Basic Statistics for Biochem Data - Biochemistry

  1. A t-test gives p=0.03p = 0.03 with significance level α=0.05\alpha = 0.05. What is the conclusion?
    Reject the null hypothesis. The difference is statistically significant at the 0.05 level because p<αp < \alpha.
  2. What information about an effect does a p-value not provide?
    The size or biological importance of the effect. With a large nn, even a trivially small difference can give a small p-value.
  3. What does a 95% confidence interval for a mean mean?
    If the experiment were repeated many times, about 95% of intervals built this way would contain the true population mean.
  4. A measurement has SD = 6 with n=4n = 4 independent replicates. What is the SEM?
    33. SEM=64=3\text{SEM} = \frac{6}{\sqrt{4}} = 3.
  5. What is the difference between biological and technical replicates?
    Biological replicates are independent samples (separate cultures or preparations). Technical replicates are repeat measurements of the same sample.
  6. What is a Type I error?
    Rejecting a null hypothesis that is actually true, a false positive. Its probability equals the chosen α\alpha.
  7. Which statistical test compares the means of three or more independent groups in one analysis?
    One-way analysis of variance (ANOVA), usually followed by post hoc tests to find which groups differ.
  8. What is the difference between standard deviation (SD) and standard error of the mean (SEM)?
    SD describes the spread of individual measurements. SEM describes how precisely the sample mean estimates the true mean.
  9. Why must p-values be corrected when many comparisons are made on one data set?
    Each test carries an α\alpha chance of a false positive, so many tests inflate the chance of at least one false positive across the family.
  10. What is the formula for the standard error of the mean?
    SEM=SDn\text{SEM} = \frac{\text{SD}}{\sqrt{n}}, where nn is the number of independent measurements.
  11. What does a p-value represent?
    The probability of obtaining results at least as extreme as those observed if the null hypothesis were true.
  12. Using a Bonferroni correction, what p-value threshold applies to each of 10 tests for an overall α\alpha of 0.05?
    0.0050.005. Bonferroni divides α\alpha by the number of tests: 0.0510\frac{0.05}{10}.

Study these flashcards