Reliability how important is it?

Sunday, February 19th, 2012

       

Reliability is one of the things that make a research worth paying attention to. When psychologists report their research as reliable, it means that no matter how many times they repeated a certain event, similar results would often occur.

However, why is reliability such an important factor when it comes to research? It is argued that reliability is important in research simply because it can tell you whether your experiment was any good or not.

An example I read on another blog which I thought was good to show reliability, was if your  weights tell you every time you are two stone lighter then the scales would be high in reliability as it consistently tells you, you are two stone lighter. This is similar to experimental research; if your experiment keeps giving you extremely different results when you tested it time and again, you would come to the conclusion that it is not a very good experiment.

There are different types of reliability researchers use in order to test their own experiments. The first one is called Internal Consistency; this means that people answering a questionnaire for example, should all answer in the same way depending on what type of group they have been put into e.g. extraverted people or introverted people. This measure is good because you can clearly see if the questionnaire is reliable and if it needs to be changed in any way. The second one is called Test – retest; this means that people might be given one test and then they repeat the test in order to gain reliable results. However, the problem with this is that if people have already been tested they might already know what to expect from the questionnaire and therefore their results might not necessarily mean that the questionnaire is reliable. The last one is called Interrater; this means that two people observed the same behaviour and they should agree on what is happening. Moreover, if the people agree on the behaviour then it can be suggested that this is reliable. However, the problem with this is that the two different people might have described the behaviour in the same way, but they might not agree on the meaning of the behaviour.

In conclusion, reliability is important when it comes to experimental research because it can show us whether what we are measuring is actually good. However, sometimes the way we test for reliability might not necessarily be reliable in itself.

Advertisements

8 comments on “Reliability how important is it?

  1. lmr92 says:

    I agree that reliability is very important in research studies. If replications of a study get different results each time, we’ll know that this study is not reliable, and therefore won’t be influenced by insignificant data. One thing i would add is that reliability is meaningless without validity. If we go back to your example about the scales consistently saying that someone was 2 stone lighter – yes i agree they do seem reliable, but if this measurement is incorrect – so not valid, then it doesn’t matter whether or not the scales give a consistent reading, because the reading is wrong.
    Overall i’d say that reliability is very important to a research study, but reliability without validity is like ben without jerry: sad, lonely and not important.

  2. psucab says:

    I completely agree that when it comes to research, reliability is up there with the most important things to keep in mind.

    To obtain good, strong, reliable results a researcher must show that the results they have found are not merely flukes, and if the experiment were to be repeated under the same conditions then others would be able to generate the same results.

    Arguably the most important part of conducting an experiment is gathering ones data, and this is where reliability is crucial. To do this we use various instruments, whether this be a stopwatch, a questionnaire etc. It’s obvious that we need to ensure that these different ways of measuring are reliable. A way to improve the reliability of results is to, again, repeat the experiment in order to ensure that chances of malfunction or general error are minimalized.

  3. repugh18 says:

    I completely agree that reliability is crucial for a good piece of research – to enable various different researchers to replicate the research and find similar results would be extremely useful in the field of psychology. Take the research of Ainsworth for example – the research that she carried out was high in validity so was therefore easily replicated world wide to show differences in cross cultural attachments. As well as reliability, validity also has to be considered as although a study may well be reliable if it isn’t valid then it may not be as good as it could possibly be.

  4. I do agree that reliability is important in research, but only quantitative research. As far as qualitative research goes, reliability is totally unnecessary. In qualitative paradigms the terms credibility, neutrality or confirmability, consistency or dependability and applicability or transferability are to be the essential criteria for quality (Lincoln & Guba, 1985). Despite ‘dependability’ corresponding closely with the quantitative concept of reliability, Stenbacka (2001) argues that “If a qualitative study is discussed with reliability as a criterion, the consequence is rather that the study is no good”, highlighting its lack of importance in qualitative research.
    However, it is possible to incorporate the qualities of quantitative research into the qualitative paradigm by using a method called triangulation. If what you are looking for is reliability and validity but are conducting a qualitative study, firstly I would say you have got the wrong idea about qualitative research but nevertheless triangulation will utilize quantitative research’s ability to gain reliability and validity of a measure. Qualitative and quantitative research methods should be treated as separate entities, like tools with different purposes. There are some occasions where one of the ‘tools’ is more appropriate than the other, for example, qualitative methods would not be the right ‘tool’ for measuring reaction times just as quantitative methods would not be appropriate for gaining in-depth insight into somebody’s feelings.
    To summarise my rant I do agree that reliability is important in research, but its importance does not cover all aspects of research; with the main example being qualitative 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

%d bloggers like this: