LEARNING OUTCOMES

Provide statistical tools relevant to specific applications and the experience of on-field experts. 

AIMS AND LEARNING OUTCOMES

At the end of the course the student will be able to:

• identify the degree of applicability of the analytical techniques
• identify the main issues in data analysis in the applied contexts listed below
• identify the adequate statistical methods
• carry out the analyses with the adequate software
• interpret the results of the analyses and assess their validity
• summarize the results in a semi-professional report that describes and provides an explanation for the results

Measurement models in psychometrics (3 CFU)
The course introduces to statistical issues in psychometric theory and to the use of statistical software (R) for carrying out basic psychometric analyses.

Demography in Italy and in the world: topics, data and measurements (1 CFU)
To illustrate via a complex example the issues related to the communication of demographic data to the general population.

Official statistics (1 CFU)
The system of official statistics.

Multiple statistical tests in biomedical research (1 CFU)
To understand the problem related to performing multiple statistical hypothesis tests and to learn how to handle it critically.

Further seminar activities (not evaluated) could be organised each year. Usually they are presented by data scientists who work in applied contexts such as companies, consumer companies, public bodies. 

TEACHING METHODS

Combination of traditional lectures and lab sessions with the softwaresMatlab and R.

SYLLABUS/CONTENT

Psychometrics

• Classical test theory
• Psychological variables or constructs
• Definition of the content domain of a construct and its operationalizations
• Measurement models in psychology: reflective indicators models and formative indicators models
• Item analysis and reliability
• Exploratory factor analysis
• Confirmatory factor analysis
• Structural Equation Models
• Item Response Theory

Applications in R (packages 'psych', 'lavaan', 'mirt') will be shown.

Demography in Italy and in the world: topics, data and measurements

The course is based on the careful reading and analysis of a highly data-driven volume. Collecting, analysing, and presenting data to help society organise itself to turn opportunities into new realities. The reference text to be analysed has changed over the years

Official statistics

Official statistics, statistical sources, official data, SISTAN, legislation on statistics, codes of statistics.

Multiple statistical tests in biomedical research

Statistical hypothesis tests are characterised by a probability of a false positive (significance level) and a probability of a false negative (1-test power). When several tests are performed, the probability that at least one true hypothesis will be rejected increases rapidly. In experimental research, this is the case, for example, when several markers are tested, a diagnostic test is repeated several times, several drugs are compared to a standard, several evaluation criteria are used to assess the effectiveness of a treatment, etc. The significance of the observed results is determined by the level of significance.

The significance of the observed results is largely overestimated if the results of the most significant tests are interpreted without taking into account the multiplicity of tests performed. This problem is even more crucial if only the most significant results are reported and disseminated.

In the case of multiple tests, several measures of error are available, in particular the probability of encountering at least one false positive (FWER, familywise error rate) and the false discovery rate (FDR). Each has a different meaning and is more or less appropriate depending on whether the purpose of the research is more exploratory or confirmatory.

Several correction methods for multiple tests are available in the statistical literature. Each correction method aims to control the FWER or FDR and each has different properties. In the context of evaluating different efficacy criteria in clinical trials, the FDA (US Food and Drug Administration) has made preliminary guidelines available.

Practical examples with R software: inflation of the significance level of multiple tests combined, correction of the significance level of multiple tests.

RECOMMENDED READING/BIBLIOGRAPHY

Psychometrics
Furr, R. M., & Bacharach, V. R. (2018). Psychometrics: An introduction. 3rd (or 4th) edition.Sage
Handouts and other teaching material (e.g., R codes) will be shared online.  

Demography in Italy and in the world: topics, data and measurements
Information about recommended readings will be provided by the teacher

Official statistics
Reading and study materials will be provided by the teacher

Multiple statistical tests in biomedical research
Dmitrienko, D’Agostino, 2013, Traditional multiplicity adjustment methods in clinical trials. Stat Med 32
Goeman J, Solari A, 2014, Multiple Hypothesis Testing in Genomics. Stat Med 33
FDA, 2017, Multiple Endpoints in Clinical Trials. Draft Guidance for Industry. UCM536750