STA304H1/STA1003H1: Surveys, Sampling and Observational Data Fall 2022

Sections L0101 & L0201

Instructor:  Dr. Samantha-Jo Caetano

Office hours:  Thursdays 11am - 12pm ET (on Zoom) and Thursdays 4pm - 5pm ET (on Zoom)

Preferred pronouns:  she/her

Course Information

Course administrative email

Use the email address [email protected] for all administrative inquiries or personal in- quiries (e.g., notifying the teaching team of a personal circumstance). To ensure your email inquiries do not go to spam/junk, please use your University of Toronto email address (i.e., email addresses from ”@mail.utoronto.ca”) Please note that the [email protected] inbox will not be monitored after December 31, 2022.

Course web page

All materials will be posted on Quercus https://q.utoronto.ca. Course materials provided on Quercus are for the use of students currently enrolled in this course only.  Distributing course materials to anyone outside of the course is considered unauthorized use.

Teaching assistants

See the course Quercus page for information about TAs, office hours and contact.

Calendar description

Design of surveys, sources of bias, randomized response surveys.  Techniques of sampling; stratification, clustering, unequal probability selection. Sampling inference, estimates of pop- ulation mean and variances, ratio estimation. Observational data; correlation vs. causation, missing data, sources of bias.

Required prerequisites

ECO227Y1 /STA255H1  /STA261H1  /STA248H1  /STA238H1  /STAB57H3 /STA258H5 / STA260H5 /ECO227Y5

Please note that all prerequisites for all STA courses are strictly enforced and your instructor cannot waive them. Any questions about prerequisites should be directed to                        [email protected].

Class format

The course is scheduled for 3 hours of lecture per week (per lecture section).  The sessions will be delivered as such:  Tuesday lectures will be synchronous and in-person, Thursdays will be a synchronous online office hour (on Zoom). We will be using a partial flipped class for most modules in this class. Most weeks, there will be a set of lecture videos or readings available mid-week (posted by Thursday morning 11am ET) to watch prior to the Tuesday synchronous in-person class. The 2 hour time slot (on Tuesdays) will be held synchronously in-person.

Suggested Weekly Routine

Note:   The  instructor will  do  her  best  to  make  all  in-person  synchronous  lectures  available virtually  too .    When  possible  the  synchronous  Tuesday  lectures  will  be  recorded  and  made available to  all students in the course .

Reference Materials

We will be relying on material from the following textbooks. Please note, that access to the textbooks are not mandatory in this course, but having them for reference is recommended.

1. Wu, Changbao and Mary E. Thompson, 2020, Sampling Theory and Practice, Springer. This is the primary reference for the course.

Available for free as in pdf through the UofT library here.

2.  Gelman, Andrew, Jennifer Hill and Aki Vehtari, 2020, Regression and Other Stories, Cambridge University Press.

3. Kohavi, Ron, Diane Tang, and Ya Xu, 2020, Trustworthy Online Controlled Experiments: A Practical Guide to A/B Testing, Cambridge University Press.

4. McElreath, Richard, 2020, Statistical Rethinking, 2nd Edition, CRC Press.

Course Content

The course will consider the following overarching themes in statistical theory and data anal- ysis:

• Designing a survey or sample that is appropriately gathering information of interest.

• Carrying out a variety of statistical analyses in R to make inference on the data collected from a survey/sample.

• Identifying and implementing different sampling techniques and different study designs and the trade-offs involved in each.

• Identifying sources of bias within a study and comment on a study’s design, including its weaknesses, strengths, and appropriate analyses.

• Clearly communicating results of statistical analyses to technical and non-technical au- diences.

• Understanding best practices and ethical considerations in survey design, sampling, and observational data collection and analysis.

We will consider various perspectives on these themes, including Bayesian, frequentist, and likelihood approaches.  We will consider methods that rely on mathematical thinking and methods that rely on computational thinking, with particular emphasis on computational approaches to analyzing data and understanding statistical methods.

Assessments

The assessments for STA304 Fall 2022 are described below. Your final grade will be calculated as the maximum of Weight Scheme A and Weight Scheme B.

Policy for Quizzes:

There will be no accommodation for missed weekly quizzes beyond the flexibility already built into the grading scheme (i.e., best 6 of 10). Quizzes will be due Mondays and are intended to be a quick way to check your knowledge and understanding of the material posted in the recent module (i.e., released the previous Thursday) before going to the Tuesday lecture.  For example, the quiz due on Monday October 3, 2022 will be about the materials released on Thursday September 29, 2022.

Quizzes will be short (no longer than 15 minutes) and will be each worth a total of 1 point. Weekly quizzes will take place on Quercus. The quizzes can consist of multiple choice, tick all that apply, matching, numerical answer, fill in the blanks, and/or short answer questions.  Please note that most of these questions will be auto-graded by Quercus in its default manner. For example, please note that for tick all that apply questions it will deduct partial points if you select incorrect options and add partial points for selecting correct answers.

For support on writing Quercus quizzes please visit the link below and read under the “Best Practices and Troubleshooting while taking a Quiz” section.

https://q.utoronto.ca/courses/46670/pages/student-guide

Policy for Assignments:

Both assignments will be open-ended reports. There will be strong focus on communication and writing throughout these assessments. Thus, it is recommended to seek writing support, if needed. Please see the “Writing Resources” section of the syllabus for more information on this.

• Assignments will be posted on Quercus at least one week in advance of the due date.

• We will NOT accept email submissions for assignments; assignments must be submitted through Quercus.

• Sometimes life throws unexpected curve balls.  We like to be as flexible as possible, within the limits that are inherent to teaching a course this large and with a team of TAs who need to balance their workloads.  Do try your best to meet the assignment deadlines.   These deadlines are planned to spread your work out and allow TAs to schedule their grading time.  For assignments, there is a 7-day grace period to submit without penalty.  You do not have to request this 7-day extension from our teaching team you just need to submit your work when you are finished (within the extra 7 days). Please note, if you submit during the grace period, you will receive your grade later than those who submit on time, because your TA will mark it when their schedule allows. If you need an extension beyond 7 days, then you will need to contact your Registrar’s Office and notify them of your personal situation and have them contact our teaching team. Assignments not submitted within the 7 day grace period will receive a grade of 0, unless your registrar’s office or the accessibility office made alternative arrangements with the STA304 teaching team.

• Since assignments are open ended and writing-based, we will be using a Plagiarism Detection Tool, OurOriginal, on the written components of the assignments. Please see the “Academic Integrity”section of the syllabus for more information on the Conditions of Use of this tool.

Policy for the Midterm Test:

The midterm test will take place in-person in lecture on Tuesday October 18.   The midterm test will be 75 minutes and you must write in your registered lecture section. If you do not write in the lecture section you are currently enrolled in, you will receive a grade of 0. If you miss the midterm, the weight will be shifted to the final exam (i.e., Weight Scheme B). There is no need to contact the instructors/teaching team in cases of a missed test.

Policy for the Final Exam:

The final exam will be scheduled by the Faculty of Arts and Sciences sometime in the final assessment period in December.  The final exam will be cumulative and will be a proctored in-person timed examination. Any scheduling conflicts or petitions to write a deferred exam must go through the Faculty of Arts and Sciences standard procedures.

https://www.artsci.utoronto.ca/current/faculty-registrar/exams-assessments

Be sure not to make travel plans until the final exam schedule has been officially released by the Faculty of Arts and Sciences

Regrade Requests:

• Mistakes occasionally happen when marking.   If you feel there is an issue with the marking of an assessment, you may request that it be re-marked.  The course re-mark policy exists to correct mistakes, and any request should clearly identify the error (for example, a question that was not marked, or a total was incorrectly calculated). Regrade requests must be made via a form (the link will be shared in a Quercus announcement when the grades are posted). Regrade requests will NOT be accepted by email and must satisfy the following requirements:

– must not be sent within the first 24 hours of the release of the assessment grade,

– must be received within one week of the date that the marks for the assessment became available,

– must include your full name and student number, (and group number if applicable), and

– must give a specific, clear, and concise reason for each request, referring to a possible error or omission by the marker.  Re-mark requests without a specific reason will not be considered.

Please note that your entire test/assignment may be remarked when submitting a remark request.  Keep in mind that it is possible for your assessment grade to go down if the regraded mark is lower than your original assessment grade.

• For the final exam, the remark process is handled by the Faculty of Arts and Sciences.

• All regrading will be done by the instructor.

Academic integrity

Academic integrity is fundamental to learning and scholarship at the University of Toronto. Participating honestly, respectfully, responsibly, and fairly in this academic community en- sures that the University of Toronto degree that you earn will be valued as a true indication of your individual academic achievement, and will continue to receive the respect and recognition it deserves.

Familiarize yourself with the University of Toronto’s Code of Behaviour on Academic Matters available at http://academicintegrity.utoronto.ca.

Discussion about lecture materials, textbook concepts and course concepts with your class- mates and the teaching team is encouraged, but it is expected that you work independently on all individual assessments. Please note, you may not submit for credit any work that was completed by another student (or person).  This includes, but is not limited to, partially or fully completed code, communication of solutions, and plagiarism. In particular, you are ex- pected to complete and submit independent work for assignments (that are not group work), the midterm, quizzes and the final exam. Specifically, you are expected to work on individual work, individually. You may discuss lecture materials and general course concepts, but it is expected that you work individually through assessments.  You may use code provided by your STA304/1003 instructors without providing a citation. If you use code from any other source, you must provide the source.  To protect yourself from potential academic integrity offences, do not share your code and written submissions.

Since assignments are open ended and writing based, we will be using a Plagiarism Detection Tool, OurOriginal, on the written components of the assignments.  The Conditions of Use is stated as follows:  “Normally, students will be required to submit their course essays to the University’s plagiarism detection tool for a review of textual similarity and detection of possible plagiarism.  In doing so, students will allow their essays to be included as source documents in the tool’s reference database, where they will be used solely for the purpose of detecting plagiarism. The terms that apply to the University’s use of this tool are described on the Centre for Teaching Support  Innovation web site ( https://uoft.me/pdt-faq).”

General Course Policy

Accessibility needs

The University of Toronto is committed to accessibility.  If you require accommodations for a disability, or have any accessibility concerns about the course, the classroom, or course materials, please contact Accessibility Services as soon as possible:                                       [email protected] or http://www.accessibility.utoronto.ca.

Computing

Computational work is a central part of developing statistical thinking and developing facility in the use of computational tools for carrying out simulations and data analysis is a core objective of this course.

We will use R, the R Studio IDE, and R Markdown. All of these are freely available. You need            to first install R, and then R Studio. R can be downloaded for free from http://cran.r-project.org. R Studio can be downloaded for free from http://www.rstudio.com/products/rstudio/download/.

Additionally, you can also use R Studio through the U of T Jupyterhub, by selecting the RStu-                 dio option and logging in with your utorID and password, available here: https://jupyter.utoronto.ca Some resources for using R and R Markdown:

• The course supplementary notes give guidelines on installing and getting started with R and R Studio.

• A short intro to R workshop is available here:

https://awstringer1.github.io/ssu-r-workshop/ssu-r-workshop.html

• Hands-On Programming with R by Garrett Grolemund, available here:

https://rstudio-education.github.io/hopr

• R for Data Science by Hadley Wickham and Garrett Grolemund, available here:

https://r4ds.had.co.nz

• An R Markdown Cheat Sheet is available at https://rstudio.com/resources/cheatsheets

Practice Problems

There will be weekly quiz problems for practice. Some recommended practice problems may be assigned from the textbooks, but these will not be graded.

Communication

Tentatively, we will be using Zoom for online synchronous meetings (e.g., office hours) in the course.  Please ensure you are able to access your UofT Zoom account as you will need to be signed in this way to enter any STA304/1003 course related online Zoom meetings. https://utoronto.zoom.us/.

We will be using Piazza as the platform for discussions related to the course material and assessments. You can find our course page at:                                                                       piazza.com/utoronto.ca/fall2022/sta304h1f.

Students can post anonymously to classmates on Piazza, but the identity of the author of all posts is viewable by instructors.

Be sure to read Piazza’s Privacy Policy and Terms of Service carefully.  Take time to un- derstand and be comfortable with what they say.  They provide for substantial sharing and disclosure of your personal information held by Piazza, which affects your privacy. When you use Piazza, only provide content that you are comfortable sharing under the terms of the Privacy Policy and Terms of Use. With that being said Piazza will still be considered a part of our class and thus all posts and conduct on Piazza must remain professional. Posts regard- ing personal matters such as inquiries about grades, reporting absences, regrade requests, etc. should be communicated via email (at [email protected]) and *NOT* be posted on Piazza. Piazza is intended for students to receive support regarding course information and content and thus should be an overall positive and professional environment.

Again, email is appropriate only for personal matters that can not be shared with the rest of the class. To be fair to all students, we are not able to answer questions about the course material by email.  These questions should be asked on the discussion forum (publicly) or during office hours.