RSS

V a l i d i t y . . .

10 Mar

Validity is the extent to which the instrument measure the variable to what it intended to measure.

I feel that validity is important because without it the research seems pointless, if the experiment did not measure what they were supposed to, what is the point? The ultimate aim is to tell people what the study has found and what it means, the accuracy and the truth about the results all come from the validity. So if any other ambiguous factors or variables that are found, this would questioned the validity of the research itself.

There are several methods for assessing the validity of measurement (Gravetter & Forzano, 2009). One of the example is called face validity which shows that when a measure (superficially) appears to measure what it claims to measure (this is the least scientific one, but simple as) said by Anastasi (1988). This can be demonstrated via the IQ test that we did for our Personality and Intelligence module, the questions included logic, reasoning, such questions appear to be appropriate for measuring intelligence which means that it has high face validity.

The relationship between validity and reliability, both are required in order to carry out quality measurement procedure, they can be independent and related. The two factors are linked together because a research cannot be valid unless it is reliable. Lets use the IQ test again to explain my point, if I was to do the IQ test twice and results came out to have a huge discrepancy (82 & 139), we would have no idea what my actual IQ is. This big gap between the results is very very unlikely, almost impossible if we are truly measuring intelligence, so I can assume that there during the test, there must be error (random/measurement/experimenter) in which that the numbers have no meaning to us, I mean how do I know if we are actually measure intelligence when using the IQ test (to use methods to assess the validity of research, see above points). On the other hand it is not necessary for a measurement to be valid for it to be reliable.  An example would be your weight claims to be a measure of your IQ, although, there is no validity, it would be classified as reliable, producing steady scores from one measurement to another.  – Just because something that is reliable does not mean it is valid. Just because something that is valid does not mean it will be reliable.

Validity should represent what the researcher tested, and should be strong in terms of that the content validity is high; clearly showing that what you have tested represents your field of study. All aspects of the research must be looked over as to how methods could invalidate findings, so no doubts are being questioned. Overall, Validity and reliability should work in hand in hand, validity alone is not nearly as strong as them two being used together.

Advertisements
 
7 Comments

Posted by on March 10, 2012 in Uncategorized

 

7 responses to “V a l i d i t y . . .

  1. psychmja1

    March 12, 2012 at 11:23 pm

    You mention both face validity and content validity in your blog and I wanted to follow on from that by mentioning construct validity and also two criterion based validity factors- concurrent validity and predictive validity.
    So, a test has construct validity if we can see an association between test scores and the prediction of a theoretical trait. For example, tests of intelligence such as the IQ test are just one example of a type of measurement that has this type of validity. It can be difficult to measure constructs, therefore it is important to demonstrate this validity.
    Criterion related validity refers to a test that has demonstrated its effectiveness in making a prediction about a criterion. Concurrent validity is the first type of criterion validity that I will mention. Concurrent validity refers to a test or measures ability to produce accurate estimations of the current state of an individual in accordance with the criterion. For example, a test that measures stress levels would have concurrent validity if a previous measure that had already been validated had found similar results. Predictive validity is demonstrated by a measure that is able to produce accurate estimations of scores on a different criterion measure. SAT scores are an example of predictive validity as tests were taken in the final year of primary school and then again in year 9. These SAT scores are predictive of performance in future tests, such as GCSE’s. 🙂

     
  2. S-s-s-statman

    March 13, 2012 at 3:02 pm

    The value of validity in an experiment was conclusively covered with very little error if any, face validity was given the most attention – and was covered nicely. This blog was concise, which always appeals to me, especially when commenting. Yet still, I felt that you addressed the topic of validity sufficiently – a big well done!

    You briefly mentioned that “there are several methods for assessing the validity of measurement” which is perfectly correct, and in a perfect world in which you were writing a textbook rather than vast numbers of blogs I am sure you would have given some explanation of the other methods. Gravetter and Forzano (2009) provide a well written narrative in regards to the different forms of validity which I am sure you are well aware of; the remaining five are: (1) Concurrent Validity – “demonstrated when scores obtained from a new measure are directly related to scores obtained from an established measure”, (2) Predictive Validity – “demonstrated when scores obtained from a measure accurately predict behaviour according to a theory”, (3) Construct Validity – “requires that the scores obtained from a measurement procedure behave exactly the same as the variable itself”, (4) Convergent Validity – “demonstrated by strong relationship between the scores obtained from two different measures” and finally, (5) Divergent Validity – “demonstrated by using two different methods of measuring two different constructs, and then convergent validity must be shown for each of the two constructs. There should be little or no relationship between the scores obtained for the two different constructs when they are measured by the same method.

    I cannot criticise this blog, which has made commenting very difficult. To reiterate, this was a very well written concise blog – you obviously ‘get it’!

    As an endnote – It is also important to also consider both internal and external validity, two forms of validity that are often overlooked when writing about validity.

     
  3. poeywycheung

    March 15, 2012 at 12:51 am

    Helloooo 😀
    It is a well-written blog and really easy to understand!
    “Any research can be affected by different kinds of factors which, while extraneous to the concerns of the research, can invalidate the findings” (Seliger & Shohamy 1989, 95).
    Just to back up, there are a few more types of validity, the very main ones are Internal Validity, External Validity. Internal validity is a measure which ensures that a researcher’s experiment design can concluded a causal relationship between the vaiables.
    External validity is about generalisation, whatever it can be applied to other populations, different times and places and still get similar result?

     
  4. instant loans

    May 21, 2013 at 9:19 pm

    irjjiky zickynj rxddpwm mfwjmkg

     

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: