Course Introduction
Interrupted time-series evaluation is a quasi-experimental approach used to assess whether a policy, programme, regulation, or institutional reform is associated with a meaningful change in an outcome over time. By analyzing repeated observations collected before and after an intervention, evaluators can examine whether the intervention coincides with a change in the outcome level, the underlying trend, or both. This course equips participants with the conceptual foundations and practical analytical skills needed to design, implement, and interpret interrupted time-series studies for public policy and development programmes. Participants will learn how to establish a credible pre-intervention baseline, identify the timing of an intervention, select appropriate outcome measures, and estimate changes beyond the patterns that would otherwise be expected.
Public policies often affect populations and institutions over extended periods, making it important to understand both immediate and longer-term changes. Examples include the introduction of a national health policy, a tax reform, a revised education curriculum, a transport regulation, an environmental standard, or a social protection programme. In these situations, randomized experiments may be impractical, politically infeasible, or ethically inappropriate, while routine administrative and monitoring data may provide valuable repeated observations. The course examines how interrupted time-series methods can use these data to investigate policy effects, identify changes in outcome trajectories, and support evidence-informed decision-making. Participants will also learn to recognize situations in which the method is unsuitable or requires additional design features to support credible conclusions.
A central component of the training is understanding the statistical structure of time-series data and the assumptions required for reliable intervention evaluation. Participants will explore secular trends, seasonality, autocorrelation, cyclical patterns, outliers, structural breaks, and changes in measurement practices. They will learn to distinguish a temporary fluctuation from a sustained change associated with an intervention and to formulate models that estimate immediate level changes and subsequent trend changes. The course covers segmented regression, model specification, time variables, intervention indicators, interaction terms, and the selection of appropriate error structures. Practical emphasis is placed on interpreting coefficients in substantive terms and understanding how analytical decisions affect estimates of policy impact.
The credibility of an interrupted time-series evaluation depends on more than the selection of a statistical model. Participants will examine competing explanations for observed changes, including concurrent policies, economic shocks, demographic shifts, reporting changes, seasonal influences, and other events occurring near the intervention date. They will learn how to assess the adequacy of pre-intervention observations, inspect graphical patterns, test model assumptions, examine residuals, and conduct sensitivity analyses. The course also introduces controlled interrupted time-series designs, in which outcomes from an appropriate comparison population or setting help account for external influences. Participants will understand how comparison series can strengthen causal interpretation while recognizing that differences between intervention and comparison settings may introduce additional analytical challenges.
The course integrates quantitative analysis with practical evaluation planning, data management, and policy interpretation. Participants will learn how to identify reliable data sources, assess observation frequency, manage missing values, document intervention dates, and construct analysis-ready datasets from administrative records, surveillance systems, surveys, and routine performance monitoring. They will explore methods for communicating uncertainty, assessing the practical importance of estimated changes, and distinguishing statistical evidence of an interruption from definitive proof that a policy caused the observed outcome. Ethical considerations, transparent reporting, reproducibility, and appropriate handling of sensitive information are addressed throughout the training to strengthen the integrity and usefulness of evaluation findings.
By the end of the course, participants will be able to design and conduct interrupted time-series evaluations that provide structured evidence on the performance of policies and development interventions. They will develop practical skills in exploratory time-series analysis, segmented regression, diagnostic testing, robustness assessment, and interpretation of intervention effects. Participants will also learn to prepare evaluation reports, visualizations, and policy briefs that communicate changes in outcome levels and trends to technical and nontechnical audiences. These competencies can strengthen policy learning, institutional accountability, programme adaptation, and resource allocation across public health, education, social protection, economic development, environmental management, and other sectors where repeated observations are available.
Duration
5 days
Who Should Attend
-
Monitoring and evaluation specialists responsible for assessing policy effectiveness and programme outcomes over time.
-
Public policy analysts evaluating reforms, regulations, government initiatives, and institutional interventions.
-
Economists and statisticians working with longitudinal, administrative, and routinely collected data.
-
Impact evaluation practitioners designing quasi-experimental studies where randomized assignment is unavailable.
-
Government planning officers responsible for performance measurement and evidence-informed policy review.
-
Public health researchers evaluating health policies, surveillance interventions, and service-delivery reforms.
-
Education specialists assessing curriculum reforms, school policies, and student performance interventions.
-
Social protection professionals evaluating cash transfers, eligibility reforms, and welfare policy changes.
-
Environmental and climate policy analysts examining regulatory changes and environmental outcome trends.
-
Transport and infrastructure analysts evaluating safety policies, operational reforms, and service performance.
-
Development consultants conducting policy evaluations and institutional performance assessments.
-
Research officers responsible for preparing datasets and statistical analyses for policy studies.
-
Donor representatives and programme managers reviewing evidence of intervention effectiveness.
-
Data analysts using statistical software to investigate changes in indicators over time.
-
Academic researchers and postgraduate students specializing in applied statistics, econometrics, and impact evaluation.
Course Objectives
-
Explain the principles of interrupted time-series evaluation and its application to estimating policy effects using repeated observations collected before and after an intervention.
-
Formulate evaluation questions, define outcome indicators, identify intervention dates, establish observation periods, and determine the level of analysis required for credible assessment.
-
Prepare time-series datasets by examining observation frequency, missing values, outliers, measurement changes, seasonality, and the consistency of pre- and post-intervention records.
-
Apply segmented regression to estimate immediate changes in outcome levels and changes in outcome trends following a policy or programme intervention.
-
Diagnose autocorrelation, seasonality, secular trends, residual patterns, and other time-series characteristics that may affect statistical inference and model reliability.
-
Assess competing explanations for observed changes, including concurrent interventions, external shocks, policy anticipation, demographic changes, and alterations in data collection practices.
-
Design controlled interrupted time-series evaluations using appropriate comparison groups or outcome series to strengthen causal interpretation and address common external influences.
-
Conduct sensitivity analyses and robustness checks using alternative intervention dates, model specifications, time windows, seasonal adjustments, and appropriate error structures.
-
Interpret effect estimates, uncertainty intervals, statistical significance, practical relevance, and limitations without overstating the causal conclusions supported by the evaluation design.
-
Communicate findings through reproducible analytical workflows, visualizations, technical reports, and policy briefs that support evidence-based decision-making and programme improvement.
Comprehensive Course Outline
Module 1: Foundations of Interrupted Time-Series Evaluation
-
Definition, purpose, and applications of interrupted time-series evaluation in policy analysis and development programme assessment.
-
Differences between simple before-and-after comparisons, interrupted time-series designs, controlled interrupted time-series designs, and randomized evaluations.
-
Understanding intervention interruptions, outcome trajectories, baseline periods, follow-up periods, and the role of repeated observations in causal assessment.
-
Sector-specific applications in public health, education, taxation, social protection, environmental regulation, transport, and public service delivery.
Module 2: Evaluation Design and Time-Series Data Preparation
-
Formulating evaluation questions, defining outcome measures, specifying intervention dates, and establishing appropriate units of analysis.
-
Assessing observation frequency, the number of pre- and post-intervention observations, data availability, and the suitability of administrative datasets.
-
Preparing analysis-ready datasets with consistent dates, indicator definitions, missing-value treatment, duplicate checks, and documented data transformations.
-
Identifying changes in reporting systems, policy implementation dates, anticipation effects, phased rollouts, and implementation delays that may affect evaluation design.
Module 3: Exploratory Time-Series Analysis and Visualization
-
Plotting outcome trends over time to identify baseline patterns, potential intervention effects, unusual observations, and possible structural breaks.
-
Examining secular trends, seasonality, cyclical behaviour, sudden fluctuations, and changes in outcome variability before model specification.
-
Using descriptive summaries, moving averages, seasonal plots, and exploratory diagnostics to understand patterns in policy outcome data.
-
Developing visualizations that clearly distinguish intervention dates, pre-intervention trends, post-intervention trajectories, and important contextual events.
Module 4: Segmented Regression and Intervention Effect Estimation
-
Understanding the segmented regression framework for estimating changes in outcome levels and trends after an intervention.
-
Constructing time variables, intervention indicators, post-intervention time variables, interaction terms, and relevant adjustment variables.
-
Interpreting immediate level changes, changes in slope, combined post-intervention trajectories, and estimated differences from the projected baseline trend.
-
Selecting appropriate regression specifications for continuous, count, binary, or other outcome measures according to the data structure and analytical assumptions.
Module 5: Time-Series Assumptions and Statistical Diagnostics
-
Understanding autocorrelation and its implications for standard errors, statistical significance, confidence intervals, and model reliability.
-
Assessing seasonality, residual patterns, heteroscedasticity, nonlinearity, outliers, and departures from the assumed outcome process.
-
Applying appropriate diagnostic plots, residual autocorrelation checks, and model assessment procedures to identify potential specification problems.
-
Selecting suitable approaches for serial dependence and seasonal adjustment while documenting the assumptions and limitations of the chosen methods.
Module 6: Controlled Interrupted Time-Series Design
-
Selecting comparison groups, geographical areas, institutions, or outcome series that can help account for external changes unrelated to the intervention.
-
Specifying controlled segmented regression models to distinguish changes associated with the intervention series from changes observed in the comparison series.
-
Assessing comparability, baseline trend similarity, differential exposure, contamination, spillovers, and other factors that may compromise the comparison.
-
Interpreting controlled intervention effects and understanding how comparison-series selection influences the strength of causal conclusions.
Module 7: Confounding, Concurrent Events, and Threats to Validity
-
Investigating concurrent policy changes, economic shocks, epidemics, environmental events, and institutional reforms occurring near the intervention date.
-
Examining history effects, maturation, regression to the mean, measurement changes, changes in population composition, and other alternative explanations.
-
Assessing whether intervention timing is clearly defined and whether implementation was immediate, gradual, staggered, or subject to anticipation.
-
Developing causal interpretation strategies that integrate contextual evidence, policy documentation, implementation records, and plausible alternative explanations.
Module 8: Robustness Checks and Sensitivity Analysis
-
Testing alternative model specifications, intervention dates, pre- and post-intervention windows, trend forms, and seasonal adjustment approaches.
-
Assessing the sensitivity of estimated effects to outliers, missing observations, influential time points, and changes in data inclusion criteria.
-
Applying placebo interruption dates and appropriate comparison outcomes to examine whether similar apparent effects occur without the intervention.
-
Evaluating the consistency of findings across reasonable analytical choices and documenting results that weaken or strengthen confidence in the estimated effects.
Module 9: Statistical Software, Interpretation, and Reporting
-
Implementing time-series data preparation, visualization, segmented regression, and diagnostic procedures using suitable statistical software.
-
Interpreting regression coefficients, effect estimates, uncertainty intervals, practical significance, and the distinction between level and trend changes.
-
Preparing clear analytical tables, time-series plots, technical documentation, and reproducible scripts for independent verification and review.
-
Communicating evaluation results, assumptions, limitations, and policy implications through technical reports, executive summaries, and stakeholder presentations.
Module 10: Applied Policy Evaluation and Decision-Making
-
Developing a complete interrupted time-series evaluation protocol for a selected policy intervention, including questions, indicators, timing, and data requirements.
-
Conducting an applied analysis that integrates exploratory plots, model specification, estimation, diagnostic testing, and sensitivity assessment.
-
Translating findings into actionable recommendations for policy continuation, modification, implementation improvement, and further evaluation.
-
Establishing procedures for integrating time-series evaluation into institutional monitoring systems, policy review cycles, learning agendas, and evidence-based planning.
Training Approach
This course will be delivered by our skilled trainers who have vast knowledge and experience as expert professionals in the fields. The course is taught in English and through a mix of theory, practical activities, group discussion and case studies. Course manuals and additional training materials will be provided to the participants upon completion of the training.
Tailor-Made Course
This course can also be tailor-made to meet organization requirement. For further inquiries, please contact us on: Email: training@upskilldevelopment.com Tel: +254 721 331 808
Training Venue
The training will be held at our Upskill Training Centre. We also offer training for a group (at a discount of 10% to 50%) at requested location all over the world. The Onsite course fee covers the course tuition, training materials, two break refreshments, buffet lunch, airport transfers, Upskill gift package, and guided tour.
Visa application, travel expenses, dinners, accommodation, insurance, and other personal expenses are catered by the participant
Certification
Participants will be issued with Upskill certificate upon completion of this course.
Airport Pickup and Accommodation
Airport pickup and accommodation is arranged upon request. For booking contact our Training Coordinator through Email: training@upskilldevelopment.com, +254 721 331 808
Terms of Payment:
Unless otherwise agreed between the two parties’ payment of the course fee should be done 3 working days before commencement of the training so as to enable us to prepare better.








