# Chapt 3 - The Measurement of Behavior-Chapter 3 Examples IQ...

This preview shows pages 1–5. Sign up to view the full content.

The Measurement of Behavior—Chapter 3 Examples IQ measurement--book; handedness and creativity--lecture Types of Measures (really covered in Chapter 4) SLOP or SOAP S elf-report S elf-report L ife/Archival data O bservational measures O bservational measures A rchival/Life data P hysiological measures P hysiological measures “Levels” of Measurement Goal of all measurement is to assign numbers to participants responses—to QUANTIFY behavior Allows us to analyze the data Rating scale Do you ever feel lonely? Not at all Somewhat Quite a bit A lot Can be converted to numbers=>assign 1, 2, 3, 4 BUT—note, not all numbers are the same Four different levels of measurement

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document
Four different LEVELS (scales) of MEASUREMENT Not ALL numbers are the same vary in degree to which the numbers used to capture participants’ behaviors correspond to the real number system That is, one extreme is “representational”, other is “real” N, O, I, R [R I O N] =>REPRESENTATIONAL ( least info, less powerful statistical tests ) NOMINAL scale = NAMING (purely representational) ORDINAL = ORDERING ranking from highest to lowest; intervals (distances) not meaningful INTERVAL = intervals are meaningful But no true zero point and can’t divide =>REAL (most info, most powerful statistics) Two reasons why level of measurement is important determines amount of info provided by a measure (more better) determines kinds of statistical tests that can be performed on data (more powerful tests better) How is that important to Your Project? Choose highest level of measurement possible
(yes/no vs. rating scale) When using Nominal—make coding (representation of constructs with numbers) meaningful Estimating the Reliability of a Measure Measurement goal is to assign numbers to a quality of interest (behavior, events) so that they correspond in some meaningful way to the attribute we are trying to measure That is, variability in the assigned numbers should as accurately as possible reflect the variability in the attribute being measured (e.g., friendliness, helping behavior, handedness, creativity) 1 = 2 + E 3 = 3 + E 3 = 2 + E 1 = 4 + E 5 = 5 + E Observed Score = True score + Measurement error # we recorded = in reality + degree to which we are off Every time we measure sth., want to know whether it is reliable (solid, trustworthy, repeatable) Important characteristic of a measure is that it is reliable RELIABILITY

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document
The consistency or dependability of a measure (or a measuring technique) Ps score on measure consists of two components True Score + Measurement Error True score = score Ps would have obtained if measure were perfect and we were able to measure without errors All measures obtain some Measurement Error Take IQ test what if anxious? Will performance
This is the end of the preview. Sign up to access the rest of the document.

## This note was uploaded on 04/22/2008 for the course PSYCH 31:043 taught by Professor Klohnen during the Spring '08 term at University of Iowa.

### Page1 / 14

Chapt 3 - The Measurement of Behavior-Chapter 3 Examples IQ...

This preview shows document pages 1 - 5. Sign up to view the full document.

View Full Document
Ask a homework question - tutors are online