1. Data and the Nature of Measurement
Graziano and Raulin
Research Methods: Chapter 4
Graziano & Raulin (2000)
Graziano & Raulin (1997)

2. Research Variables
Variable: Any characteristic that can take on more than one value
Examples: speed, level of hostility, accuracy of feedback, reaction time
Research is the study of the relationship between variables
Therefore, there must be at least two variables in a research study (or there is no relationship to study)
Graziano & Raulin (2000)

3. Measuring Variables in Research
Measurement: A process by which we assign numbers to indicate the amount of some variable present
Sometimes the number assignment is easy to understand (e.g., speed measured in number of seconds)
Sometimes it is more arbitrary (e.g., 1 for male and 2 for female)
Graziano & Raulin (2000)

4. Scales of Measurement
Based on how closely the scale matches the real number system
Scales of Measurement (as proposed by Stevens)
Nominal
Ordinal
Interval
Ratio
Graziano & Raulin (2000)

5. Nominal Scales A naming scale Produces nominal or categorical data
Each number reflects an arbitrary category label rather than an amount of a variable
Examples: diagnostic categories, political affiliations, preference for consumer products
Produces nominal or categorical data
Has mathematical property of identity
Graziano & Raulin (2000)

6. Ordinal Scales A scale that indicates rank ordering
Reflects the order, but not the amount of a variable
Examples: order of finish in a race, class rankings
Produces ordered data
Has mathematical properties of identity and magnitude
Graziano & Raulin (2000)

7. Interval Scales A scale that has equal intervals Produces score data
The scale indicates amount, but there is no zero point on the scale
Examples: temperature on the Celsius scale, most psychological tests
Produces score data
Has the mathematical properties of identity, magnitude, and equal intervals
Graziano & Raulin (2000)

8. Ratio Scales A scale that fits the number system well
The scale has a true zero and equal intervals, just like the real number system
Examples: time, distance, number correct, weight, frequency of behavior
Produces score data
Has the mathematical properties of identity, magnitude, equal intervals, and a true zero
Graziano & Raulin (2000)

9. Controlling Measurement Error
Measurement error decreases the accuracy of measurement
Anything that lead to inconsistency in the measurement process can produce measurement error
Response set biases (e.g., social desirability) will add measurement error to any self-report measure
Graziano & Raulin (2000)

10. Operational Definitions
The specific procedures by which the researcher measures and/or manipulates a variable
In research, every variable should be operationally defined
The more careful and complete the operational definition, the more precise the measurement of the variable will be
Graziano & Raulin (2000)

11. Reliability Reliability refers to the consistency of measurement
Types of reliability
Interrater reliability: degree of agreement between two independent raters
Test-retest reliability: degree of consistency over time
Internal consistency reliability: degree to which the items of a measure are in agreement
Graziano & Raulin (2000)

12. High Reliability
High reliability occurs when the rank ordering of individuals is very similar in the separate measures
High test-retest reliability is illustrated in the figure at the right
Graziano & Raulin (2000)

13. Low Reliability
Low reliability occurs when the rank ordering of individuals is very different in the separate measures
Low test-retest reliability is illustrated in the figure at the right
Graziano & Raulin (2000)

14. Effective Range
The effective range of the measure is the range in which it can give an accurate indication of the level of the variable
All measures have a limit to their effective range, so it is critical to select measures whose effective range will pick up the variability expected in your study
Example: Using a calculus test to measure math skills in second graders would NOT work
Graziano & Raulin (2000)

15. Validity
A scale is valid if it measures what it is supposed to measure
Validity also refers to how well a scale predicts other variables (e.g., an IQ test is likely to be a reasonably valid predictor of grades in school)
When used this way, the scale is called the predictor measure and the measure predicted is called the criterion
Graziano & Raulin (2000)

16. High Validity
High validity is indicated when the rank ordering of test scores is very similar to the rank ordering on the criterion measure
High validity in indicated in the figure on the right
Graziano & Raulin (2000)

17. Low Validity
Low validity is indicated when the rank ordering of test scores is very different from the rank ordering on the criterion measure
Low validity in indicated in the figure on the right
Graziano & Raulin (2000)

18. Scale Attenuation Effects
If the effective range is insufficient, scores will will cluster at the top or bottom of the scale (termed scale attenuation effects)
Floor effect: insufficient range at the bottom of the scale, so most low scorers are bunched together
Ceiling effect: insufficient range at the top of the scale, so most high scorers are bunched together
Scale attenuation effects distort score, thus reducing both reliability and validity
Graziano & Raulin (2000)

19. Objective Measurement
Objective measures are the hallmark of science
Produce the same result no matter who does the measurement
Therefore, scientific principles will apply no matter who tests these principles
Objective measures reduce biases that could distort results (see Chapters 8 and 9)
Graziano & Raulin (2000)

20. Summary Measuring variables is central to research
Several scales of measurement exist
Reduce measurement error with carefully developed operational definitions
Want to enhance reliability and validity of your measures
The goal is to produce objective, accurate measures of your variables
Graziano & Raulin (2000)