QUANTITATIVE
MEASUREMENT
MEASUREMENT
The assignment of numbers to
observations according to a set of
rules
Examples of Measurement
HAVE YOU EVER HEARD?
• Stanford Binet IQ test to measure intelligence
• Poverty Line to measure whether one is poor.
• Aptitude test
• GRA
• TPA
• SQ , EQ
• What else?
Why Measure?
We use many measures in our daily activity:
- To measure blood pressure; to measure our body
temperature; to check gas gauge; to check the speed by
watching speedometer; etc.
- We measure physical world.
- We measure social world.
Measurement also extend our senses (exp.using
telescope or microscope).
Measurement yields exact information.
Social measures provide precise information about
social reality.
In Measurement, some of things are easily to see, but
some cannot be directly observed.
• When social researchers test a hypothesis,
they measure concepts or
variables.
• Quantitative researches are more concerned
about measurement issues than are
qualitative researches.
• Measurement is distinct step in the research
process that occours prior to data collection.
The differences between quant.and
qual.measurement.
• Involving time.
- Quantitative: think about variables and convert them
into specific action and separate them from gathering
and analyzing data.
- Qualitative: measurement occurs during the data
collection process.
• Involving the data itself.
Quantitative: develop technique that produces
quant.data, to construct measurement technique that
bridge concepts and data.
Qualitative: more often in the form of spoken and written
words, actions, sounds, symbols, physical objects, or
visual images (e.g., maps, photographs, videos, etc.)
Part of the Measurement.
• Deductive route: begin with abstract idea, follow with
measurement procedure, and end with empirical data
that represent the idea.
• Uses two processes: conceptualization and
operationalization in measurement.
Conceptualization: the process of taking a
construct and refining it by giving it a conceptual or
theoretical definition.
Conceptual definition is a definition in abstract,
theoretical terms.
Operationalization links a conceptual definition
to a specific set of measurement or procedures, the
construct’s operational definition.
KONSEP DAN PENGUKURAN
Imagery (mis.modernisasi)
Concept specification (terpaan media,
urbanisasi, melek huruf, empathy)
Selection of indicators (mis. terpaan media:
frekuensi terpaan, intensitas, resensi terpaan)
Formation of indices (skala atau sub skala untuk
mengukur indikator)
Strategy for Testing the Hypothesis
THEORY
HYPOTHESIS
CONCEPT CONCEPT
INDICATORS INDICATORS
ITEMS ITEMS
QUESTIONNAIRE
Conceptualization and
Operationalization: Abstract construct to
Concreate Measure.
Indep.Var. Dep.Var.
Abstract Construct Abstract Construct
Conceptual Conceptual
Definition Definition
Indicator or Indicator or
measure measure
Items question Items question
Every item should be valid
and reliable
Reliability and
Validity
• Reliability and validity are salient because
constructs in social theory are often
ambiguous, diffuse, and not directly
observable.
• Perfect reliability and validity are virtually
impossible to achieve.
VALIDITY
Validity: how well an idea about reality
“fits” with actual reality.
How well the social reality being
measured through research matches
with the constructs researchers use to
understand it.
We measure what is being measured
RELIABILITY
Reliability: measure has
dependability or consistency
of result.
Operation of the same
concept yield the same
results
How to improve reliability?
• Clearly conceptualize all constructs.
- single construct or subdimension of a
construct is measured.
- Each measure should indicate one and
only one concept.
• Increase the level of measurement.
• Use multiple indicators of a variable.
• Use pretest, pilot study, and replication.
How to Measure Reliability?
1. Test Retest Method.
2. Parallel Form/Multiple Form.
3. Split-half Method.
Developed by Spearman and Brown.
4. Item Analysis.
Developed By Kuder Richardson.
Two formulas: KR 20 and KR 21.
5. Cronbach Alpha.
Types of measurement Validity
1. Face Validity.
It is a judgment by the
scientific community that the
indicator really measures the
construct.
2. Content Validity.
It address the question, is
the full content of definition
represented in a measure?
It is also called curricular
validity.
Content validity involves three steps:
(a). Specify fully the entire content in
construct’s definition.
Example:
Feminism: a person’s commitment to set
of beliefs creating full equality between
men and women in areas of arts,
intelectual pursuits, family, work, politics,
and authority relations.
(b). Samples from all areas
of definition.
(c) . Develop indicators that
taps all of the part of
the definition.
3. Criterion Validity.
To compare with another measure of the
same construct that is widely accepted.
Also called as External Validity.
logically related to construct that is being
measured. There two subtypes of criterion
validity:
There two subtypes of criterion
validity:
1. Concurrent validity: associated with
preexisting indicator that is judged to be
valid.
2. Predictive validity. Indicator should be
able to predict future event.
SUMMARY OF MEASUREMENT VALIDITY
TYPES.
Face : in the judgment of others.
Content : captures the entire meaning.
Criterion : agrees with a preexisting
measure.
(1) Concurrent : agrees with
preexisting measure.
(2) Predictive : agrees with future
behavior.
DISCUSSION:
1. How is the relationship between reliability
and validity.
2. What is internal and external validity in
relation to experimental study?
BROAD CLASSES/LEVEL OF
MEASUREMENT
1. Nominal: kategoris, Diskrit/terpisah. Tidak
diberi nilai tetapi kode. Mis.variabel agama,
jenis kelamin, pekerjaan.
2. Ordinal: berjenjang, diberi nilai secara
arbritrary, tidak memiliki nol mutlak. Mis.
Variabel sikap, pendapat, persepsi.
3. Interval: memiliki nilai, memiliki jarak yang
tetap, tidak memiliki titik nol mutlak.
4. Ratio: spt.data interval tetapi memiliki nilai nol
mutlak. Contoh: penghasilan, pemilikan benda.
Level of Measurement
• Continuous and discrete variables.
• Four level of measurement.
Characteristics of the level of measurement.
Level Different Ranked Distance between
True
Categories categories zero
Nominal yes no no no
Ordinal yes yes no no
Interval yes yes yes no
Ratio yes yes yes yes
Scale and Index
Researchers have created thousand of scales and
indexes to measure social variables.
What is the difference between scale and idex?
Scale: is a measure in which researcher captures
the intensity, direction, level, or potency of
variable construct.
Index is a measure in which a researcher adds or
combines several distinct indicators of a
construct into a single score.
Example of index: HDI
Scales: the purpose.
1. Scales help in the conceptualization and
operationalization process.
Scales show the fit between a set of
indicators and a single construct.
2. Scales produce quantitative measures
and can be used with other variables to
test the hypotheses.
Commonly used scales.
1. Likert Scales.
- widely used and very common in survey
research.
- Intended to measure people’s attitude. -
Moving from “Strongly disagree” to
“Strongly agree” with 5 categories.
Other modifications are possible.
Minimum uses two categories, such
as “agree” and “disagree”.
Examples of Type of Likert Scales
• The Rossenberg Self Esteem Scales:
All in all, I am inclined to feel that I am
a failure:
1. Almost always true.
2. Often true.
3. Sometimes true.
4. Sedom true.
5. Never true.
• A Student Evaluation of Instruction
Scale.
Overall, I rate the quality of instruction in
the course as:
( ) Excellent
( ) Good
( ) Average
( ) Fair
( ) Poor
• A Market Research Mouthwash Rating
Scales.
Brand Dislike Dislike Dislike Like Like Like
Com- Some- a little a little some- Com-
pletely what what
pletely
X ------- ------- -------- ------- ------- -------
Y ------- ------- -------- ------- ------- -------
• Work Group Supervisor Scale.
My Supervisor:
Never Seldom Some Often Always
times
Lets member know what 1 2 3 4 5
is expected of them
Is friendly and 1 2 3 4 5
approachable
Treats all unit members 1 2 3 4 5
as equals
2. Bogardus Social Distance scale.
3. Sematic Differential.
4. Guttman Scaling.
• Lanjutkan ke
CONSTRUCTING THE QUESTIONNAIRE
