SMC Measurement in Research and Its Importance Questions

1. Define what is a measurement.2. List and describe the two types of measurements researchers use.3. Describe the three levels of continuous level data.4. Explain the difference between validity and reliability. Why are these two concepts important formeasurement in quantitative research?5. List and define the different types of validity.6. Describe what is a reliability coefficient.7. List and define the different types of reliability. Measurement
Chapter 5
Copyright © 2019 McGraw-Hill Education. All rights reserved. No reproduction or distribution without the prior written consent of McGraw-Hill Education.
The use of numbers as a tool for
identifying and presenting
information
Measurement
The process that links the
conceptual to the empirical
Necessary to conduct quantitative
research
Copyright © 2019 McGraw-Hill Education. All rights reserved. No reproduction or distribution without the prior written consent of McGraw-Hill Education.
2
Numbers measure value, intensity,
degree, depth, length, width, and
distance
Descriptive and evaluative device
Measurement
Principles
 Numbers have no value until we provide
meaning
 Measurement includes everything the
researcher does to arrive at a number
 Measurement provides the
operationalization of the variable
Copyright © 2019 McGraw-Hill Education. All rights reserved. No reproduction or distribution without the prior written consent of McGraw-Hill Education.
3
Data are discrete or continuous
Levels
of
Measurement
 Each produces a different kind of data
 How data are collected determines
how they can be used in statistical
analyses
Copyright © 2019 McGraw-Hill Education. All rights reserved. No reproduction or distribution without the prior written consent of McGraw-Hill Education.
4
The presence or absence of some
characteristic
 Also known as nominal or categorical
data
Discrete
Data
Categories
 Reflect different types, not differing
amounts
 Have no inherent value
 Must be mutually exclusive,
exhaustive, and equivalent
Copyright © 2019 McGraw-Hill Education. All rights reserved. No reproduction or distribution without the prior written consent of McGraw-Hill Education.
5
Reveals quantity, intensity, or magnitude
Continuous
Levels
of
Measurement
 Values that differ in degree, amount, or
frequency can be ordered on a continuum
Three types

Ordinal data

Interval data

Ratio data
Copyright © 2019 McGraw-Hill Education. All rights reserved. No reproduction or distribution without the prior written consent of McGraw-Hill Education.
6
Rank elements in logical numerical order
 1st, 2nd, 3rd, and 4th
 Sequence suggests value of data
Ordinal
Data
Ranking positions are relative
Distance between ranks is unknown
Zero does not exist
Copyright © 2019 McGraw-Hill Education. All rights reserved. No reproduction or distribution without the prior written consent of McGraw-Hill Education.
7
Identify highest, next highest, and so on
 4, 8, 10, 14, 22, 25, 29, 31, 32, 37, 41, 42
Interval
Data
Identify exact difference between and
among scores
Acknowledge zero
Copyright © 2019 McGraw-Hill Education. All rights reserved. No reproduction or distribution without the prior written consent of McGraw-Hill Education.
8
Likert-type scales
Types
of Interval
Scales, 1
Strongly
disagree
Disagree
Undecided
Agree
Strongly
agree
1
2
3
4
5
Copyright © 2019 McGraw-Hill Education. All rights reserved. No reproduction or distribution without the prior written consent of McGraw-Hill Education.
9
Semantic scale
 Participants are asked to locate the
meaning they ascribe to a stimulus,
using a stimulus statement
Types
of Interval
Scales, 2
 The response scale is anchored by two
opposites, usually bipolar adjectives
 Can be used singly, or a series of
semantic differential scales can
represent a variable
Copyright © 2019 McGraw-Hill Education. All rights reserved. No reproduction or distribution without the prior written consent of McGraw-Hill Education.
10
Have all of the characteristics of interval
data
Ratio
Data
Zero is absolute
 Indicates a complete lack of the variable
measured
Provide a measure of the degree to which
something actually exists
Copyright © 2019 McGraw-Hill Education. All rights reserved. No reproduction or distribution without the prior written consent of McGraw-Hill Education.
11
Measurement has validity to the extent that
it measures what you want it to measure
and not something else
Validity is a matter of degree
Validity
 Internal validity
 Face validity
 Content validity
 Criterion-related validity
 Concurrent or predictive
 Construct validity
Copyright © 2019 McGraw-Hill Education. All rights reserved. No reproduction or distribution without the prior written consent of McGraw-Hill Education.
12
Degree of consistency among similar items
Reliability coefficient: 0.0 to 1.0
Reliability
 Closer to 1.00 indicates a greater degree of
reliability
 Generally, above 0.70 is acceptable
Internal reliability
 Multiple items invoke similar responses
Copyright © 2019 McGraw-Hill Education. All rights reserved. No reproduction or distribution without the prior written consent of McGraw-Hill Education.
13
A measurement should be both valid and
reliable
The
Relationship
between
Validity
and
Reliability
Validity and reliability are connected in
fundamental ways
 Reliable measurements can be obtained
without validity
 When validity is achieved, reliability is
presumed
Copyright © 2019 McGraw-Hill Education. All rights reserved. No reproduction or distribution without the prior written consent of McGraw-Hill Education.
14
Issues of data collection
 Internal validity
Threats
to
Validity
and
Reliability
 Reliability over time
Issues of sample representativeness
 External validity
 Ecological validity
Do alternative explanations exist?
Copyright © 2019 McGraw-Hill Education. All rights reserved. No reproduction or distribution without the prior written consent of McGraw-Hill Education.
15
Researchers are responsible for:
 Collecting data accurately and ethically
Issues
of
Data
Representation
 Interpreting and reporting data responsibly
Quality of data interpretation cannot be
better than quality of data collected
 Measurement is central to the quality of
outcomes and links to theory
Copyright © 2019 McGraw-Hill Education. All rights reserved. No reproduction or distribution without the prior written consent of McGraw-Hill Education.
16