Data and Measurement: Foundations for Business Statistics

Study Guide - Smart Notes

Tailored notes based on your materials, expanded with key definitions, examples, and context.

Data and Measurement in Business Statistics

Introduction

Understanding data and measurement is fundamental to statistical analysis in business. This section covers the sources and types of data, the measurement process, and the evaluation of measurement procedures, providing a foundation for statistical thinking and data-driven decision making.

Data Sources: Finding and Producing Data

Data and Its Sources

Variable: An attribute of the population or process under study that can assume more than one value.
Data: Observations of a particular variable.
Primary Data: Data collected directly by the researcher or analyst for a specific study.
Secondary Data: Data originally collected for another purpose but used in the current study.

Examples of Data Sources

Primary Data	Secondary Data
Survey responses from a sample of consumers	Statistical Abstracts or Survey of Current Business
Annual inventory data collected by a company	Textbooks summarizing the work of others

Advantages and Disadvantages

	Advantages	Disadvantages
Primary Data	Control over data production; tailored to the question of interest	May be expensive; time-consuming
Secondary Data	Low cost; quick to obtain	May not meet the data producer's needs; may be inappropriate for the question of interest

Producing Data: Study Types

Observational vs. Experimental Studies

Observational Study: Measures variables of interest without disturbing the units under study.
Experimental Study: Manipulates independent variables to observe their effect on dependent variables.

Examples

Observational: Polling likely voters to determine candidate preference.
Experimental: Measuring employee productivity before and after a training program.

Advantages and Disadvantages

	Advantages	Disadvantages
Observational Study	Less expensive; possible with large data sets; applicable to many questions	Cannot manipulate units; only weak evidence of causation
Experimental Study	Can study specific causal relationships; provides better evidence of causation	Expensive; may produce small data sets; applicability may be limited

Measurement: Concepts and Process

Definition and Goal

Measurement: The process by which numbers are assigned to phenomena occurring in the real world.
Goal: To translate characteristics and properties of real-world phenomena into analyzable forms.

Components of Measurement

Empirical Events: Observable characteristics of objects, individuals, or organizations.
Numbers: Symbols used to designate empirical events.
Mapping Rules: Statements dictating how numbers are assigned to events.

The Measurement Process

Identify the variable of interest and develop a conceptual definition.
Develop an operational definition for the variable.
Develop or obtain a measurement instrument.
Evaluate the instrument for validity, reliability, and practicality.
Use the measurement instrument.

Conceptual vs. Operational Definitions

Conceptual Definition: Relates the concept to other concepts, similar to a dictionary definition.
Operational Definition: Specifies how the concept is measured, including mapping rules and conditions.

Measurement for Managerial Decision Making

Scientific Measurement: Focuses on how well the number reflects the 'real' nature of the object or event.
Managerial Measurement: Focuses on how the number relates to users and their purposes, impacting attention, problem solving, and performance evaluation.

Measurement Procedures

Methods of Measurement

Observation: Analyst or machine observes and records data.
Personal Interview: Analyst obtains data by asking individuals.
Self-Enumeration: Individuals respond to questionnaires without analyst present.

Advantages and Disadvantages

Procedure	Advantages	Disadvantages
Observation (Human)	Completeness, removes distortion	May influence activity, subjectivity
Observation (Machine)	Objectivity, reliability	May not observe beliefs, limited scope
Personal Interview	Responsive questioning, detailed data	Expensive, interviewer bias
Self-Enumeration	Low cost, wide coverage	Respondent misunderstanding, limited control

Evaluating Measurement Procedures

Accuracy in Measurement

Unbiased: Does not systematically overstate or understate the true value.
Reliable (Precise): Repeated measurements on the same unit give similar results.

Target Shooting Analogy

Not Accurate, Biased, Reliable
Not Accurate, Unbiased, Unreliable
Accurate, Unbiased, Reliable

Key Concepts in Evaluation

Validity: Measures what it claims to measure.
Reliability: Produces consistent results over time.
Sensitivity: Discriminates along the dimension of concern.
Practicality: Economical, convenient, and interpretable.
Influentiality: Provides influential information.

Errors in Data

Sources of Measurement Error

Instrument Error: Numbers may not represent the property accurately.
Measurer Error: Incorrect application or recording of measurement.
Respondent Error: Intentional or unintentional misreporting due to various factors.

Definitional Error

A discrepancy between the data needed for decision making and the data collected.

Measurement Scales

Types of Measurement Scales

Nominal: Categorizes items without implying order or arithmetic meaning. Example: Defective = 1, Nondefective = 0
Ordinal: Ranks items in order, but differences between ranks are not meaningful. Example: Ranking inspectors by sensory capability
Interval: Equal intervals between values, but no true zero. Example: Temperature in Celsius or Fahrenheit
Ratio: Equal intervals and a meaningful zero; ratios are meaningful. Example: Time, weight, number of errors

Comparison Table: Measurement Scales

Scale	Order	Equal Intervals	True Zero	Arithmetic Operations
Nominal	No	No	No	None
Ordinal	Yes	No	No	Ranking only
Interval	Yes	Yes	No	Add/Subtract
Ratio	Yes	Yes	Yes	All (Add, Subtract, Multiply, Divide)

Types of Data (Variables)

Quantitative vs. Qualitative Data

Quantitative Data: Uses mapping rules with fixed units (e.g., meters, seconds). Includes interval and ratio scales.
Qualitative Data: Uses mapping rules without fixed units. Includes nominal and ordinal scales.

Quantitative Variables: Discrete and Continuous

Discrete: Can assume a finite or countably infinite number of values. Example: Number of defective items produced today
Continuous: Can assume an infinite number of values within a range. Example: Length of bolts, time between breakdowns

Original Data

Preserving Evidence

Symbols generated by measurement (raw data)
Conditions under which measurements were made
Human element: who designed, performed, and collected the data
Order of measurements

Net Promoter Score (NPS)

Definition and Calculation

NPS: Measures customer loyalty and satisfaction.
Calculation:

Classification of Respondents

Promoters (9-10): Loyal enthusiasts
Passives (7-8): Satisfied but unenthusiastic
Detractors (0-6): Unhappy customers

Example Table: NPS Calculation

NPS	% Promoters	% Detractors
-100	0	100
0	50	50
50	75	25
100	100	0

Questions to Ask of Data

Why were data collected?
Who collected the data?
How were the data collected?
When were the data collected and in what order?
What is the operational definition of measured concepts/properties?
What type of measurement scale generated the data?
Was the measurement instrument under statistical control?
Was the measurement instrument accurate?
What types of measurement errors might be reflected in the data?

Additional info: These notes provide foundational concepts for Ch. 1 and Ch. 2 of a Statistics for Business course, including definitions, examples, and practical applications relevant to business decision making and statistical analysis.