Multivariate Analysis Techniques

Course objectives

2 Obiettivi formativi 2.1 Obiettivi generali
 The course aims to let students practice and familiarize with advanced data analysis models. Each data analysis model will be considered with respect to the underlying statistical theory, its assumptions and its basilar principles. Students will learn in which case it will be more appropriate to use a particular model of analysis as well as to interpret and correctly communicate and criticize the results. Students will also learn to use statistical software for running data analysis and for writing and communi-cate results for scientific or non-scientific contexts. 2.2 Obiettivi specifici 2.2.1 Conoscenza e capacità di comprensione (knowledge and understanding). The course will extends and generalize the basilar concepts on descriptive statistics and on inferential principles that students have to acquire before the course begins. At the end of the course, students should be able to read and/or write a scientific report and/or conduct the data analysis concerning the ad-vanced statistical models considered in the course. 2.2.2 Capacità di applicare conoscenza e comprensione (applying knowledge and under-standing). The advanced statistical models considered in the course shares a common background with other scien-tific disciplines other than psychology (like neuroscience, biology, sociology, economics etc.). So the knowledge of these advanced statistical models is also an occasion for cross-breeding and let psychologi-cal research hypothesis to be investigated with methods and analysis that cross scientific disciplines. 2.2.3 Autonomia di giudizio (making judgements). The laboratory is an hands-on experience that has the aim to let the student practice and familiarize with statistical analysis of simulated datasets of real scientific reports by using a statistical software. In particu-lar students will acquire the ability to run a specific statistical analysis and to read and to critically discuss the output for investigating with respect to a specific research hypothesis. Having the possibility to com-pare their results and their comments with those reported in the scientific report will help students to be more incisive and to acquire independence of judgment in concluding that a specific research hypothesis has been (scientifically) proved. 2.2.4 Abilità comunicative (communication skills). During the course, the use of different communicative styles are encouraged. In particular during lectures the discussion of formal and academic aspects of statistical models and principles will be considered. While during practice with statistical software, procedural and pragmatic aspects of research investigation are stressed (i.e. how to use different multivariate models for investigating different research hypotheses). Finally, students are encouraged to critically draw conclusions about whether statistical results support or not research questions and their limits (i.e. whether findings are generalizable to other samples, context or not). 2.2.5 5) Capacità di apprendimento (learning skills). At the end of the course, students will know how to read statistical results of a scientific article, will be able to run the needed statistical multivariate models for investigating specific research hypotheses, and will be able to write down and appropriately comment findings according to the investigated research hypotheses. These skills will give students easy access to scientific literature regular update and the practice with the statistical package will open the possibility to work in the marketing research area.

Channel 1
FABIO PRESAGHI Lecturers' profile

Program - Frequency - Exams

Course program
Lectures (6 CFU, 48 hours): - Research designs: non-, quasi- and true experiments; threats to the validity of experiments (8 hours) - Reliability of measures (4 hours) - Multiple Linear Regression Analysis: Least square method of estimation, fit index (R, R-squared, differential R-squared), regression coefficients, and standardized regressing coefficients. Direct and interaction effects (12 hours) - Analysis of Variance (ANOVA): factorial between-subjects design, factorial within-subjects design, and mixed or split-plot design. Main effects, interaction effects, post-hoc effects for main and interaction effects. Effect size and power of the test. (8 hours) - Exploratory Factor Analysis: Thurstone fundamental equation, estimation methods, extraction methods, rotation methods. (8 hours) - Introduction to Multiple Logistic Regression Analysis: fit indexes, odds, odds-ratio, regression coefficients (4 hours) - Introduction to the Cluster Analysis: cluster types, clustering algorithms fit indexes (measures of distance and association), hierarchical and non-hierarchical clustering. (4 hours) Laboratory (3 CFU, 36 hours): - Introduction to the statistical software, organizing the data into a dataset, screening of data (2 hours) - Computing composite indexes and measures, reliability estimation, and analysis (2 hours) - Exercise: chi-square test, z-test, student-t-test, correlations, and graphs (4 hours) - Exercise: ANOVA: factorial between-subjects design, factorial within-subjects design, and mixed or split-plot ANOVA design (12 hours) - Exercise: Multiple Regression Analysis (8 hours) - Exercise: Exploratory Factor Analysis (8 hours)
Prerequisites
Descriptive statistics and Inferential Statistics
Books
For the Multivariate data analysis: Tabachnick, B. G., & Fidell, L. S. (2013, 6th ed.). Using multivariate statistics, 5th. Needham Height, MA: Allyn & Bacon. For Research design: Anderson-Cook, C. M. (2005). Experimental and quasi-experimental designs for generalized causal inference. For reliability and validity of measures: Crocker, L., & Algina, J. (1986). Introduction to classical and modern test theory. Holt, Rinehart and Winston, 6277 Sea Harbor Drive, Orlando, FL 32887.
Teaching mode
Lectures will consider statistical theory underlying multivariate models and their assumptions. Lectures will be paralleled by exercise with a statistical software. During exercise, students will learn how to run statistical models for investigating specific research hypotheses (concerning both difference among means or association among multiple variables) by analyzing simulated data from real scientific research articles. In this way students have the occasion to learn how authors of the article presented results and how they commented whether findings support their hypotheses.
Frequency
Lectures: optional Exercise: mandatory
Exam mode
Written examination: Students must respond to 30 questions with closed responses in 45 minutes. Exercise, 1 hour: students have to investigate a series of research hypotheses by running data analysis with a real dataset and have to write a report of their findings in terms of results and comment on the results Oral examination, about 30 minutes
Lesson mode
Lectures will consider statistical theory underlying multivariate models and their assumptions. Exercises with statistical software will parallel lectures. During the training, students will learn how to run statistical models to investigate specific research hypotheses (concerning differences among means or associations among multiple variables) by analyzing simulated data from real scientific research articles. In this way, students can learn how the article's authors presented results and how they commented on whether the findings support their hypotheses.
  • Lesson code1044869
  • Academic year2025/2026
  • CoursePsychology of Communication and Marketing
  • CurriculumSingle curriculum
  • Year1st year
  • Semester1st semester
  • SSDM-PSI/03
  • CFU9