Chapter: Measurement and Scaling Techniques

Measurement
Measurement is the process of assigning numbers or symbols to characteristics, properties, or attributes of
objects or phenomena according to specific rules. This process transforms qualitative observations into
quantitative data, making them analyzable and comparable. Measurement helps researchers, scientists, and
practitioners obtain objective, reliable, and valid data about what is being studied, allowing them to quantify,
compare, and evaluate various aspects systematically.

Purposes of measurement
✓ Quantifying Abstract Concepts: Measurement allows researchers to quantify abstract concepts such
as attitudes, satisfaction, intelligence, or job performance. By assigning numerical values, abstract
concepts become observable and analyzable.
✓ Facilitating Comparisons: With standardized measurements, it’s possible to compare different
entities, groups, or conditions. This comparison can be across time, locations, or groups, enabling
researchers to detect trends, differences, or similarities.
✓ Enhancing Objectivity: Measurement provides a standardized, objective approach that minimizes
personal bias and subjectivity. By following set criteria, researchers ensure consistent evaluation
across subjects or conditions.
✓ Improving Decision-Making: Quantitative data from measurements allow for informed decision-
making. Organizations use measurement results to develop strategies, allocate resources, or improve
processes based on objective data.
✓ Supporting Reliability and Validity: Consistent measurement methods enhance the reliability
(repeatability) and validity (accuracy) of results. Well-designed measurement tools ensure that
researchers and practitioners are obtaining accurate reflections of the phenomena being studied.

Types of measurement
✓ Nominal Scale: The nominal scale is the most basic level of measurement, used for labeling or
categorizing without implying any quantitative value or order. Examples: Gender (male/female), types
of cuisine (Italian, Chinese, Indian), eye color (blue, green, brown).
✓ Ordinal Scale: The ordinal scale allows for categorization and rank ordering of items, but the intervals
between ranks are not necessarily equal or known. Examples: Education levels (high school,
bachelor’s, master’s, PhD), customer satisfaction ratings (satisfied, neutral, dissatisfied),
socioeconomic status (low, middle, high).
✓ Interval Scale: The interval scale offers both order and meaningful, equal intervals between values,
but it does not have an absolute zero point (meaning “zero” does not indicate the absence of the
attribute).
✓ Ratio scale: Ratio scale represents the actual amounts of variables. Measures of physical dimensions
such as weight, height, distance, etc. are examples. Generally, all statistical techniques are usable with
ratio scales and all manipulations that one can carry out with real numbers can also be carried out with
ratio scale values.

Reliability
Reliability in research refers to the consistency and stability of a measurement tool or instrument over time.
If a measurement is reliable, it produces similar results under consistent conditions, meaning that if the
measurement process is repeated, the outcomes should be similar, assuming no changes in what is being
measured. Reliability is essential for ensuring that the data collected is dependable and that results are
reproducible. Using methods like Cronbach’s Alpha, which indicates how well items on a test measure the
same concept.
Key Aspects of Reliability
✓ Consistency: Reliability implies that the measurement yields consistent results across repeated trials,
showing that the measurement instrument is stable.
✓ Precision: Reliable measurements are precise and free from random errors.
✓ Dependability: Reliability allows researchers and practitioners to trust that their findings are a true
reflection of the concept being measured.

Validity
Validity in research refers to the extent to which an instrument, test, or measurement accurately measures
what it is intended to measure. It’s a crucial aspect of any research method because it ensures that conclusions
drawn from the data are accurate and meaningful. If a test or instrument is valid, it accurately reflects the
underlying concept being measured, enhancing the credibility and applicability of research findings.

Types of Validity
➢ Content Validity: Assesses whether the instrument covers all relevant aspects of the concept it aims
to measure. Example: In a math test, content validity ensures that all topics in the curriculum are
adequately represented.
➢ Construct Validity: Examines whether the test accurately measures the theoretical construct it claims
to measure.
➢ Convergent Validity: Ensures the measure correlates well with other measures of the same construct.
➢ Discriminant Validity: Ensures the measure does not correlate too closely with different, unrelated
constructs.

Scaling
Scaling in research refers to the process of assigning numbers or symbols to represent the intensity, direction,
or level of a particular attribute or characteristic. Scaling allows researchers to measure qualitative concepts
(like opinions or satisfaction) quantitatively, enabling them to analyze data more easily. Different types of
scales are used to measure variables, each with unique characteristics based on the nature of the data and the
research objectives.

Likert Scaling
Likert Scaling is a commonly used approach in survey research to measure attitudes, opinions, or perceptions.
Named after psychologist Rensis Likert, this scale provides a range of response options for respondents to
indicate the intensity of their feelings or opinions. It typically presents a statement and asks respondents to
rate their level of agreement or disagreement, helping researchers quantify subjective attitudes.

Key Features of Likert Scaling

✓ Range of Responses: Likert scales generally include 5, 7, or 9 points to capture varying degrees of
agreement or disagreement. A typical 5-point Likert scale might look like this: 1 = Strongly Disagree,
2 = Disagree, 3 = Neutral, 4 = Agree, 5 = Strongly Agree
✓ Symmetrical Design: Likert scales are usually symmetrical, with an equal number of positive and
negative options around a neutral midpoint. This structure helps reduce response bias by offering
balanced choices.
✓ Ordinal Scale: Likert scaling provides an ordinal measurement, allowing researchers to rank
responses based on the degree of agreement. However, while responses indicate order, the intervals
between points may not be equal.
 ✓ Composite Scoring: Likert scales are often used with multiple items (questions) measuring the same
concept. Researchers add up responses across these items to create a composite score representing the
respondent's overall attitude.
State the parameter used in 5 points, 7 points, and 9 points Likert Scaling
In Likert scaling, the number of response options can vary, with 5-point, 7-point, and 9-point scales being
among the most common. Here’s a breakdown of the parameters used in each of these Likert scales:

5-Point Likert Scale

Scale Options:
1 = Strongly Disagree 4 = Agree
2 = Disagree 5 = Strongly Agree
3 = Neutral
Parameters:
Neutral Option: Present (3) Positive Options: 2 (4 and 5) Negative Options: 2 (1 and 2)

7-Point Likert Scale

Scale Options:
1 = Strongly Disagree 5 = Slightly Agree
2 = Disagree 6 = Agree
3 = Slightly Disagree 7 = Strongly Agree
4 = Neutral
Parameters:
Neutral Option: Present (4)
Positive Options: 3 (5, 6, and 7)
Negative Options: 3 (1, 2, and 3)

9-Point Likert Scale

Scale Options: 5 = Neutral
1 = Strongly Disagree 6 = Slightly Agree
2 = Disagree 7 = Moderately Agree
3 = Moderately Disagree 8 = Agree
4 = Slightly Disagree 9 = Strongly Agree
Parameters:
Neutral Option: Present (5)
Positive Options: 4 (6, 7, 8, and 9)
Negative Options: 4 (1, 2, 3, and 4)
The choice of scale (5-point, 7-point, or 9-point) often depends on the research objectives, the complexity of
the attitudes being measured, and the desire for granularity in responses. More points can capture subtle
differences in opinion but may also lead to decision fatigue for respondents.