Inferential statistics

HideShow resource information
Preview of Inferential statistics

First 471 words of the document:

Inferential statistics
Used to see if your results are significant or if they are due to chance. It looks at
the size of the sample and the size of the difference between the 2 groups to
estimate the likelihood that the 2 groups are really different or linked. Each
statistical test will work out if there is a significant difference. For most
investigations the accepted level of significance is p<0.05 (5%) which means 95%
of the time the results will be significant and there is a 5% chance that the results
were due to chance and not the manipulation of the IV.
Choosing a statistical test
Spearman's rank is used it the study is investigating a correlation. If a
difference was investigated then the level of data must be considered
(nominal or ordinal).
Chi squared is used if the study is looking for a difference and if the level
of data was nominal (named categories) as nominal data is grouped into
categories. The score of each category are described as frequencies.
Mann Whitney U test is used if the study was looking for a difference, if
the level of data was ordinal and if the type of experimental design was
independent groups.
Wilcoxon T test is used if the study was looking for a difference, if the
level of data used was ordinal and if the type of experimental design was
repeated measures.
Spearman's rank
Looks at a correlation co-efficient called Rho: this is between +1 and -1. To see if
the Rho is significant a significance table is used long with the number of
participants (N) and the type of hypothesis (one or two tailed). To be significant
the Rho needs to be greater than the critical value which means that we can then
reject the null hypothesis and accept the alternative hypothesis and the
significance of the study's results. The critical value is found with the Rho,
number of participants and the type of hypothesis. The + or ­ symbol in the value
of Rho is ignored when finding if the Rho is significant (e.g. +0.58 or -0.58).
Chi squared test
The calculated value of chi needs to be compared to the critical value. If chi is
greater than the critical value then we reject the null hypothesis and there is a
significant difference. Findings from a investigation involving chi squared are
represented in a contingency table. To calculate the degree of freedom the
number of rows minus 1 is multiplied by the number of columns minus 1. This is
used to find the critical value by referring to a critical value table to see if the

Other pages in this set

Page 2

Preview of page 2

Here's a taster:

If it is then there is a significant
difference between the 2 groups.
Mann Whitney U test
A value for U is calculated. The sample sizes of the 2 groups are needed along
with the type of hypothesis to look up the critical value to see if there is a
significant difference. In order to be significant the value of U must be lower
than the critical value.
Wilcoxon T test
The value of T is calculated.…read more

Comments

No comments have yet been made

Similar Psychology resources:

See all Psychology resources »See all resources »