Statistical Testing Flashcards
two types of statistics
descriptive statistics such as measures of central tendency and dispersion
inferential statistics — involves testing a hypothesis of a group of people and inferring that the same is true for all people belonging to that group (e.g. women)
how to make an inference?
we have to use statistical tests designed to work out the probability of whether a particular set of data could simply have occurred by chance
a difference may have been found between two samples but this difference must be tested to see whether it is big enough to be of significance
what is probability?
probability = a numerical measure of the likelihood or chance that certain events will occur
what is significance?
indicates whether or not the research findings are sufficiently strong for us to accept the research hypothesis under test
the sign test
a statistical test to determine the significance of a sample of related items of data
when to use a sign test…
used when looking at paired all related data
the two related pieces of data could come from a repeated measures design in which the same person is tested twice
can also be used with matched pairs design because the participants are paired and therefore count as one person tested twice
how to do the sign test…
- state the hypothesis
- record the data and work out the sign — subtract one score from the other, if the result is a positive number record a plus (+) and if the result is a negative number record a minus (-)
- find the calculated value — the test statistic is calculated by adding up the pluses and adding up the minuses and selecting the less frequent value, for instance if there was 10 pluses and 3 minuses, the less frequent sign is minus so the calculated value is 3
- find the critical value — N is the total number of scores, ignoring any zero values. locate the column headed 0.05 for whichever test is relevant, go down the column until you reach the row which begins with the N value, the number in this box is the critical value
- is the result in the right direction? — if hypothesis is directional, we have to check that the result is the expected direction, we can accept the hypothesis if it is
how to report the conclusion that can be drawn…
a table of critical values is used to judge significance
the calculated value is compared to the the critical value to see if the calculated value is significant
if the calculated value is equal to or less than the critical value, the result is significant and the hypothesis can be accepted
what test to use and when?
one-tailed test is used with a directional hypothesis
two-tailed test is used with a non-directional hypothesis
what significance level to use? and what does it mean?
always use the 5% or 0.05 significance level unless the question states otherwise
a 0.05 level means that the hypothesis is true 95% of the time but there is a 5% probability that the result would have occurred anyway even if the IV did not have an effect
what is a test statistic?
test statistic = a statistical test is used to calculate a numerical value, for each test this value has a specific symbol such as S for the sign test
why do researchers sometimes use the 0.1 significance level?
to show that the results are highly significance
sometimes researchers wish to be more certain so use the 0.01 level of significance
especially where certainty about the results are highly important, like when testing the effect of a drug in treating a disease
define the critical and calculated value
calculated value = the value of a test statistic calculated for a particular data set
critical value = in a statistical test, the critical value is the value of the test statistic that must be reached to show significance
