Clinical stats 4 Flashcards
(21 cards)
You want to compare the mean pain score between two independent groups: smokers vs non-smokers.
Independent T-Test
You compare the mean age across three diet groups. Which test?
ANOVA (Analysis of Variance)
A dentist records pain scores before and after treatment in the same group of patients. Which test?
Paired T-Test
A study measures BP at 0, 10, and 20 mins after taking medication. You want to analyse trends over time.
Repeated Measures ANOVA
You want to see if age and systolic BP have a linear relationship. Which test?
Pearson’s Correlation
You’re predicting pain score based on age and weight.
Linear Regression
Comparing pain score between males and females with non-normal data?
Wilcoxon Rank Sum Test
Comparing means of 3+ groups, but the data isn’t normal.
Kruskal-Wallis Test
You compare implant success rate between smokers and non-smokers. All counts >5.
Chi-Square Test
ou compare implant success rate between smokers and non-smokers. All counts < 5.
Fisher’s Exact Test
You compare success of two fillings done in the same patient.
McNemar’s Test
Comparing denture fit success for upper vs lower dentures in the same patients. Which test?
McNemar’s Test
What test compares means between 2 related groups?
Paired T-Test
What test compares proportions between 2 independent groups?
Chi-Square Test
What’s the non-parametric alternative to:
a) Paired T-Test
b) T-Test
c) ANOVA
a) Wilcoxon Signed-Rank
b) Mann-Whitney U
c) Kruskal-Wallis
🧪 What are the 3 assumptions for a T-test?
✅ Normal distribution
📊 Continuous data
👯♀️ Independent samples (for independent t-test)
What do you use to test linear relationship between 2 continuous variables?
Pearson’s Correlation
🧠 What’s the main difference between Chi-Square and Fisher’s Exact Test?
Chi-Square is used when all expected counts ≥5,
Fisher’s is used when any expected count <5 🔢
📊 You want to compare before and after cholesterol levels in the same group of patients.
The data is not normally distributed.
Wilcoxon Signed-Rank Test ✍️
⚖️ What’s the difference between linear regression and logistic regression?
Linear: Outcome = continuous
Logistic: Outcome = binary (yes/no) 📉🔁
💡 What kind of variable is “pain score on a scale of 0–10”?
Continuous 📏
