Stockholm School of Entrepreneurship (SSES) in Stockholm
The course is taught as a combination of physical and online meetings. The physical meetings are concentrated to a two-day session in Stockholm.
September 9th: online
September 22nd–23rd: SSES Office (10-17 on Thursday, 9-16 on Friday)
October 7th: online
October 27th-28th: online
ABOUT THE COURSE
This course provides an overview of quantitative research methods with a particular focus on applications in studies of entrepreneurship and innovation. A basic understanding of quantitative methods, including training in statistics and probability theory, is an expected prerequisite.
Participants are given an opportunity to broaden and deepen their understanding of what tools are available for common research problems, with an emphasis on specification, estimation, interpretation and valuation of various forms of mathematically formulated models. The overall ambition is for course participants to gain a useful overview of how data can be analysed using econometric models and machine learning tools, with emphasis on methods that are used in contemporary state-of-the-art research on entrepreneurship and innovation.
The central sections of the course are constituted by presentations and exercises introducing important concepts and techniques. Prioritization of what techniques are treated at greater depth is to some extent determined by participants’ interests and needs.
After having completed the course, participants should gain the ability to:
- identify appropriate quantitative methods in addressing different types of research questions
- make informed methods choices given the available data and the data-generating process
- assess methodological choices in contemporary quantitative research papers
- conduct research on entrepreneurship and innovation using quantitative methods
The course is taught in modules. The first four modules are structured around lectures and exercises. In the last third of the course, the content will be prioritized according to participant demand and interests. These ’on demand’ modules are run in seminar form.
Each module contains a combination of lectures and assignment work. Exercises held during the course are designed to help participants develop the necessary expertise for the final, summative assessment. In lectures, the statistical software STATA (latest edition) will be used to illustrate and demonstrate course content. When working with assignments, however, participants may choose to work in R or Python instead.
- What do we need quantitative methods for?
– Cause and effect
– Models as tools
– Moderation, mediation, etc
– Mixed methods approaches
- Econometric toolbox ”basics” (two classes)
– Regression models for binary outcomes (Logit, Probit, LPM)
– Regression models for censored and truncated data
– Regression models for multi-item discrete choice outcomes (multinomial logit, nested logit)
– Pooled Cross-sections and Panel Data Models
– Overview of complementary models (survival analysis and duration models, structural equation models, factor analysis)
- Towards causal analysis
– Granger causality
– Matching – Sample Selection and Heckman Models – Endogeneity and Instrumental Variables Estimation
- The language of prediction
– Training/evaluation sample split
– Machine learning techniques
MODULES ON DEMAND
- Methods for generating data
– Text analysis
– Survey methodology
– Experimental approaches
- Overview of data sources common for research on E&I
– International surveys
– Patent data
– Register data
- Data, law and ethics
– What data is sensitive (GDPR etc)?
– Obtaining ethical approval for quantitative research
– Storing data securely
- Research ethics in quantitative research
– The research replication crisis
Attendance is mandatory for all participants. The Course Director assesses if and how absence may be compensated.
Participants are expected to hand in four short assignments, corresponding to the four basic modules of the course. The first criteria for finishing the course is to have handed in four acceptable assignments.
The second and final part of the examination takes the form of a data hackathon, where participants do their best to turn information into knowledge. Participants will frame a research question, create and implement a research design, mobilize data resources and present findings in the form of a 3,000-word research paper. Participants may use a dataset that they intend to use in their own research.