Reliability.

Reliability.

1. Reliability is based on how consistent an experiment is, therefore linking to the replicability also. Research is reliable if it produces the same results each time.

2. Internal Reliability: Consistency within the method. This is the concern of consistency within the specific test.

3. External Reliability: Consistency between uses of the method. This is the ability to produce the same findings every time a test is carried out.

4. Measuring Internal Reliability:

   • Split-half technique- This is used specifically with a test whereby half the scores such as the even numbered questions are correlated with the odd numbered questions to ensure that they are similar in terms of difficulty for example. A high degree of similarity would give the study high internal validity.
   • Inter-Rater Reliability- This is used to test the accuracy of the observations. If the same behavior is rated the same between two different observers then the observations are externally reliable. The observers must be well trained and have clear observation criteria to ensure the rating is consistent.

5. Measuring External Reliability:

   • Test-Retest- This involves testing once and then testing again at a later date, so replicating the original research. Strong consistency between the two different findings shows it is reliable.

6. Improving Reliability:

   • Inaccuracy can be reduced by: 1) taking more than one measurement from each participant, e.g. calculating a score over a number of trials. 2) Pilot studies can be used to check that the measurement works properly. 3) If there is more than one researcher they should be trained (inter-rater reliability). 4) Ensure that the measurement is as objective as possible when regarding what to do.