Schedule

Fall 2026

The schedule is the main course hub. Each session lists the required preparation, class focus, and principal product or lab. Because this course includes a live Univision polling project, production milestones may shift. Changes will be announced in class and posted here.

Access: Open-access readings are linked directly. For items marked UT Libraries/Canvas, sign in through UT Libraries or use the course copy posted on Canvas. Page ranges refer to the editions listed in the complete reading reference.

Foundations: Evidence, error, and measurement

Session 1 · August 24

Why polls fail and what good survey design requires

Required preparation

In class: Complete the Initial Survey Autopsy diagnostic, establish a shared language for evaluating survey evidence, and introduce the Univision partnership and R/Quarto workflow.

Session 2 · August 31

Research objectives, stakeholders, ethics, and total survey error

Required preparation

The Groves total survey error review is recommended rather than required on this lab day.

In class: Translate stakeholder questions into researchable objectives and identify threats to the intended claims.

Lab 1: RStudio projects, Quarto, importing survey data, reading a codebook, missing values, recoding, summarizing, and reproducible workflow. Lab guide · Student template

September 7

Labor Day: no class

Session 3 · September 14

Concepts, measurement, and questionnaire architecture

Required preparation

  • Groves et al. 2009. Chapter 7, “Questions and Answers in Surveys,” pp. 217–253. Focus on constructs, measurements, responses, and response error. UT Libraries/Canvas.
  • Krosnick, Jon A., and Stanley Presser. 2010. “Question and Questionnaire Design.” In Handbook of Survey Research, 2nd ed. Read pp. 263–286, through the end of the section on open versus closed questions. Course PDF on Canvas.
  • Pew Research Center. “Writing Survey Questions.” Read “Question development,” “Measuring change over time,” and “Open- and closed-ended questions.” Open access.

In class: Move from abstract concepts to indicators, distinguish measurement from operational convenience, and design the overall flow of a questionnaire.

Principal product: Construct map.

Session 4 · September 21

Question wording and response design

Required preparation

In class: Examine comprehension, retrieval, judgment, response mapping, acquiescence, social desirability, question order, response scales, and treatment of uncertainty.

Principal product: Draft question module.

Session 5 · September 28

Testing and multilingual equivalence

Required preparation

  • Willis, Gordon B. 2005. Cognitive Interviewing: A Tool for Improving Questionnaire Design, Chapter 1, pp. 1–15. UT Libraries/Canvas.
  • Survey Research Center. 2016. “Translation: Overview.” Read Guidelines 1–4 and their rationales. Open access.

In class: Conduct and evaluate cognitive interviews, identify systematic problems, and revise English and Spanish modules for conceptual equivalence.

Lab 2: Distributions, proportions, cross-tabulations, grouped summaries, measurement, scales, and accurate visualization with ggplot2. Lab guide · Student template

Representation and survey production

Session 6 · October 5

Sampling and representation

Required preparation

  • Groves et al. 2009. Chapter 3, “Target Populations, Sampling Frames, and Coverage Error,” pp. 69–94. UT Libraries/Canvas.
  • Groves et al. 2009. Chapter 4, “Sample Design and Sampling Error,” pp. 97–118, stopping before “Systematic Sampling.” UT Libraries/Canvas.
  • Pew Research Center. 2023. “Comparing Two Types of Online Survey Samples.” Read the overview and “How the probability and opt-in samples compare.” Open access.

In class: Define the inferential population, compare sampling frames and recruitment sources, and assess probability and nonprobability approaches.

Principal product: Sampling assessment.

Session 7 · October 12

Modes, recruitment, and nonresponse

Required preparation

In class: Analyze mode effects, coverage, contact strategies, incentives, accessibility, response rates, and nonresponse bias.

Assignment: Individual Poll Autopsy assigned. Due Sunday, October 18, at 11:59 p.m.

Session 8 · October 19

Programming and quality assurance

Required preparation

  • U.S. Census Bureau. 2026. Design Guidelines for U.S. Census Bureau Web Surveys and Censuses. Read “Overall Look and Feel,” “Screen Elements,” and the entries for single-select, multi-select, grids, and open-text questions. Open access.
  • Dillman, Don A., Jolene D. Smyth, and Leah Melani Christian. 2014. Internet, Phone, Mail, and Mixed-Mode Surveys, Chapter 9, “Web Questionnaires and Implementation.” UT Libraries.
  • Course instrument QA protocol. Read the complete checklist posted on Canvas and bring it to class.

In class: Integrate modules, review logic and randomization, test mobile and bilingual displays, and maintain a formal issue log.

Principal product: Programmed instrument and QA log.

Session 9 · October 26

Lab 3: From a Sample to an Estimate

In class: Sampling variability, standard errors, margins of error, confidence intervals, and subgroup comparisons using the October Univision/YouGov Texas Poll.

Lab 3: Lab guide · Student template · Texas poll data

Analysis and communication

November 2

No class — additional lab recording online

There is no in-person class. An additional lab recording will be posted online for review and practice.

No submission is required. Use the recording to review the Lab 3 material, practice the workflow, and come back ready to work with the Texas poll.

Session 10 · November 9

Fieldwork and data quality

Required preparation

In class: Establish fieldwork monitoring rules, examine paradata and respondent-quality indicators, and distinguish defensible quality controls from post hoc deletion.

Principal product: Fieldwork and data-quality protocol.

Session 11 · November 16

Weighting and poll estimates

Required preparation

In class: Compare weighted and unweighted estimates from the Texas poll, introduce the logic of survey weights, examine effective sample size and design effects, and produce defensible toplines.

Lab 4: Weighting and poll estimates using the Texas poll.

November 23

Fall break: no class

Session 12 · November 30

From results to claims

Required preparation

In class: Move from estimates and subgroup comparisons to defensible public claims. Evaluate uncertainty, choose which findings matter, build clear graphics, and make sure the language of each claim matches the strength of the evidence.

Principal product: Priority findings, claims, and graphics for the Univision report.

Session 13 · December 7

Analyzing the Univision poll: report workshop

Required preparation: No new reading. Bring your working analysis, tables, graphics, and draft findings.

In class: Analyze the Univision poll in teams, refine subgroup comparisons and interpretations, resolve remaining methodological questions, and work directly on the final report.

Principal product: Substantive analysis and report draft.