Psychology 105

Advanced Research Methods

Spring 2021


Syllabus for Spring 2021

Instructor:

Jack L. Vevea (psyc105ucm@gmail.com)
(Please note that this is a special email address for this class; we will not monitor it after the conclusion of the class. My regular email is jvevea@ucmerced.edu.)
Social Science and Management Building (SSM) 306 A
Office hours: Wednesdays 2:00-4:00, or by appointment.
Telephone: (209) 658-1706 (although email is usually a more effective way to contact me)

Teaching Assistant:

Kyle Hamilton (psyc105ucm@gmail.com)
Office hours: Thursdays after class.

Textbook and equipment:

No text is required for this class. It will be useful for you to have access to an introductory statistics text. If you no longer have one, contact me. You will need a simple calculator with basic mathematical functions like logs and square roots. We will be holding class on line, so I assume that you will have access to a computer during class.
We will be making heavy use of a statistical program called R in this class. Here is a link to a guide that covers some important R basics.

Meeting times:

We will meet Tuesdays and Thursdays from 11:00 A.M. to 12:15 A.M. at a Zoom link that has been distributed via Catcourses.

Course description:

Psychology 105 will focus on description and inference in the context of the general linear model. We will approach this subject from both the frequentist and the Bayesian perspectives.

Course learning goals:

In the class, you will:
  • learn about the basics of the statistical program R;
  • review simple descriptive and inferential statistical techniques;
  • review the concepts of random variables and probability;
  • review the concept of the sampling distribution;
  • gain a deeper understanding of correlation and linear regression with one predictor;
  • learn enough about regression with multiple predictors to facilitate discussion of linear inferential models;
  • review a conventional approach to the various forms of the analysis of variance (ANOVA);
  • investigate the mathematics that operates behind the scenes in ANOVA;
  • learn about the importance of assumptions in statistical inference;
  • learn about power analysis;
  • learn about computer-intensive methods;
  • learn about Bayesian methods.
  • Course learning outcomes:

    By the end of the class, you will be able to:
  • use graphical methods and descriptive statistics to characterize simple and conditional distributions (demonstrated in homework and exams);
  • apply probabilistic reasoning techniques (rules for combining probabilties, Bayes' theorem) to real problems (demonstrated in homework and exams);
  • demonstrate an understanding of the concept of sampling distributions (assessed on the first exam);
  • use computer simulation to assess characteristics of probability distributions (demonstrated in homework);
  • use the computer to perform and interpret tests involving means and conditional means, from both frequentist and Bayesian perspectives (demonstrated in homework by production and interpretation of computer analyses; demonstrated on exams by interpretation of computer analyses and by hand computations using partially complete computer output);
  • use the computer to perform confidence intervals about means and differences between means, as well as Bayesian credibility intervals and high density intervals (demonstrated in the same manner as skills involving the corresponding tests);
  • identify and evaluate assumptions relevant to simple tests and confidence intervals (demonstrated in the same manner as the previous two learning outcomes);
  • compute and interpret correlations and simple linear regressions (demonstrated in homework and exams);
  • compute and interpret diagnostic information about the fit and stability of regression models (demonstrated in homework through production and interpretation of computer analyses, and on exams by interpretation of computer output);
  • perform inference and confidence intervals for correlation coefficients, regression parameters, and regression predictions, using both frequentist and Bayesian perspectives (demonstrated in homework by production and interpretation of computer analyses; demonstrated on exams by interpretation of computer analyses, and by hand computations based on partially complete computer output);
  • compute and interpret multiple regression models (demonstrated in homework by production and interpretive discussion of computer analyses; demonstrated on exams by interpretation of computer output);
  • produce and interpret added-variable plots (demonstrated in homework);
  • demonstrate an understanding of the logic behind ANOVA (assessed in homework in which you will compute a simple ANOVA using inefficient formulas that highlight the concepts);
  • compute and interpret pre- and post-hoc contrasts and comparisons (demonstrated in homework and exams);
  • implement manual coding of ANOVA problems using a multiple regression approach (demonstrated in homework and exams);
  • conduct power analyses associated with the statistical inference approaches covered in the class, using both R and specialized software (demonstrated in homework assignments; knowledge of what affects power will be assessed on exams);
  • use computer-intensive methods as an alternative to conventional approaches to statistical inference (demonstrated in homework);
  • perform simple Bayesian analysis using Markov chain Monte Carlo methods (demonstrated in homework).
  • General comments on the purpose of the class:

    While you should not think of this class as a class in statistical computing, we will use statistical software (specifically, R, OpenBugs, and G*Power) frequently throughout the quarter. Ordinarily, students can learn R comfortably from classroom work and posted transcripts. However, some of you may find this introduction useful.

    The overall goal of this course is not to offer a sequential presentation of all the basic statistical techniques you might need for simple analyses of psychological data. Rather, it is to teach the skill of thinking statistically, and to foster a deeper understanding that will enable you to learn and apply new analytic techniques independently.

    Prerequisites:

    Completion of Psychology 10 and Psychology 15 (or equivalent). The Psychology 15 prerequisite may be waived under some circumstances.

    Although the course does not emphasize mathematics, you should know something about the basics of algebra (the ideas of equations and manipulation of variables) and geometry (plotting points on a plane, the equation of a line). If you feel ill-prepared in any of those areas, a quick review might be in order.

    Evaluation:

    Ordinarily for this class, grading is based on a combination of attendance and class participation, written homework, a midterm exam, and a comprehensive final exam. Attendance and participation normally count for 20% of your final grade; homework normally count for 50% of your final grade, and each exam normally counts for 15%. However, because of the logistical difficulties of administering exams remotely, this semester grading will be based entirely on homework and attendance and participation.

    The attendance and participation component will be based on your participation when called on to ask a question. Each student should come to every class with well-defined questions about recent class content. We will periodically choose a random student to ask a question. This will happen at least four times for each student during the semester; your participation grade will be A if you are present with a question prepared on all four occasions, B for three, C for two, D for one, and F if you were never present on a day you were selected. This does not imply that questions from anyone are not welcome at any time; rather, you are encouraged to ask as many questions as you find useful. But you must be present and prepared on the days you are called to get participation credit. Your attendance and participation score will count for 25% of your final grade.

    There will be five homework assignments. Your lowest score will be dropped before grades are calculated. Further, when a homework assignment is returned to you, you may resubmit it for up to 10 days following the date it was returned. All homework scores improved in that way will replace the original score. Your homework score will count for 75% of your final grade.

    Those components make up the final grade in the following manner. First, each component (attendance and participation, homework) gets a grade point value: A+ = 4.3, A = 4.0, A- = 3.7, B+ = 3.3, B = 3.0, B- = 2.7, and so on. The weighted average of the grade points from the two components determines your final grade. The following table shows the mapping of grade point averages to letter grades:

    Grade Point Range Letter Grade
    GPA = 4.15 A+
    3.75 < GPA < 4.15 A
    3.50 < GPA < 3.75 A-
    3.25 < GPA < 3.50 B+
    2.75 < GPA < 3.25 B
    2.50 < GPA < 2.75 B-
    2.25 < GPA < 2.50 C+
    1.75 < GPA < 2.25 C
    1.50 < GPA < 1.75 C-
    0.75 < GPA < 1.50 D
    GPA < 0.75 F

    Academic Integrity

    Students should be familiar with University policies on academic honesty. You will find relevant information here. In the overall context of that policy, the following information is specific to this class:

  • Cooperative work on the computational aspects of homework assignments is strongly encouraged, but you are expected to work independently on discussion and interpretation. The words you submit in your written assignments should be entirely your own.
  • Students with Disabilities

    UC Merced has a variety of services available to accommodate students with disabilities. Information is available here.

    How to submit homework assignments

    You should submit homework assignments through CatCourses. After you navigate to the CatCourses page for this course, click on the "Assignments" button in the list on the left. Then, click on the link to the specific homework assignment you are submitting. Make sure to combine everything into one document because CatCourses will allow only one file. That file must have a .doc, .docx, or .pdf file extension (no .odt, .zip, .pages file extensions). All deadlines are at midnight on the due date, and the CatCourses system will not accept submissions after that time.


    Course Outline

    Review and Introduction to Software

    January 19-21
    Initial class meeting: introduction, using the class web page. Obtaining and using R. R basics: reading in data, simple functions. Graphical methods. Some technical vocabulary. Understanding empirical distributions.
    January 26-28
    Measures of central tendency. Measures of variability. Some subtleties of graphing. R as a tool for describing distributions.
    February 2-4
    Other aspects of shape. Putting it all together: the use of graphics and descriptive statistics to describe distributions.
    Homework One is available (due February 11). You can find a document that contains an example of how you might approach the homework here.

    Probability and Sampling Distributions

    February 9-11
    Review of probability. Random variables and probability distributions. The frequentist approach to understanding probability. The distinction between discrete and continuous random variables. Computer simulation as a tool for understanding probability distributions.
    Bayes' theorem. Statistics: a special kind of random variable. Sampling distributions. Empirical approximations to sampling distributions.

    Models

    February 16-18
    Models and conditional distributions. Introducing simple linear regression.
    February 23-25
    Simple linear regression: inference and assumptions. The decomposition of the sum of squares.
    Homework Two is available (due March 7).
    March 2-4
    Regression diagnostics. Wrapping up simple linear regression.
    March 9-11
    Review of the class so far; catching up.
    Homework Three is available (due March 18). You can find a document that contains an example of how you might approach the homework here.

    Estimation

    March 16-18
    Principles of estimation. The likelihood. Maximum likelihood estimation.
    March 23-25
    No class: Spring break. Please stay safe!
    March 30 - April 1
    Bayesian estimation and inference. Markov chain Monte Carlo; OpenBugs.

    Common Inferential Techniques and the Linear Model

    April 6-8
    Conditional means with a simple binary predictor. Effect sizes and confidence intervals. The t test as a linear model: dummy coding, effects coding, nonsense coding. Comparing means, a Bayesian approach.

    Homework Four is available (due April 20). Note that this deadline is five days later than normal; that's because this is a somewhat longer than normal assignment. Get an early start.

    Multiple Regression and ANOVA

    April 13-15
    An introduction to multiple linear regression.
    April 20-22
    Analysis of variance (ANOVA): conceptual approach. ANOVA and the linear model.
    April 27-29
    Contrasts and comparisons.
    Homework Five is available (due May 6).

    Power Analysis

    May 4-6
    Power analysis for one- and two-sample tests. Power analysis for ANOVA.