39844827
| Question | Answer |
| What is the Null hypothesis (H0)? | The null hypothesis tests for NO DIFFERENCE between the groups. If we test each group with a lot more data then there is a lower probability that the differences are due to chance. f.e.: There is no difference between the recovery time of patients taking drug A or B. We can reject or Fail to reject the 0 Hypothesis |
| What is the alternative hypothesis (HA)? | The alternative hypothesis tests for DIFFERENCE between the groups. f.e.: HA = There is a difference between the recovery time of patients taking drug A or B. |
| What is the hypothesis-testing protokoll? | 1. collect Data + form a hypothesis (Ho, HA) 2. Statistical test (with a certain P-Value) 3. Decision to reject or fail to reject the null hypothesis |
| What is the P-Value? | A P-value represents the probability of our result assuming the NULL hypothesis is true (Area under the graph) high-> H0 is right low-> HA is right commonly accepted P-Value is 0.05 or 5% |
| What is the difference between a Parametric and a non-Parametric test? | parametric test-> data comes from a normal distribution (t-test) non-parametric test-> data isn’t normally distributed or it cannot be approximated by a normal distribution (Wilcoxon) |
| What can be assumed of the t-test? | A t-test assumes the two samples come from the same population And calculates the probability of the observed difference in means The P-value is simply the shaded area below the t- distribution to the left of -ts and the right of ts. You can always look up the values of the ts in tables and the corresponding P-value. |
| What can you test with a Shapiro-WIlk test? | tells us if our data could come from a Normal distribution or not. • The test statistic is W. • If the P-value associated with W is lower than 0.05 then our data is most likely NOT normal. |
| What can be said about the Wilcoxon test? | Samples need to be independent • Advantage: Since the test is based on ranks it is not affected by outliers. Data doesnt have to be normally distributed • Disadvantage: Less powerful than a parametric test. |
| How to you go about the Wilcoxon test? | To calculate it, we first order the data from lowest to highest for both groups. • Then, for each value, we count the number values of the other group that are smaller. • We then sum each new column and obtain a K1 for group 1 and K2 for group 2. The test statistic Us is defined by the largest K. resulting p-value tells us if Ho or H1 is plausable |
| What can be said about the Chi-squared test? | Data is categorical (imagine we sample many individuals from a population and store genotype information. This information isn’t numerical.) Compare the observed #genotypes with the expected #genotypes. |
| What is the Chi-squared test protokoll? | 1. convert raw data to a count table 2. compare observed vs. expected frequencies in categorical data. For each category (AA,AB): (observed-expected^2/ expected) 3. compare the test statistics (x^2) to the 5% significance |
| what can be said about the Fishers exact test? | - This test is an alternative to the chi-squared test. • It works only for 2x2 contingency tables • It can also be used when dealing with small samples. • In genetics, it is particularly useful to quantify the overlap between two sets of genes. |
Want to create your own Flashcards for free with GoConqr? Learn more.