Not the second choice: supporting students to do secondary analysis in their dissertations

Jen Buckley Jen Buckley highlights some of the tools the User Support and Training team at the UK Data Service have been developing to make it easier for students to do secondary data analysis as part of their dissertations or projects.

 

Doing a dissertation or project can be a challenging and rewarding experience for undergraduate students. At this culmination of their studies and opportunity to experience research, secondary analysis can be a great option. However, from our Higher Education Teaching Consultation 2021, plus our own teaching experience, we know students may not consider or reject the idea of secondary analysis for their projects.

Why secondary analysis?

Students can access most data available via the UK Data Service using their university or college login. This relatively easy and free access to data gives students opportunities to do research projects they either could not or would struggle to achieve with primary research.

One of the most notable opportunities is the option to access data from large nationally representative social surveys. For instance, data from the annual British Social Attitudes (BSA) gives topical information on the attitudes of the British public towards a wide range of social issues. Or there is the Crime Survey for England and Wales (CSEW), which can be used to explore experiences of crime and attitudes towards the police and the criminal justice system. These are just two examples from a wide range of social surveys, a range that means there is a good chance students can find nationally representative data relevant to their interests.

Most social surveys are repeated regularly, giving the chance to examine different time points. There are also longitudinal studies such as the Millennium Cohort Study (MCS), which is following a representative sample of people born 2000-02.

Other forms of data offer opportunities too. Students interested in the characteristics of areas can use aggregate data from the UK Census or socio-economic time series data aggregated to a country or regional level. And while secondary analysis of quantitative data might be more common, qualitative secondary analysis offers students opportunities to explore data at a scale or with groups that would be out of the reach of a dissertation project.

Why not?

From both consultations with lectures and our dissertation award, we know that many students do dissertations based on secondary analysis. However, those seeking to promote secondary analysis, can find resistance among students.

‘I want to collect my own data’

Some students are attracted to the idea of collecting their own data, which is understandable when they spend time learning about data collection. Some lecturers also talked about a preference for qualitative research, where secondary analysis is less common in general.

Clearly some research questions will require primary data collection. However, in some cases, a student’s preference for collecting their own data might mean they do not adopt the best approach. For instance, a student doing their own survey can struggle with sample size and bias – all after having spent time designing and conducting their surveys.

‘It will affect my marks’

A further reason that emerged in our discussions with lecturers is that students can be unaware that secondary analysis is an option at all or that it is equal to primary research.  For instance, in relation to this problem, one lecturer described how they need to

“overcome the barrier of students worrying it will not be as good to use existing data or that they will be marked down for not collecting data when the opposite is probably true”.

‘I can’t find any data’

A further – more practical – barrier to secondary analysis are challenges students can encounter during the process. For instance, in the early stages, students need to find relevant data for their project. This step can be difficult at the best of times but especially difficult for anyone unfamiliar with relevant data sources or who are less clear about their research questions or the data they need.

‘It’s hard to understand the data’

Secondary analysis requires students to understand datasets that are often quite complex. And to understand these datasets, they need to navigate documentation that can be far from user friendly. Then, once they do find data, students can face a large amount of work to prepare the data for analysis.

Supporting increased student use of secondary data

The UK Data Service appreciates its role in supporting students to access and use data for research projects.

Dissertation Award

Starting in 2016, our dissertation award aims to promote and recognise dissertations based on secondary analysis. Each year, we review some excellent dissertations based on the analysis of data from the UK Data Service. Most entries use data from one of the national or cross-national social surveys such as Understanding Society and the Labour Force Survey. Notable examples include students comparing data before and after the pandemic to understand the impact of social environment and applying for special license data to link survey data to other sources. But overall, it is the variety of data and research questions that is impressive as well as the students who clearly articulate why the data they have used was the best option for answering their research question.

New student resources

From last year’s consultation with higher education teaching staff, repeated messages were the need to get students to think of secondary analysis as an option in the first place and help them find and access relevant data. In response to these message, we have recently launched a new student section in our Learning Hub.

The section is designed to connect students to all our student relevant information and resources. There is basic information about the data that is available and students can watch a video talking about secondary analysis, find out about our dissertation award, and access dissertation case studies.

To tackle the challenges around students knowing what data is available – and how they might find and use it – we created a range of new dissertation resources. Using survey data showcases social surveys and how survey data might be used including example research questions and datasets.

New quick start guides have been designed to help students get familiar with the most student friendly surveys. We have started with  British Social Attitudes (BSA) and the Health Survey for England (HSE) but more will be added by the end of the year. These pages resulted from specific suggestions for short orientation guides to the major surveys likely to be used in dissertations.

Finding and accessing data for your project is a resource to help with the challenge of finding relevant data for a project. It includes information, videos, and worksheets, covering topics such as searching for data, evaluating the usefulness of data, and understanding data access conditions.

What’s next

We hope these resources, which are freely available on our website with a Creative Commons License, help those supervising and doing dissertations. Videos and worksheets can be accessed directly from our site or embedded in existing dissertation resources.

But we also think there is more we can do. At present, we are looking at three further areas of work.

First, we are seeking to develop a broader range of dissertation case studies that include student voices and present secondary analysis as something achievable to all students.

We are looking at extending our resources that help students identify relevant data for their project through resources that introduce data for research on popular topics.

And to help those that do take the plunge, there will be resources on common challenges involved in using data such as practical guides for dealing with missing data or managing large files.

But also, we want the conversation to continue and to hear from lectures and students what they think can help future students. If you have any thoughts, ideas or queries please email us at ukdstraining@manchester.ac.uk.


About the Author

Dr Jennifer Buckley is a Research Associate at the University of Manchester and part of User Support and Training for the UK Data Service. She develops training to support researchers and teachers with a special interest in learning resources that support the use of data in undergraduate social science teaching.

Leave a Reply

Your email address will not be published. Required fields are marked *