Analysis

Last Updated:

Quantitative and qualitative analysis for M&E programs. Cleaning, coding, descriptive and inferential statistics, thematic analysis, mixed methods, and how to turn collected data into defensible findings.

How Do I Choose?

Checking AI-Assisted Qualitative Coding: Trace Every Suggestion to an ExcerptAn AI tool can suggest codes and themes quickly, but each suggestion must be traced to the excerpt it relies on. A fictional example shows the checks, the failures to expect and what to record.Choosing a Qualitative Analytic Tradition: Match the Method to the QuestionThematic, framework, content, grounded theory, narrative and rapid analysis answer different questions and carry different quality checks. Choose by question, time and stance, then use the checks that fit.From Interview Notes to Findings: A Traceable Worked CaseA fictional worked case follows six interviews from excerpts to codes, memos and claims, keeping the cases that do not fit and stating what the findings cannot say.How to Clean Your Dataset Before Analysis: A Step-by-Step M&E ChecklistA step-by-step checklist for cleaning M&E data after collection. Duplicate detection, outlier identification, skip logic validation, consistency checks, and cleaning log documentation.Joint Displays and Conflicting Results: Keep the Disagreement VisibleA joint display lines up survey and interview findings by question, population and period. A fictional case shows how to read agreement, silence and conflict without forcing a single story.Qualitative vs Quantitative vs Mixed MethodsQualitative, quantitative, and mixed methods are not a quality ranking. They answer different questions. Here's when to use each, how to combine them, and what integration actually looks like.Surveys vs Interviews vs Focus GroupsThe three most common M&E data collection methods, compared. Surveys tell you how many, interviews tell you why, focus groups tell you what people agree on.Summarizing a Distribution: Mean, Median, Shares and SpreadTwo fictional villages share the same average spending and very different households. A worked case shows which summary to report, and what weights and small samples change.