Sie sind auf Seite 1von 22

2

1
Scaling & Measurement

4
0011 0010 1010 1101 0001 0100 1011

Narender Singh
RBS New Delhi
Espire Campus
Scaling
• Scaling is the branch of measurement that involves the construction of an
0011 0010 1010 1101
instrument that0001 0100 qualitative
associates 1011 constructs with quantitative metric
units. In many ways, scaling remains one of the most hidden and
misunderstood aspects of social research measurement.

2
• And, it attempts to do one of the most difficult of research tasks -- measure

1
abstract concepts.
• Scaling describes the procedures of assigning numbers to various degrees
of opinion, attitudes and other concepts.

4
• It is applied to the procedures for attempting to determine quantitative
measures of subjective abstract concepts.
This can be done in two ways:
1. Making a judgment about some characteristic of an individual and then
placing him directly on a scale that has been defined in terms of that
characteristics, and,
2. Constructing questionnaires in such a way that a score of individual’s
responses assigns him a place on a scale.
Types of data
• Nominal Data: when we are collecting information on a
variable
0011 0010 that, 0001
1010 1101 naturally
0100 or by design, can be grouped into two
1011
or more categories that are mutually exclusive and collectively
exhaustive. The numerical symbols assigned are recognized as

2
labels only and have no quantitative value or no order or no

1
distance relationship.

4
• Ordinal Data: it includes the characteristics of Nominal data
plus an indicator of order. i.e. a>b>c..
• Interval data: it has the power of Ordinal Data plus it also
incorporates the concept of equality of interval.
• Ratio data: it includes all the powers of previous data plus the
provision of absolute Zero or Origin. In research it can is used
to measure money values, population counts, distances, return
rates, productivity rates, etc..
Characteristics of a good
measurement
0011 0010 1010 1101 0001 0100 1011

Tool: a measurement tool has to satisfy certain criteria;


• Uni-dimensionality: the scale should measure one

2
characteristic at a time.

1
• Linearity: scoring system should be devised, preferably one
based on interchangeable units.

4
• Validity: This refers to the ability of a scale to measure what it
is supposed to measure.
• Reliability: The scale should give consistent results.
• Simplicity: A scale should be as simple as possible.
• Practicability: The tool should be easily administrable,
contain proper instructions, and should be easily understood
and conveniently arranged for easy completion.
Characteristics of a good
measurement
0011 0010 1010 1101 0001 0100 1011

• Validity: it refers to the effectiveness of an instrument in


measuring the specific property that it intends to measure.

2
The measurement of abstract properties like attitude, morale,

1
motivation, perception, etc.
• The degree of validity of an instrument is determined

4
through the application of logic and/or statistical procedures.
• Types of validity: It is concerned with different aspects of
the measurement process. Each of these types uses;
1. logic,
2. statistical verification or both to determine the degree of
validity, and
3. has special value under consideration
Characteristics of a good
measurement
• Content validity: The content validity of a measuring instrument is the
0011 0010 1010
extent 1101 it0001
to which 0100adequate
provides 1011 coverage of the investigative
questions guiding the study. This is of two forms
1. Face validity is a logical type resting on investigator’s individual

2
evaluation as to the validity of a measuring instrument. It is often a matter
of opinion.

1
2. Sampling validity is for the representing character of the content of the
instrument. The measuring instrument’s must contain a representative

4
sample of the universe. It is also decided by a panel of judges.

• Predictive validity: It reflects the success of measures used for prediction


or estimation. This refers to the empirically measured association between
the result produced by instrument and subsequent outcomes.
• Construct validity: A measure is said to be possess construct validity to
the degree that it confirms to predicted correlations with other
propositions. Attitude scales and aptitude and personality test generally
fall under this category. In order to determine the construct validity, the
investigator has to find out if these relationships exists.
Characteristics of a good
measurement
0011• 0010 1010 1101 0001
Reliability: It is 0100 1011
the degree that a measure supplies
consistent results. It is concerned with the estimates of the
degree to which a measurement is free of random or unstable

2
error. The coefficients of reliability are:

1
1. Stability: It is the reliability of a test or instrument inferred

4
from examinee scores securing consistent results.
2. Equivalence: Degree to which alternatives forms of the
sample measure produce error results.
3. Internal consistency: Degree to which instrument items are
homogenous and reflect the same underlying constructs.
• Practicality: It is defined as economy, convenience,
interpretability.
Scale classification bases
0011 0010 1010 1101 0001 0100 1011

The number assigning procedures or the scaling

2
procedures may be broadly classified on one or more

1
of the following bases:

4
1. Subject orientation
2. Response form
3. Degree of subjectivity
4. Scale properties
5. Number of dimensions
6. Scale construction techniques
Subject orientation
Here, a scale may be designed to measure
0011 0010 1010 1101 0001 0100 1011
characteristics of the respondent who completes it or
to judge the stimulus object which is presented to the

2
respondent.

1
• First, the stimulus is assumed to be sufficiently

4
homogeneous so that the between-stimulus variation
is small as compared to the variation among
respondent.
• In the latter approach, we ask the respondent to
judge some specific object in terms of one or more
dimensions.
Response form
0011 0010 1010 1101 0001 0100 1011

• Here we can classify the scales as categorical and


comparative.

1
2
• Categorical scales are also known as rating scales.
• These scales are used when a respondent scores

4
some object without direct reference to other
objects.

• Comparative scales is known as ranking scale,


where the respondent is asked to compare two or
more objects.
Degree of subjectivity

• With this basis the scale data may be based on whether we measure
subjective personal preferences or simply make non-preference
judgments.
0011 0010 1010 1101 0001 0100 1011
• Respondent is simply asked to judge which person is more
effective in some aspect or which solution will take fewer

2
resources without reflecting any personal preferences.

Scale properties:

1
4
• Considering scale properties, one may classify the scales as
nominal, ordinal, interval and ratio scales.
• Nominal: classify without indicating order, distance or unique
origin.
• Ordinal: indicate magnitude relationships of ‘more than’ or ‘less
than’.
• Interval: have both order and distance values
• Ratio: it possess all the above features.
Number of dimensions:
0011 0010 1010 1101 0001 0100 1011

• Scales can be classified as unidimensional

2
and multidimensional.

1
• Unidimensional measures only one attribute

4
of the respondent or object,
• whereas multidimensional scaling recognizes
that an object might be described better by
using the concept of an attribute space of ‘n’
dimension.
Scale construction techniques
0011 0010 1010 1101 0001 0100 1011
• Arbitrary/ adoc approach: Here scale is prepared on ad-hoc basis. It is
presumed that such scale measures the concepts for which they have been

2
designed.

1
• Evaluation /Consensus approach: A panel of judges evaluate the items chosen
for inclusion in the instrument in terms of whether they are relevant to the
topic area and unambiguous in implication.

4
• Item analysis approach: Individual items are developed into a test which is
given to a group of respondents. After the test, the total scores are calculated
for every one. Then individual items are analyzed to determine which items
are responsible for variation of highest score and lowest score.
• Cumulative approach: This is chosen on the basis of their conforming to some
ranking of items with ascending and descending discriminating power.
• Factor scales: It may be constructed on the basis of inter-correlation of items
which indicate that a common factor accounts for the relationship between
items.
Different scales
0011 0010 1010 1101 0001 0100 1011

1. Arbitrary approach • Arbitrary scales

2
2. Consensus scale approach • Differential scales

1
3. Item analysis approach
4. Cumulative scale

4
• Likert scales
approach
5. Factor analysis approach
• Gutman’s scalogram

• Osgood’s semantic
differential scale, multi-
dimension scaling.
Arbitrary scales
0011 0010 1010 1101 0001 0100 1011

• Developed on ad-hoc basis.

2
• Deigned by researchers own subjective selection of items.

1
Process:

4
1. Researchers collects few statements or items which they believes are
unambiguous and appropriate for a given topic.
2. Some of these are selected for inclusion in the measuring instrument ,and
3. Then people are asked to check in the list on which they agree.

Here researcher has to rely mostly on his intution and competence.


Thurstone-type scales
• L.L.thrustone was associated with differential scales which have been
0011 0010developed
1010 1101using consensus
0001 scale approach.
0100 1011
Selection of items is by a panel of judges who evaluate the items in terms
of whether they are relevant to topic and unambiguous in implication.
Procedure:

2
Researcher gather a large number of statements, usually more than 20.

1
1. These statement are submitted to a panel of judges, each of whom
arranges them in groups or piles ranging from one extreme to another.

4
Generally the number of piles is eleven.
2. Each judge yields a composite position for each of the items. In case of
disagreement among judges on particular item, it is discarded.
3. Items retained are given value from 1 to 11 as per group.
4. Final selection of statements in then made by selecting a sample of
statement scores which are evenly spread from one to another. The
statement so selected will be the final scale for respondent.

After the scale is developed, respondents are asked to check the statement
with which they agree. The median value of statement that they check is
worked out and quantify their opinion.
Likert-type scales
0011 0010 1010 1101 0001 0100 1011
• It is based on summated scales / item analysis approach, where a particular item is
evaluated on the basis of how well it discriminates between those person whose

2
total score is high and whose is low.
• Thus it consist of number of statements which express either favorable or

1
unfavorable attitude towards a given object to which the respondent is asked to
react. Each response in given a numerical value, indicating the favorableness and

4
unfavourableness, the scores are totaled to measure the respondent’s attitude.
• Procedure:
• The researcher collects a large number of statement which are relevant to attitude
being studied.
• A trial test is administered to the number of subjects.
• The response of various statement are scored .
• The total score of the respondent is obtained by adding his scores that he received
for separate statement.
• The statement is found out which have high discriminatory power.
• Only those statement which consistently correlate with total test is retained in the
final instrument and others are discarded.
Guttman’s scalogram
0011 0010 1010 1101 0001 0100 1011

• Constructed by Louis Guttman.

2
• Like other scales , it also consists of series
of statement to which a respondent
1
4
expresses his agreement or disagreement.
• The special feature is to form a cumulative
series.
0011 0010 1010 1101 0001 0100 1011

• The purpose of Guttman’s scaling is to establish a one-


dimensional continuum for a concept you wish to measure.

2
• Essentially, we would like a set of items or statements so

1
that a respondent who agrees with any specific question in
the list will also agree with all previous questions.

4
• Put more formally, we would like to be able to predict item
responses perfectly knowing only the total score for the
respondent.
• For example, imagine a ten-item cumulative scale. If the
respondent scores a four, it should mean that he/she agreed
with the first four statements.
• If the respondent scores an eight, it should mean they agreed
with the first eight.
0011 0010 1010 1101 0001 0100 1011
• Develop the Items.
• You would develop a large set of items that reflect the concept. You would

2
want to be sure to specify in your definition whether you are talking about

1
any type of immigration (legal and illegal) from anywhere (Europe, Asia,
Latin and South America, Africa). Let's say you came up with the
following statements:

4
• I would permit a child of mine to marry an immigrant.
• I believe that this country should allow more immigrants in.
• I would be comfortable if a new immigrant moved next door to me.
• I would be comfortable with new immigrants moving into my community.
• It would be fine with me if new immigrants moved onto my block.
• I would be comfortable if my child dated a new immigrant.
• Of course, we would want to come up with many more statements
0011 0010 1010 1101 0001 0100 1011

• Rate the Items.

2
• Next, we would want to have a group of judges

1
rate the statements or items in terms of how
favorable they are to the concept of immigration.

4
• They would give a Yes if the item was favorable
toward immigration and a No if it is not.
• Notice that we are not asking the judges whether
they personally agree with the statement. Instead,
we're asking them to make a judgment about how
the statement is related to the construct of interest.
0011 0010 1010 1101 0001 0100 1011
• Develop the Cumulative Scale. The key to Guttman’s scaling is in the
analysis.

2
• We construct a matrix or table that shows the responses of all the respondents

1
on all of the items.
• We then sort this matrix so that respondents who agree with more statements
are listed at the top and those agreeing with fewer are at the bottom.

4
• For respondents with the same number of agreements, we sort the statements
from left to right from those that most agreed to those that fewest agreed to.
• We might get a table.
• Each scale item has a scale value associated with it (obtained from the
scalogram analysis). To compute a respondent's scale score we simply sum the
scale values of every item they agree with.

Das könnte Ihnen auch gefallen