Sunteți pe pagina 1din 7

A Change From Traditional Methods

Terri-Lynn Jones-Wood
July 14, 2015

Human creations are constantly evolving, from the creation of the Internet to the
creation of the first iPhone. When new substances are created, whether it is a
psychological assessment or the latest gaming system, consumers look to the validity of
these creations to determine whether or not they are worth consuming. Psychological
assessments are particularly important when considering the level of validity because
they can push forward progress for an individual or hinder their progress. There are
many different processes that can be used when determining validity, however on in
particular, the Process Focused (PD) model stands out because it is variably different
from the traditional processes.
Assessments have been used consistently throughout history, however validity
came into existence long after. Validity was not a key factor until the mid 20th century,
when psychologists Crohnbach and Meehl published Construct Validity in
Psychological Test. The other crucial paper was Convergent and Discriminant
Validation by the Multitrait-Multimethod Matrix, published by Campbell and Fiske.
Studies over the past 50 years have based their starting points on these two crucial papers.
This new factor of assessments allowed psychologists to determine a coefficient to
determine the relationship that has developed as a criterion that could potentially affect a
test score. An example of this would be a variable on a test that creates some type of
error.
There are innumerable variables that can impact assessment scores. Despite their
magnitude, these variables are can easily be analyzed, which was originally the purpose
of the traditional method. In the traditional method, the heart of validity lies in outcome
in the relation of predictor scores to some criterion measure. (Bornstein, 2011).

However, the PF model changes the emphasis of validity from outcome to process as well
as focusing less on correlation and more on experimentation.
Traditionally, validity has focused largely on construct validity, content validity,
and criterion-related validity. Construct validity is when an unobservable construct is
obsessed and can be broken down into convergent validity and discriminant validity.
Convergent validity allows for a theoretical relationship between a variable and a test
score. Discriminant validity, on the other hand, allows for a theoretical view when there
is either minimal to no relationship between a test score and a variable.
The traditional process of validity uses this process to recognize that the criterion
can function as concurrent validity or predictive validity. Concurrent validity is an index
of the degree to which a test score is related to some criterion measure obtained at the
same time, or concurrently (Cohen, Swerdlik, & Sturman. 2013). Predictive validity is
an index of the degree to which a test score predicts some criterion measure (Cohen,
Swerdlik, & Sturman. 2013). Concurrent validity deals with things occurring at the
same time and involves present time. Predictive validity is associated with the future. A
major flaw in using the traditional method is this dichotomy between concurrent and
predictive validity. Concurrent validity, although dealing with the present, could
potentially restart itself in the predictive future without knowing the precise point of
change or occurrence.
In contrast, the PF Model is intended to use every aspect of the observable
criterion as a calculating tool. The PF Model will identify the variables of the criterion
that can affect the test score of whatever assessment tool is being used. Then, it will

predict the outcome of that relationship and how it was created. This will allow for the
individual administering and analyzing the test to give the test taker options to cover a
particular variable. This was created after constant research and analysis related to
cognitive modeling. Since Embretsons introduction of the concept of construct
representation, numerous researchers have used these techniques to deconstruct the
cognitive processes that occur when respondents engage items from various measures of
aptitude, intelligence, and mental ability, adding expert ratings of item content and causal
modeling techniques as additional methods for evaluating construct representation
(Bornstein, 2011).
The traditional method focuses on enumerating the relationships between criterion
and the test takers scores. The PF Model is structured to begin in a similar manner, but
extends it. Simply put, the PF model is a more enhanced process because it simplifies,
while still enhancing researchers understanding of test bias and test score misuse. This
allows for modifications to be made depending on the assessment. This model can be
transformed into one that is well adjusted. The PF model conceptualizes validity as the
degree to which respondents can be shown to engage in a predictable set of psychological
processes during assessment, with those processes dictated a priori by the nature of the
instrument(s) used, and context in which testing takes place (Bornstein, 2011). The PF
Model is simply the more effective and enhanced process than the traditional method.
In considering a chosen construct to assess validity, teen substance abuse is
something considerably important, while using the Alcohol Dependence Scale (ADS) as a
method of assessment. Using Bornsteins four-step process can aid in assessing validity
for the Alcohol Dependence Scale. For the first step in the PF Model, a psychological

process needs to be specified along with context variables that can alter this process. The
psychological is using the ADS to provide a quantitative measure of the severity of
alcohol dependence consistent with the concept of the alcohol dependence syndrome.
The assessment consists of 25 items that cover multiple topics directly related to alcohol
use. The process can be altered through changing the timeframe that it is given or giving
instructions that are slightly different than the normal ones used. Telling test takers that
the assessment measures something slightly different could also be a way to alter the
process or not telling them at all. Since this is typically a test given to adults, altering the
test by administering it to teens is also another way to alter the process.
The second step requires simple evaluation of the results of the assessment. The
test is only a five-minute test and an evaluation of the test can be used to determine
alcohol dependence. A score of 9 or more on the assessment is highly predictive of a
diagnosis of alcohol dependence. This allows for psychologists to determine the level of
intervention for the test taker. The information can also be used for basic research
studies regarding the severity of alcohol dependence.
The third step requires determining flaws in the experimental design of the
assessment. There are certain variables that can alter the validity of the assessment. This
particular study deals with substance abusers, which can be very problematic when it
comes to administering the test. The test taker may be inhibited at the time to testing,
causing answers and responses to be inaccurate. The test taker could also be nervous and
anxious in giving their answers and may not answer truthfully. There is also a possibility
of error on the test administrator part. The test administrator could be unclear with
questions or directions given during the test. The final step in this process is to test in

various populations to aid in determining validity amongst various groups and settings.
This step represents a more long-term goal and requires additional research and studies.

References
Cohen, R., & Swerdlik, M. (2013). Psychological testing and assessment: An
introduction to tests and measurement (8th ed.). New York, NY: McGraw-Hill.
Bornstein, R. (2011). Toward a Process-Focused Model of Test Score Validity: Improving
Psychological Assessment in Science and Practice. Psychological Assessment, 23(2),
532-534.

S-ar putea să vă placă și