Just another WordPress.com site

Reliability in science

Reliability offers importance in psychological experiments such as quality of measurement this is defined as the extent to which you can measure a participant’s true score, it is something that every scientist must be aware of, especially in social sciences and biological experiments. True score is defined as a person’s general ability to take part in a study to their full potential. Reliability is also how repeatable the experiment is… is this really important when it comes to scientific tests, and why?

Reliability is important when it comes to scientific tests, it is essential for experiments to have a measure of quality so that other psychologists can carry out the tests again and investigate whether their results correspond to those that have been published, the consistency of the experiment must be maintained, a researcher must use as many repeat sample groups as possible, to reduce the chance of an abnormal sample group skewing the results. For example, if you use the same samples in each manipulation of an experiment and one generates results that are completely different from the others, then you know that there may be faults within the experiment, this is also where pilot studies can be useful, as a smaller scale experiment can reduce the amount of abnormalities found in the proper experiment as these have already been altered.

In relation to the real world, reliability is essential when it comes to competition of products being sold, if a company presents something about their product which makes it stand out from the rest of the competition, and this is then proven to be wrong, the company loses the reliability it has been working so hard to gain, negative customer satisfaction also loses reliability as they will not give positive feedback to others. Their reputation will also be lost, as customers will go to companies whose products have never failed in reliability.

In conclusion, I believe that reliability is essential in scientific experiments and is something that all scientists do and should take into consideration within their experiments. Through research I have seen that reliability is also important within the consumer world, if a company or product lacks reliability then this can discredit their product and make it less sellable to the public.

Advertisements

Comments on: "Week 5- Reliability in Science" (3)

  1. I agree that reliability is an essential part of research; good measures and observations must be consistent. As well as being important in research and the consumer world have you considered how important reliability is in the applications of psychology, for example diagnosis of disorders?

    Reliability is extremely important in the classification and diagnosis of disorders such as schizophrenia, depression, phobic disorders and obsessive-compulsive disorders. With reference to psychological disorders reliability refers to the extent to which psychiatrists can agree on the same diagnosis when assessing patients independently, this seems to create much of a debate in psychology as to whether the diagnosis is reliable.

    As a method of increasing reliability in the diagnosis and classification of disorders the Diagnostic and Statistical Manual of Mental Disorders (DSM-III, 1980)* was introduced. Carson (1991)** claimed the DSM-III had overcome the issues of reliability once and for all. Think about it, psychiatrists now had a fairly reliable classification system so this should have led to a greater agreement of who had schizophrenia and depression and who didn’t. Despite people claiming the DSM-III increased reliability there is still little evidence that clinical psychologists routinely use DSM with high levels of reliability. Inter rater reliability correlations in the diagnosis of schizophrenia were as low as 0.11 (Whaley, 2001)***.

    Maybe when discussing reliability in the future we shouldn’t be so naïve by arguing how reliability is just a method to meet scientific criteria. Reliability stems much further than the discussion of science and has real world important applications in the clinical practice. It seems to me that reliability isn’t just about manufactures having credibility and selling more, reliability is much more important in ensuring patients receive the most sound diagnosis leading to the most effective treatment or therapy.

    * American Psychiatric Association. (2000). Diagnostic and statistical manual of mental disorders (4th ed., text revision). Washington, DC: Au.
    ** Carson, R. C. (1991). Dilemmas in the pathway of the DSM-IV. Journal of Abnormal Psychology, 100, 302-7.
    *** Whaley, A. L. (2001). Cultural mistrust and clinical diagnosis of paranoid schizophrenia in African-America patients. Journal of Psychopathology and Behavioural Assessment, 23, 93-100.

  2. I think both the actual blog and the above comment are very good, using real-world situations to portray the need for reliability. If we are to believe and accept the results of a study, no matter what the findings are, we are less likely to believe it if it is the only study to have found such results. However, if several studies have investigated the same area and found similar results, we’re likely to believe it. Reliability like this leads to paradigms, as the increased reliability in one area, and decrease of reliability in another leads to a change in belief. For example, Freud though that personality was a result of inner unconscious conflicts, and in the era of his reasoning, he had much support. However, as scientific research became more widely used and accepted, people found that genetics and upbringing were also key to personality development.

    However, how can we be sure of the reliability of a study which is un-replicable due to ethical constraints, or simply because the variables cannot be made to the same degree as in the original study. Take Milgram’s (1963) experiment. The reasons for the inability to replicate this study in modern society are obvious – deception and participant protection to name a few – so how can we be sure of the reliability of it’s findings? Yes, studies have attempted to replicate the study, within the ethical guidelines, but we cannot know for sure without almost exact replication that the study was reliable.

    Also, take studies conducted years ago. They may well have had a lot of research support at the time of investigation, but moving over 50 years ahead, society has changed, and so will have the results. This means the study was reliable at some point, but relating it to psychology of today, it isn’t reliable at all!

    Groundbreaking studies also have an issue in terms of their reliability. Because they are, arguably, the first research conducted in their field, are they reliable, bearing in mind we have nothing to test it’s re-testability? Can we assume something is reliable, until proven otherwise, or should we always question the research?

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

Tag Cloud

%d bloggers like this: