Beyond descriptive statistics

Beyond descriptive statistics

1. Hypothesis

   • A statement or prediction about what will happen in your study
   • Some studies may not have a hypothesis, or may have more than 1
   • Null hypothesis significance testing: We can discover an alternative hypothesis through tests, or find a null hypothesis
     • Null hypothesis: no difference or relationship between groups
     • WE only mention the alternative hypothesis - NEVER the null
   • Non-directional hypothesis - don’t hypothesis a direction = two tailed tests
   • Directional hypotheses - hypothesise a direction = one tailed tests

2. Statistical significance

   • refers to the likelihood of observing the same, or more extreme difference between groups
   • Type 1 error: the probability of rejecting the null hypothesis when it is actually true

3. Inferential statistics

   • Samples represent populations
   • Inferences for population from sample
   • Inferential stats give the probability that these inferences are correct - estimate what the mean of the ‘real’ population is

4. Sampling

   • Probability sampling: every member of population has an equal chance of being selected
   • Simple random sampling: select names from population at random
   • Systematic sampling: pick every 10th, 50th, 100th etc name depending on proportion to be sampled
   • Stratified sampling: Divide population into strata on important characteristics, use random sampling within each
   • Cluster sampling: when groups are naturally clustered, sample randomly within each cluster e.g. level of study
   • Non-probability sampling: not every member has an equal chance. More practical, less likely to be representative
   • Opportunity sampling: use whoever is available
   • Quota sampling: decide quota sizes in advance, recruit opportunistically within each quota e.g. 10 males ages 18-25, 20 left handed women

5. Court

   • Truth and Errors
     • Defendant = guilty, jury finds them innocent
     • Defendant = innocent, jury finds them innocent
     • Defendant = guilty, jury finds them guilty
     • Defendant = innocent, jury finds them innocent

6. Type 1 error example:

   • Sending an innocent person to prison
   • we say there is an effect of our manipulation when there isnt one

7. Type 2 error example:

   • Releasing a guilty person
   • we say there is no effect of our manipulation when there is one

8. Slide 42/62