Course Syllabus

Research Methods II – Quantitative Methods for Research on Mass Media, Communication Studies 783

Instructor: Prof. Stuart Soroka, Rm. 5338 North Quad, ssoroka@umich.edu.

Objectives: Communication Studies 781 and 783 form the research methodology graduate sequence for PhD students in Communication Studies; in addition, we require all students to take Psychology 613. These courses are intended to teach fundamental techniques for conducting valid scholarly research using a range of methodologies employed in the discipline.

In COMM 783 the goal is to present the background and basics of quantitative analysis, in enough depth for you to be able to use some basic statistics in your own research, and to engage with scholarship employing these methods.  This class is taught simultaneously, alongside Psychology 613.  613 typically provides detailed training in ANOVA, correlation and regression, and general linear models.  We do not cover these subjects in any detail here.  We do spend a fair bit of time on some of the critical theoretical background to statistics taught in 613, however; and we supplement the statistics taught there with discussion of measurement and validity, univariate and bivariate analyses, data visualization, content analysis, probability and hypothesis testing.

For students undertaking statistical work in their theses, additional methodology courses will be required.  At a minimum, you will take the second course in the Psychology series, 614; but there will be additional courses available in Communication Studies and other departments as well.

Course Design & Readings: The following texts are required (and available through Amazon, either in hard copy or electronically):

Sean Gailmard. 2014. Statistical Modeling and Inference for Social Science. New York: Cambridge University Press.

Nathan Yau. 2011. Visualize This: The FlowData Guide to Design, Visualization and Statistics. New York: Wiley.

These texts will be used alongside other readings, made available through Canvas.  We will also draw on a number of papers written by communications colleagues here at Michigan, including:

Sonya Dal Cin et al. 2008. “Youth exposure to alcohol use and brand appearances in popular contemporary movies,” Addiction 103: 1925-1932.

Kristen Harrison and Veronica Hefner. 2014. “Virtually Perfect: Image Retouching and Adolescent Body Image,” Media Psychology 17: 1-20.

Scott W. Campbell and Nojin Kwak. 2011. “Political Involvement in ‘‘Mobilized’’ Society: The Interactive Relationships Among Mobile Communication, Network Characteristics, and Political Participation.” Journal of Communication 61: 1005-1024.

Allison Allison Harell, Stuart Soroka and Kiera Ladner. 2013. “Public Opinion, Prejudice and the Racialization of Welfare in Canada.” Ethnic and Racial Studies 37(14): 2580-2597.

These papers will also be made available through Canvas; so too will the datasets they employ.  We will using these to explore topics in texts on statistics, data visualization, and content analysis.  So we might consider bar graphics alongside the impact of mobile communication on political participation, for instance; or think about experimental design as part of a discussion about image retouching and body image.  The goal is to link our thinking about methods and statistics with recent research in communication studies; and not just to think about how to conduct research, but to practice by replicating and re-testing some of these analyses. 

Seminars will proceed on the assumption that students have read all required readings beforehand, and participation grades are determined based on students’ discussion, command and critique of both methods and issue-related readings.

Additional Readings: In addition to the weekly readings listed below, students may want to read other material — relating to statistics, and/or the use of R — from either the course notes (available on Canvas) or through online textbooks.

There are many more that may be useful intermittently, both in the course and afterwards. Some of my favorites are:

For general statistics texts: John Fox’s Applied Regression Analysis and Generalized Linear Models and An R Companion to Applied Regression; Peter Kennedy’s A Guide to Econometrics; StatSoft Electronic Textbook; HyperStat Online. The 'little green books' from Sage are also available at http://srmo.sagepub.com.proxy.lib.umich.edu.

On R: Verzani’s, simpleR — Using R for Introductory Statistics; Venables et al., An Introduction to R; the Comprehensive R Archive; R Tutor; Quick R.

On data visualization: Jacoby’s Statistical Graphics for Visualizing Multivariate Data

Software: Students are strongly encouraged to bring laptops to class. This is not a requirement, but it is generally useful to not just to watch but also do what we talk about in class. 

We will be exploring data, in ways that hopefully illustrate the material in the text, using R. You will need to have installed R on a computer before class begins. To do so, please visit this website: http://www.r-project.org. R itself is not especially user-friendly, so after installing R, I strongly recommend that you install RStudio, which you get here: http://www.rstudio.com. Both R and RStudio are free.

Requirements: Weekly assignments will comprise 80% of your grade. All assignments are submitted electronically. The other 20% will be based on your participation in the seminar. Students are of course expected to complete all readings before class begins each week.  Lateness in assignments, or in attendance, will not be tolerated; which is to say that participation and/or assignment grades will be penalized accordingly.

Further details on assignments will be made available in class. This syllabus is subject to change, based on student interest and course progress.

Click here for a Statement on Students with Disabilities, Statement on Academic Misconduct, or Statement on Sexual Misconduct.

 

Outline (in progress, as of Sept 8)

Note that we take a somewhat different approach to the study of methods and background theory – rather than start with theory and move to methods, we start with some basic methods (as a motivation to understand the theory) before turning to theory.  This is the approach in the Gailmard text; the important additions here include further discussion of measurement, of data visualization, and of content analysis.

Sept 9:  Introduction

Gailmard, Ch 1

Sept 16:  Social Science

Brian Fay and J. Donald Moon. “What Would an Adequate Philosophy of Social Science Look Like?” Ch 2 in Michael Martin and Lee C. McIntyre, eds., Reading in the Philosophy of Social Science (MIT Press, 1994).

Simon, H. A. “On judging the plausibility of theories” (pp. 25-48). In H. A. Simon, Models of Discovery (Dordrecht, Holland: D. Reidel Publishing, 1977).

Scott W. Campbell and Nojin Kwak. 2011. “Political Involvement in ‘‘Mobilized’’ Society: The Interactive Relationships Among Mobile Communication, Network Characteristics, and Political Participation.” Journal of Communication 61: 1005-1024.

Sept 23: Measurement, Reliability and Validity

Edward Carmines and Richard Zeller. 1979. Reliability and Validity Assessment, Sage, chapters 1-2.

Gailmard, Ch 2.1

Kristen Harrison and Veronica Hefner. 2014. “Virtually Perfect: Image Retouching and Adolescent Body Image,” Media Psychology 17: 1-20.

Sept 30:  Univariate and Bivariate Statistics

Gailmard, Ch 2.2, 2.3

Allison Allison Harell, Stuart Soroka and Kiera Ladner. 2013. “Public Opinion, Prejudice and the Racialization of Welfare in Canada.” Ethnic and Racial Studies 37(14): 2580-2597.

Oct 7:  Data Visualization

Yau, Chapters 3-6.

Dal Cin et al. 2008. "Youth exposure to alcohol use and brand appearances in popular contemporary movies." Addiction 103: 1925-1932. 

Oct 14: Text as Data: Surveying the Possibilities

Klaus Krippendorff and Mary Angela Bock, eds., The Content Analysis Reader Sage, 2009, Chapters 1.2, 1.7, 2.3, 2.4, 2.7, 3.4, 3.9, 5.8, 7.7.

Oct 21: Text as Data: Turning Words into Numbers

Klaus Krippendorff and Mary Angela Bock, eds., The Content Analysis Reader Sage, 2009, Chapters 4.4, 4.5, 6.2

Stuart Soroka, Lori Young and Meital Balmas. 2015. “Bad News or Mad News? Sentiment Scoring of Negativity, Fear, and Anger in News Content,” AAPSS 659(1): 108-121.

Oct 28: Data Generating Processes, Probability Theory

Gailmard, Ch 3 and Ch 4

Nov 4: Expectation and Moments, Probability and Models

Gailmard, Ch 5 and Ch 6

 

Nov 11: Sampling, and Sampling Distributions

Gailmard, Ch 7

Nov 18: Hypothesis Testing

Gailmard, Ch 8

Dec 2: Estimation

Gailmard, Ch 9

Dec 9: Causal Inference

Gailmard, Ch 10