Lecture Communication research: Asking questions, finding answers (4e) Chapter 5: Measurement

Lecture Communication research: Asking questions, finding answers (4e) Chapter 2: The research process: Getting started. After reading this chapter, you should be able to: Describe what a theory is and its role in communication research, explain why the research process starts with identifying a research problem, develop a preliminary question from a topic or issue, explain why a preliminary question is superior to a topic in conducting... Đề tài Hoàn thiện công tác quản trị nhân sự tại Công ty TNHH Mộc Khải Tuyên được nghiên cứu nhằm giúp công ty TNHH Mộc Khải Tuyên làm rõ được thực trạng công tác quản trị nhân sự trong công ty như thế nào từ đó đề ra các giải pháp giúp công ty hoàn thiện công tác quản trị nhân sự tốt hơn trong thời gian tới.

Chapter 5

MEASUREMENT IS

 The use of numbers as a tool for identifying

and presenting information

 The process that links the conceptual to the

empirical

 Necessary to conduct quantitative research

MEASUREMENT PRINCIPLES

 Numbers measure value, intensity, degree, depth,

length, width, distance

 Descriptive and evaluative device

 Numbers have no value until we provide meaning

 Includes everything the researcher does to arrive at a number

 Details the operationalization of the variable

LEVELS OF MEASUREMENT

 Data are discrete or continuous

 Both can represent communication phenomena

 Each produces different kind of data

 How data are collected determines how they can be used in statistical analyses

DISCRETE DATA

 The presence or absence of some characteristic

 Also known as nominal or categorical data

 Categories

 Reflect different types not differing amounts

 Have no inherent value

 Must be mutually exclusive, exhaustive, and equivalent

CONTINUOUS LEVELS OF MEASUREMENT

 Reveals quantity, intensity, or magnitude

 Values that differ in degree, amount, or

frequency can be ordered on a continuum

 Three types

1 Ordinal data

2 Interval data

3 Ratio data

ORDINAL DATA

 Ranks elements in logical numerical order

 1 st , 2 nd , 3 rd , 4 th

 Sequence suggests value of data

 Ranking positions are relative

 Distance between ranks is unknown

 Zero does not exist

INTERVAL DATA

 Identifies highest, next highest, and so on

 4, 8, 10, 14, 22, 25, 29, 31, 32, 37, 41, 42

 Identifies exact difference between and among scores

 Acknowledges zero

 Allows meaningful comparisons

2 TYPES OF INTERVAL SCALES

 Likert-type scales

 Semantic differential scales

Strongly disagree Disagree Undecided Agree Strongly agree

Not at all

RATIO DATA

 All of the characteristics of interval data

 Zero is absolute

 Indicates complete lack of the variable measured

 Provides measure of degree to which

something actually exists

 0, 4, 8, 10, 14, 22, 25, 29, 31, 32, 37, 41, 42

 Extent to which it measures what you want it to

measure and not something else

 Validity is a matter of degree

 Degree of consistency among similar items

 Reliability coefficient – 0.0 to 1.0

 Closer to 1.00, the greater the degree of reliability

 Generally, above 70 is acceptable

 Internal reliability

 Multiple items invoke similar response

VALIDITY AND RELIABILITY

 A measurement should be both valid and

reliable

 Validity and reliability connected in

fundamental ways

 Reliable measurements can be obtained without

validity

 When validity is achieved, reliability is presumed

THREATS TO VALIDITY AND RELIABILITY

 Issues of data collection

 Internal validity

 Reliability over time

 Issues of sample representativeness

 External validity

 Ecological validity

 Do alternative explanations exist?

ISSUES OF DATA REPRESENTATION

 Researchers responsible for

 Collecting data accurately and ethically

 Interpreting and reporting data responsibly

 Quality of data interpretation cannot be better than

quality of data collected

 Measurement is central to quality of outcomes and

links to theory