operationalizing variables

Variables are really important in a psychological research as they do specify our results. How we manipulate our variables (such as behaviour, characteristics) will have a result on the outcome of the research. In order to gather the desired outcome we need to operationalise those variables. This means that we need to define in depth our independent variables and dependent variables. In other words we have to say how we want to measure the behaviour. For example Bandura (1961) intended to measure aggression. We cannot simply just measure aggression as there is different types of it. Therefore he specified the inter rater observer validity: such as kicking, punching the doll where the behaviour perceived as aggressive. There were three conditions in which children could either see a model who was behaving aggressively do the doll or not. The children were them observed if they would repeat aggressive behaviour or not to the doll.

 

Some of the behaviours cannot be seen and therefore we have to operationalise them in order to investigate the reason underneath. For example thoughts in depression cannot be overtly observed. Therefore we need to give for example a valid questionnaire which measure thoughts (independent variable) to see what thoughts people with depression have it (dependent variable). If the research operationalise variable in a right way it allows him/her to exclude any extraneous variables/confounding variables. This means that all the unwanted variables that could affect the results are controlled and do not affect the explanation of the behaviour. For example if our independent variable is a memory task, other variables such as: participant’s tiredness, mood etc could have an impact on the results (dependent variable). Therefore it is extremely important to strictly operationalise our independent variable. Good control over the variables is also beneficial to the research as it shows that the research is high in internal validity.

 

On the other hand operationalisation of the variables could be a bit subjective in some methods. Take into account natural experiment or observational studies. Yes you define the inter rater reliability (you operationalise what you will measure by specifying the patterns of behaviour that you look for) however every researcher have got they own scale and reasoning on what they perceive as right. This could lead to the researcher bias. For example, Rosehan (1973) conducted a research about people behaviours after labelling mentally ill people with different disorders. He and his colleagues were admitted to the psychiatric hospital on the basics of schizophrenia disorder. In fact they were perfectly fine they were just faking the symptoms of that disorder. The aim of this study was to observe how the medical staff treats people based on the label imposed. Obviously the behaviour that they reported was only observed by them and therefore the independent variables was not really operationalized. Clearly this threats internal validity of any research.

 

In conclusion the purpose of every research is to operationalise it variables in order to minimanise any errors that could possibly affect the findings and give wrong results. Researchers strive to make their research looks high in internal validly(good control) by sacrificing external validity (generalisability) .  Every research design have main two important variables that are being defined (operationalsied) . Independent variable is being manipulated to see the effect on the dependent variable. Different factors investigated will have different operationalisation.