Welcome

Welcome!

  • This is Psychology 2812B, “Statistics for Psychology II”
  • My name is Paul Gribble
  • I’m a Professor in the Dept. Psychology and also in the Dept. Physiology/Pharmacology
  • My research is on neural control of voluntary movement and motor learning in humans — https://gribblelab.org
  • I don’t hate statistics!
    • I hope after this course neither will you

Administrivia

  • Lectures are once per week (check Student Centre for day/time/location)
  • Labs are each week (check the OWL site for your section’s day/time/location)
  • Course resources and weekly schedule are on the course website:

Grades

  • 20% Weekly homework assignments
    • 10 homeworks are assigned (2% each)
  • 40% Midterm Exam
    • in regular class time (Feb 28)
  • 40% Final Exam
    • date/time/location TBA

Textbook

  • You do not have to buy a textbook
  • We will be using free books and resources on the internet
  • Assigned readings will be posted on the course website

R/RStudio books

Statistics books

Answering questions with data

by Matthew J. C. Crump, Danielle J. Navarro, & Jeffrey Suzuki

https://crumplab.com/statistics/

Statistics books

Learning Statistics with R

by Danielle Navarro

https://learningstatisticswithr.com

Laptop

  • You should bring a laptop to lecture
  • You should bring a laptop to labs
  • You will need to install on your computer:
    1. R
    2. RStudio

R

RStudio

  • RStudio
    • integrated development environment (IDE) for R
    • free & open-source
    • available for macOS, Windows, GNU/Linux
    • download here

Lectures

  • Lectures are where I present the logic & rationale of the weekly topic
  • I may (or may not) show an example in RStudio
  • You will get more out of class if you complete the assigned readings before the lecture

Homework

  • homework assignment is based on weekly topic
  • you will write code in RStudio
  • what you need to do to be prepared for the homeworks:
    • complete the assigned readings
    • try out the example code in readings
    • attend lectures
    • attend labs

Late Homework

  • Homework is due at 5:00 pm on Fridays
  • A homework solution is posted on the course webpage the following Monday at 12:00 pm
  • For each 24-hour period (or portion thereof) that your homework is late until Monday at 12:00 pm, it will incur a penalty of 0.5% (each homework is worth 2%)
  • If your homework has not been completed by the time the sample solution is posted (Monday at 12:00 pm), it will receive a score of 0

Labs

  • Each homework assignment will have two components
    1. Lab Component to be done together with the TA in the lab session
    2. Home Component to be done on your own
  • Both components are to be handed in and graded
  • Lab attendance is optional
    • but don’t come to the TA or Professor for help on the homework if you chose to skip the lab

Labs/Homeworks

  • in lecture and in your readings we will focus on the logic & rationale of the statistical tests and a high-level overview of how to use RStudio to perform them
  • in the labs and homeworks you will learn the hands-on details of how to use RStudio to perform the tests & procedures we discuss in class

Exams

  • some multiple choice
  • some short-answer
  • mainly organized around:
    • logic and interpretation of data & data visualizations
    • logic, meaning, interpretation and procecures for statistical tests
    • written interpretation of findings
  • I won’t ask you to generate R code on the exams

Exams

  • The midterm exam will be in-class during regularly scheduled lecture time
  • It will be composed of short answer & multiple choice questions, involving definitions of terms, understanding concepts, and analysing and interpreting data
  • I will post sample questions on the course website

Exams

  • You don’t have to memorize equations
  • You do have to be familiar with any equations we discuss in class and that are discussed in the mandatory readings
  • “familiar with” means knowing how to use the equation to calculate the thing of interest

Exams

  • closed book, no laptop, no calculator
  • pencil/pen and paper
  • one double-sided “cheat sheet” is permitted, standard letter sized paper (8.5 x 11”)
    • but it must be handwritten not typed

Syllabus

  • Full details about the course policies & procedures can be found in the course syllabus
  • syllabus can be found on OWL and on the course website
  • read the syllabus

Course Goals

  • understand the logic & rationale behind statistical tests
  • learn some common statistical tests & procedures
    • common thread: linear models of data
  • learn to use RStudio for:
    • data analysis & visualization
    • statistical tests
    • writing reports

A Pep Talk

  • statistics has a bad reputation
  • courses can be boring, stressful, confusing
  • often focused on rote memorization of recipes & procedures
  • you end up with little undertsanding of the how & why

Our Goals

  • lectures
    • establish an understanding of the rationale and procedure for various statistical tests
  • labs & homework assignments
    • you will build up your own repertoire of statistical approaches using RStudio
  • start simple and build on each subsequent idea
    • requires you to keep up with the course material!

What is Statistics?

  • statistic is not math (it makes use of some math)
  • statistics is not calculation (calculations are performed)
  • statistics is not (just) different ways of describing data
  • statistics is a logical framework for interpreting data
  • statistics helps you answer the question:
    What do my data mean?

What is Statistics?

  • We’re not really (usually) interested in our dataset on its own
  • We are actually interested in what our dataset says about how the world works
  • we use inferential statistics to make conclusions about how the world works (poplulation) based on our data (sample)

Elements of a statistical framework

  • fundamentals of probability (previous courses)
  • sampling theory (previous courses)
  • Null hypothesis significance testing (previous courses and this course)
  • Linear models of data (this course)
  • Competency with modern tools for analysis & visualization (this course: RStudio)
  • Bayesian approaches (future courses)

Our approach in this course

  • linear models of data to
    describe phenomena
  • null hypothesis significance testing for
    inferences about the world (population)
    based on our data (sample)
  • RStudio as a modern tool to do both of the above

Course Webpage

https://gribblelab.org/2812

  • assigned readings
  • lecture slides
  • homework assignments

Psychology 2812B FW23