PSY2 – Research Methods: Lesson 07

Reliability and validity

PSY2 RM PowerpointTeachers or students who want the original PPT please tweet @psychopepper

Learning Objectives

  • To KNOW and UNDERSTAND what reliability is, both internal and external.
  • To ANALYSE ways in which we can test the reliability of a measure, both internal and external.
  • To KNOW and UNDERSTAND what validity is, both internal and external.
  • To ANALYSE ways in which we can test the validity of a measure, both internal and external.

Lesson Outline:

Tasks:

  1. Take notes on the definition of reliability in booklet understanding that consistency is the key word (sub in to check meaning in written work).
  2. To reflect on the different ways we can test for and improve reliability (A Level).
  3. Take notes on the definition of validity in booklet understanding that accuracy is the key word (sub in to check meaning in written work).
  4. To reflect on the different ways we can test for and improve validity(A Level).
  5. To discuss how the terms will be used throughout the course to evaluate studies.
  6. To create two foldable flowers as a revision aid of the key terms associated with reliability and validity.
  7. Utilise the target boards to discuss and visualise that the two terms are not interchangeable.

Content Recap:

Reliability: How consistently a method measures something. It relates to our ability to repeat a study and obtain the same results. It is linked to replication but is not the same.

External Reliability: How reliable (or consistent) a measure is over time i.e. does it give the same results time after time.

Test-Retest Method: A way of testing the external reliability (or consistency) of a measure (normally a self report). The same measure is given to the same participants again and you are looking for a high degree of similarity (0.8 or 80% correlation is normally accepted as reliable).

Replication: A way of testing the external reliability (or consistency) of a measure (normally an experiment). Do the experiment again with exactly the same standardised procedure and you are looking for a high degree of similarity (0.8 or 80% correlation is normally accepted as reliable).

Internal Reliability: How reliable (or consistent) a measure is within itself i.e. does it give the same results throughout.

Split Half Method: A way of testing the internal reliability (or consistency) of a measure (normally a self report). The measure is split in half (odd questions and even questions) and you are looking for a high degree of similarity in the score in each half (0.8 or 80% correlation is normally accepted as reliable).

Inter-Rater Reliability: A way of testing the internal reliability (or consistency) of a measure when outside raters are being used. Instead of one person making judgements several are used and their judgements are compared to each other. You are looking for a high degree of similarity (0.8 or 80% correlation is normally accepted as reliable). This is also called inter-interviewer (when it is an interview) and inter-observer (then it is an observation).

Validity: How accurately a method measures something. It relates to how truthful, ‘real’ or ‘honest’ the study is and if it is a fair reflection of reality.

External Validity: How valid (or accurate) a measure is over time i.e. does it give the same results (generalise) in different scenarios.

Ecological Validity: The ability to generalise the results to different places/setting (outside the lab). The effect is still accurate when the high level of control in the lab is removed.

Population Validity: The ability to generalise the results to different people. The effect is still accurate when you use different types of people.

Temporal Validity: The ability to generalise the results to a different time/era (sometimes called historical validity). The effect is still accurate when conducted in modern times.

Internal Validity: How valid (or accurate) a measure is within itself i.e. how confident we can be that we measured what we intended to measure.

Control: The amount of restrictions placed. In order to have a lot of control a research study needs to identify any potential influences on the result and make sure they are kept constant. If this happens then we know that what we did caused the changes we see. (A positive of laboratory experiments).

Mundane Realism: The extent to which an experiment reflects the real world. Mundane: meaning boring, an experiment should reflect what happens in our day to day lives. (A negative of laboratory experiments because they don’t have it).

If you would like to access PsychoPepper’s Glossary of Terms please click here

Resources used

Reliability and Validity Foldable Flowers

Study Notes (TBC):

If you would like to download a set of study notes for this lesson, you can do so here. If you would like access to the original word file please tweet @psychopepper

TEXTBOOKS OR EXTRA READING:

I would suggest that you have a copy of one of the textbooks which will allow you to read around the subject matter, pre-read ahead of lessons or even take extra notes/practise questions afterwards. I would recommend the following (you do not need to replicate books, one of each type is plenty!)

Extra ReadingElephants on Acid or Hippo Eats Dwarf

Please follow and like us:

Leave a Reply

Your email address will not be published. Required fields are marked *