Contents

Minute Course description.

Statistics can be frightening but you really should not worry for this course. This course is built around applying techniques rather than deriving the mathematical rationale behind each technique. By the end of this course you will understand the key assumptions to commonly applied methods in psychology. You will also be able to run those techniques and interpret the outputs of those techniques in R. This course will also lay the foundation for your work on programming experiments as well as you thesis. This course is, however, not a programming course or a maths course, so no need to panic! More information can be found on the elearning portal. Note that this course previously went by PY0782, so don’t be surprised if you see that module code pop up. This course is one component of PY0794. The 2nd component, Qualitative Research methods runs in semester 1 and will have its own manual - it is led by Dr. Stephen Dunne.

Course Load.

This section of the course counts for 20 credits. 1 credit corresponds with around 10 hours learning time. Given that we have 2 contact hours, you should count to spend up to 8 hours each week working towards this course. This means that outside of the contact hours, you should count to be around a day full-time engaged with this course.

Course objectives.

There are four key course objectives.

  1. KU3: Formulate balanced judgements with regard to complex, incomplete, ambiguous or sensitive data.
  2. KU 4 - Contribute to the creation of new knowledge and practical applications within the discipline through a critical understanding of the processes through which knowledge is created
  3. IPSA 2 - Use a variety of techniques, advanced research methods and technological skills applicable to psychological enquiry
  4. PVA 1 - Apply relevant ethical, legal and professional practice frameworks (e.g., BPS), and maintain appropriate professional boundaries.

In addition, you will learn:

  • Conducting common statistical tests in R.
  • Understanding key assumptions to statistical tests.
  • Applying statistical tests and interpreting them.
  • Summarising them in tables.
  • Make (beautiful) statistical reports.

Course work and assessment.

  • Detailed information on assessment can be found in the electronic learning platform.
  • You will write two assignments for the Quant. Part of the module. These assignments are logical extensions of what we have done in class. If you have done the exercises, at the end of each class you should have no problem with the assignments.
  • Each assignment counts for 30% of your total mark (2x30% = 60%). The Qual. Part accounts for the remaining 40%. The guidelines are provided on the elearning portal.
  • Style and presentation are important. Please closely follow the guidelines.
  • At the end of each session, you should apply what you have learnt in your assignment. Do not leave it to the last weeks!
  • For fairness and transparency, I will only answer questions on the assignments via the elearning portal. When posting a question make sure it is clearly phrased. Ensure that the problem is not due to your machine, i.e. ensure that you have run your code on a different machine as well. Contrary to popular belief there are bad questions, rather bad ways in which to ask for help. When asking a question on the forum, clearly indicate what you have already done to tackle the problem. For example, include a link to a post with a similar problem to indicate that you have searched for a solution, but that your problem still remains. I also encourage you to find help out there in the real world (e.g., post a question on www.stats.stackexchange.com). Also, make sure your problem is reproducible (indicate which packages are loaded, which system you are on, etc.). I expect you to have worked through the exercise corresponding to the lecture before asking the question. If you have not attempted to complete the exercise, then my first response to your question will be, have you tried working through the exercise? Note that when I answer, I might not give you the direct answer as then you would just copy that, rather I will point to the resources which will likely allow you to resolve your issue. I hope you understand, that there is little value in me just giving you the code, rather than you working through the solution and finding out why your solution did not work. There might be some issues which you cannot resolve, even after help, you should then consider moving on to the next item in the assignment, after you have given it your all. There are plenty of items on the list and you can get a very good grade, even if you are unable to solve a particular question.
  • If you run into issues please post them on the discussion board as **early as you can, I might not be able to reply if you leave it right up to the deadline.** You should count on 2, max. 3 working days for an initial reply in the forum (if I am in office).

Attendance.

  • Attendance is required as the class is run in workshop format. The knowledge you gain is cumulative, and therefore it is important that you do not miss sessions, as future classes depend on what you have previously learned.
  • The assignments are tied in with every class, therefore attendance will highly correlate with your study success.
  • Please email me, if you are going to miss a class. You should work through the materials and exercises. Before asking a question about materials, you should indicate what you have done so far.
  • Further information regarding attendance and policy can be found in the MRes. handbook.
  • Please contact me if you have special requirements for the class and I will do the utmost to accommodate your needs. If you have very specific needs also contact Ask4Help or the handbook to find out which resources are available to you.

Prerequisites.

I assume that you have some basic knowledge on statistical concepts. There is no need to have but it would be really helpful if you revisit your undergraduate statistics notes. I will assume very little prior knowledge take it slow but you should be familiar with some basic mathematical and statistical concepts (e.g., what is a mean, standard deviation, what is a p value, what is a t-test, what is a logarithm, what is an exponential function, if you do not know any of these concepts, re-read your undergraduate or A-level notes on statistics/maths.).

Induction.

You can find materials relating to induction here.

Lectures.

The academic calendar is in your programme handbook. For the Quant. Part of the module in semester 1, we will work through 11 lectures in 12 weeks, notice that there will be an academic development week with no lecture after week 5. Depending on timing the content of this schedule might shift a bit.

  1. 30/1. Introduction. R Studio and R environment. Basics of Markdown.
  2. 6/2. Data visualisation. (+ basic tests)
  3. 13/2. Basic ANOVA + Extensions (ANCOVA / MANOVA)
  4. 20/2. Correlation and multiple regression. Logistic Regression.
  5. 27/2. Moderation analyses.
  6. 12/3. Path analysis and ‘Basic Mediation’ (Sobel test / ‘Preacher & Hayes method’).
  7. 19/3. Exploratory Factor Analysis.
  8. 16/4. Confirmatory Factor Analysis CFA I
  9. 23/4. Confirmatory Factor Analysis II and SEM. Mediation via SEM.
  10. 30/4. Multilevel models I + II
  11. 7/5. Buffer / Revision week / queries.

The handouts for every course lecture will be available via the electronic learning portal. There is no need to take extensive notes, as all the key information will be on the sheets.

Every lecture will have some additional exercise(s) at the end, which will prepare you for the assignments. The course is cumulative, so you should really aim to complete the exercise at the end of each week.

Inevitably, there are things which I cannot cover (but which are very interesting) but feel free to ask me about those. If we have time left, I can cover those. If there is sufficient demand I can also looking into hosting a workshop on these techniques. Similarly, I can point you in the right direction should you need those techniques for your project or internship. Bottom line: R can do all these amazing things:

  • Bayesian statistics / Equivalence testing.
  • Statistical simulation. Permutation testing.
  • Time series analysis. Cox Regression / Growth curve modelling.
  • Text mining / emoji analysis
  • Non-linear modelling / Ordinal logistic models / Probit models / curve fitting.
  • Power analysis.
  • Cluster analysis.
  • Item response theory.
  • Machine learning / Data mining.
  • Graph theory / Directed acyclic graphs / Social network analysis.
  • Generate art
  • Meta-analysis but see here.

Similarly, I will not cover other software packages (e.g., AMOS, MLWin, Gephi, JASP) but again feel free to ask me about those.

Bring your PC / Macbook to class!

  • Rstudio is available on the PCs but you will benefit if you have it on your on machine! I would encourage you to install this before the first class.
  • Install RStudio desktop and R. More detailed How to and/or step 1 and 2 from here
  • I expect you to turn off email, messaging, etc. and only be engaged with classroom materials. (#no9gag)

Contact.

  • Discussion board. This is the preferred method, as several of your colleagues will have either the same or very similar queries!
  • Ask me any comprehension questions during class breaks or send me an email. Note that I might refer you to discussion board.
  • Additional 1 to 1 meetings are by appointment only (email). I am in ‘CoCo’ which is in the Northumberland Building (NB165) but meetings might also be virtual. These meetings are by exception only as you should raise any queries in class or via the discussion board.
  • email: thomas.pollet@northumbria.ac.uk
  • twitter: @tvpollet.
  • Phone details are on the learning platform.

I typically reply within max. 3 business days (unless you receive an out of office). Note that I will have limited availability during the christmas break. So please post questions early on the blackboard

Resources.

Check the RStudio website.

Give swirl a go!

I have compiled a (non-definitive) reading list. The information on my personal website is more useful.

Books and other stuff:

I will post additional resources via the elearning portal. So watch that space. Also see the above mentioned reading list.

You can view some videos on youtube on R, including by yours truly. Start here

Blogs to follow.

People to follow on Twitter / Mastodon / Bluesky.

  • Nathan Yau
  • Hadley Wickham
  • Edward Tufte
  • Jenny Bryan
  • Mara Averick
  • Danielle Navarro
  • Julia Silge

Changes made in response to student feedback.

I take on board relevant study feedback. Below I only list the most meaningful changes.

2023-2024

  • Semester 2 module. Now preceded by programming course.
  • Short video(s) on theming
  • Revised and updated course content

2022-2023

  • Shorter videos on Rstudio and R to be released.
  • Move in time table.
  • Revised and updated some course content.

2021-2022

  • Positive reception of worksheets and ‘flipped classroom’, so implemented for this year.
  • Positive feedback on assistants who will continue to be available. They also wrote ‘cheat sheets’ which will be available.
  • Revised and updated some course content.

2020-2021

  • Supply of additional materials
  • two assistants now available for more sessions.
  • Revised and updated course content. Checked functionality on R 4.0+

2019-2020

  • Two assistants are available to aid with some exercises.
  • RStudio + R available in Library commons.
  • Revised and updated course content