National Centre for Research Methods

This online course introduces innovative sampling methods that are intuitive and feasible to apply. Rather than sampling from a predefined list, which is often not available, we sample from network connections instead, even if the connections are not known in advance. 

Whether you are a statistician, data scientist, biologist, or sociologist, you may find yourself studying data that has an inherent network structure. Have you ever opened a standard textbook on sampling, but could not find any methods that seem to help you? Or have you wondered how your data might be generalised to the real-world population it represents?

For example, you may be interested in social networks, but you are not sure how to collect the data in a way that capitalises on the links among the individuals. Or, during an epidemic, one knows one must track the people in contact with the confirmed cases, but you may have wondered what one needs to do to be able to estimate the case prevalence at the same time. For a third example, you may have heard that Google calculates PageRank based on randomly crawling the hyperlinks from one webpage to another, but you might want other measures of connectivity, such as small cliques of webpages with mutual hyperlinks to each other. Or perhaps you work with spatial data and are used to treat space continuously. Here, it is also possible to introduce a network structure among all the actual locations and represent the spatial data on a graph. 

If your problem resembles any of these examples, this is the course that will help you. Unlike the standard sampling methods, where units are selected from a list at each stage, we will introduce you to novel approaches where the sampling is based on the connections among the units, i.e. network-driven, even when the connections are not known in advance. You will be introduced to many such sampling methods as well as the unifying ideas that enable them, which have applications to various socio-economic, technical, biological or environmental phenomena.

Each method will be explained using small datasets that can be easily visualised so that the level of abstraction is kept to a minimum, and the accompanying computer workshops shall allow one to explore the underlying ideas directly through numerical results without having to labour with the relevant mathematical formulae and equations. 

The course covers the following topics, organised across 5 sessions.

• Multiplicity sampling, such as indirect sampling or line intercept sampling. Some examples: customer-to-product transaction data, sampling siblings via households for gene study, line-in-space transecting animal-tracks for the purpose of wildlife conservation.

• Adaptive network sampling, where a connection is sampled only if it satisfies a threshold condition. Some examples: persons-in-contact during an epidemic, contiguous cells-in-space in species abundance study.

• Probability snowball sampling, where the sample grows wave-by-wave by the new connections encountered. Some examples: sampling friendship networks, actor-actor co-casting for equitable professionalism. 

• Walk sampling by random, targeted and lagged walks. Some examples: Google PageRank, efficient sampling of spatial objects, dynamic monitoring of a specific subpopulation.

• Graph sampling as the unifying framework, where the interest may be any specific subgraphs such as nodes, edges, triangles, squares, K-stars. Generally applicable sampling strategies.

By the end of the course participants will:

• Have been introduced to a range of network-driven sampling methods
• Have obtained first-hand experience with various associated probability-based estimators 
• Be aware of the unifying ideas of graph sampling strategies applicable to a variety of situations

THIS COURSE IS BEING TAUGHT EVERY THURSDAY OVER 5 WEEKS FROM 18 FEBRUARY - 18 MARCH (10:00-12:00, hour break 13:00-14:00) AND EQUATES TO TWO TEACHING DAYS FOR PAYMENT PURPOSES.

The size, structure and dynamics of the population in which we live provide some of the most fundamental facts about our society. This course aims to give participants a practical introduction to the techniques of demography, including methods for dealing with each of the components of population change (mortality, fertility, and migration). The course will include hands-on computer workshops using the R programming language.

This course is intended as an introduction for and a companion to a related NCRM course Population Estimation and Forecasting, which will run from 7-9 April 2027. Participants may choose to attend one or both courses as their prior experience and time allows. 

The course covers: 

• Introduction to demographic measurement

• Sources of demographic data

• The life table; stable population theory

• Approaches for dealing with unreliable data

• The analysis of fertility 

• Migration estimation

By the end of the course participants will:

• Understand the basics of demographic measurement and the key sources of demographic data
• Understand and employ lifetables and the Kaplan-Meier estimator
• Understand and employ methods for the analysis of migration
• Be able to apply these approaches to a range of demographic data across different contexts and at different scales

Pre-requisites

This practical course will include workshops in the R programming language. Basic knowledge of R is a requirement. If you do not currently use R, but would still like to take the course, pre-course material will provide links to free introductory online courses to aid preparation. 

Target Audience

• The  course is of interest to anybody with an interest in quantitative methods from across disciplines (e.g. social sciences, economists etc), sectors and from across the career life-course (student, junior or senior researchers). 

The  course is particularly suited to:  

• Analysts working at or with, for example, National Statistical Offices, International Non-Governmental Organisations, and Government Departments with an interest in population estimation and projection. Examples include the ONS, the DHCLG; local and devolved government bodies; UN agencies. 

• Employees of private sector organisations who have an interest in population change:

  Consumer-focussed industries; 

  Retailers

  Market research agencies

  Utility companies

  Construction companies

• PhD students working on population-related topics including demography, population health, human geography, etc.

THIS COURSE IS BEING DELIVERED IN-PERSON AT THE UNIVERSITY OF SOUTHAMPTON (BUILDING 39, ROOM 2015) AND WILL RUN FROM 10:00 to 17:00 ON 5 AND 6 APRIL. IT EQUATES TO TWO TEACHING DAYS FOR PAYMENT PURPOSES.

An introduction to machine learning using Scikit Learn in Python.  Learners will be introduced to several machine learning techniques including regression, clustering, dimensionality reduction, and neural networks.  The course also includes a brief overview of the ethics and implications of machine learning.

The course covers:

• Introduction to machine learning
• Regression
• Introducing Scikit Learn
• Clustering with Scikit Learn
• Dimensionality reduction
• Neural networks
• Ethics and implications of machine learning 

By the end of the course participants will:

• Gain an overview of what machine learning is and the techniques available.
• Understand how machine learning and artificial intelligence differ.
• Be aware of some caveats when using Machine Learning.
• Apply linear regression with Scikit-Learn to create a model.
• Measure the error between a regression model and input data.
• Analyse and assess the accuracy of a linear model using Scikit-Learn’s metrics library.
• Understand how more complex models can be built with non-linear equations.
• Apply polynomial modelling to non-linear data using Scikit-Learn.
• Use two different supervised methods to classify data.
• Learn about the concept of hyper-parameters.
• Learn to validate and cross-validate models
• Understand the difference between supervised and unsupervised learning
• Identify clusters in data using k-means clustering.
• Understand the limitations of k-means when clusters overlap.
• Use spectral clustering to overcome the limitations of k-means.
• Recall that most data is inherently multidimensional.
• Understand that reducing the number of dimensions can simplify modelling and allow classifications to be performed.
• Apply Principle Component Analysis (PCA) and t-distributed Stochastic Neighbor Embedding (t-SNE) to reduce the dimensions of data.
• Evaluate the relative peformance of PCA and t-SNE in reducing data dimensionality.
• Understand the basic architecture of a perceptron.
• Be able to create a perceptron to encode a simple function.
• Understand that layers of perceptrons allow non-linear separable problems to be solved.
• Train a multi-layer perceptron using Scikit-Learn.
• Evaluate the accuracy of a multi-layer perceptron using real input data.
• Understand that cross validation allows the entire data set to be used in the training process.
• Consider the ethical implications of machine learning, in general, and in research.

IMPORTANT: Please note that this course includes computer workshops. Before registering please check that you will be able to access the software noted below. Please bear in mind minimum system requirements to run software and administration restrictions imposed by your institution or employer with may block the installation of software.

Pre-requisites:

Participants will need a basic understanding of Python. You will need to know how to write a for loop, if statement, use functions, libraries and perform basic arithmetic. 

Payment using the Online Store can only be completed via Visa and Mastercard Credit/Debit Card or PayPal.  AMEX is not accepted.
If you have not previously created an account for the Online Store, you will need to create an account to make a booking.

Within current contexts of heightened political polarisation, evolving digital environments and growing scrutiny of research on contentious issues, social science researchers are increasingly required to navigate complex safeguarding considerations. Exposure to distressing and sometimes hateful content can be emotionally taxing and can impact on researchers’ mental health. Researchers who work on sensitive and potentially ‘controversial’ subjects have to contend with the risks of doxxing and online/offline harassment, especially when choosing to become professionally visible in/on (social) media. Risks are significantly heightened for minoritised and marginalised researchers, including female(-presenting), gender non-conforming, neurodiverse, disabled, LGBTQ+ and racialised researchers.

The aim of this course is to support participants in understanding and addressing the ethical, safeguarding, emotional and psychological dimensions of their work, as well as the challenges that come with safely navigating professional visibility – paying particular attention to how risks intersect with identity and career stage. The course will further provide participants with practical guidance how to protect themselves both online and offline, how to create and establish support networks and communities, and to engage constructively with their institutions, including ethical review processes. We will further discuss how researchers can look after their wellbeing, set boundaries and engage in self-care whilst doing high risk and sensitive research.

Please note that this course will be facilitated on Microsoft Teams and will run over two consecutive mornings with the afternoon of Day 1 being used for an individual task to feed back to the group on Day 2. 

THE COURSE EQUATES TO ONE TEACHING DAY FOR PAYMENT PURPOSES.

At any time horizon, policy depends upon estimates of population size and structure. Without population forecasts, planning for pension spending, housing demand, healthcare provision, and school place requirements becomes impossible. To know what the size and structure of our population will look like in the future, we need to have the tools to make demographic projections confidently. 

This course introduces the methods for forecasting the key components of population change, dealing in turn with mortality, fertility and migration. The course also includes an introduction to matrix models for multi-population forecasting is provided, together with a discussion of how to validate and assess uncertainty in demographic projections.

The course will include hands-on computer workshops using the R programming language.

This course is intended to follow on from a related NCRM course Introduction to Demographic Methods, which will run on 5-6 April 2027. Participants may choose to attend one or both courses as their prior experience and time allows. 

The course covers: 

• The cohort component projection model

• Matrix population models 

• Principal component methods for population estimation and forecasting

• Validating and assessing uncertainty in population forecasts

By the end of the course participants will:

• Appreciate contemporary issues in demographic forecasting
• Be able to produce projections of each of the demographic components (mortality, fertility, and migration)
• Produce full population projections using the cohort component model
• Use matrix models for multi-population projection

Pre-requisites

This practical course will include workshops in the R programming language. Basic knowledge of R is a requirement. If you do not currently use R, but would still like to take the course, pre-course material will provide links to free introductory online courses to aid preparation. 

Knowledge of basic demographic techniques is recommended, including: 

• The distinction between rates and probabilities in a demographic context
• The life table
• Basic fertility measures

Completion of the companion course Introduction to Demographic Methods would be advantageous but is not required.

Target Audience

• The  course is of interest to anybody with an interest in quantitative methods from across disciplines (e.g. social sciences, economists etc), sectors and from across the career life-course (student, junior or senior researchers). 

The  course is particularly suited to:  

• Analysts working at or with, for example, National Statistical Offices, International Non-Governmental Organisations, and Government Departments with an interest in population estimation and projection. Examples include the ONS, the DHCLG; local and devolved government bodies; UN agencies. 

• Employees of private sector organisations who have an interest in population change:

  Consumer-focussed industries; 

  Retailers

  Market research agencies

  Utility companies

  Construction companies

• PhD students working on population-related topics including demography, population health, human geography, etc.

THIS COURSE IS BEING DELIVERED IN-PERSON AT THE UNIVERSITY OF SOUTHAMPTON (BUILDING 39, ROOM 2015) AND WILL RUN FROM 10:00 to 17:00 ON 5 AND 6 APRIL. IT EQUATES TO TWO TEACHING DAYS FOR PAYMENT PURPOSES.

Qualitative Diary Methods (QDMs) are increasingly recognised as a valuable and important method in social science research, due to concern across disciplines with an overreliance on cross-sectional research, a lack of focus on temporality, and the need to capture evolving processes and the daily dynamics of phenomena. 

QDMs offer a range of innovative approaches and tools for social science researchers that enable us to capture and subsequently begin to understand, how phenomena are experienced in the moment, as well as how they evolve over time. However, they remain a methodological blindspot in much postgraduate research training.

This workshop will provide researchers with a new range of methods to add to their methodological toolkit, as well as support and guidance in managing some of the challenges associated with these methods, including insights into qualitative diary (longitudinal) analysis approaches. 

The course covers: 

• Intro to Qualitative Diary Methods (QDMs) 

• When and why to use different qualitative diary methods

• Addressing challenges associated with QDMs

• Ethical considerations when using QDMs

• Designing pen and paper diaries

• Designing audio and video diaries

• Designing photo-elicitation diaries

• Designing qualitative ‘App’ diaries

• Analysing data from QDMs: Moving beyond thematic analysis

By the end of the course participants will:

• Have a comprehensive understanding of the benefits and challenges of QDMs, and how to manage these challenges
• Be able to evaluate when different QDMs should be used and why
• Be able to design appropriate QDM research projects to address particular research questions
• Have a critical awareness of particular ethical issues associated with the use of QDMs
• Have a comprehensive understanding of analytical approaches suited to data gathered using QDMs

Please note that this is a full day event running from 10:30am to 4:30pm.

This online course on questionnaire design, explores question wording issues and the questionnaire as a whole with a focus on web surveys and mobile-friendly web surveys. The course is full of practical advice. It also provides tips for anyone moving from interviewer-administered surveys to web surveys. Mirroring in-person training, there course will be interactive. There will also be 6 small group workshops to facilitate putting the course concepts into practice.

Questionnaire Design

The popularity of web surveys
Getting started with a new questionnaire
Trade-offs – short and simple versus clearThe popularity of web surveys
Four cognitive stages a respondent goes through in answering a survey question
Solutions to ambiguous term, understanding recall error and reducing question sensitivity
Question wording guidelines - This about the do's and don'ts of writing survey questions for any context.
Workshop 1 - Critiquing a survey question
Some additional issues with factual questions
Highlights from mini appendix: Demographic questions are always the most difficult to write
Workshop 2 - Writing a survey question
Mini appendix on actual versus usual behaviour
Highlights from mini appendix on some additional issues with subjective questions
Know the deeper issues with open and closed questions
Problematic question formats to be aware of or avoid (eg. agree / disagree)
Mini appendix on other problematic formats (satisfaction, tick all that apply, ranking and hypothetical questions)

Web surveys

Don't rely on web survey software templates
Workshop 3 - Critiquing web survey software templates
Modes of quantitative data collection

Modes of quatitative data collection: Mixing modes
Modes of quantitative data collection: Overall mode differences (the obvious ones)
Mini Appendix on mode effects due to satisficing
Workshop 4 - Interpreting data from a mixed mode experiment

Back to web surveys

Designing the web survey itself
Day 2 appendix - 8 question testing methods for web surveys
The special things that web surveys can do, but should we?
Visual versus not visual
Mini appendix on tips for paper self-completion
Workshop 5 - Visual problems
Web surveys for mobile phones - earlier evidence, current thinking
Highlights from mini appendix on data collection differences: What should you do?
Mini appendix on "push to web"

Back to questionnaire design

Examples of question revisions based on testing results
Workshop 6 - Revising survey questions
Highlights from mini appendix on extra tasks on mobile phones 
By the end of the course participants will:

Have greater questionnaire design skills in general and the ability to critique existing web survey software templates
Have the ability to create effective web survey questionnaires as well as mobile-friendly ones
Have better knowledge about questionnaire-related mode differences and effects
This course is for anyone interested in questionnaire design for web and mobile web surveys. Ideally participants need some familiarity with surveys and questionnaire design.

Preparatory Reading (desirable):

Link 1

Link 2

PLEASE NOTE THIS COURSE IS TAUGHT OVER THREE DAYS (10:00-15:00), AND EQUATES TO TWO TEACHING DAYS FOR PAYMENT PURPOSES.

Payment using the Online Store can only be completed via Visa and Mastercard Credit/Debit Card or PayPal.  AMEX is not accepted.
If you have not previously created an account for the Online Store, you will need to create an account to make a booking.

Why do our friends often know one another, even though we are only a few connections away from people in very different social circles? Why does information spread easily through some networks, while changes in behaviour require repeated encouragement? Who should be targeted in an intervention, and how does the answer depend on what we want to achieve?

Social network analysis offers a theoretical framework and a methodology to answer questions like these by focusing on the relationships between and among social entities. A network is a structure composed of units and the relationships that connect them. The position of these units and the overall structure help us understand how social ties form, how information and behaviour spread, and how individuals gain access to resources and opportunities.

The focus of the course is on the connection between the mathematical foundations of networks and the social phenomena we want to understand. We will introduce concepts from matrix algebra and graph theory across both days, and use empirical examples and practical exercises to examine what these concepts tell us about social life. We will also discuss how network data are collected and what evidence is needed to distinguish between different explanations for the patterns we observe.

The course covers:

• Mathematical foundations of social networks: matrix algebra and graph theory;

• Network structure, cohesion and models of network formation;

• Centrality, betweenness and brokerage;

• Contagion, diffusion and network interventions;

• Network data, sampling and research design

Learning outcomes:

• Develop an understanding of social phenomena through social network theory;
• Learn how to use and interpret graph-theoretic and matrix algebra concepts with real-world data;
• Understand how local patterns of connection contribute to the overall structure of a network;
• Examine how network position and structure affect social processes;
• Acquire knowledge of network data collection and the implications of sampling and missing data;
• Assess competing explanations for observed network patterns and identify the evidence needed to investigate them.

The course is aimed researchers and analysts interested in using social network analysis to understand social phenomena, including students, academics, government researchers and researchers in third sector organisations. The course is relevant to work in sociology, survey research, public health and related disciplines.

Pre-requisites

Basic familiarity with data matrices and introductory statistical concepts. No previous training in social network analysis is required. Mathematical concepts will be introduced as they are used throughout the course. Practical exercises and demonstrations will accompany the theoretical sessions. Software requirements will be circulated before the course.

THIS COURSE IS BEING DELIVERED OVER TWO DAYS.

The course aims to provide participants with an understanding of the key issues involved in the design, implementation and analysis of sample surveys, with particular emphasis on sample design, data collection procedures, non-response, questionnaire design and testing, and the handling of missing data. Participants will gain practical insights into survey design and management, as well as the strengths and limitations of probability-based and non-probability-based surveys.

The course is designed for a broad range of professionals, including data analysts working across different disciplinary fields, government analysts, policy professionals, survey designers, researchers and PhD students. It will be particularly valuable for anyone involved in analysing survey data, designing or commissioning surveys, interpreting survey findings, or developing policies and making decisions that rely on statistical analysis of survey data. The course will help participants critically assess survey quality, understand the limitations of survey evidence and make more informed use of survey data in research, policy development and practice.

The course covers: 

• Total Survey Error Framework

• Survey sampling

• Survey non-response

• Probability-based vs non-probability-based surveys

• Survey mode

• Questionnaire design and testing

By the end of the course participants will:

• Demonstrate understanding of the key principles and practices involved in survey design, implementation and data analysis.
• Explain and evaluate the principles of questionnaire design and testing, drawing on relevant survey methodology literature and best practice.
• Identify, describe and critically evaluate a range of sampling designs, including probability-based and non-probability-based surveys, and understand their respective strengths and limitations.
• Identify and evaluate potential sources of survey error, including issues relating to sampling, non-response and data collection, and understand approaches to assessing and minimising their impact on survey quality.
• Demonstrate understanding of approaches to identifying and handling missing data in survey analysis, and assess their implications for the interpretation of survey findings.
• Critically assess the quality, limitations and suitability of survey data for particular research questions, policy development and decision-making.
• Interpret survey findings critically and understand the implications of survey design and data quality for statistical analysis and evidence-based policy.

Target Audience

The course is designed for a broad range of professionals, including data analysts working across different disciplinary fields, government analysts, policy professionals, survey designers, researchers and PhD students. It will be particularly valuable for anyone involved in analysing survey data, designing or commissioning surveys, interpreting survey findings, or developing policies and making decisions that rely on statistical analysis of survey data.

THIS COURSE WILL TAKE PLACE ON 24 MAY (10:00-17:00), 25 MAY (9:00-17:00) AND 26 MAY (09:00-16:00) AND EQUATES TO THREE TEACHING DAYS FOR PAYMENT PURPOSES.