DESIGN OF RESEARCH STUDIES
Planning and preparation is necessary if a research study is to provide appropriate and informative data.
Aims hypotheses and operationalisation
AIMS: El objetivo, es el preciso enunciado de porque se realiza la investigación. Deben considerar que es lo que se estudia y que se intenta lograr. Once the general aims have been established, a muchmore precise hypothesis can be developed.
HYPOTHESIS: Predice que es lo que pasará en el estudio. Es mas precisa que el aim. El IV es manipulada por el experimenter, y las confounding variables para asses the effects de esta manipulacion en la VD. Se muestra así una relación de causa-efecto entre las 2 variables. It is the precise prediction of what the result of the manipulation of the IV willbe.
- Alternative (or experimental): alternative with any study but experimental with experiment study. All experimental hypothesis are alternative but not all alternative are experimental.
o Directional: The direction of the difference or relationship is predicted. Are used when there is some evidence.
o Non-directional: predicts that there will be a difference between conditions or adifference between variables, but give no indication as to the nature of the difference or relationship.
- Null : predicts that there will be no difference between conditions, or no relationship between variables.
OPERATIONALISATION: In order to develop a hypothesis, the researcher needs to use a precise definition of the variables in terms of how they are o be used in a particular study.Experimental design We have to decide which kind of design is more appropriate.
Independent groups design
Different participants take part in different conditions and are randomly allocated to conditions. In this design, there could be experimental conditions receiving variations in experimental treatment. It is also possible to have one or more experimental conditions together with a control condition,where participants are given no particular treatment. Since participants are only exposed to one condition, this design eliminates possible order effects (when carrying out a task in one condition may affect performance in the other condition(s)). However there is a problem with individual differences, because the participants in one condition may be different in some important respect from thosein the other condition(s), which would bias the results.
Repeated measures design
Each participant provides data in both (or all) conditions, and his or her performance in each condition is compared. An advantage of the design over an independent group design is that any effects of individual differences are eliminated, because the performance of each participant in one condition is comparedwith that same person´s performance in the other condition(s). The main problem with the design is possible order effects. These effects include practice, boredom and fatigue. It is sometimes possible to minimize these effects by counterbalancing (by varying the order in which participants do the tasks). Another drawback is that more materials may need to be prepared when repeated measures designis used than when the design uses independent groups.
One of the criteria for a study to be an experiment is that there should be an element of random allocation to avoid the possibility of bias in the way in which the experiment is carried out. In repeated measures design, random allocation is not possible, as participants take part in both or all the conditions. But the order in which theycarry out each condition should be randomized, within the constraints of counterbalancing.
Matched pairs design
This design aims to exploit the advantages of both independent groups design and repeated measures design, while avoiding some drawbacks of both. Each participant is matched with another participant on characteristics that are thought to be relevant to the experiment, such as age, sex,...