Professional Documents
Culture Documents
Concept of Measurement Report
Concept of Measurement Report
LEARNING OBJECTIVES:
To understand the concept of measurement.
To learn about the measurement process and how to develop a good measurement scale.
To understand the four levels of measurement
To explore the concepts of reliability and validity.
MEASUREMENT
The process of assigning numbers or labels to persons, objects, or events in accordance with specific
rules for representing quantities or qualities or attributes.
Measurement, then, is a procedure used to assign numbers that reflect the amount of an attribute
possessed by a person, object, or event. Note that it is not the person, object, or event that is being
measured, but rather its attributes. A researcher, for example, does not measure a consumer per se
but rather measures that consumers attitudes, income, brand loyalty, age, and other relevant
factors.
The concept of rules is key to measurement. RULE: The guide, method, or command that tells a
researcher what to do.
Rules for assigning labels to properties of variables are the most important component of
measurement, because poor rules can make the outcome meaningless. Accurate measurement
requires rules that are both clear and specific.
It is a category of thought used to group sense data together as if they were all the same.
Constructs are invented for theoretical use and thus are likely to cut across various preexisting
categories of thought. The value of specific constructs depends on how useful they are in explaining,
predicting, and controlling phenomena, just as the value of everyday concepts depends on how
helpful they are in everyday affairs. Generally, constructs are not directly observable. Instead, they
are inferred by some indirect method from results such as findings on a questionnaire.
Constructs aid researchers by simplifying and integrating the complex phenomena found in the
marketing environment.
i.e., Brand loyalty, social class, personality, purchasing power
Constructs that are associated with a scientific theory are required to be defined in the above
mentioned terms. Hence, any constructs that possess constitutive meaning can be used in the
theories.
OPERATIONAL DEFINITION- specifies what the researcher must do to measure the concept
under investigation.
This CONCEPT refers to the measurement of observable characteristics and the process that will
be used for assigning value to the concept.
A scale can be defined as a set of number or symbols which are constructed in such a way that these
numbers or symbols can be assigned using a rule to the individuals (attitudes or their behaviors) to
whom the scale is applied.
Hence, this stage of the measurement process is completely about developing such a scale which
fulfills the requirement of the study and the process.
Here, a formula is used to measure the reliability and validity of measurement considering M=A.
M refers to measurement while A is used for complete accuracy. Hence, the last step of the
measurement process is about evaluating and assessing the reliability and validity of measurement
results.
where M = measurement
A = accuracy
E = errors- (random or systematic)
There are four basic levels of measurement: nominal, ordinal, interval, and ratio. These levels differ
as to how closely they approach the structure of the number system we use. It is important to
understand the level of measurement of variables in research, because the level of measurement
determines the type of statistical analysis that can be conducted, and, therefore, the type of
conclusions that can be drawn from the research.
1. Nominal:
Scales that partition data into mutually exclusive and collectively exhaustive categories.
Example:
Gender: (1) Male (2) Female
Geographic Area: (1) Urban (2) Rural (3) Suburban
The numbers assigned to objects or phenomena have no true numerical meaning; they are
simply labels and nothing else. The term nominal means name-like.
Ordinal numbers are used strictly to indicate rank order. The numbers do not indicate
absolute quantities, nor do they imply that the intervals between the numbers are equal.
Example:
A controversial (and common) use of ordinal scales is to rate various characteristics. (1) Very
Certain, (2) Certain, (3) Neutral, (4) Uncertain, (5) Very certain
3. Interval:
Scales that have the characteristics of ordinal scales, plus equal intervals between points to
show relative amounts; they may include an arbitrary zero point.
Interval scales enable the researcher to discuss differences separating two objects. They are
amenable to computation of an arithmetic mean, standard deviation, and correlation
coefficients.
4. Ratio:
Scales that have the characteristics of interval scales, plus a meaningful zero point so that
magnitudes can be compared arithmetically.
Ratio scale reflects the actual amount of a variable. Physical characteristics of a respondent
such as age, weight, and height are examples of ratio-scaled variables.
Other ratio scales are based on area, distance, money values, return rates, population counts,
and lapsed periods of time.
Example: The actual weight of a hamburger or serving of fries is measured in ratio and is
meaningful to the research.
Every Statistical procedure we will use and model we build depends on the nature of the type of the
data. Datasets are formed by two types of data Qualitative data and Quantitative data.
Ordinal qualitative or quantitative data that can be arranged in some order. It generally
does not make sense to do computations with ordinal data.
Interval quantitative data which intervals are meaningful, but ratios are not. Data at this
level have an arbitrary zero point.
Ratio quantitative data which both intervals and ratios are meaningful. Data at this level
have a true zero point.
Though reliability and validity are different from each other, they are still related. In order for
research data to be of value and of use, they must be both reliable and valid.
Reliability:
Degree to which measures are free from random error and, therefore, provide consistent data. The
extent to which the survey responses are internally consistent.
Validity:
Degree to which what the researcher was trying to measure was actually measured.
TESTING RELIABILITY
There are three ways to assess reliability:
There are two problems with equivalent forms that should be noted. First, it is very difficult, and
perhaps impossible, to create two totally equivalent forms. Second, if equivalence can be
achieved, it may not be worth the time, trouble, and expense involved.
3. Internal Consistency:
The ability of an instrument to produce similar results when used on different samples during
the same time period to measure a phenomenon.
Split Half Technique:
A method of assessing the reliability of a scale by dividing the total set of measurement items
in half and correlating the results.
TESTING VALIDITY
Validity addresses whether the attempt at measurement was successful. The validity of a measure
refers to the extent to which the measurement device or process is free from both systematic and
random error. Types of validity include face, content, criterion related, and construct validity.
1. Face:
The degree to which a measurement seems to measure what it is supposed to measure. It is the
weakest form of validity
2. Content:
The representativeness, or sampling adequacy, of the content of the measurement instrument.
3. Criterion Related:
The degree to which a measurement instrument can predict a variable that is designated a
criterion. Two subcategories of criterion-related validity are predictive validity and concurrent
validity.
Predictive:
The degree to which a future level of a criterion can be forecast by a current measurement
scale.
Concurrent:
The degree to which another variable, measured at the same point in time as the variable of
interest can be predicted by the measurement instrument.
4. Construct Validity:
The degree to which a measurement instrument represents and logically connects--via the
underlying theory--the observed phenomenon to the construct. Two statistical measures of
construct validity are convergent and discriminant validity.
Situation 1 shows holes all over the target, which could be caused by the use of an old rifle, being a
poor shot, or many other factors. This complete lack of consistency means there is no reliability.
Because the instrument lacks reliability, thus creating huge errors, it cannot be valid.
Situation 2 denotes a very tight pattern (consistency), but the pattern is far removed from the bulls-
eye. This illustrates that an instrument can have a high level of reliability (little variance) but lack
validity. The instrument is consistent, but it does not measure what it is supposed to measure. The
shooter has a steady eye, but the sights are not adjusted properly.
Situation 3 shows the reliability and validity that researchers strive to achieve in a measurement
instrument; it is on target with what the researcher is attempting to measure.