assessing reliability?

what does the test-retest theory of assessing reliability test?

the consistency of measures when the same test is administered to the same person twice.

when is test-retest used?

experiments

how do you use a test-retest to assess reliability?

1. conduct experiment
2. wait for a period of time
3. conduct experiment again - same participants
4. conduct a correlation between participants score on first and second go. If correlation +0.80 then the experiment is reliable.

what does the inter-observer theory of assessing reliability test?

it measures the degree of agreement between different people observing or assessing the same thing

when is inter-observer used?

observations

how do you use an inter-observer to assess reliability?

1. conduct observation
2. observers sit apart
3. do a correlation between observer 1 scores (tally) and observer 2 (tally). Correlation co-efficient should be +0.80 or over to state that your observation is reliable.

how do you observe an observation - content analysis?

1\. content (decide on the content for your content analysis ie text, video etc)

2\. categories (behavioural- decide on them)

3\. code (operationalise (make them measurable) the behavioural categories)

4\. count (tally the behaviours you see – each observer sits apart)

5\. correlation/compare (must be +0.8 correlation coefficient to be a strong positive correlation between the 2 observers (inter-rater reliability)