Reliability vs. Validity

‘Reliability’ and ‘validity’ – two words that almost always crop up when discussing and analysing scientific research. So, why are they so special? This week I’m going to be discussing the importance and possible flaws surrounding these two fundamental aspects of psychological research.

Firstly I will define each term, since I know how easily they can be confused and mixed up! Reliability refers to the consistency of a measurement. For example, you step on your bathroom scales and read that you have lost 5lbs since last week, you think back to the amount of Mcdonalds you stuffed yourself with that week and so step back onto the scales to double check. Now you read that you have in fact gained 2lbs (which sounds a bit more reasonable) and realise that you can’t really trust your scales, you need one that is more reliable.  Validity, on the other hand is the extent to which the procedure measures what it intends to measure. There are many different types of validity, including external validity which is the extent to which the results can be generalised, and internal validity which is when a study produces one single explanation for the relationship between the variables.

Within psychological research it is fully understood that achieving perfect reliability is next to impossible since many error sources will be having an impact on the consistency of results. For a start, psychological research usually involves humans and the use of humans generally leads to inconsistency. We get tired, we daydream and most of us get bored of continuing repetitive tasks. Environmental changes can also have an impact on results, for example, the time of day, temperature and lighting conditions. However, there are ways to assess and improve issues of reliability. The test-restest method is a good example of reliability assessment, this is when the study is carried out then repeated after a suitable interval. This method was used when assessing the reliability of the Beck Depression Inventory. Beck et al. (1996) studied the responses of 26 outpatients on two separate therapy sessions one week apart, they found a correlation of .93 therefore demonstrating high test-restest reliability of the depression inventory¹. This is an example of why reliability in psychological research is necessary, if it wasn’t for the reliability of such tests some individuals may not be successfully diagnosed with disorders such as depression and consequently will not be given appropriate therapy.

Research strives to have high validity in order to achieve valid conclusions from studies. Results of a study must be valid to be accurately applied and interpreted. Many variables are very difficult to study in psychological research, such as hypothetical constructs as they cannot be directly observed or measured. For example, how do we know if we are actually measuring intelligence when using IQ tests? Therefore, several ways of assessing the validity of research have been established. One example is concurrent validity. This is when validity is established by comparing one method with another, previously validated method. Mattick and Clarke (1998) illustrated that their Social Phobia Scale correlated well with older, accepted measures².

Reliability and validity are both very important criteria for analysing the quality of measures. Although they are independent aspects, they are also somewhat related. A measurement procedure cannot be valid unless it’s reliable however, a measurement can be reliable without being valid (someone could measure your height and say they are measuring intelligence, they would get a consistent and reliable result each time but this would NOT be valid since height is not an indicator of intelligence). In psychological research there is rarely an established standard for measurements. For instance it is hard to analyse how accurate a depression measure is since there is no official unit of depression, therefore validity and reliability substitute this problem. Without reliability and validity it would be very difficult to decide which research should be trusted and which should be completely disregarded.

In all, reliability and validity are important factors of psychological research studies. They allow us to gain firm and accurate results, as well as helping us to generalise our findings to a wider population and, in turn, apply research results to the world to improve aspects of people’s lives.

¹. Beck, A. T., Steer, R. A., and Brown, G. K. (1996). Manuel for the Beck Depression Inventory.

². Mattick, R. P, and Clarke, J. C. (1998) Development and validation of measures of social phobia scrutiny fear and social interaction anxiety.

About these ads

4 responses to “Reliability vs. Validity

  1. There are some interesting points raised and some good references, i too used the Beck et al. (1996) example as this does clearly demonstrate the importance of reliability and that it is essential, particularly in our field of psychology, in enabling us to go on and do further research in the safe knowledge that t a type of measurement procedure is reliable.
    Both reliability and validity, as you have said are important for psychological research but perhaps something could be further discussed is how reliability and/or validity are used in the everyday aspect, as there are many companies claiming that their products are reliable and this is important to us as the consumer, rather than the psychologist.

  2. Pingback: Homework for my TA…. Week 4/5 « psuc27

  3. Another way to get more reliable results would be to use better measurement methods to show measurements more realistically e.g. using lbs instead of stones. This allows us to see our weight in a better light. For example if we just uses stones it may show that every month we weigh the same but if we used lbs we may see, for example that we have been putting weight on.
    I liked the way that you explained the 2 different types of validity and used an example that we can all relate to from having to do the IQ tests in Personality and Individual Differences. I also thought that it was good that you also had a study to use as evidence. It is always good to have evidence to support your statement and to explain how it benefits us now in real life.
    I thought that your explanation about how a measurement cannot be valid unless it’s reliable but a measurement can be reliable without being valid was clear and explained the fact well.

  4. Pingback: Homework for my TA Week 9 « hls92

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