Jamie Robins and I have written a book that provides a cohesive presentation of concepts of, and methods for, causal inference. Much of this material is currently scattered across journals in several disciplines or confined to technical articles. We expect that the book will be of interest to anyone interested in causal inference, e.g., epidemiologists, statisticians, psychologists, economists, sociologists, political scientists, computer scientists… The book is divided in 3 parts of increasing difficulty: causal inference without models, causal inference with models, and causal inference from complex longitudinal data.
The Data Playbook Beta is a recipe book or exercise book with examples, best practices, how to's, session plans, training materials, matrices, scenarios, and resources. The data playbook will provide resources for National Societies to develop their literacy around data, including responsible data use and data protection. The content aims to be visual, remixable, collaborative, useful, and informative. There are nine modules, 65 pieces of content, and a methodology for sharing curriculum across all the sectors and networks. Material has been compiled, piloted, and tested in collaboration with many contributors from across IFRC and National Societies. Each module has a recipe that puts our raw materials in suggested steps to reach a learning objective. To help support you in creating your own recipe, we also include a listing of 'ingredients' for a topic, organised by type:
research on health comm messaging effects
Breakthrough ACTION has distilled guidance on social and behavior change (SBC) monitoring methods into a collection of technical notes. Each note provides an overview of a monitoring method that may be used for SBC programs along with a description of when to use the method and its strengths and weaknesses.
We applied a Hidden Markov Model* (see Figure 1) to examine how and why behaviours did or did not change. The longitudinal repeated measure design meant we knew about food waste behaviour at two points (the amount of food wasted before and after the program), changes in the amount of food wasted reported over time for each household (more or less food wasted) and other factors (e.g. self-efficacy). By using a new method we could extend our understanding beyond the overall effect (households in the Waste Not Want Not program group wasted less food after participating when compared to the control group).
The aim of this study was to establish if distinct segments were evident in a sexual health context drawing from measures sourced from four segmentation bases extending application of segmentation to all recommended bases . This study indicates how researchers can use two-step cluster analysis to identify segments, which are represented by a group of individuals who share similar characteristics that differ from other groups in the larger heterogeneous target audience. Further, this study demonstrates how available information can be used delivering a dashboard to inform program design and planning.
•Despite its sequential nature, healthcare seeking is often analysed as single event. •We demonstrate the value of sequential healthcare data analysis. •Descriptive analysis exposes otherwise neglected behavioural patterns. •Sequence-insensitive indicators can be inconsistent and misleading. •Sequence-sensitive evaluation hints at adverse behaviours of wealthy patients.
Effective visualizations communicate complex statistical and quantitative information facilitating insight, understanding, and decision making. But what is an effective graph? This cheat sheet provides general guidance and points to consider.
The next time you find yourself stating that a deal or other business outcome is “unlikely” or, alternatively, is “virtually certain,” stop yourself and ask: What percentage chance, in what time period, would I put on this outcome? Frame your prediction that way, and it’ll be clear to both yourself and others where you truly stand.
Typically, cascades are based on HIV treatment moni-toring data, which focus on getting people living with HIVto a point of viral suppression. HIV prevention cascadesfocus on the steps required to prevent HIV infection andsuccessfully implement HIV prevention programs. Preven-tion cascades include demand-side interventions that focuson increasing awareness, acceptability and uptake of pre-vention interventions, supply-side interventions that makeprevention interventions more accessible and available, andadherence interventions thatsupport ongoing adoption andcompliance with prevention behaviours or products...
Small, medium or large — what sample size of users fits your study is a composite question. The magic number of 5 users may work magic in some studies while in some it may not. It depends on the constraints put on by project requirements, assumptions about problem discoverability and implications to the design process. Assess these factors to determine the number of users for your study: What’s the nature and scope of research — is it exploratory or validatory? Who and what kind of users are you planning to study? What’s the budget and time to finish the study? Does your research involve presenting statistically significant numbers or inferring behavioural estimates for the problem statement?
Lesson: Use "commitment" question instead of attention check questions.
For fields where the threshold for defining statistical significance for new discoveries is P < 0.05, we propose a change to P < 0.005. This simple step would immediately improve the reproducibility of scientific research in many fields. Results that would currently be called “significant” but do not meet the new threshold should instead be called “suggestive.”
Web Metrics, YouTube Basics and Mobile Metrics Guides