amw1992

Can psychologists measure everything?

with 10 comments

 As psychologists our primary interests are looking into behaviour and social interactions, these can measured in two ways: via overt behaviours and covert behaviours. Overt behaviours are behaviours that can be observed and directly measured such as how many times someone hits someone. Covert behaviours are behaviours which cannot be directly measured, such as self-esteem and fear. These types of behaviours are said to be hypothetical constructs. To measure such behaviours we use operational definitions, these are things like physiological measures such as heart-rate and sweat, or IQ tests, they are an indirect measure of the construct we wish to measure. For example in order for psychologists to assess how depressed a person may be (depression being a covert behaviour) they ask the individual to fill out the Beck Depression Inventory or BDI ( Dr. A, T. Beck), this is a 21 question multiple choice, self-report inventory. As you can imagine this open to social desirability or can be manipulated by the individual to reach the outcome they desire, for example if someone is seeking attention and being ‘depressed’ will gain them that attention then they are more likely to up play being depressed when in fact they aren’t. Thus showing that the use of operational definitions is open to convergent validity problems, this is the idea that the measure you are using is similar to the desired behaviour in which you wish to measure.

            It would seem that psychologists can measure both overt and covert behaviours via the use of direct observation and operational definitions. However the use of operational definitions to measure covert behaviours is a less valid and reliable technique than direct observation, and as suggested by behavioural psychologists should not be measured. However clinical psychologists disagree as they measure covert behaviours in order to carry out assessments on an individual. So to conclude, yes psychologists can measure everything to do with behaviour, all be it on method is less reliable than the other.

 

Written by amw1992

February 19, 2012 at 12:12 pm

Posted in Uncategorized

«
»

10 Responses

Subscribe to comments with RSS.

  1. Thank you for this wonderful and brilliant article. The measurement of OVERT and COVERT behaviours has been given me some stress but now I am cleared.

    Monday Meesua Gentle

    February 19, 2012 at 2:42 pm

    Reply

  2. I agree with above comment, thank you so much for clearing up the difference between the two, they can be so confusing because the two words sound so similar ( why does this always happen in psychology?).

    You raise a good point about how the use of operational definitions to measure covert behavior are less valid and reliable than measuring overt behavior. Even as early as 1943 Start Dodd was concerned with making operational definition more reliable. There are however ways of making sure that the particular operational definition that you use is more reliable than another. For example interrater reliability (which is is the degree of agreement among raters). Dodd (1943) noted the more complex the behavior that is recorded, the more difficult it is to achieve good interrater reliability, so therefore this would suggest the less complex the operational definition the more reliable it is.

    prpsjj

    February 20, 2012 at 12:49 pm

    Reply

    • I believe you have clearly stated your point of view and supported it with strong evidence. And indeed, self-report often raises the issue of a lack of convergent validity. However, should covert behaviour even be measured at all if it is not as good as overt behavioural measures? I agree with the previous comment that there are ways of preventing a problem in validity. By definition it cannot be directly measured and violates the assumption of objectivity. Yet behaviour analysts seek social validity and therefore subjectivity (http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1311293/ ). I’d like to expand on your statement about overt behaviour measurement as it has generally been shown to be highly reliable and valid. OBS(Overt Behaviour Scale) was introduced as a measurement method for the challenging behaviours of individuals who suffered brain injury. It measures verbal and physical aggression towards the self, towards others and towards the environment. (http://www.tbims.org/combi/obs/index.html).This illustrates just how important observing overt behaviour can be for understanding and intervening in behaviours with problematic outcomes. But measuring covert behaviour is equally crucial for the perpetuation of the discovery of cognitive functionning. For instance, a state of depression characterised by feeling low, having a diminished appetite etc can only really be measured by asking the patient. Although the results may not be entirely valid, they will be as accurate as if someone else interpreted his/her feelings. While it is possible to measure expression of those feelings, feelings remain personal and covert. But have you considered the possibility of combining both direct observation and self-report measures? In the case of a child with Prader-Willi syndrome (http://www.medicinenet.com/prader-willi_syndrome/article.htm) for instance, keeping a record of whether he/she is eating and when she/he feels hungry is measuring both overt (eating) and covert (hunger) behaviours and bringing both together could lead to a better dealing strategy for living with this incurable disease.

      psuc6b

      February 21, 2012 at 5:28 pm

      Reply

  5. I think that psychologists can not measure everything. Some constructs, for example love, are difficult to define and therefore difficult to measure. How could we measure love? Should we measure how much time partners spend together, or how many times a week she makes a delicious dinner and ions his shirts. There are some good operational definitions which are reliable and can be used even by people who do not have anything to do with psychology (Howitt & Cramer, 2011). Measures of stress, for example Student Academic Stress Scale (SASS), have been found to be reliable (Busari, 2011). But, there are also many operational definitions of one construct and we do not know if they are valid, for example measures of schizophrenia (Murray, 1997).

    vanilla85

    February 22, 2012 at 8:42 pm

    Reply

  7. You have made a strong article involving the differences between both the observations which I at times find hard to differentiate and your blog has helped a lot with this. Regarding your point on the self report measurement on the beck depression inventory I agree it is open to social desirability biases. There is an article (http://www.sciencedaily.com/releases/2007/09/070905203624.htm) to do with research on seeing if there is a noticeable difference in activity between clinically depressed and healthy control subjects. They found that after a 5 day period of monitoring both groups motions using an accelerometer showed a difference between the groups using a scaling parameter. It is a type of probability distribution whereby the larger the scale parameter, the more spread out the distribution. The results found that for patients with major depression, the scaling parameter is smaller than it is for the healthy subjects. This would mean that there might be a significance in activity levels for those suffering with depression as the distribution is smaller and therefore more concentrated. The implication of this is that applying statistical analyses could possibly provide a more accurate measurement for measuring depression and possibly other illnesses.

    tommywiseau

    February 22, 2012 at 9:22 pm

    Reply

  9. I’d like to think that psychologists can measure everything… eventually. Just look how elusive body language and lying was until Paul Ekman came and said “relax! I got this.”. Sure it’s still unreliable, but it’s valid enough to be useful in forensic sciences. I’d like to think that as we dig deeper, into the head, into the neuron, there won’t be any secrets left someday.

    Frank M.G. & Ekman, P.
    Nonverbal Detection of Deception in Forensic Contexts (2003)
    Handbook of Forensic Psychology. Academic Press.

    aglinskas

    February 23, 2012 at 12:08 am

    Reply

Leave a comment