This module will not be running in the academic year 2022-23.
As individuals and scholars, we are frequently confronted with the claim that A causes B, or the requirement to verify whether a relationship between A and B exists. While anecdotal accounts can help inform our opinion, it is dangerous to rely on one-off observations to verify more general relationships. Equally, even where we believe there may be a relationship, we can be misled by confounding influences which obscure or mislead us.
This is where quantitative approaches can help us untangle the relationships we observe around us, and help us answer question of whether these relationships hold in the wider population. The skills acquired on this module will be invaluable for any student wishing to pursue research which involves large numbers of participants, or which involves the analysis of datasets from official sources.
This module uses an innovative Problem-Based Learning approach to teaching intermediate quantitative concepts which promotes self-directed and reflective learning. Through tackling multifaceted and complex social issues, students will begin to generate, appreciate and understand broader, underlying, conceptual problems around why quantitative approaches are relevant, and to uncover the appropriate methodologies. In addition to improved learning outcomes, this approach also aligns learning activities with the processes of independent research — effectively preparing students for independent project work or modules which encourage individual enquiry.
Through group discussion and research, students will begin building their knowledge and confidence in plotting and estimating bivariate relationships, uncover the core technical approaches we use for this, and the conditions under which these approaches are appropriate. They will build on existing knowledge of distributions to study the principles of hypothesis testing to understand how we can use results based on a sample to make inferences about the wider population. As the course develops, problems will move toward the requirement to understand more complex multivariate relationships, the importance of control variables in reducing 'noise' in our models, and finally extensions which allow us to use our frameworks to plot non-linear relationships.
The combination of PBL discussion classes, and practical workshops will build students' confidence at using statistical computer packages to put into practice the concepts they uncover through their research, and to take their first steps in statistically modelling the relationship between two or more variables. This module builds on the introductory understanding students acquired in IP110 Quantitative Methods for Undergraduate Research.
Principal Learning Outcomes
On completion of this course, students will be able to:
- Demonstrate an understanding and usefulness of the key concepts in describing relationships in data, the meaning and of descriptive statistics used to describe such relationships, and the generation and application of such statistics using real-world data.
- Use regression analysis to evaluate linear bi-variate relationships in real-world data, understand issues in data and its collection which impact on the analysis of relationships, and demonstrate an understanding of the conditions under which using this approach is appropriate and how we verify that these conditions are met.
- Demonstrate an understanding of how estimated relationships based upon sample data can be used to make inferences about relationships in the wider population, and the associated principles of distributions and hypothesis testing.
- Employ multivariate regression techniques to investigate real-world data to evaluate linear and non-linear relationships, and demonstrate understanding of the conditions under which such approaches are appropriate.
- Use statistical computer software packages to manage data and perform data analysis tasks.
This is an indicative module outline only to give an indication of the sort of topics that may be covered. Actual sessions held may differ.
The following outline represents the core knowledge and competency gain associated with the course activities. In order to facilitate the acquisition of knowledge and competency, the course is delivered via Problem-Based Learning which emphasizes student knowledge acquisition via efforts to understand the parameters of, and solutions to, complex real-world situations. Study will encourage engagement in contemporary challenges across themes such as education, deprivation, social statistics, climate and environment, and health and wellbeing.. The course will be based around four key conceptual problems around the use of quantitative research to elicit relationships:
- Why employ quantitative approaches to relationships, and why not? Which will allow students to develop knowledge and understanding around:
- Core statistical concepts used to describe relationships
- Differences between quantitative and qualitative approaches and their respective strengths
- Why correlation isn't causality
- How can we model the strength of a relationship? Which will allow students to develop knowledge and understanding around:
- Core principles of linear regression modelling
- Estimating straight line relationships using ordinary least squared (OLS) approaches
- The required assumptions of ordinary least squared modelling and what happens when we ignore them
- Using regression analysis of data samples to make inferences about the population
- How can we deal with complex multifaceted relationships? Which will allow students to develop knowledge and understanding around:
- Control variables, confounding influences, and the usefulness (and shortcomings) of multivariate modelling
- Estimating linear relationships using OLS when we have many variables
- The required assumptions of multivariate ordinary least squared modelling and what happens when we ignore them
- Using multivariate regression analysis of data samples to make inferences about the population
- What happens when we don't have a linear relationship? Which will allow students to develop knowledge and understanding around:
- The role of interactions in data
- Inclusion of data transformations in regression models and what they mean
- Interpreting binary variables in the context of statistical models
The case study structure of the course means that each problem will be associated with in-depth news articles, data, academic journal articles, and other source materials relating to a contemporary issue (possible themes include the study of issues and relationships in education, deprivation and social exclusion, social justice, climate and environment, and health and wellbeing).
It is essential to the course that students have access to one or more well-chosen technical references in order to assist in the technical side of their learning. Such technical sources might include (or be similar to):
Angrist, J.D, and Pischke, J.-S. Mostly Harmless Econometrics: An Empiricist's Companion. Princeton University Press, 2009
Gujarati, D. Econometrics by Example. 2nd Edition. MacMillan Higher Education, 2014
Contextual sources will be confirmed based upon relevant and current topics used as examples, and may include both book chapters and academic journal articles
1 x 1,250-word technical report (30%)
1 x 15-minute in-class group presentation (15%)
1 x Group technical report (15%)
|1 x 90-minute computer-based exam (40%) (questions seen 7 days before the test)|
- Skills in data analysis using a range of methods
- Skills in using statistical computer software packages to manage data and perform data analysis tasks
- Problem solving
- Information technology
- Oral and written communication
- Digital literacy
Through this module, you will develop different skills that are sought by employers which will support your professional development. We have highlighted this to enable you to identify and reflect on the skills you have acquired and apply them throughout your professional journey including during the recruitment process whether this is in a CV/application form or at an interview.
- Data analysis: Qualitative and quantitative analysis techniques and evaluation methods using tools such as Excel, STATA.
- Teamwork: Collaborating with peers and multiple partners on project briefs involving sharing ideas, knowledge and best practice.
- Time and self-management: Developed through planning and managing weekly tasks, working towards agreed group schedules, as well as on your own initiative without supervision.
- Research: Developed through carrying out research using samples from real-world data involving formulating research questions, conducting literature reviews, identifying appropriate measurement variables, and analysing and interpreting results. Additionally, writing and communicating the research in an appropriate manner.
Please note: Module availability and staffing may change year on year depending on availability and other operational factors. The School for Cross-faculty Studies makes no guarantee that any modules will be offered in a particular year, or that they will necessarily be taught by the staff listed on these pages.