Syllabus

PSY 652 · Methods of Research in Psychology I · Colorado State University · Fall 2026

Note

Download a copy: PSY 652 Syllabus — a single self-contained file that keeps this page’s formatting and opens in any browser, online or off. The same syllabus is posted on Canvas.

Course information

PSY 652: Methods of Research in Psychology I

Semester Fall 2026
Credits 4
Course format In person, with an asynchronous online component
Prerequisites Graduate standing in Psychology or a related department. You should be comfortable with basic algebra and complete the prerequisite Module 1 (M01) before the first pre-study of the semester.
Course website https://psy652.colostate.edu

Meeting times and location

Day Time What Where
Monday 10:00–11:30 am Lecture Behavioral Sciences Building 357
Wednesday 3:30–4:45 pm Lab Behavioral Sciences Building 357
Friday 10:00–11:00 am Pre-study — asynchronous, no synchronous meeting Online

Welcome to PSY 652. This is a course in applied statistical reasoning, but it is also a course in how researchers actually work with data: asking clear questions, documenting decisions, writing reproducible code, checking what you have done, and explaining what the evidence can — and cannot — support.

How the course week runs. Each module follows the same Friday → Monday → Wednesday rhythm. Before Monday, you read the module and work through the pre-study on your own. Monday’s lecture assumes that preparation and pushes the ideas further through examples, discussion, and activities. Wednesday’s lab is where those ideas become practice: you write the code yourself, with help close by when you need it. The Friday hour is reserved for the asynchronous pre-study and, during project weeks, may also be used for team meetings, project work, or occasional scheduled sessions.

What a normal week requires

When What
Before Monday Read the module and work through the Friday pre-study
Monday Lecture — usually built around an activity rather than a talk. Bring your laptop.
Before Wednesday Revisit whatever did not land in lecture
Wednesday Lab — write the code yourself, with help in the room. Bring your laptop.
By Tuesday midnight Render your lab and submit it — before the next lab
During project weeks Make at least one meaningful contribution to your team’s repository and push it to GitHub

That rhythm is the backbone of the course. If you keep up with those steps, you will arrive at each class ready for the work we are doing there — and the material will build in the way it is meant to.

Instructor information

Instructor: Kimberly L. Henry, Ph.D. Department of Psychology & Colorado School of Public Health

Email kim.henry@colostate.edu
Office Behavioral Sciences Building 220
Office hours Monday 11:30 am–12:30 pm (immediately following lecture), and by appointment

Lab Instructor: Allie Kom

Email allie.kom@colostate.edu
Office Behavioral Sciences Building 325
Office hours Tuesday 12:30–1:30 pm

GitHub usernames. For the group projects you will add us both as collaborators on your team repository: KimberlyHenry (instructor) and alliekom (lab instructor).

Communication policy

Email is the best way to reach either of us outside class. Please include “PSY 652” in the subject line so your message is easy to spot. We aim to reply within two business days; messages sent over a weekend may not receive a response until Monday.

Our office hours are placed where they are for a reason. The instructor’s come immediately after Monday’s lecture, while the ideas are still fresh; the lab instructor’s are on Tuesday, the day your lab is due. If a concept is still fuzzy, a project decision feels uncertain, or a notebook simply will not render, bring it with you. You are also welcome to ask for an appointment outside those hours.

If you begin to fall behind, reach out early. The material in this course compounds. A small gap caught in Week 3 is usually easy to repair; the same gap carried quietly into Week 10 is much harder. Asking for help early is part of doing quantitative work well.

Course description

PSY 652 is the first course in the graduate methods sequence. Its goal is to give you a durable foundation for applied data analysis in the behavioral sciences: how to describe and visualize data, reason with probability, understand statistical inference, conduct common hypothesis tests, and build and evaluate linear regression models.

The course is intentionally both conceptual and computational. We will care about getting the number right, but just as much about understanding where it came from, what assumptions support it, and what claim it actually licenses. Along the way, you will become increasingly fluent with R, RStudio, Quarto, Git, and GitHub — not as separate technical skills, but as the tools that make a modern analysis transparent and reproducible.

The course is organized into three Parts:

  • Part 1 · Describing data (M01–M05) — measurement, tools, visualization, wrangling, descriptive statistics
  • Part 2 · Inference (M06–M09) — probability, confidence intervals, the logic of hypothesis testing, common tests
  • Part 3 · Modeling (M10–M12) — simple and multiple linear regression, assumptions and diagnostics

Learning outcomes

Upon completion of this course, students will be able to:

  1. Describe a dataset using appropriate numerical summaries and truthful, readable graphics, and justify the choice of summary for a given variable type.
  2. Transform raw data into a documented, analysis-ready dataset, tracking denominators, missingness, and every derived variable.
  3. Explain probability, sampling variability, and the sampling distribution, and interpret a confidence interval in terms of the procedure that produced it.
  4. Select, conduct, and interpret the null hypothesis significance test that matches a given research design, and report the result in APA style with an effect size and an interval estimate where appropriate.
  5. Fit, interpret, and evaluate simple and multiple linear regression models, including checking assumptions and applying remediation when they fail.
  6. Produce reproducible analyses organized in a documented project repository — project-relative paths, traceable data preparation, code-generated results, and a report another analyst can rebuild from a clean session.
  7. Scope conclusions to what the design and data actually support, naming the specific limitations of a given sample and measurement strategy.

Required materials

  • This course website (https://psy652.colostate.edu) — the home for every module, pre-study, lab, project brief, and course resource.
  • R (version 4.4 or later), RStudio, Quarto, and GitHub Desktop — all free. M02 walks you through the setup step by step; you are not expected to arrive already knowing these tools.
  • Posit Cloud — browser-based RStudio that can serve as a backup if your local installation gives you trouble.
  • A laptop — bring it to both Monday lecture and Wednesday lab, and to the Part B portion of Exams 2 and 3. Lectures are built around activities you run on your own machine, so you will need it every class meeting, not just on lab days.
  • R for Data Science (R4DS), by Hadley Wickham, Mine Çetinkaya-Rundel, and Garrett Grolemund, 2nd edition (2023). It is available free online at https://r4ds.hadley.nz.

Grading

Component Weight
Lab assignments (weekly) 20%
Engagement 10%
Exam 1 — after Part 1 · take home 10%
Exam 2 — after Part 2 · in class 15%
Exam 3 — after Part 3 · in class 15%
Group Project 1 — Pew Data Brief 15%
Group Project 2 — NHST Reproduction 15%
Total 100%

Your final course grade is determined by the components above and the scale below. CSU does not assign grades of C−, D+, or D−. Final percentages are rounded to two decimal places before the scale is applied; no additional rounding is used, so an 89.99 remains a B+.

Grade Range
A+ 100% to 96.67%
A <96.67% to 93.33%
A− <93.33% to 90.0%
B+ <90.0% to 86.67%
B <86.67% to 83.33%
B− <83.33% to 80.0%
C+ <80.0% to 76.67%
C <76.67% to 70.0%
D <70.0% to 60.0%
F <60.0% to 0.0%

Feedback turnaround. You should receive feedback while it can still help with what comes next. Labs and exams will normally be graded and returned within one week, and group projects within two weeks. If circumstances ever make that timeline impossible, we will let you know and give you a revised date.

Keep your own copies. Technology occasionally fails at exactly the wrong moment. Keep a copy of everything you create for the course, including work submitted through Canvas. Most of your labs and projects will also live in Git repositories that you push to GitHub, so you will naturally build a second copy as you work — one practical benefit of the workflow we use.

Exams

There are three exams, one at the end of each Part of the course. Each has the same two components because the course asks you to develop two related abilities: reasoning about a method and carrying that reasoning out in R.

Component What it covers Resources
Part A · written Concepts, interpretation, reading output, choosing an approach and justifying it. Paper and pencil. None. No notes, no website, no devices.
Part B · coding A live R assessment on a WebR page like the ones you use for pre-study. You will run a series of tasks and then answer questions about the results you get. The course website is permitted.

The split is deliberate. Part A asks whether you understand what a method is doing, when it applies, and how to interpret what it produces. Part B asks whether you can turn that reasoning into a working analysis in R. In real research, nobody earns points for remembering the spelling of every function argument; looking things up is part of the work. The harder skill is knowing what to do and why.

No AI assistance is permitted on any exam, in either part. This includes agents, chat assistants, in-editor code completion, and any tool that generates or explains code for you.

Exam 1 · take home

Released Monday, September 21 — the Monday after Module 5 finishes — and due Monday, September 28, one week later. Both parts are completed on your own time.

Exam 1 is deliberately lower-stakes than the two that follow. Its real job is orientation: it lets you see the shape of an exam in this course, experience what Part B feels like, and learn how much time the two parts take while the stakes are still modest. Treat it as a genuine assessment, but also as a rehearsal for the rest of the semester.

How the take-home version works. You will receive both parts as hard copies in class and return them completed one week later. Part A is closed-resource on your honor — no notes, website, or AI — just as it will be for Exams 2 and 3; Part B still permits the course website. You do not need to complete the exam in one sitting, and there is no timer beyond the one-week window.

The work must be entirely your own, and so must the conversation about it. You may not discuss Exam 1 — the questions, your approach, or your answers — with anyone other than Kimberly L. Henry, Ph.D. and Allie Kom. That includes classmates, students who have already finished, study groups, tutors, and online forums, and it applies for the full week the exam is out. If something is unclear, ask one of us; that is exactly what we are here for, and a question asked is always better than a guess.

Exams 2 and 3 · in class

Both run across the two class meetings of their week:

Exam 2 · after Part 2 Exam 3 · after Part 3
Part A · written Monday, October 19 Monday, November 30
Part B · coding Wednesday, October 21 Wednesday, December 2

Part A is the one exception to the bring-your-laptop rule. Bring a pencil and a standard (basic) calculator on Monday; you will not need a laptop. Bring it as usual on Wednesday for Part B.

Labs

Labs are where the course becomes practice. Each is graded out of 10 points on the finished product — the rendered HTML you submit to Canvas.

The scale is deliberately coarse. Every lab tells you what to build, walks you through the major steps, and ends with a Double check list naming the pieces that should be present before you submit. There is no hidden standard to reverse-engineer. If you work through the lab carefully, write the requested interpretations in your own words, render successfully, and submit on time, you should expect full marks. These grades are meant to reward serious practice, not to rank you against your classmates.

Points What it looks like
10 Every step attempted. Every written section from the Double check list is present and in your own words. The notebook renders cleanly with no unresolved errors.
8–9 Substantially complete, with a small gap: one written response is a single line where the lab asked for a few sentences, or one step is missing, or an error is left visible in the output.
6–7 The code is largely there but the writing is not — interpretations skipped or reduced to a phrase. Analysis without interpretation is half the assignment.
4–5 Roughly half the lab, or code with essentially no written work.
1–3 A fragment. Little evidence of engagement with the material.
0 Not submitted, or the file does not render.

A note on that last row. Rendering is part of the assignment, not an administrative step after the analysis is finished. If a notebook will not render, another person cannot reliably read or reproduce it. Render early, render often, and leave yourself enough time to fix the final version before the deadline.

The lab debrief

Every lab ends with a five-minute debrief. We stop coding, look up from our screens, and pull the session together: what was difficult, what helped, which statistical idea the code actually made visible, and where that idea goes next. The prompts are printed at the end of each lab page, so you can glance at them before we begin.

The debrief carries no points of its own and is not part of what you submit. Its purpose is to turn a sequence of commands into an idea you can carry forward.

Stay for it, and take part. You do not need to have the cleverest answer in the room; a useful question, a point of confusion, or an observation about what finally clicked all count. Your engagement in these debriefs is one of the things we consider in the Engagement grade.

Working together is encouraged; submitting together is not. Talk through the labs with classmates as much as you like. The code you submit and the interpretations you write must be your own.

Engagement

This course works best when everyone arrives ready to do something with the material rather than encounter it for the first time in the room. The Module material, paired with the Friday pre-study, gives you the first pass; Monday’s lecture can then spend its time on the ideas that deserve discussion, examples, and a little struggle; Wednesday’s lab assumes those ideas are familiar enough to put into code. Preparation is what lets each part of the week do its own job.

Engagement is worth 10% of your grade. At the end of the semester, the instructor and lab instructor assign this score together as a holistic judgment using the scale below. We are looking for a pattern across the semester, not a perfect record in every single week.

Pre-study activities are not graded separately. They carry no points of their own and nothing is submitted; the questions on those pages are there so you can check your own understanding as you go. Completing them before Monday and coming to class prepared is one of the clearest forms of engagement we can see, and it gives you a much better chance of getting something useful from lecture.

What we are looking for

  • You come to Monday lecture having read the module and completed the Friday pre-study. Not mastered — read and worked through. Lecture is where the hard parts get resolved, and that only works if you arrive knowing which parts were hard.
  • You have real familiarity with the week’s ideas. You do not need to have mastered them, but you should know what the module covered and be able to identify what still feels uncertain.
  • You participate in lecture activities and discussions. Many lectures are built around an activity rather than a long presentation; they work only when people actually try the problem and contribute to the conversation around it.
  • You work constructively with a partner in lab. Most labs ask you to reason through a question with someone else. Being a good partner counts.
  • You ask questions when you have them. A thoughtful question that shows where your reasoning got stuck is every bit as valuable as a confident answer.

This is not a measure of how much you talk. A quiet student who arrives prepared, works thoughtfully with a partner, and speaks when there is something worth saying can earn the highest engagement score. We are evaluating preparation, contribution, and effort — not volume or personality.

Points What it looks like
10 Consistently prepared and engaged. Arrived having done the reading and pre-study, contributed to activities and discussions, and was a reliable partner in lab.
8–9 Generally prepared and engaged, with occasional off weeks.
6–7 Present, but frequently underprepared — attending without having done the reading or pre-study, so unable to engage with the session’s actual work.
4–5 Often absent, or present and consistently disengaged.
1–3 Rarely present, or present with essentially no engagement across the semester.

You will get a status check partway through the semester. After Part 1, we will give each of you a brief sense of where you currently stand on this scale. The point is not to surprise anyone; it is to make sure you have plenty of time to adjust if your preparation or participation has been uneven.

An approved absence is not disengagement. Absences covered by a Student Disability Center accommodation, a University-sanctioned activity, or a documented emergency are not counted against you here. Neither is using an accommodation to participate in the way that works for you.

Group projects

Both project briefs are available from the first day of the semester, and they are worth reading before you need them. Several rubric dimensions reward habits — documenting provenance, keeping a readable Git history, building reproducibly — that cannot be added convincingly the night before a deadline.

You will complete two group projects in teams of two to three. Each serves as a capstone for the part of the course it follows and asks you to put several course skills together around a real analytic problem. Each project produces both a polished Quarto report and a shared GitHub repository that shows how the work was built.

Project 1 · Pew Data Brief Project 2 · NHST Reproduction
Follows Part 1 (M01–M05) Part 2 (M01–M09)
What you do A focused descriptive analysis of a Pew American Trends Panel wave your team chooses Reproduce one published inferential result from a paper with open data
Report due Friday, October 23, 5 PM Friday, December 4, 5 PM
Presentation Week of October 26–30 Week of December 7–11

Each project ends with a 30-minute team presentation — roughly 20 minutes of prepared content and 10 minutes for questions. The aim is not simply to show the final product, but to explain the reasoning behind it: what you chose, what you learned, and what the analysis can and cannot establish.

Both projects are graded out of 100 points using the same six dimensions. Five are team-level. Peer accountability is scored individually, because collaboration should broaden what each person understands rather than divide the project into isolated pieces. Any team member may be asked to explain any part of the analysis during questions, so everyone should understand the whole project even when responsibilities are divided.

Full instructions and the rubric are on the course website:

Attendance

Plan to be in class. Monday lectures and Wednesday labs are where the material is worked through rather than simply read, and neither is recorded. Lab is especially difficult to recreate on your own: it is the time when you write the code yourself with classmates and instructors nearby, so a problem that might take an hour alone can often be resolved in a few minutes.

There is no separate attendance score and no roll taken simply for being in the room. Still, attendance matters in practice. A pattern of unexplained absence will also affect the Engagement grade because participation depends, in part, on being present for the work.

If you need to miss a class, let us know in advance when you can, and connect with a classmate about what you missed before the next meeting. If illness, an emergency, or another circumstance keeps you away for more than one session, contact the instructor so we can make a plan before a temporary disruption becomes a larger gap.

Late work policy

The default course policy is straightforward: late work is not accepted for labs, group projects, or other graded assignments. Labs are due at 11:59 PM on the Tuesday after the lab — the night before the next lab meets — so you have the Wednesday lab session plus almost a full week to finish, render, and submit. A lab submitted after that deadline receives a zero. The two group projects are graded from the state of the team’s GitHub repository at the stated deadline; the project rubric explains how that works in practice.

The policy is firm because the course moves forward every week. Labs are intentionally frequent and relatively low-stakes, so one missed assignment does not determine your grade; carrying unfinished work into the next module, however, can quickly make the course harder than it needs to be. The goal of the deadline is to keep the practice close to the material it is meant to reinforce.

This policy is, of course, subject to University-required exceptions. It does not override approved Student Disability Center accommodations (including approved deadline-extension or attendance-flexibility accommodations), participation in University-sanctioned activities covered by CSU attendance policy, or any other exception CSU requires. If an SDC accommodation applies, your accommodation letter is sufficient documentation; you do not need to explain or justify the underlying circumstances. When possible, let us know early enough that we can plan with you.

Extensions will also be granted without penalty for documented medical or family emergencies. Contact the instructor as promptly as you reasonably can so we can work out what needs to happen next.

Academic integrity

Academic integrity in this course means that the work you submit reflects your own understanding and decisions. Collaboration on pre-study activities and labs is encouraged — talking through a difficult idea with someone else is part of learning it — but the code you submit for a lab and the interpretation you write must be your own. Exams are completed independently.

The course follows CSU’s academic integrity policy and Student Conduct Code. Please review the Academic Integrity — Student Resources page; if you are ever unsure whether a particular kind of help is allowed, ask before you submit the work.

Using AI

On exams: none, in either part. See Exams above.

Everywhere else, the baseline standard is simple: you must be able to explain and defend every line you submit. If you cannot say what a line does, why it is there, and what would change if you removed it, that line is not yet ready to be part of your work — whether it came from a classmate, a forum answer, an author’s published script, or an AI assistant.

That standard is not about detection. It is about learning. The point of the course is not merely to produce code that runs; it is to become someone who can reason through an analysis, debug it, explain it, and decide whether the result makes sense. Code you have not understood cannot yet do that work for you.

Used well, an AI assistant can support that learning. Asking it to explain an error you have already read, clarify an unfamiliar argument, or suggest a cleaner way to express code you wrote can help you become more fluent. Asking it to decide the analysis for you or write the interpretation you are meant to reason through substitutes for the learning rather than supporting it.

In the group projects, this standard is assessed directly. You will put your code on screen and answer questions about the choices behind it. If you use AI, note briefly in the project README.md what you used it for. Appropriate use is not something to hide; being able to explain the work is what matters.

Project 2 adds one requirement to that baseline. In a reproduction, figuring out what analysis a published paper actually requires is itself part of the skill being assessed. For that reason, you are asked to attempt each analytic step yourself before seeking help of any kind — from the authors’ script, a classmate, or a model. Once you have tried, those resources can help you diagnose and improve your reasoning; reading the solution first removes the part of the task that teaches you the most. The Project 2 page explains this expectation in full.

If there is reason to believe the policy has not been followed — for example, AI use on an exam or submitted work that you cannot explain — the matter will be handled under the CSU Academic Integrity Policy and Student Conduct Code. University procedures include notifying you of the concern and giving you an opportunity to respond before an academic penalty is assigned. A finding of academic misconduct may result in a grading penalty, up to and including a zero on the affected assessment, and will be reported as University policy requires.

The instructor’s use of AI in this course

AI is not used to grade your work. Your labs, exams, and projects are read and evaluated by the instructor and lab instructor.

CSU resources on AI and integrity

Accessibility and accommodations

Colorado State University is committed to equitable access to learning. If you have a disability — visible or not — and need accommodations, please register with the Student Disability Center (https://disabilitycenter.colostate.edu) and share your accommodation letter with me as early in the semester as you can, ideally within the first two weeks. Early contact gives us the best chance to have what you need in place before you need it.

If something about the format of the course creates a barrier for you, please tell me even if you are not registered with the SDC. Much of our work happens through a website and through code you run on your own machine; those tools should support your learning, not become an obstacle you have to work around silently.

Additional university policies and student resources

CSU maintains a central page that brings together University policies and resources for challenges students may encounter during the semester — academic, financial, health, personal, and otherwise.

https://col.st/2FA2g

The linked page is the best place to start when you are unsure which University resource or policy applies. Standard CSU policies are also available in the Passport to Canvas course, where they are updated if University policy changes; log in with your NetID to view them.

Schedule overview

The table below gives the semester at a glance. The full week-by-week schedule — including every module, pre-study, lab, exam, and project deadline — is maintained on the course website at https://psy652.colostate.edu/site-guide/whats-coming-up.html.

Weeks Part Modules Assessments
1–4 Part 1 · Describing data M02 Tools · M03 Visualization · M04 Wrangling · M05 Description Weekly labs
5 Exam 1 released Sep 21, due Sep 28
5–8 Part 2 · Inference M06 Probability · M07 Confidence intervals · M08 NHST logic · M09 Common tests Weekly labs
9 Exam 2 · Oct 19 & Oct 21 · Project 1 report due Fri Oct 23
10 Project 1 presentations · October 26–30
11–13 Part 3 · Modeling M10 Simple regression · M11 Multiple regression · M12 Diagnostics Weekly labs
14 Thanksgiving break
15 Exam 3 · Nov 30 & Dec 2 · Project 2 report due Fri Dec 4
16 Project 2 presentations · December 7–11

Two weeks are intentionally heavier than the others. In Week 9, Exam 2 falls on Monday and Wednesday and the Project 1 repository is due Friday; Week 15 repeats that pattern with Exam 3 and Project 2. Those deadlines are shown together above so they do not arrive as a surprise. Build both projects steadily across their assigned weeks: the final week should be for checking, polishing, and rehearsing — not for beginning the analysis.

Module 1 (M01) is a prerequisite module. Please complete it before the first pre-study of the semester. Week 1 begins with M02, and the rest of the course assumes that M01’s measurement vocabulary is already familiar.