CLASSIFICATION AND ASSESSMENT:Reliability, Test retest, Split Half

<< CAUSES OF ABNORMAL BEHAVIOR ETIOLOGICAL FACTORS OF ABNORMALITY

DIAGNOSING PSYCHOLOGICAL DISORDERS:The categorical approach, Prototypical approach >>

Abnormal Psychology PSY404

LESSON 15

CLASSIFICATION AND ASSESSMENT

Assessment is the process of gathering information from a new patient. It is the systematic collection and

analysis of information about a person's characteristics and behaviors.

A classification system consists of a list of various types of problems and their associated symptoms.

In Diagnosis we identify or recognize a disorder on the basis of its characteristic symptoms.

A classification system is used to subdivide or organize a set of objects. Classification system can be based

on various principles, and its value will depend primarily on the purpose for which they were developed.

Different classification systems are not necessarily right or wrong; they are simply more or less useful.

Clinical assessment and diagnosis are centered to the study of the psychopathology. Clinical assessment

refers to systematic evaluation and measurement and psychological, biological and social factors in an

individual presenting with a possible psychological disorder. Diagnosis is the process of determining

whether the particular problem that the individual has needs all the criteria as given in DSM-IV-R in the

classification of disorders.

Example A

Suppose your class fellow experiences sensations that make her believe she is having a heart attack.

Difficulty in breathing, rapid heart beat and burden on her chest. She is taken to the emergency of a hospital

and she is told that the problem is psychological and physically she is al right.

Example B

Your aunt is depressed, she has lost her husband. She does not eat, does not sleep, and does not go to her

work. You are worried. You want her to return to normal.

Example C

A teacher observes that one of her students is disruptive, unpopular with the class. What should be done

and how the problem should be treated.

When we frequently come across medical problems, psychological problems, social problems or a

combination of either of the two, we frequently ask how can we decide? How can we be sure, what

treatment is needed? How can we differentiate between different types of psychological disorders?

In the mental health field, we describe, classify, explain, select, predict, plan and evaluate to do all these

tasks; we need procedures and methods to measure and define psychological disorders.

Tools for Assessment

Assessment is the systematic collection and analysis of information about a person's characteristics and

behaviors. There are several assessment procedures, such as:

1. Interviews

2. Questionnaires

3. Psychological tests

4. Rating Scales

5. Observation

6. Behavior samples

Each assessment procedure is to judge according to the following criteria which includes

Reliability

Validity

Standardization

Utility

Abnormal Psychology PSY404

1. Reliability

Reliability refers to consistency or repeatability of the results. Reliability is computed by several statistical

procedures. Reliability is expressed as a matter of degree. Usually, on a continuum of 0 to 1 where one

means perfect reliability, this is a rare thing. There are three types of reliability.

a. Test Retest

b. Internal consistency

c. Inter rater reliability

Test retest: is the consistency of a test results over time. The same test questionnaire or an interview

should yield the same results, when used on the same person twice (tested on two different occasions). This

type of reliability is important when compulsive behavior is being measured or anxiety is being measured.

Split Half (Internal consistency) A type of reliability is internal consistency or correspondence

(correlation) between test items. A questionnaire intended to measure potential for child abuse so we focus

on the concept of child abuse, now every item of the questionnaire should relate and measure the concept

of child abuse. If all the items on the questionnaire contribute to identify this concept then individual item-

item correlation will be high and individual item to total item score will be high.

Inter rater reliability A type, of reliability is Inter rater reliability or consistency among scorers or

observers. Independent judges, who are observing a person's behavior, come to the same conclusion. This

kind of reliability evaluates the agreement between two raters administering the same interview, rating the

same video of a person's behavior or observing a person's behavior in a particular setting. High inter rater

reliability increases the confidence that the procedure is measuring, what it is suppose to measure. It is clear

that an instrument measuring a behavior should be high in reliability if we want to draw conclusions from it.

For example: an intelligence test demonstrates low test retest reliability. It clearly shows that it is not

measuring intelligence.

While two observers, observe the same child in the classroom and agree in their ratings that his

intelligence should be high.

2. Validity

Validity is a method, which means does the test measure, what it has been designed to measure i.e. an

intelligence should measure intelligence, a personality test should measure personality then it is a valid

measure and it will give valid and accurate results. Suppose that a bag of sugar when put on the scale should

read its weight, every time the same bag of sugar is put on the scale should give the same reading. Then the

weighing scale is valid.

Kinds of Validity:

a. Face validity

b. Criterion (Predictive validity and Concurrent validity)

c. Content validity

d. Construct validity

Face validity does not by itself establish the test's trust worthiness. It simply conveys that the test and its

items should appear making sense to the test taker. This is not validity in the real sense. It's simply means

that a test on depression should include questions about how often a depressed individual cry or weeps. So

face validity is the apparent sense the test makes to the person who is taking it.

Predictive validity is a test ability to predict a person's future characteristics or behavior. We could

establish predictive validity by administering a test to a group of school students and predict their

performance for the future senior school i.e. predictive validity makes prediction about the individual's

future behavior based on his present behavior. When we ask questions like, Is an individual likely to become

anxious or depress in future? We are dealing with the concept of predictive validity.

Concurrent Validity: A test designed to measure student's present or current anxiety state e.g. should

produce anxiety scores that agree with school counseling records and parent's reports.

Abnormal Psychology PSY404

Content validity: A test that displays high content validity reflects that it measures all important aspects of

the behavior, skill or quality that it is measuring. All achievements test and intelligence test and all teacher

made classroom test should have high content validity. All entrance exams and admission tests should have

high content validity (THEY SHOULD BE CONTENT BASED).

Construct validity: Construct validity measures what they are intended to measure and not something else.

Do achievement tests measure ability in a given subject area or do they measure something else? Some

students do very well on an achievement test and others do very poorly on multiple choice tests. Before any

test can be useful, it must meet the requirements of standardization and utility criteria as well.

3. Standardization

Standardization is process by which a certain set of standards or norms is determined for a technique in

order to make its use consistent across different measurements, e.g. the assessment might be given to a large

number of people who differ on important factors, such as age, race, gender, socio economic status and

diagnosis, where scores would then be used as a standard or norm for comparison purposes, e.g. if you are a

Pakistani, 19 years old male from a middle class background on your score on a psychological test should be

compared to the scores others like you.

4. Utility

A final criterion for deciding that an assessment procedure is worth employing is its utility or usefulness.

To be useful, the assessment procedure should be valid, reliable, standardized and useful.

Table of Contents: