The Journal
On getting statistics right: choosing the correct test, keeping every number real, and writing findings you can defend.
Statistics for nursing research: the daily workflow, covered end-to-end
Pre-post pilots, scale validations, cohort survival, satisfaction surveys. Here's the daily statistical workflow of nursing research mapped onto the tests thericerca actually runs, plus the honest note on what lies beyond.
Read the piece →Statistics for UX research: the quantitative workflow, in one place
Task times, error rates, SUS scores, A/B tests, survey scales. Here's the daily quantitative UX workflow mapped onto the tests thericerca actually runs, plus the honest note on where large-scale product experimentation platforms take over.
Read the piece →Statistics for marketing and consumer research: survey work, end-to-end
Brand trackers, driver analyses, concept tests, segmentation. Here's the daily quantitative marketing workflow mapped onto the tests thericerca actually runs, plus the honest note on where conjoint and choice modelling take over.
Read the piece →Statistics for people analytics and HR research: your workflow, one tool
Engagement scales, turnover models, promotion-time analyses, pay-equity comparisons. Here's the daily people-analytics workflow mapped onto the tests thericerca actually runs, plus the honest note on where the specialist tools take over.
Read the piece →Teaching undergraduate research methods: a full-semester toolkit
Every test in a standard Stats 101 syllabus, visible assumption checks, APA output, and reproducibility exports that let students see the code without needing to write it. A field guide for instructors thinking about tooling.
Read the piece →Learning biostatistics for clinical trials: the statistics you need first
Clinical trial biostatistics has a big vocabulary and a small window for error. Here's a staged path from the statistical foundations through trial methodology to the specialist tools, with an honest note on which parts thericerca handles.
Read the piece →Learning epidemiology from scratch: the statistics you need first
Epidemiology is where measurement meets consequence. Here's a staged path from the statistical foundations through study-design vocabulary to causal inference, and an honest note on which parts thericerca handles.
Read the piece →Learning meta-analysis: the statistics you need first
A good meta-analysis is where evidence stops being a list and starts being a conclusion. Here's a staged path from effect-size fluency through the pooling models to network meta-analysis, with an honest note on our role in the picture.
Read the piece →Learning psychometrics: the statistics you need first
Psychometrics is where measurement stops being a hunch and starts being a science. Here's a staged path from the statistical foundations through measurement theory to the specialist tools, with an honest note on which parts thericerca handles.
Read the piece →Learning pharmacometrics: the statistics you need first
Pharmacometrics has a steep on-ramp and scattered resources. Here's a staged path through the statistical foundations, the PK/PD theory, and the specialist tools, with an honest note on which parts thericerca helps with.
Read the piece →When a good study meets the wrong test
Clean design, real data, honest questions, and still the wrong conclusion. Here's where analyses go quietly wrong, and how thericerca keeps the choice of test right.
Read the piece →How to choose the right statistical test
The right test isn't a matter of taste; it follows from three properties of your data. A field guide to choosing correctly, and to knowing when your case is genuinely unusual.
Read the piece →What a p-value actually means (and what it doesn't)
It's the most cited number in science and the most misread. What a p-value genuinely tells you, the misconceptions to unlearn, and why it should never travel alone.
Read the piece →t-test or ANOVA? Comparing groups the right way
Two groups or ten, the question is the same, but the tool isn't. Why three groups don't mean three t-tests, and how ANOVA's two-stage logic prevents false findings.
Read the piece →Parametric vs non-parametric: when your data breaks the rules
How to actually check normality, why n > 30 isn't the free pass everyone claims, the full map of swaps, what Mann–Whitney really tests (it isn't medians), and how to report a rank-based result.
Read the piece →Statistical power and sample size, explained
Too small a study doesn't just miss real effects; its 'significant' results are more likely to be flukes. How the four levers of power fit together, and how to size a study right.
Read the piece →Effect size: the number your p-value forgot
With enough data, everything is significant. Effect size is the antidote: the measure of how big a finding actually is, and the number that lets research build on itself.
Read the piece →