Standardized testing is used to identify specific psychological characteristics in individuals. It is commonly used in the diagnosis of clients, for example to help diagnose language delays and disorders, and to assess cognitive skills. Tests such as the Wechsler Adult Intelligence Scale (2008) which measures intelligence are widely used. However, despite universal acceptance, can standardized tests such as these accurately identify characteristics in every individual, and subsequently be used to influence the work of psychologists?
A standardized test is a measure that is scored and administrated in a consistent manner. For every standardized test there is a set of normative data obtained from samples of the population, which is used to compare the test participants to. This sets standards by which the individual client can be compared with. These tests allow psychologists to work from the same baseline, and know that they are adhering to the same criteria.
However, what happens when these ‘norms’ are not as representative as initially suggested? An example of this appeared in the Minnesota Multiphasic Personality Inventory (Hathaway and McKinley, 1943), which was revised in 1989 to improve its validity. The original sample was found to have lacked representation of ethnic minorities; therefore the revised version used a larger sample that was representative of U.S census figures of 1980.
Reviewing standardized tests should be a necessity, as this allows any flaws to be smoothed over and improved upon. Tests which are absent from any form of review may not be reliable and valid. For example, many of the height and weight charts still used today have not been amended since as early as the 1950s. This ignores the changes of our modern lifestyle, which have led to taller, more overweight adults and children. Bundred , Kitchiner Buchan (2001) found there was a highly significant increase in weight and BMI under the age of four from 1989 to 1998. Results such as this will skew the upper distribution; therefore how can the results from these tests accurately reflect the size of individuals in modern society? Height and weight charts are commonly used to monitor development in childhood, and if a child scores on the upper centile for weight and is perceived as being overweight, this can lead to anxiety in parents.
We must also consider if it is acceptable, ethical and accurate to imply a certain test score indicates some kind of impairment. For example Naremore, Densmore and Harman (1995) suggested issues with the use of standard scores in language impairment. A standard score indicates how many standard deviations a piece of data is above or below the mean. Language tests in particular are often standardized using only children of‘normal’ language ability. Therefore, when deciding a ‘cut-off’ point for language impairment, some of these children will appear below the line. This will mean that any other children of a normal language ability who subsequently take the test could also appear below this statistical line. It will therefore be difficult to distinguish between those children who actually have language impairment and those who do not.
Inevitably, although standardized tests are used often in Psychological research, we must remember that reliability and validity are not in-built qualities of these measures. When assessing individuals, it is vital that these levels are high, as the effect that these tests can have on an individual can significantly impact on their future, and can have life-long implications. Precision is vital, as an incorrect diagnosis can incorrectly colour perceptions of those around the client, and will determine the treatment given. However as long as we are aware of these tests’ limitations, in the differential diagnosis of conditions, the use of them is not significantly detrimental on society.