or in student office hours. However, please do not post answers or code that would be close to giving the answer as this would contravene the University policy on plagiarism. Policies regarding how to use the forums can be found on the main Moodle pages. If a member of the team needs to see the code you are trying, they will ask you for it.

• To avoid confusion: you are allowed to post code when asking questions about a task in the quantitative fundamentals data skills book, but you should not post code relating to a question about either of the data skills assignments unless asked to by a member of staff. Again, this is because posting code related to an answer in an assignment could count as plagiarism.

• When submitting the file, make sure that you submit the completed .Rmd file for that assignment only. You must include your GUID in the filename, followed by ‘MSc_RM1’ and the Assignment Number. The file that you are given will have most of this information already but when submitted it should read, for example, ‘9804672_MSc_Rm1_Assignment1.Rmd’.

• Please note that these are individual assessments and not group work. While we encourage students to work together and learn the skills together, copying and passing each other code and answers by any means will be considered plagiarism and can result in you being reported to Senate.

• Further information about the feedback you will receive for this assignment can be found in the Feedback Information Sheet section below.

7.4 How to do well in these assignments

• Completing all formative and summative work on the practical data skills. All the assignments build on each other and information within the formative assignments is needed to complete the summative assignments.

• Use the quantitative fundamentals data skills book activities to guide you. In almost all cases, the answers to the assignments require using skills that you have already covered in these materials.

• Avoid making changes to the .Rmd file other than to provide answers and your GUID (e.g., avoid deleting backticks, changing code chunk names, not using the file provided).

• Try to use the code that we teach in the book.

• Read the question carefully and ensure that you provide exactly what is asked for (e.g., code or a single value). Do not deviate from what the question asks. You are more than encouraged to practice different approaches in your own time, but for the assignments you should only do what the questions ask.

• Follow the instructions for how to fill out the .Rmd file carefully, making sure you do not change anything in the file you should not.

• Please pay special attention if the question asks for a specific output, e.g. value or a tibble/data_frame. For example, when asked for a value as an output, make sure it is a single value and not a value stored in a tibble. Finally, when altering code inside the code blocks, do not re-order, rename, or remove the code blocks (e.g. T01, T02, etc.). If you do, this will impact your grade and may result in you receiving a H grade for this assignment! If you are unsure about any of these points, please ask a member of staff.

• Test that the .Rmd file you are submitting is reproducible after completing it, by "knitting" it as a html file. This also shows what you have accomplished and allows you to spot potential errors. If your code does not knit, this may result in you receiving an H grade for this assignment, so please check carefully.

• A great test of your code is to close R Studio, restart it, open and knit your code, see that it runs, and go through it to make sure it is correct. This will test whether you have remembered to include essential elements, such as libraries, in your code. This does not necessarily mean that your code is correct, only that your code runs. You need to check the output and compare to the question to see that it matches.

• When loading in data, you must use a relative path as opposed to an absolute path. Relative would mean having the data file in the same folder as the .Rmd file and just calling the data file when reading it in, e.g. read_csv("data.csv"). Absolute would be using a path specific to the folder structure of your computer, e.g. read_csv("C:/PhilsComputer/PhilsFolder/data.csv"). Only use relative paths and under no circumstances should you rename the data file from the one we tell you to use. Note that “data.csv” is a different name from “data (1).csv” and different from “data_1.csv”. Only use the name we give you.

• The following functions are considered forbidden and must not appear anywhere in your code at all. Inclusion of them in your code may result in you receiving no marks for the assignment. The functions are: install.packages(), help(), vignette(), setwd(),help.start(), or View().

• Please read the feedback we give you in past assignment to recognise what you need to work on in future assignments.

• Collate your notes and learn from your mistakes. These assignments are weighted low to allow you the space to make mistakes and to not have a major impact on your overall course grade. It is imperative that you take this opportunity.

• Refer back all the time to what you have done previously. The skills stay the same; only the data changes.

7.5 Common Mistakes

• Changing the .Rmd file to something other than providing your answers and your GUID.

• Changing the name of the data file to something other than the one we provide – e.g. “data.csv” and “data (1).csv” are two different files.

• Using code that is very different to that taught in the book. We will check any incorrect answers and if you have used different code that produces what is asked you will still get the mark, however, using different code from the book increases the risk that you will produce a different output to the one intended.

• Failure to follow instructions (e.g., writing code when a single value was requested).

• Including any forbidden code in your .Rmd file, e.g., install.packages() (you should never write code that would change something on someone else’s machine as it causes issues and it is impolite!).

• We always ask students to use the function read_csv() to load in data, yet students often use read.csv(). There is a difference between the output these create, so always do what the questions ask.

• Be very careful with spelling. For example, UK is not the same as Uk.

• Never do more than what the question asks. We check the output of a task to see that it matches what we expect. If it does not match what we expect, then it is wrong. Remember the point is about being reproducible and replicable. Think of this in terms of you are sending your work to a colleague who expects a certain thing, and if your code works differently, then that is not good practice.

• Try things out in the console and in your own scripts, but only put your final answers in the .Rmd.

• Failing to read your feedback from previous assignments.

• Failing to check that your code knits before submitting it.

• Failing to check that you are submitting the right file. Only one file is allowed per submission.

7.6 How is the assessment related to the lectures for this course?

These practical data assignments assess your ability to wrangle and visualise data in an open and reproducible way, to interpret and work with data, to understand analysis and research methods, and to present studies in a professional manner, as discussed in the lecture series. While these assignments are of course most closely related to the Research Methods 1 lectures, understanding data and analysis will help with your critical evaluation of research in all fields of Psychology. By understanding where the results come from, it will give you a much deeper approach to critically assessing the claims of research.

7.7 Why am I being assessed like this?

The practical skills taught through these assignments are critical for psychological researchers to develop to understand analysis, methods, and research. Your skills will progress throughout your degree, but the best approach to developing these skills is through regular continuous assessment. Much like learning a language, more practice over a longer period of time will give you better results than cramming at the last minute. Continual regular practice is better than one big chunk. As such, these assignments ensure that you are maintaining a steady and solid rate of progress throughout the course.

7.8 How does this relate to previous work I have completed?

• The individual and generic feedback from previous assignments, and solutions for these assignments will help you complete these assignments. In addition, the data skills book activities will really help you in completing these assignments. You should not overlook the importance of these formative tasks.

• Do not forget that asking questions on the forums, dropping into GTA session, and speaking with the teaching team in labs is also giving you feedback and feedforward. All these sources of support will help you complete this assessment.

7.9 Academic Integrity

Please note that when submitting your work for assessment we accept it on the understanding that it is your own effort and work and unique to the set assignment.

To support you in understanding what plagiarism is and in avoiding it, please read the following resources that the University provides:

• SRC Advice and Support

• Code of Student Conduct and Plagiarism Statement

• Avoiding plagiarism and engage in good academic practice (a Moodle course you can self-enrol in)

• Academic Advice in MVLS

University statement on AI: The advent of free AI tools is transforming our world and offers many opportunities to help us deal with large amounts of information. While we support students in learning how to use these tools to help them study, you should not submit for assessment something generated by an AI tool as though it is your own work. Please carefully read the University’s position on AI.

Statement on groupwork: We encourage students to form a study group and peer feedback groups. However, this assignment is not a group work assignment, so your work must be your own individual contribution. If you make a study group or a peer review group, avoid sharing final drafts or near final drafts of your work.

7.10 Feedback information

7.10.1 How is this assessment graded?

Each summative practical data skills assignment will have 11 questions with each question being worth either 1, 2, or 3 points. Your grade will be converted into an alphanumeric grade on the standard 22-point scale (e.g., a score of 16 = B2).

Please note that it is imperative to the grading that you do not amend any of the code chunk names in the submission file. If you are unsure what these are then please ask. If you amend the code chunk names then your code will fail the first step in being reproducible and you will receive an H.

7.10.2 How will the feedback from this assessment help me in the future?

The practical data skills assignments will help you develop your transferable data skills to complete future work that requires working with data, such as your dissertation, and courses or jobs that require such skills. Additionally, as you will see, these skills can be used for tasks such as writing reports, creating reference sections, conducting text analyses and building websites, as well as a wealth of other tasks and is therefore an extremely useful transferable skill. These skills also give you a much deeper understanding of data and research.

7.10.3 What type of feedback will I receive for this assessment?

For these assignments, you will get a feedback sheet for each assignment that tells you whether you received full, partial, or no credit for each task/question within the assignment. You will receive individual feedback on each task within an assignment, telling you where your answer was acceptable or telling you what was incorrect about your answer. You will also see the correct answer to each task within an assignment along with a short piece of generic feedback for that task. We will also make a solution to the assignment available that you should compare to your own work to see alternative approaches, or to see where you have gone wrong, and to see what you have done correctly.

7.10.4 Who assessed my work?

The assignments will be graded by an experienced member of staff using computer-assisted marking. Note that this is not automated marking as the member of staff has the final say on all submitted answers and checks the process at every step. In the first instance, we compare the output of your answer to the expected output or the code used to the code requested. If these comparisons match then you are given full credit for the task. If your answer does not match what was asked for in the task, then the member of staff manually checks your output to the expected output, or your code to the expected code, and to the question asked, to see if your code/output still answers the question or not. A decision is then made on whether to award full, partial, or no credit for that task.

7.10.5 Can I get more feedback?

Yes! We encourage you discuss your assessment first with the marker or course lead (regardless of what grade you received) and you can do so either by messaging the marker or course lead directly to arrange an appointment. You can also speak to the GTA team about your coding skills for further development. All staff contact details can be found on the University of Glasgow directory.

7.10.6 Can I have my work regraded?

Further feedback meetings with the person who marked your assignment is purely about additional information to help you improve and is not about changing your grade or having your work regraded. That said, even if you are unhappy with your grade, your first point of contact should be to arrange an additional feedback meeting with your marker for further discussion to help explain your feedback and grade. Following this, if you still have concerns you should consult the guidance from the SRC which provides a clear explanation of the University appeals procedures.