<?xml version="1.0" encoding="UTF-8"?><rss version="2.0"><channel><title>Celeus blog</title><description>Worked statistical analyses for researchers, each with the R script and output shown.</description><link>https://celeus.ai/blog/</link><item><title>Confounding variables and how adjustment works</title><link>https://celeus.ai/blog/confounding-variables/</link><guid isPermaLink="true">https://celeus.ai/blog/confounding-variables/</guid><description>What a confounding variable is, how adjusting for one changes an odds ratio, and which variables to leave out of the model, worked on the birthwt data.</description><pubDate>Fri, 23 Oct 2026 00:00:00 GMT</pubDate></item><item><title>Correlation matrix in R with p-values and intervals</title><link>https://celeus.ai/blog/correlation-matrix-in-r/</link><guid isPermaLink="true">https://celeus.ai/blog/correlation-matrix-in-r/</guid><description>Build a correlation matrix in R with a p-value and confidence interval for every pair, adjust for testing many pairs at once, and read the table honestly.</description><pubDate>Fri, 23 Oct 2026 00:00:00 GMT</pubDate></item><item><title>Correlation vs causation in research reports</title><link>https://celeus.ai/blog/correlation-vs-causation/</link><guid isPermaLink="true">https://celeus.ai/blog/correlation-vs-causation/</guid><description>Why a strong correlation does not show cause: a seeded simulation where a third variable drives two measures, and a checklist of what a design needs.</description><pubDate>Fri, 23 Oct 2026 00:00:00 GMT</pubDate></item><item><title>Cox regression in R: coxph() with the exact script</title><link>https://celeus.ai/blog/cox-regression-in-r/</link><guid isPermaLink="true">https://celeus.ai/blog/cox-regression-in-r/</guid><description>Fit a Cox model in R with coxph() on the survival lung data, read the output line by line, run cox.zph for proportional hazards, and see the script.</description><pubDate>Fri, 23 Oct 2026 00:00:00 GMT</pubDate></item><item><title>How to interpret a hazard ratio in Cox regression</title><link>https://celeus.ai/blog/hazard-ratio/</link><guid isPermaLink="true">https://celeus.ai/blog/hazard-ratio/</guid><description>What a hazard ratio of 2 does and does not mean, how the reference group and the unit change it, and how to report it with a confidence interval.</description><pubDate>Fri, 23 Oct 2026 00:00:00 GMT</pubDate></item><item><title>Inter-rater reliability: kappa and alternatives</title><link>https://celeus.ai/blog/inter-rater-reliability/</link><guid isPermaLink="true">https://celeus.ai/blog/inter-rater-reliability/</guid><description>Cohen&apos;s kappa computed by hand and in R on a two-rater table, weighted kappa for ordered grades, and why high agreement can still give a low kappa.</description><pubDate>Fri, 23 Oct 2026 00:00:00 GMT</pubDate></item><item><title>Multicollinearity and VIF: a worked example in R</title><link>https://celeus.ai/blog/multicollinearity-vif/</link><guid isPermaLink="true">https://celeus.ai/blog/multicollinearity-vif/</guid><description>When correlated predictors matter: how to compute and read the variance inflation factor, what it does to confidence intervals, and when to leave it alone.</description><pubDate>Fri, 23 Oct 2026 00:00:00 GMT</pubDate></item><item><title>Regression to the mean in pre-post studies</title><link>https://celeus.ai/blog/regression-to-the-mean/</link><guid isPermaLink="true">https://celeus.ai/blog/regression-to-the-mean/</guid><description>Why extreme baseline scores drift back toward the average with no treatment at all, shown in a seeded simulation, and how a control group and ANCOVA remove it.</description><pubDate>Fri, 23 Oct 2026 00:00:00 GMT</pubDate></item><item><title>Residual standard error: how to read it in R</title><link>https://celeus.ai/blog/residual-standard-error/</link><guid isPermaLink="true">https://celeus.ai/blog/residual-standard-error/</guid><description>Where the residual standard error sits in R&apos;s lm() output, what it means in your outcome&apos;s units, and how to read it alongside the residual plot.</description><pubDate>Fri, 23 Oct 2026 00:00:00 GMT</pubDate></item><item><title>Sensitivity vs specificity in a diagnostic study</title><link>https://celeus.ai/blog/sensitivity-vs-specificity/</link><guid isPermaLink="true">https://celeus.ai/blog/sensitivity-vs-specificity/</guid><description>How sensitivity, specificity and predictive values are estimated from a 2x2 table, why prevalence moves the predictive values, and how to report intervals.</description><pubDate>Fri, 23 Oct 2026 00:00:00 GMT</pubDate></item><item><title>Standard error vs margin of error vs SD, explained</title><link>https://celeus.ai/blog/standard-error-vs-margin-of-error/</link><guid isPermaLink="true">https://celeus.ai/blog/standard-error-vs-margin-of-error/</guid><description>Standard deviation, variance, standard error, margin of error and IQR: what each measures, how they connect, and which to report, with a worked example.</description><pubDate>Fri, 23 Oct 2026 00:00:00 GMT</pubDate></item><item><title>Type I and type II errors in hypothesis testing</title><link>https://celeus.ai/blog/type-i-and-type-ii-errors/</link><guid isPermaLink="true">https://celeus.ai/blog/type-i-and-type-ii-errors/</guid><description>Null and alternative hypotheses, alpha, type I and type II errors and power, shown by counting false positives and misses across 2,000 seeded studies.</description><pubDate>Fri, 23 Oct 2026 00:00:00 GMT</pubDate></item><item><title>ANOVA test explained: one-way, Welch and two-way</title><link>https://celeus.ai/blog/anova-explained/</link><guid isPermaLink="true">https://celeus.ai/blog/anova-explained/</guid><description>What an ANOVA test does, how to choose one-way, Welch, two-way or repeated-measures ANOVA, and a worked Welch example in R with eta squared and its interval.</description><pubDate>Thu, 22 Oct 2026 00:00:00 GMT</pubDate></item><item><title>Breusch-Pagan test in R: what it tells you</title><link>https://celeus.ai/blog/breusch-pagan-test/</link><guid isPermaLink="true">https://celeus.ai/blog/breusch-pagan-test/</guid><description>How the Breusch-Pagan test checks a regression for non-constant spread, how to read it in R, and why large samples flag changes that may not matter.</description><pubDate>Thu, 22 Oct 2026 00:00:00 GMT</pubDate></item><item><title>Chi-square test explained, with a worked R example</title><link>https://celeus.ai/blog/chi-square-test-explained/</link><guid isPermaLink="true">https://celeus.ai/blog/chi-square-test-explained/</guid><description>What the chi-square test of independence computes, how to read expected counts, Cramer&apos;s V and residuals, and a worked R example on real survey data.</description><pubDate>Thu, 22 Oct 2026 00:00:00 GMT</pubDate></item><item><title>Cohen&apos;s d vs Hedges&apos; g vs Glass&apos;s delta</title><link>https://celeus.ai/blog/cohens-d-vs-hedges-g/</link><guid isPermaLink="true">https://celeus.ai/blog/cohens-d-vs-hedges-g/</guid><description>Which standardised mean difference to report, Cohen&apos;s d, Hedges&apos; g or Glass&apos;s delta, by sample size, unequal spread and purpose, with a worked R example.</description><pubDate>Thu, 22 Oct 2026 00:00:00 GMT</pubDate></item><item><title>How to write a confidence interval conclusion</title><link>https://celeus.ai/blog/confidence-interval-conclusion/</link><guid isPermaLink="true">https://celeus.ai/blog/confidence-interval-conclusion/</guid><description>What a correct confidence interval conclusion says, built from one worked example in R, with common incorrect wordings side by side and how to fix each.</description><pubDate>Thu, 22 Oct 2026 00:00:00 GMT</pubDate></item><item><title>Cramer&apos;s V and phi: effect sizes for chi-square</title><link>https://celeus.ai/blog/cramers-v/</link><guid isPermaLink="true">https://celeus.ai/blog/cramers-v/</guid><description>What Cramer&apos;s V and phi measure, why their small, medium and large cut-offs depend on table size, the bias-corrected V, and a worked example in R.</description><pubDate>Thu, 22 Oct 2026 00:00:00 GMT</pubDate></item><item><title>Data cleaning checklist before a statistical test</title><link>https://celeus.ai/blog/data-cleaning-checklist/</link><guid isPermaLink="true">https://celeus.ai/blog/data-cleaning-checklist/</guid><description>An ordered data cleaning checklist for statistical analysis: structure, codes, duplicates, ranges, units, missing data and outliers, run on R&apos;s airquality data.</description><pubDate>Thu, 22 Oct 2026 00:00:00 GMT</pubDate></item><item><title>Dissertation statistics: a design-first guide</title><link>https://celeus.ai/blog/dissertation-statistics/</link><guid isPermaLink="true">https://celeus.ai/blog/dissertation-statistics/</guid><description>Choose the statistics for a thesis from the study design: five common designs, the test, the assumption checks and the effect size to report for each.</description><pubDate>Thu, 22 Oct 2026 00:00:00 GMT</pubDate></item><item><title>Dunn&apos;s test after Kruskal-Wallis, explained</title><link>https://celeus.ai/blog/dunns-test/</link><guid isPermaLink="true">https://celeus.ai/blog/dunns-test/</guid><description>What Dunn&apos;s test compares after Kruskal-Wallis, how Holm and Bonferroni adjust its p-values, and a worked R example on six insecticides with effect sizes.</description><pubDate>Thu, 22 Oct 2026 00:00:00 GMT</pubDate></item><item><title>Effect sizes for Mann-Whitney and Wilcoxon tests</title><link>https://celeus.ai/blog/effect-size-mann-whitney/</link><guid isPermaLink="true">https://celeus.ai/blog/effect-size-mann-whitney/</guid><description>The rank-biserial correlation for Mann-Whitney and Wilcoxon tests: what it measures, how its interval works, why it is not Z over root N, how to report it.</description><pubDate>Thu, 22 Oct 2026 00:00:00 GMT</pubDate></item><item><title>Effect sizes: Cohen&apos;s d, eta squared, OR and r</title><link>https://celeus.ai/blog/effect-sizes-explained/</link><guid isPermaLink="true">https://celeus.ai/blog/effect-sizes-explained/</guid><description>What an effect size is, which one goes with which test, the SD each version of Cohen&apos;s d uses, and a worked R example with d and its 95% interval.</description><pubDate>Thu, 22 Oct 2026 00:00:00 GMT</pubDate></item><item><title>Eta squared vs partial eta squared: which to report</title><link>https://celeus.ai/blog/eta-squared/</link><guid isPermaLink="true">https://celeus.ai/blog/eta-squared/</guid><description>Eta squared, partial eta squared and omega squared differ only in the denominator. A two-way ANOVA in R shows which to report and the interval to give.</description><pubDate>Thu, 22 Oct 2026 00:00:00 GMT</pubDate></item><item><title>F-test vs t-test: what each one compares</title><link>https://celeus.ai/blog/f-test-vs-t-test/</link><guid isPermaLink="true">https://celeus.ai/blog/f-test-vs-t-test/</guid><description>Which F-test you mean decides the answer: the ANOVA F on two groups equals Student&apos;s t squared, while the variance-ratio F compares spreads, not means.</description><pubDate>Thu, 22 Oct 2026 00:00:00 GMT</pubDate></item><item><title>Heteroscedasticity: how to detect it and fix it</title><link>https://celeus.ai/blog/heteroscedasticity/</link><guid isPermaLink="true">https://celeus.ai/blog/heteroscedasticity/</guid><description>What non-constant spread does to a regression, how to see it in a residual plot, and when a log transform or robust standard errors is the right fix.</description><pubDate>Thu, 22 Oct 2026 00:00:00 GMT</pubDate></item><item><title>Homogeneity of variance and what to do about it</title><link>https://celeus.ai/blog/homogeneity-of-variance/</link><guid isPermaLink="true">https://celeus.ai/blog/homogeneity-of-variance/</guid><description>What unequal group variances do to Student&apos;s t and classic ANOVA, why Welch beats testing first, and a worked Welch ANOVA in R with effect sizes.</description><pubDate>Thu, 22 Oct 2026 00:00:00 GMT</pubDate></item><item><title>How to report chi-square test results in APA style</title><link>https://celeus.ai/blog/how-to-report-a-chi-square-test/</link><guid isPermaLink="true">https://celeus.ai/blog/how-to-report-a-chi-square-test/</guid><description>A complete results paragraph for a chi-square test of independence: counts, expected counts, the statistic, Cramer&apos;s V and the odds ratio with intervals.</description><pubDate>Thu, 22 Oct 2026 00:00:00 GMT</pubDate></item><item><title>How to report a t-test in APA style, with examples</title><link>https://celeus.ai/blog/how-to-report-a-t-test/</link><guid isPermaLink="true">https://celeus.ai/blog/how-to-report-a-t-test/</guid><description>Fill-in APA 7 sentences for Welch, Student, paired and one-sample t-tests, the formatting rule behind each symbol, and a worked example from a real R run.</description><pubDate>Thu, 22 Oct 2026 00:00:00 GMT</pubDate></item><item><title>The independence assumption: clustered data</title><link>https://celeus.ai/blog/independence-assumption/</link><guid isPermaLink="true">https://celeus.ai/blog/independence-assumption/</guid><description>What repeated or clustered observations do to standard errors, shown on real tree growth data in R with and without a random intercept, and how to spot it.</description><pubDate>Thu, 22 Oct 2026 00:00:00 GMT</pubDate></item><item><title>The IQR rule: upper and lower fences for outliers</title><link>https://celeus.ai/blog/iqr-rule-upper-lower-fence/</link><guid isPermaLink="true">https://celeus.ai/blog/iqr-rule-upper-lower-fence/</guid><description>How to compute Tukey&apos;s upper and lower fences by hand and with R&apos;s boxplot.stats(), why the two can disagree, and what to do with a value beyond a fence.</description><pubDate>Thu, 22 Oct 2026 00:00:00 GMT</pubDate></item><item><title>Analysing Likert scale data: items and scales</title><link>https://celeus.ai/blog/likert-scale-data-analysis/</link><guid isPermaLink="true">https://celeus.ai/blog/likert-scale-data-analysis/</guid><description>How to analyse Likert data in a thesis: when a single item needs an ordinal method, when a summed scale can take a t-test, and a worked ordinal regression in R.</description><pubDate>Thu, 22 Oct 2026 00:00:00 GMT</pubDate></item><item><title>Linear regression assumptions and how to check them</title><link>https://celeus.ai/blog/linear-regression-assumptions/</link><guid isPermaLink="true">https://celeus.ai/blog/linear-regression-assumptions/</guid><description>The four assumptions of linear regression, the R check and the fix for each, and which failures bias the estimates and which only distort the standard errors.</description><pubDate>Thu, 22 Oct 2026 00:00:00 GMT</pubDate></item><item><title>Logistic regression assumptions, real and invented</title><link>https://celeus.ai/blog/logistic-regression-assumptions/</link><guid isPermaLink="true">https://celeus.ai/blog/logistic-regression-assumptions/</guid><description>Which assumptions logistic regression really makes, which ones it never made, and how to check each one in R, worked through on the birthwt data.</description><pubDate>Thu, 22 Oct 2026 00:00:00 GMT</pubDate></item><item><title>One-sample t-test: comparing a mean to a reference</title><link>https://celeus.ai/blog/one-sample-t-test/</link><guid isPermaLink="true">https://celeus.ai/blog/one-sample-t-test/</guid><description>When a one-sample t-test fits, where the reference value has to come from, and a worked R example on Newcomb&apos;s 1882 light measurements with Cohen&apos;s d.</description><pubDate>Thu, 22 Oct 2026 00:00:00 GMT</pubDate></item><item><title>How to handle outliers before a statistical test</title><link>https://celeus.ai/blog/outliers-before-statistical-test/</link><guid isPermaLink="true">https://celeus.ai/blog/outliers-before-statistical-test/</guid><description>A decision rule for outliers before a test: correct errors, exclude by a written criterion, keep genuine extremes, with seeded data run with and without them.</description><pubDate>Thu, 22 Oct 2026 00:00:00 GMT</pubDate></item><item><title>How to find a p-value from a test statistic</title><link>https://celeus.ai/blog/p-value-from-test-statistic/</link><guid isPermaLink="true">https://celeus.ai/blog/p-value-from-test-statistic/</guid><description>Turn a reported t, z, F or chi-square statistic into its p-value by hand and with R&apos;s pt(), pnorm(), pf() and pchisq(), compared with full test output.</description><pubDate>Thu, 22 Oct 2026 00:00:00 GMT</pubDate></item><item><title>P-values and confidence intervals: report both</title><link>https://celeus.ai/blog/p-values-and-confidence-intervals/</link><guid isPermaLink="true">https://celeus.ai/blog/p-values-and-confidence-intervals/</guid><description>What a p value and a confidence interval each tell you, how they relate, how to read p = 0.06, and how to report both, with a worked example in R.</description><pubDate>Thu, 22 Oct 2026 00:00:00 GMT</pubDate></item><item><title>Paired vs unpaired t-test: which design you have</title><link>https://celeus.ai/blog/paired-vs-unpaired-t-test/</link><guid isPermaLink="true">https://celeus.ai/blog/paired-vs-unpaired-t-test/</guid><description>How to tell whether your two sets of measurements are paired or independent, the designs people misclassify, and what ignoring the pairing costs, shown in R.</description><pubDate>Thu, 22 Oct 2026 00:00:00 GMT</pubDate></item><item><title>Parametric test assumptions: checks and fallbacks</title><link>https://celeus.ai/blog/parametric-test-assumptions/</link><guid isPermaLink="true">https://celeus.ai/blog/parametric-test-assumptions/</guid><description>The assumptions behind t-tests, ANOVA and regression, how to check each without the pre-test traps, and what to do when one fails, with a worked Welch ANOVA.</description><pubDate>Thu, 22 Oct 2026 00:00:00 GMT</pubDate></item><item><title>ANOVA post hoc tests: Tukey, Games-Howell, Dunn</title><link>https://celeus.ai/blog/post-hoc-tests-after-anova/</link><guid isPermaLink="true">https://celeus.ai/blog/post-hoc-tests-after-anova/</guid><description>Which post hoc test follows which ANOVA: Tukey after the classic F-test, Games-Howell after Welch, Dunn after Kruskal-Wallis, all three run on one dataset.</description><pubDate>Thu, 22 Oct 2026 00:00:00 GMT</pubDate></item><item><title>Pretest posttest design: change scores or ANCOVA?</title><link>https://celeus.ai/blog/pretest-posttest-design/</link><guid isPermaLink="true">https://celeus.ai/blog/pretest-posttest-design/</guid><description>How to analyse a two-group pretest posttest design: change scores, ANCOVA or a mixed model, when each one fits, and a worked ANCOVA example in R.</description><pubDate>Thu, 22 Oct 2026 00:00:00 GMT</pubDate></item><item><title>Multiple linear regression, and when to switch</title><link>https://celeus.ai/blog/regression-analysis-guide/</link><guid isPermaLink="true">https://celeus.ai/blog/regression-analysis-guide/</guid><description>Choosing a regression model from your outcome type (linear, logistic, Poisson, negative binomial or ordinal), with a multiple linear regression worked in R.</description><pubDate>Thu, 22 Oct 2026 00:00:00 GMT</pubDate></item><item><title>APA style statistics reporting: every common test</title><link>https://celeus.ai/blog/reporting-statistics-apa-style/</link><guid isPermaLink="true">https://celeus.ai/blog/reporting-statistics-apa-style/</guid><description>The APA 7 rules for reporting statistics (italics, leading zeros, p values, degrees of freedom, intervals) and a fill-in template for each common test.</description><pubDate>Thu, 22 Oct 2026 00:00:00 GMT</pubDate></item><item><title>Survival analysis: Kaplan-Meier, Cox and AFT in R</title><link>https://celeus.ai/blog/survival-analysis-guide/</link><guid isPermaLink="true">https://celeus.ai/blog/survival-analysis-guide/</guid><description>Survival analysis in R on public lung cancer data: censoring, Kaplan-Meier, log-rank, Cox regression with its proportional hazards check, and AFT.</description><pubDate>Thu, 22 Oct 2026 00:00:00 GMT</pubDate></item><item><title>T-tests explained: one-sample, paired and Welch</title><link>https://celeus.ai/blog/t-tests-explained/</link><guid isPermaLink="true">https://celeus.ai/blog/t-tests-explained/</guid><description>What a t-test does, how to choose the one-sample, paired, Student or Welch version, and a worked Welch example in R with Cohen&apos;s d and its interval.</description><pubDate>Thu, 22 Oct 2026 00:00:00 GMT</pubDate></item><item><title>Do you need to test for normality before a t-test?</title><link>https://celeus.ai/blog/test-for-normality-before-t-test/</link><guid isPermaLink="true">https://celeus.ai/blog/test-for-normality-before-t-test/</guid><description>Should you test normality before a t-test? Usually not. Why a Shapiro-Wilk pre-test misleads at small and large n, what to check instead, and an R example.</description><pubDate>Thu, 22 Oct 2026 00:00:00 GMT</pubDate></item><item><title>Two-sample t-test: formula, assumptions, example</title><link>https://celeus.ai/blog/two-sample-t-test/</link><guid isPermaLink="true">https://celeus.ai/blog/two-sample-t-test/</guid><description>The two-sample t-test formula worked by hand and confirmed in R, its assumptions, and Student&apos;s versus Welch&apos;s version on the same unequal-sized groups.</description><pubDate>Thu, 22 Oct 2026 00:00:00 GMT</pubDate></item><item><title>Welch&apos;s t-test: when to use it instead of Student&apos;s</title><link>https://celeus.ai/blog/welchs-t-test/</link><guid isPermaLink="true">https://celeus.ai/blog/welchs-t-test/</guid><description>Why Welch&apos;s t-test is the safer default for two groups, how its degrees of freedom work, and a worked R example where group sizes and spreads both differ.</description><pubDate>Thu, 22 Oct 2026 00:00:00 GMT</pubDate></item></channel></rss>