Master 2020/2021

Quantitative Data Analyses

Category 'Best Course for Career Development'

Category 'Best Course for Broadening Horizons and Diversity of Knowledge and Skills'

Category 'Best Course for New Knowledge and Skills'

Type: Elective course (Modern Social Analysis)

Area of studies: Sociology

Delivered by: Department of Sociology

Where: Saint-Petersburg School of Social Sciences

When: 1 year, 3 module

Mode of studies: offline

Instructors: Viyaleta Korsunava

Master’s programme: Modern Social Analysis

Language: English

ECTS credits: 9

Contact hours: 40

Course Syllabus

Abstract

The course covers different types of regression modeling, including further insights into linear regression and diagnostics for linear models, binary, multinomial, ordered, Poisson regression, along with multilevel data analysis. Students are assumed to have basic knowledge of statistics and be familiar with several conventional statistical methods, most importantly linear regression, and with the R programming environment.

Learning Objectives

• The main objective of the course is to give an introduction to a variety of extensions of linear regression analysis widely used in modern social sciences, as well as implementations of the respective methods in R, a popular, free programming language for statistical computing. By the end of the course, students will be able to choose relevant methods of analysis, implement necessary techniques, and interpret the results of modeling.

Expected Learning Outcomes

• Able to read and understand most academic social sciences articles that use quantitative approach
• Able to design a quantitative social study
• Able to use R programming language for complex statistical computations
• Able to choose statistical methods appropriate to their data and substantive research problem

Course Contents

• Advanced analysis with linear regression
  Interaction terms. Variable transformations. Quadratic terms, logarithms. Regression model diagnostics. The linearity of the data. Homogeneity of variance, non-constant error variance. variance inflation factors. Non-independence of Errors. Outliers, hat values, and high leverage points. Standardized residuals (studentized residuals), Cook’s distance, variance inflation factors, Durbin-Watson test.
• Binary logistic regression
  Variables with binary outcomes. Standard logistic function. Bivariate and multiple logistic models. Logit and probit link functions. Latent variable interpretation. Interpretation of β-coefficients, odds-ratio. Model fit: pseudo R2, PCP, ePCP.
• Multinomial logistic regression
  Variables with multiple outcomes, the difference between ordered and unordered responses. Contrasts in outcome variables. Interpretation of β-coefficients in multinomial models.
• Ordered logistic regression
  Cumulative probability. Latent variable interpretation, thresholds. Odds-ratio and the interpretation of the coefficients. Model selection and assumptions.
• Models for count data
  Poisson regression. The difference between count and ordered responses. Poisson distribution. Interpretation of the coefficients, incident ratios. Overdispersion, offsets, excessive zeroes.
• Introduction to multilevel modeling
  Hierarchical linear models. Hierarchy in the data structure. Intraclass correlation coefficient. Fixed and random effects. Random intercepts and random slopes. Linear models for hierarchical data.
• Multilevel logistic regressions
  Different types of generalized linear models with a hierarchical structure ICC in logistic models. Random intercepts and random slopes in logistic models. Multilevel models for binary, ordinal, and count responses.

Assessment Elements

• Test 1
• Test 2
• In-class assignments
  An unweighted average of grades for in-class assignments.
• Final exam
  The exam is held online (in Skype) in the form of a test covering all topics.

Interim Assessment

• Interim assessment (3 module)
  0.2 * Final exam + 0.4 * In-class assignments + 0.2 * Test 1 + 0.2 * Test 2

Bibliography

Recommended Core Bibliography

• Agresti, A., & Finlay, B. (2014). Statistical Methods for the Social Sciences: Pearson New International Edition (Vol. Pearson new international ed., 4. ed). Harlow England: Pearson. Retrieved from http://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=nlebk&AN=1418314
• Field, A. V. (DE-588)128714581, (DE-627)378310763, (DE-576)186310501, aut. (2012). Discovering statistics using R Andy Field, Jeremy Miles, Zoë Field. Retrieved from http://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=edswao&AN=edswao.363067604
• Smith, R. B. (2011). Multilevel Modeling of Social Problems : A Causal Perspective. Dordrecht: Springer. Retrieved from http://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=edsebk&AN=371921

Recommended Additional Bibliography

• Chatterjee, S., Hadi, A. S., & Ebooks Corporation. (2012). Regression Analysis by Example (Vol. Fifth edition). Hoboken, New Jersey: Wiley. Retrieved from http://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=edsebk&AN=959808
• Freund, R. J., Wilson, W. J., & Sa, P. (2006). Regression Analysis (Vol. 2nd ed). AMsterdam: Academic Press. Retrieved from http://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=edsebk&AN=320724
• Jiang, J. (2007). Linear and Generalized Linear Mixed Models and Their Applications. New York: Springer. Retrieved from http://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=edsebk&AN=212826
• Upton, G. J. G. (2016). Categorical Data Analysis by Example. Hoboken, New Jersey: Wiley. Retrieved from http://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=edsebk&AN=1402878