# Reliability

## Reliability is the stability, consistency, and precision of a given measure

Reliability is an empirical concept derived from

**Classical Test Theory**(**CTT**). This theory postulated that an individual's**true score**on a given psychological instrument was a function of their**observed score (O)**on the instrument plus (**+**) the inherent systematic and unsystematic**error**that exists when measuring human psychological constructs (**E**). In essence, the**true score**is the individual's "**real**" score that we**cannot**obtain due to measurement error.Measurement error plays a central role in understanding reliability. A fundamental assumption of reliability is that measurement error is

There are statistical measures that assess the amount of measurement error in a survey, the

**dispersed randomly within a given population**. This is assumption is fundamental because it allows researchers to surmise that increases or decreases in observed scores are caused by measurement error that is**completely independent**of the individual's true score or ability. As scores fluctuate across homogeneous and heterogeneous populations, the fluctuation of scores occurs at**random and is not fully indicative of the individual's true ability**. Thus, as measurement error increases, our ability to discern a precise measure of the individual's true ability is decreased.There are statistical measures that assess the amount of measurement error in a survey, the

**reliability coefficient**and the**standard error of measurement (SEM)**. The SEM is like a "**bell-curve**" for an individual person's true ability that is based on the amount of error associated with the measure. As reliability increases, the SEM will**constrict**and we can become**more confident**that an individual's observed score on the survey is indicative of their true score or ability. The inverse holds true as well. As reliability decreases, the SEM will**widen**and we become**less confident**that an individual's observed score is indicative of their true score.### Types of reliability

There are

**four**primary types of reliability:**Internal consistency****, test-retest, alternate/parallel forms, and inter-rater**. Click on a button below to learn more about that type of reliability. The methods for conducting and interpreting each type of reliability in SPSS will be presented.Survey items are written to cover a content area in the literature related to

**one construct**and the items should be**correlated at a theoretical or conceptual level**.Survey items are written to assess the

**stability**of a trait, characteristic, behavior, emotional state, or construct across time..Researchers are writing

**two different survey instruments that cover the same content area**.A panel of experts are

**rating the performance**of a behavior(s), task(s), or construct in either a simulated or natural environment.## Hire A Statistician - Statistical Consulting for Students

**DO YOU NEED TO HIRE A STATISTICIAN?**

Eric Heidel, Ph.D.

**will provide the following statistical consultingservices for undergraduate and graduate students at $50/hour. Secure checkout is available with Stripe, Venmo, Zelle, or PayPal.**

- Statistical Analysis
- Research Design
- Sample Size Calculations
- Diagnostic Testing and Epidemiological Calculations
- Survey Design and Psychometrics