Chapter 1: Psychometrics & the importance of psychological measurement

Why does psychological testing matter?

Decisions are based upon results of psychological tests

Good behavioral research depends on measuring behavior well

In general, your life is affected by psychological testing

Psychological tests measure

observable events, like behavior

Why measure behavior?

The behavior might be important & it may reflect an unobservable psychological attribute like anxiety and depression

A psychological test is

a systematic procedure for comparing the behavior of two or more people

Tests must be capable of comparing the behavior of different people

interindividual differences

Tests must be capable of comparing the the behavior of the same person at different points in time or different circumstances

intraindividual differences

Types of psychological tests

Content

Response formal

Methods of administration

Norm-referenced versus criterion-referenced

Power versus speeded tests

Reflective/effect indicators versus formative/causal indicators

Vary in content:

achievement tests, aptitude tests, intelligence tests

Response format:

open-ended (think essay) and closed -ended (think true/false or multiple choice)

Norm referenced

usually used to understand how a person compares with other people done by comparing a person's score with scores from a reference sample (a reference sample or normative sample is a sample of people who complete a test and the sample is thought to represent some well-defined population)

Criterion referenced-

used in settings in which a decision must be made about a person's skill level set a cutoff score divide into two groups -those that exceed the score and those who do not

Power tests -

not time limited and test subject is expected to answer all of the questions

Speeded tests -

time limited and subject is not expected to finish the test -score is the number of questions answered in the allotted time

Indicators -

with a hypothetical construct such as intelligence determines in part a person's responses and the responses are seen as indicators

Intelligence would be a reflective/effect indicator -

subject's response caused by level of intelligence

Socioeconomic status is an example of a formative causal indicator -

quantify SES by combining indicators like income, educational level, etc.

indicators not viewed as being caused by the person's SES but they define what SES is

What is psychometrics?

The principles and procedures for evaluating the attributes of a psychological measure (i.e., for understanding whether a test provides information that is reliable and meaningful).

Challenges to Psychological Measurement

Participant reactivity

Observer expectation and bias

Composite scores

Score sensitivity

Lack of awareness of psychometrics

Participant reactivity-

the act of measurement can influence the psychological state or process being measured

The subjecy changes behavior to accommodate the researcher

demand characteristics

The subject tries to impress the person doing the measurement

social desirability

The subject changes behavior to convey a poor impression

malingering

Observer bias

subtle unintentional distortions of the researchers observations

Composite scores:

Psychological tests consist of several questions or statements and the subject's score is based on the sum of all the questions; a composite

Score sensitivity -

the ability of a measure to discriminate adequately between meaningful amounts of the dimension that is being measured. Procedures that measure psychological attributes may not be sensitive enough

Lack of awareness

most classroom instructors give class examinations but most of the time lack information about the psychometric properties of their exam

The purpose of measurement in psychology is to

identify and quantify psychological differences that exist between people over time or across conditions. These differences contribute to differences in test scores.

Measures in psychology require that we obtain behavioral samples. Useful psychometric information about samples can be obtained only if

people differ with respect to the behavior we are sampling.

If we think that a particular behavioral sampling procedure is a measure of an unobservable psychological attribute, we must be able to argue that

differences in the scores are related to differences on the attribute.

Psychometric quality should be a concern for

anyone measuring a psychological phenomenon (e.g., behavior, state, characteristic, process, ability, reaction, etc.)