statistical testing

statistical testing

1. the concept of significance

   • e.g.
     • if people drank a fizzy drink they spoke more words
       • = sell to millions because it cures shyness
         • but is it a significant difference
           • might be a fluke, coincidence or happen by chance
             • to find out we do a statistical test

2. the sign test

   • a simple statistical test
   • how to do it
     • 1) look for a difference rather than an association
       • 2) need to have used a repeated measures design
         • 3) data organised into categories (nominal data)

3. the concept or probability

   • all studies have to check the significant differences
   • the accepted level of probability
     • is 0.05
       • the level that the researcher accepts the research hypothesis
     • if the hypothesis is excepted
       • means there is a less than 5% chance that the results happened by chance
         • changed because of the manipulation of the iv
     • stricter then it's 0.01
       • a more stringent significance level
         • may involve a human cost
           • e.g. a drug trial
             • can't do it in the future - not ethical

4. the critical value

   • a number after the statistical test has taken place
     • is called the calculated value
       • this is compared to the critical value
         • to see whether the result is significant or not
   • given in a table of critical values
     • follow the steps
       • 1) the significance level desired (usually 0.05)
         • 2) number of participants
           • 3) directional or non directional hypothesis
       • locate the critical value for your data
   • for a sign test
     • the calculated value has to be equal to or lower than the critical value
       • then it is significant