inferential statistics

• difference between null and alternate hypothesis

  A null hypothesis claims that there is no effect in the population while an alternative hypothesis claims that there is an effect.
• The problem with descriptive statistics is that they may be too simplistic. so we use inferential
• eg. difference in memory between 16 year olds and 25 year olds.  You make the following hypotheses: 

  • Null Hypothesis: There will be no significant difference between the number of words correctly recalled on a memory test between 16 year olds and 25 year olds.  
  • Alternate hypothesis: There will be a significant difference between the number of words correctly recalled on a memory test between 16 year olds and 25 year olds.  
• An inferential statistics test helps us decide whether the result was due to chance or whether we should accept null or alternate
•  E.g. 0.00 = 0% chance, 1.00 = 100% chance it is due to chance
• In general, psychologists use the 0.05 level of significance; meaning that there is a 5% chance that the results would have occurred this way even if there was no real difference between the groups. This probability of 5% is recorded as p=0.05
• Type 1 error

  False positive 

  Accept alternate hypothesis although really the null hypothesis was true.  (Reject null hypothesis when actually it was true).  
• Type 2 error

  False negative 

  Accept null hypothesis although really the alternate hypothesis was true.  (Accept null hypothesis was actually it was false).