Basic: Für alle Neueinsteiger
Wenn’s um den Bachelor, Master oder Dr. geht:
- Checkliste Dateneingabe mit Excel Excel ist das beliebteste Werkzeug, um Daten für kleine Studien aufzunehmen. Aber wie formatiere ich die Daten, damit ein Statistikprogramm sie auch lesen kann?
- Vorzeichentest Zeigt ein Fußballmatch mit 7:1 die eindeutige Überlegenheit der Gastgeber? Zählstatistik funktioniert auch bei der Auswertung von Studien.
- Vorher-nachher-Studie Excel Was mache ich, wenn ich mit Excel eine Studie auswerten will, bei der Vorher-Nachher-Ergebniss für mehrere Behandlungsmethoden gemessen wurden?
Intermediate: Methods
- Pairwise differences What to do when the boss asks for p-values for all pairwise combinations of multiple treatments arms (she will always do). You might just say “no”.
- Ballot and bazaar In studies with multiple treatments, it is worthwhile to perform an analysis for the entire population rather than calculating each pair of treatments separately.
- Crossover study design You are planning a study with multiple arms and unequal number of patients per arm: here is the GUI for your PC, Mac or Linux.
- Regression mysteries Can you swap the axis in a regression analysis? Spoiler: no! Can you swap for correlation? Read on.
- Read SQLite from Excel SQLite is a fantastic database for small to medium-sized studies, but getting data from it into Excel requires some configuration.
Advanced: Gastric Emptying
Analysing gastric emptying studies is my bread-and-butter, so some of the items might be a bit special. * Area-under-curve considered harmful The area under the curve (AUC) is a popular measure because it is robust and insensitive to the shape of the time series. At what cost? Might be interesting even for researchers not involved in gastric emptying studies * Does the 13C breathtest measure gastric emptying? No, it’s biased by pharmacodynamics. And yes, the meal intake protocol can be improved. A quest for sample data. * Box-Cox tranformation for gastric emptying More of a reminder to myself what to do when gastric emptying data are awkwardly distributed. * Fitting 13C breath test data Estimating the half-emptying time of gastric emptying using breath test data can be awkward, when some time series are exotic. Borrowing power of population fits comes to the rescue!