psych301 study ch-3

# psych301 study ch-3 - Chapter 3 Measurement of behavior...

This preview shows pages 1–3. Sign up to view the full content.

Chapter 3 Measurement of behavior Reliability and Validity Key terms Observational measures Physiological measures Self-report/Questionnaire measures Converging operations Scales or measurement Ratio scale Interval scale Ordinal scale Nominal scale Observed scores True scores Measurement error Total variance True score variance Error variance Reliability Test-retest reliability Interitem reliability Item-total correlation Split-half reliability Cronbach’s alpha reliability Inter-rater reliability Validity Face validity Content validity Construct validity Convergent validity Discriminant validity Criterion related validity Concurrent validity Predictive validity Test bias

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

View Full Document
Scales of Measurement Ratio – this scale has an absolute zero point and equal intervals between the scale scores correspond to equal intervals in the variable being measure. For example, think of money. You can have absolutely no money, so there is a true zero point. Also, if you have \$1 and then a day later you have \$2, you have increased by \$1. If someone else has \$6,482,291 and then a day later has \$6,482,292, then they have also increased by exactly \$1. So the intervals stay equal across the range of possible scores. Other examples are weight and height. Interval – this kind of scale has equal intervals between the scale scores which correspond to equal intervals in the variable being measured. However, there is no true zero point. Think about the days of a calendar. The amount of time that passes from Feb 2 nd at 6am to Feb 6 th at 6am is exactly equal to the amount of time that passes from November 4 th at 6am to November 8 th at 6am. But the calendar date has no true zero point. Other interval scales are degrees Celsius and degrees Fahrenheit. Ordinal – Ordinal scales only tell us the order of variable values. Imagine you have three animals – an elephant, a rhino, and a flea, and you are measuring their weights using an ordinal scale, with the heaviest animal getting a 1 and the lightest animal getting a 3. These three animals would receive the scores 1, 2, and 3 in an ordinal scale. However, the difference between 1 and 2 (elephant and rhino) is much smaller than the difference between 2 and 3 (rhino and flea). Therefore, the intervals are not equal. Nominal – For a nominal scale, all we know is that the “scores” are different, but the “scores” have no mathematical meaning except that they are not equal to each other. Imagine we have four nationalities: Canadians, Americans, Mexicans, and Cubans and we assign them the values 1, 2, 3, and 4. We know that 1 is not 2, and 2 is not 3, and 3 is not 4, and 1 is not 3, and (etc, etc), but we cannot say that 1<2, or 2<3, or 3<4, or etc, etc. So the numbers are mathematically meaningless, except to indicate the value of the variable to which the number pertains. We would be just as well off to use letters A, B, C, and D instead of the numbers 1, 2, 3, and 4. Types of measures and converging operations
This is the end of the preview. Sign up to access the rest of the document.

## This note was uploaded on 09/25/2011 for the course PSYCH 301 taught by Professor Bonett during the Spring '08 term at Iowa State.

### Page1 / 7

psych301 study ch-3 - Chapter 3 Measurement of behavior...

This preview shows document pages 1 - 3. Sign up to view the full document.

View Full Document
Ask a homework question - tutors are online