Why do we need inferential 
   statistics?
    Typically, we are interested in the 
     population, not the sample
      –When we study an intervention, for 
       example, we want to be able to generalize 
       to the larger group (the population) 
      –But we usually can’t gather the whole 
       population of scores
   Why do we need inferential 
   statistics?
    Variability
      –Remember that measurements in the 
       sciences are variable; they change from 
       observation to observation
      –We need inferential statistics to assess 
       this variability and aid in our decision 
       making
   What do we learn from 
   inferential statistics?
    Inferential statistics provides us with 
     educated guesses about quantitative 
     characteristics of populations 
     (“parameters”)
    For example, is the central tendency of 
     one group different than the central 
     tendency of a second group
   Varieties of Inferential 
   Procedures
    Parameter estimation – using data 
     from a random sample to estimate a 
     parameter of the population from which 
     the sample is drawn
    Hypothesis testing – formulating 
     opposing hypotheses and determining 
     from samples which is most likely 
     correct
   Hypothesis Testing
    This may seem to be overly complicated, but
    It provides an elegant way of answering 
     research questions
    For example, we may want to determine 
     which of these two hypotheses is correct:
     – 1. The whole language teaching method 
      improves reading scores
     – 2. The whole language teaching method does 
      not improve reading scores
   We may be able to learn how
   best to teach children to read.