20659 - DATA ANALYSIS FOR MANAGERIAL DECISION MAKING
Course taught in English
Go to class group/s: 31
GIADA DI STEFANO
Class 31: GIADA DI STEFANO
Nowadays the use of data across firms is pervasive: A recent survey by PwC of more than 2,100 executives reveals that most of them consider their organization as either highly data-driven (39%) or somewhat data-driven (53%). Interestingly, these companies claim to use analytics mostly for descriptive and diagnostic purposes, rather than for predictive and prescriptive purposes. Being able to predict and prescribe may require managers to get their hands dirty, and collect themselves the data they need to answer the questions they have in mind. To put remedy to the passive use of data analytics inside firms, it is necessary to learn how to craft a research question, design a study, collect the data, and analyze them. This is exactly the spirit of our course.
The main goal of our course is to provide students with a comprehensive understanding of research methods based on primary data, i.e. data collected first-hand for a specific purpose. We would like to focus our attention on both qualitative (observation, interviews) and quantitative (surveys, experiments) data. In particular, we discuss:
- Statement of a problem / Framing a hypothesis.
- Research Design / Primary vs. Secondary data / Causality / Issues of validity and reliability.
Why should managers go qualitative? Introduction and overview of the method.
Research design – Role of the researcher, ethical considerations, sample selection, etc..
Data analysis – What we can learn from qualitative data, issues of generalizability, etc..
Introduction to NVivo.
Why should managers run surveys? Introduction and overview of the method.
Research design – Measurement, scales, etc..
Data analysis – Sampling, non-response, single method, etc..
Introduction to Qualtrics.
Why should managers run experiments? Introduction and overview of the method.
Research design – Threats to internal validity, manipulation checks, noncompliance issues, experimental mortality, interference between experimental units.
Designing an experiment.
Analyzing experimental data.
Group project presentations.
- Know what primary data are.
- Know the main terminology and concepts associated to the research methods used to deal with such data.
- Know the strengths and limitations of each method.
- Know the statistical techniques and software used to analyze different types of primary data.
- Formulate a research question.
- Choose the most appropriate method to analyze the research question at hand.
- Design an efficient protocol that avoids the most common pitfalls.
- Analyze the collected data with the most appropriate techniques.
- Present their results in an effective way in both written and oral form.
- Face-to-face lectures
- Guest speaker's talks (in class or in distance)
- Exercises (exercises, database, software etc.)
- Case studies /Incidents (traditional, online)
- Group assignments
- Interactive class activities (role playing, business game, simulation, online forum, instant polls)
The course leverages a blend of methods aimed at complementing each other and optimizing the learning experience:
- Lectures are used to discuss the theoretical and technical aspects associated to the collection and analysis of different types of primary data. During such lectures, students also have the chance to work with case studies, interactive class activities, as well as short individual and group exercises that help them understand the peculiarities associated with each type of data.
- Practice sessions provide students with a hands-on experience of the research methods we discuss in class. Those practice sessions focus on issues related to both research design and data analysis.
- Guest lectures expose students to the practices currently used in some firms.
- Finally, students also put their knowledge in practice by participating to a group project. This allows them to experience first-hand the challenges associated with designing a qualitative data collection, a survey, or an experiment.
|Continuous assessment||Partial exams||General exam|
Attending students are evaluated based on the following three criteria:
- In-class participation (10% of final grade) aimed to test the students’ ability to interact in a constructive way and present their points of view in an effective way.
- Group project (40% of final grade) aimed to test the students' ability to formulate a research question, choose the most appropriate research method to analyze the research question at hand, and design an efficient protocol to study it. Moreover, the group project allows to test students' ability to present their results in an effective way in both written and oral form.
- Written exam (50% of final grade). The exam includes both open- and close-ended questions, aimed to test students' knowledge of the main terminology and concepts associated to the research methods used to deal with primary data, the strengths and limitations of each method, as well as the statistical techniques and softwares used to analyze different types of primary data.
Non-attending students are evaluated based on a written exam that includes both open- and close-ended questions.
- The exam tests students' knowledge of the main terminology and concepts associated to the research methods used to deal with primary data, the strengths and limitations of each method, as well as the statistical techniques and softwares used to analyze different types of primary data.
- Moreover, through the open-ended questions, we intend to test the students' ability to formulate a research question, choose the most appropriate research method to analyze the research question at hand, and design an efficient protocol to study it.
Slides and articles distributed in class and posted on Bboard.