Hello, dear friend, you can consult us at any time if you have any questions, add WeChat: daixieit

STA304H5F LEC9103

Surveys, Sampling and Observational Data

Course Outline - Fall 2021

Course Description: The sample survey is a widely used technique for obtaining information about a large population at relatively small cost. Only probability samples can provide both an estimator and a measure of sampling error from the data itself. In addition to sampling error, non-sampling errors (refusals, not-at-home, lies, inaccuracies, etc.) are always present, and can produce serious biases. The course covers: design of surveys, sources of bias, randomized response surveys. Techniques of sampling; stratification, clustering, unequal probability selection. Sampling inference, estimates of population mean and variances, ratio estimation, observational data; correlation vs. causation, missing data, sources of bias. [36L, 12T]

Priority is given to students enrolled in Statistics Specialist or Major programs.

Exclusion: STA304H1

Prerequisites: STA221H5 or STA258H5 or ECO227Y5

Recommended Prep: STA260H5 or STA261H1 strongly recommended

Students who lack a pre/co-requisite can be removed at any time unless they have received an explicit waiver from the department. The waiver form can be downloaded from here.

Learning Outcomes: By the end of this course, students should be able to:

1. Design a survey or sample that is appropriately gathering information of interest.

2. Carry out a variety of statistical analyses to make inference on the data collected from a survey/sample.

3. Identify and implement different sampling techniques and different study designs and the trade-offs involved in each.

4. Identify sources of bias within a study and comment on a study's design, including its weaknesses, strengths, and appropriate analyses.

5. Clearly communicate results of statistical analyses to technical and non-technical audiences.

Course Materials

● Required Textbook

Elementary Survey Sampling, 7th edition, by Scheaffer, Mendenhall, Ott & Gerow

 Publisher: Cengage

 ISBN-10: 0840053614 | ISBN-13: 9780840053619

 Publication Date: 2012

 Available as an E-book

● Additional References

 Sampling: Design and Analysis, 2nd edition, by Sharon L. Lohr. Cengage, ISBN 9780495105275

 Complex Surveys: A Guide to Analysis Using R, by Lumley. Wiley, ISBN 978-0-470-28430-8

● Calculators: Handheld, non-programmable calculators may be used during quizzes and tests. Any calculator that has a logarithm, square root, and one memory button will suffice for this course, so there is no need to buy an expensive calculator.

● Lectures: It is important to attend the lectures (even though they are online), follow the topics, and not leave any questions behind. This is a higher-level course, and students will need all the lecture time and regular practices to learn the material well. Crunching before tests/exams is unlikely to be helpful.

● Tutorials: Tutorials will start on the second week (i.e., Sep. 14th) and run until the last week of class. Selected problems from the textbook will be posted on Quercus and will serve as an appropriate preparation for the tutorial discussions. Attending tutorials will help students with the course material, particularly with problem solving.

● Piazza: This is for student-led discussion. All questions about course material should be posted here or asked during instructor/TAs office hours. The instructor and TAs will monitor the board and will help answer questions but students are encouraged to answer posts and help their fellow classmates.

● Statistical Computing: This course uses the statistical package R. R is free statistical software and it can be downloaded from http//cran.rproject.org/. Various introductions to R are posted in the portal. Students should use R for assignments.

Assessments and Deadlines

Type	Due Date	Weight
Quizzes	Oct. 1, Oct. 8, Nov. 19, Dec. 3	15% (top 3 quizzes)
Assignment 1	Oct. 21	10%
Midterm	Nov. 2	25%
Assignment 2	Nov. 23	10%
Final	TBA	40%

Quizzes

● The quizzes are designed to help you keep up with course material.

● Quizzes will be administered using Quercus.

● The top 3 out of 4 quizzes will be considered for the final course mark. Thus, you are allowed to miss only one quiz. Any other missed quiz will be reported zero. No other adjustments will be made for missed quizzes. There are no make-up quizzes. There are no exceptions to this policy.

● Quizzes are open book (Elementary Survey Sampling & Lecture notes); however, you are advised to be informed about Academic Integrity and to know what constitutes Academic Dishonesty. Students failing to abide by these regulations will be subject to penalties as laid out in the university’s academic dishonesty policy.

● Students have 24-hour period (between Friday 10:00 AM and Saturday 10:00 AM) to start and finish a Quiz. Once a student starts a Quiz, he/she has 30 minutes to finish the quiz.

● Exact dates of quizzes will be posted on Quercus.

● More details about each quiz such as coverage/submission will be posted on Quercus.

Midterm

● The midterm test starts at 4:10 pm and end at 6:00 pm on November 2. This includes the time needed to upload your solutions. Late submissions will not be accepted.

● Coverage will be posted on Quercus.

● The midterm test will be held on crowdmark. Details will be posted on Quercus. Submission via another tool such as email/Quercus will not be accepted.

● The midterm is open book (Elementary Survey Sampling & Lecture notes); however, you are advised to be informed about Academic Integrity and to know what constitutes Academic Dishonesty. Students failing to abide by these regulations will be subject to penalties as laid out in the university’s academic dishonesty policy.

● You cannot base your justification on any procedure/fact that is not covered in the lecture notes/textbook.

● Missed Midterm Policy:

○ If you miss the midterm for a valid excuse, send email to Luai about it, in advance if possible. You must also declare yourself absent on ACORN on the day of the missed test or the day after at the latest. 0% will be recorded for the missed midterm otherwise.

○ What is a valid excuse?

Medical excuses are usually acceptable, even though they are difficult to check in the current situation.

In case you experience a technical problem, you may report yourself as being absent in ACORN. In this case, we will check your activity on crowdmark to spot all inappropriate claims. If the claim appears to be inadequately substantiated, the mark assigned for this test will be zero.

○ If your excuse is approved by your instructor, your mark on the final exam (out of 25) will be substituted for the missing mark.

Assignments

● There will be 2 assignments that must be completed individually. These are opportunities to practice and receive feedback on theoretical and applied problems.

● Since assignments are to be done individually, it is not appropriate to post questions about how to do or approach assignment questions on the Piazza. However, you may ask for clarification, or general content.

● It is an academic offence to present someone else's work as your own, or to allow your work to be presented as someone else's.

● Assignments will be submitted through crowdmark, meaning you will need to upload PDF, PNG or JPEG versions of your assignment answers.

● Students are supposed to solve all questions. However, NOT all the questions will be marked.

● Late submissions will not be accepted. Be sure to submit your work before it is due.

● Any missed assignment will be reported zero.

● You will be given ample time to complete assignments so in general you cannot ask for extensions. Under extenuating circumstances (e.g., hospitalization, not being sick for just one or two days), you may request (with supporting documentation) to the instructor an extension on the assignment. This documentation must be presented at least 2 days before the assignment due date and must specify the exact period during which you were unable to carry out your academic work. The instructor will consider it and make a final decision.