Variables and experimental desighn
- 1. Research Methods - Variables and Experimental Design
- 2. Variables A variable is a factor that has the capacity to increase or decrease in amount or kind.
- 3. What is an example of a variable?
- 4. Independent Variable (IV) An Independent variable (IV) refers to the variable that is systematically manipulated, changed, or varied in some way by the experimenter in order to assess its effect.
- 5. Dependent Variable The dependent variable (DV) shows any effects of the independent variable that can be observed or measured as being varied in some way as a result changes in the independent variable.
- 6. Extraneous variables Variables, other than (extraneous to) the independent variable, which may influence the dependent variable and therefore the results of the experiment are called extraneous variables.
- 7. Extraneous Extraneous variables can be divided into two broad categories: Participant variables Experimenter variables
- 8. Participant Variables • Sex Age • Ethnic background • Abilities • Mood • Self-esteem Intelligence
- 9. Experimenter Variables Personal characteristics, such as age, sex, and ethnic background Experimenter expectancies – behaviors that may influence unintentionally the results due to inaccurate observation Experimenter bias – experimenter behaving in a way that shows what their expectancies to participants
- 10. Uncontrolled If an extraneous variable is not controlled for (held constant) its effects become confused with or confounded with the effects of the independent variable. It then has an unwanted effect on the dependent variable and is termed a confounding variable . The researcher can no longer draw conclusions regarding the causal relationship that exists between the independent variable and the dependent variable.
- 11. Experimental Design In order to minimise the effects of extraneous variables, researchers may choose the following experimental designs. Independent groups design Matched pairs design Repeated measures design
- 12. Independent Groups Design Participants are randomly allocated to either the experimental group or control group. A limitation of the independent groups design is the potential of variables other than the independent variable because of differences among the participants in the groups
- 14. Matched pairs design Participants are matched as pairs on the basis of relevant variables. One of the pair is allocated to the experimental group and the other to the control group A limitation of this experimental design is that participants can never be perfectly matched in every respect. The process of matching participants is also very time consuming and can be difficult to achieve
- 16. Repeated measures design Participants are used in both the experimental group and the control group A limitation of this design is that it may create a ‘practice’ effect or ‘order’ effect where performance from the first occasion may improve or disrupt performance on the second occasion.
- 18. Researchers may therefore use the experimental procedure of counterbalancing to eliminate possible order effects such as fatigue, practice effects and boredom. Thus the experimental tasks are performed in different sequences by different participants or by the same participants at different times.
- 19. Placebo Effect A placebo effect refers to the participants’ response being influenced by their expectations of how they should behave.
- 20. Single Blind Procedure The placebo effect can be minimised by using a single blind procedure when allocating participants to groups. In a single blind procedure, the participants do not know whether they are in the control group or experimental group condition. This prevents the participants’ knowledge or expectations of the experiment impacting on the results.
- 21. Experimenter Effect This refers to the way an experimenter (rather than the IV) might influence the dependent variable, and therefore the results of an experiment One way is through experimenter bias , also referred to as experimenter expectancy effect. This is where the experimenter has certain expectations about the outcome of the experiment, which can subtly alter his/her behaviour and bias the results.
- 22. DOUBLE BLIND PROCEDURE In order to minimise experimenter effects, researchers may use a double blind procedure. In a double blind experiment, neither the experimenter nor the participants know whether the participants are assigned to the experimental or control group until after testing or the conclusion of the experiment A third person controls the placement of groups and never has contact with the participants. This controls the Placebo effect and the Experimenter effect.