Quantitative methods: Multivariate methods for PhD students

Ph.D. -course

Course description

Course content

General content

This course comprises three topics, as follows: Multilevel analysis, exploratory and confirmatory factor analysis and structural equation models (SEM).

The present course is sub-course C; one of three sub-courses of the overarching course on quantitative methods that covers a broad range of quantitative research methods and statistical analysis techniques.

This sub-course covers theoretical and methodological approaches to dealing with multivariate data structures and research questions.

Type of course

Methods

Learning outcomes

General learning objectives

This course seeks to promote and facilitate the rigorous use of quantitative approaches and good practices in designing and analyzing psychological research questions.

Knowledge

After completion of the course, the candidate

- Has knowledge of various approaches of data collection using multivariate designs.

- Has knowledge of various approaches of analyzing multivariate data structures.

Skills

After completion of the course, the candidate

- Can evaluate and select appropriate data analytical tools and software solutions.

- Can evaluate and select appropriate analysis in multivariate research.

- Can draw appropriate conclusions from multivariate research.

General competence

After completion of the course, the candidate

- Can master multivariate research designs.

- Can reflect upon different approaches to quantitative research methodology

Study period

Fall 2018, 24- 25- 26 September.

Credits (ECTS)

1 ECT

Course location

Faculty of Psychology, Christiesgt. 12. All days at room 006.
Language of instruction
English
Course registration and deadlines
Deadline for registration is September 10. Only by e-mail to maria.luttges@uib.no
Pre-requirements

Master Degree in disciplines relevant to educational sciences, psychology and public health.

  1. Participants must answer a brief questionnaire when signing up for the course, in order to provide information for adjusting the course topics and depth.
  2. Participants must bring their laptop computer with STATA and MPlus (trial version) installed.

STATA (Windows/Mac) can be downloaded from UiBs collection of site licensed programs: (https://tjinfo.uib.no/program).

A trial version of MPlus (Windows/Mac) can be downloaded from:

(http://www.statmodel.com/demo)

Compulsory Requirements
Participation and assignments to be completed on individual basis or in small groups. Plenary presentations.
Form of assessment

Active participation and presentations in class. 80% attendance required.

Pass or fail.

Who may participate
The course targets primarily PhD candidates at the Faculty of Psychology, UiB. Other groups can apply for participation.
Supplementary course information

Topic 1 (September 24, 2018): Multilevel analysis

- Students will be introduced to contextual models and its associated terminology including intra-class correlations, fixed vs. random effects, cross-level interactions, and intercepts/slopes-as-outcomes models.

- Data will be provided for the students to practice on in STATA.

Topic 2 (September 25, 2018): Exploratory factor analysis (EFA) and Confirmatory factor analysis (CFA)

- This part will cover the procedures involved in running exploratory factor analysis and will touch on topics such as extraction of factors, type of rotation and identification of latent constructs.

- This part will also cover the procedures involved in running a confirmatory factor analysis and will touch on topics such as model identification and model assessment.

- Data will be provided for the students to practice running EFA and CFA in STATA.

Topic 3: (September 26, 2018): Structural equation modelling

- This part comprises an introduction to analysis of multivariate statistics by means of structural equation modelling, including mediation- and moderation analysis.

- Data will be provided for the students to practice on the abovementioned analyses in MPlus.

There will be lectures and hands-on application of statistical analysis on the basis of data provided by the lecturers. Students may also use their own data for statistical analyses.

Academic responsible
Gisela Bøhm, Department of Psychosocial Science.
Lecturers
Sigurd Hystad, Tony Leino and Guy Notelaers, Department of Psychosocial Science, UiB.
Reading list

Rabe-Hesketh, S., & Skrondal, A. (2012). Random intercept models. In S. Rabe-Hesketh, & A. Skrondal, Multilevel and Longitudinal modeling using STATA (123-172). Texas: Stata press.

Rabe-Hesketh, S., & Skrondal, A. (2012). Random coefficient models. In S. Rabe-Hesketh, & A. Skrondal, Multilevel and Longitudinal modeling using STATA (pp. 181-216). Texas: Stata press.

Mehmetoglu, M., & Jakobsen, T. G. (2017). Exploratory factor analysis. In M. Mehmetoglu & T. G. Jakobsen, Applied statistics using STATA (pp. 270-290). London, UK: Sage Publications.

Mehmetoglu, M., & Jakobsen, T. G. (2017). Exploratory factor analysis. In M. Mehmetoglu & T. G. Jakobsen, Applied statistics using STATA (pp. 294-322). London, UK: Sage Publications.

Mackinnon, D. P. (2008). Introduction to Statistical Mediation Analysis. New York, Lawrence Erlbaum Associates. Chapter 5: Multiple Mediator Models & Chapter 6: Path Analysis Mediation.

Geiser, C. Data-analysis with Mplus. New York, The Guilford Press. Chapter 3: Linear Structural Equation Modelling.

Additional required and recommended literature will be announced after registration. To the extent possible, copies will be placed on the Mitt UiB-site for the course. Participants will be expected to have read a large portion of the recommended literature before the course starts.