Module - 5

Norms
Unit 1: Partition values:
Quartiles
 Quartile are values that divide a complete given set of observations into four equal
parts – Q1, Q2, Q3, Q4
 Quartiles are three data points that divide an observation into four equal groups or
quarters.

N
−cum . f
4
Q1=l+i
fq

where,
 Q1 = First quartile or 25th percentage.
 L = the lower limit of the interval in which N/4 falls
 i = width of the interval
 cum.f = the cumulative frequency of the interval just below L
 fq = Frequency of the interval that contains the N/4
Likewise, Q3 or the 75th percentage may be calculated by the following equation
N
3 −cum . f
4
Q3=l+i
fq

where,
 Q3 = third quartile or 75th percentage.
 L = the lower limit of the interval in which 3N/4 falls
 i = width of the interval
 cum.f = the cumulative frequency of the interval just below L
 fq = Frequency of the interval that contains the 3N/4
Now Q, may be estimated by the equation
 Q 3−Q 1
Q=
2

Decile
 Deciles are those values that divide any set of a given observation into a total of
ten equal parts.
 A decile is a quantitative method of splitting up a set of ranked data into 10 equally
large subsections.
 This type of data ranking is performed as part of many academic
and statistical studies in the finance and economics fields.
 The data may be ranked from largest to smallest values, or vice versa.
 A decile consists of nine data points that divide a data set into 10 equal parts.
 Eg. splitting the entire S&P 500 Index companies into deciles (50 firms in each
decile)

Percentle
 Percentile or centile divides any given observation into a total of 100 equal parts.
 Percentile scores are expressed in terms of the percentage of persons in the
standardized sample who fall below a given raw score.
 Shows individual’s relative position.
 Can also be seen as ranks in a group of 100, except that in percentiles we start
counting from the bottom. Lower the percentile, poorer the ind. Standing.
 The 25th percentile is also called the first quartile.
The 75th percentile is also called the third quartile.
 Eg. A raw score of 15 corresponds to a p28, means 28% of people obtained scores
lower.

Advantage
1. Its easy to compute
2. easily understandable.

Drawback
To reduce this drawback, the use of standard scores have grown

Unit 2: Norms
Development of norms - Steps
Developing norms is certainly a very difficult task. However, this difficulty can be
minimized if we follow the proper steps in developing norms. The following are the three
important steps in developing norms.
1. Defining the target population
2. Selecting the sample from the target population
 3. Standardizing the conditions

1. Defining the Target Population:

 The first step in developing norms is to define the target group's composition.
 The group is determined by the intended use of the test.
 For example, TOEFL is intended for students whose native language is not English but
who plan to study in English-speaking countries.
 The target population for TOEFL includes these students, while a population of PhD
candidates would be inappropriate for this test.

2. Selecting the Sample from the Target Population:

 After defining the target population, the test constructor selects a representative
sample.
 A cross-sectional representation of the target population is needed, meaning people
from all sections of the group should be included.
 Various sampling techniques are used to ensure a representative sample.
 Larger samples are preferred for norm construction, with cluster sampling or its
variations being more commonly used than completely random sampling due to
practicality concerns

3. Standardizing Conditions for Proper Test Implementation:

 Standardization of test conditions is crucial for valid and accurate comparisons of test
scores to norms.
 Factors such as sound control, lighting, ventilation, and temperature must be properly
managed.
 Key elements like test timing, test security, following test-manual directions, and
ensuring examinees work on the correct sections are critical.
 Without these standardization procedures, norms cannot be effectively used for
comparison.
These are the important steps in developing norms of a test. For norms to be a useful
comparative device, these steps must be covered thoroughly.

Unit 3: Types of norms

 Derived scores are typically divided into four common types: age scores, grade
scores, percentile scores, and standard scores.
 Based on these scores, there are four corresponding types of norms: age norms,
grade norms, percentile norms, and standard norms.

Age equivalent norms

1. Age-equivalent norms represent the average performance of a specific age group on a
particular trait or ability.
Example: Measuring the weight of 10-year-old girls in Bihar to determine age norms for
their weight.
2. These norms are best suited for traits that increase systematically with age, such as
physical traits (weight, height) and cognitive abilities (general intelligence) during
childhood and adolescence.

Disadvantages of Age Norms:

 Lack of Standard Growth Units: The rate of growth for traits like intelligence or
physical traits differs across age ranges, making the uniformity of age norms
questionable.
 Non-Comparable Growth Rates: Some traits, like maze learning, stop progressing
after adolescence, while others, like vocabulary, continue developing. This makes
comparisons of age norms for different traits problematic.
 Non-Progressive Traits: Traits like vision acuity or certain personality traits don't
show a consistent, age-related progression, making age norms unsuitable for them.

Grade equivalent norms

1. Grade-equivalent norms are defined as the average performance of a representative
sample from a specific grade or class.
2. The test is administered to a sample from multiple grades, and average performance
is calculated for each grade. Grade equivalents for in-between scores are determined
using interpolation.
3. For example, if the average performance in arithmetic for the sixth grade is 30 correct
items, a raw score of 30 becomes a grade norm of six.

Limitations of Grade-equivalent Norms:

 Non-comparability Across Subjects: The grade-equivalent in one subject (e.g.,
social studies) cannot be compared with the grade-equivalent in another subject (e.g.,
arithmetic) due to different factors influencing knowledge acquisition.
 Assumption of Similar Curriculum Experiences: These norms assume all
students in a grade have similar curriculum experiences, which is often true in
elementary grades but less so in higher grades.
 Unsuitable for Subjects with Different Growth Rates: Grade-equivalent norms
are not ideal for subjects with rapid growth in early grades but slow growth in later
grades, like spelling or arithmetic.

Common Usage:
Despite these limitations, grade-equivalent norms are widely used, especially in
achievement and educational tests, as well as intelligence tests.

Percentile norms
A percentile rank indicates the percentage of cases in a distribution that have scores at
or below a given score.
 Related terms:
 o Centiles: Divides a distribution into 100 equal parts.
o Deciles: Divides a distribution into 10 equal parts (e.g., 7th decile = 70th
percentile).
o Quartiles: Divides a distribution into 4 equal parts (e.g., 1st quartile = 25th
percentile).
 Percentile norms are widely used in psychological and educational tests to indicate
how an individual’s score compares to a standardization sample.

Interpretation of Percentile Norms:

 For example, if Mohan scores 53 on a test and 70% of the class scores the same or
lower, his score is at the 70th percentile (PR70).
 Percentile norms show the relative standing of an individual’s score within a
standardization sample.
 Percentile norms should be based on a homogeneous sample, considering factors like
age, grade, sex, and occupation.

Advantages of Percentile Norms:

 Easy to construct and understand.
 Useful even for untrained individuals to interpret.

Limitations of Percentile Norms:

1. Confusion Between Percentile and Percentage Score: Percentile is a derived
score based on the percentage of people who scored lower, while percentage score
refers to the number of correct items on the test.
2. Inequality of Percentile Scale: Raw score differences in the middle of the
percentile scale (e.g., between PR 40-60) can lead to exaggerated percentile
differences, while differences at the extremes (e.g., PR 90 or PR 10) result in minimal
percentile changes. This makes percentile ranks a nonlinear transformation of raw
scores.
3. Limited Information: Percentile norms only show an individual’s relative position,
without providing information about the actual difference in scores between
individuals.

Unit 4: Norm-referenced
Definition: A method where an individual's performance is compared to a specific
group’s performance (the norm group).
Purpose: To classify individuals from low to high based on a continuum of ability or
achievement.
Norm Group: A representative sample of individuals whose performance is used as a
reference for comparison.
Raw Scores: Transformed into "derived scores" for meaningful comparisons.
Interpretation: Derived scores help compare performances across different tests and
show the individual's relative position.
Usage: Often used in selection for specialized programs or remedial courses.
Example: Standardized tests like SAT or IQ tests.

Criterion referenced Tests

Definition: A method where an individual's performance is compared to a predefined
standard or criterion, not other individuals.
Purpose: To assess whether the examinee has achieved a specific level of performance
or skill.
Criterion: The predetermined standard, such as passing a certain number of correct
answers or demonstrating specific skills.
Raw Scores: Not compared to other test-takers but to a fixed standard.
Features:
Focuses on the examinee's mastery of specific competencies.
Performance is linked to actual behavioral measures or outcomes.
Items typically represent narrow, real-world skill sets.
Usage: Common in educational systems for measuring student proficiency.
Example: A driving test or certification exam.
Norm-Referenced Criterion-Referenced Tests:
Compare performance to others. Compare performance to a predetermined
standard

Broad domain of skills with indirect Narrow domain of skills with real-world
relevance relevance

Scores expressed as standard scores, Scores are percentages with a predefined

percentiles, or grade equivalents. passing level.

Items vary in difficulty. Items are generally of similar difficulty

Unit 5: Standard score norms

 Standard score norms are based on standard scores, which are derived scores with a
fixed mean and standard deviation.
 Standard score norms are preferred over percentile norms because they ensure equal
units across the scale, avoiding the inequality of units found in percentile norms.
 A standard score has a defined mean and standard deviation, making it consistent
throughout the scale.
 Types of standard scores include the z score (most popular), T score, stanine score,
and deviation IQ.
 While "standard score" is often used interchangeably with the z score, the z score is
just one specific type of standard score.
Why standard score norms?
1. Comparison Across Different Tests:
Standard scores are essential when comparing a person’s performance on different
tests. Raw scores cannot be directly compared, especially if the tests have
different means and standard deviations. By converting raw scores into standard
scores, such as the z score, comparisons can be made meaningfully, even across
different distributions.
2. Equal Units of Measurement:
Standard scores offer a consistent unit of measurement, meaning the size of the
score units remains the same across different distributions. This consistency makes
standard scores more useful for interpreting and comparing scores from different
tests or distributions.

Transformation into Standard Scores:

 Linear Transformation:
A raw score can be converted into a standard score through linear transformation, which
retains the original characteristics of the data. The z score is an example of this, which
expresses how many standard deviations a score is away from the mean. The formula for
the z score is:
X− M
z=
σ
Where:
 X is the raw score
 M is the mean of the distribution
 σ\sigma is the standard deviation of the distribution
Application of Standard Score Norms: Standard scores, especially z scores, are
helpful in comparing scores from different distributions. For example, if Bobby scored 60
in psychology and 56 in geography, it is not enough to look at the raw scores alone. By
converting these scores into z scores, we can determine how well Bobby performed
relative to her peers in each subject, and which test she performed better in.

Merits of z Scores:
1. Precision:
Z scores represent the most precise way of indicating a person's position within a
distribution, as they provide a standardized measure of how far a score is from the
mean in terms of standard deviations.
2. Legitimacy in Comparing Scores:
Unlike ranks or percentiles, z scores can be averaged. For instance, if a student has
 a z score of 1.25 in mathematics and -0.5 in verbal comprehension, the average of
these z scores would be a valid, standardized measure (0.375 above the mean).

Limitations and Solutions for Using z Scores:

1. Negative z Scores:
Z scores can be negative, which can be cumbersome to handle. To avoid this, a
constant (like 50 or 100) can be added to each z score, ensuring that no z score is
negative.
2. Decimal Points:
Z scores are often expressed as decimal points, making them harder to interpret.
To resolve this, the z score can be multiplied by a constant (e.g., 10 or 20), which
transforms the z score into a new standard score with more easily interpretable
units.

Normalised standard scores

 Normalized Standard Scores: These scores are adjusted to produce a normal

distribution and have a preassigned mean and standard deviation. They allow for the
comparison of scores from different types of distributions.
 Comparison with Z-Scores: Unlike raw z-scores, normalized standard scores are
adjusted to fit a normal distribution, making them comparable across different
distributions.
Types of Normalized Standard Scores:
 T-Scores: Mean of 50, standard deviation of 10. Derived from z-scores and can be
transformed from raw scores using a specific formula.
 Stanine Scores: A 9-point scale with a mean of 5 and a standard deviation of about
2. The scores are arranged according to the normal distribution curve.
 Deviation IQ: A normalized standard score with a mean of 100 and standard
deviation of 15 or 16. Used for intelligence tests, it differs from traditional IQ scores,
which are based on the ratio of mental age to chronological age.
T score
 They range from 20 to 80, with the transformation of raw scores resulting in a
distribution approximating a normal curve.
 Based on a mean of 50 and a standard deviation of 10.
 Derived from the z-score scale, with values ranging from 20 to 80 in most
distributions.
 Formula
t score =( z score ×10)+50

 T-scores transform raw scores into a distribution that approximates a normal curve.

Stanine score
 Range from 1 to 9, with a mean of 5 and a standard deviation of approximately 2.
 Developed during World War II by the U.S. Air Force.
 Scores are distributed across the nine-point scale based on the percentage of cases in
a normal distribution.
 Distribution: 4% (1st stanine), 7% (2nd stanine), 12% (3rd stanine), 17% (4th
stanine), 20% (5th stanine), 17% (6th stanine), 12% (7th stanine), 7% (8th stanine),
and 4% (9th stanine).
 Formula
stanine score =(z score ×5)+ 5

Deviation IQ
 Mean of 100 and standard deviation of 16 (may vary from 12 to 18 depending on the
test).
 Indicates the number of standard deviations a score is above or below the mean.
 Unlike the traditional IQ, it is not calculated by dividing mental age by chronological
age.
 The Wechsler Intelligence Scale was one of the first to use this system.
 The deviation IQ is a normalized standard score, not a traditional IQ.

Sten
 A variant of the stanine scale with 10 units, 5 above and 5 below the mean.
 Developed by Canfield (1951).
 Similar to the stanine scale but with an additional point, making it a 10-point scale.
 Provides a condensed standardization similar to stanine, but with a broader
distribution of scores.
 Formula
stenscore =(z score ×5)+5.5
 Unit 6 : Test Manual
A manual is a comprehensive, structured document that provides detailed information,
guidelines, and instructions about a specific tool, process, or system. In the context of a
psychological test, a manual serves as an official guide designed to ensure the
standardized, reliable, and ethical use of the test.

Key Characteristics of a Manual

1. Purpose: Explains why the test or tool exists and its intended applications.
2. Audience: Tailored for users such as psychologists, educators, or researchers.
3. Content: Includes instructions, technical details, scoring methods, and
interpretative guidelines.
4. Authority: Developed by experts or test authors, ensuring it reflects scientific
rigor and ethical standards.
5. Comprehensiveness: Covers all aspects necessary for correct administration and
interpretation.

Use
1. Standardized Test Administration
 Provides detailed instructions to ensure the test is administered uniformly across
settings and examiners.
 Minimizes variability caused by differences in test administration, preserving the
reliability and validity of the test.
2. Guidance on Scoring
 Explains how to calculate scores accurately, whether manually or via software.
 Describes how to handle scoring errors or ambiguities, ensuring consistency.
3. Interpretation of Results
 Offers clear guidelines on how to interpret raw and derived scores (e.g.,
percentiles, T-scores).
 Helps test users understand the implications of scores in the context of the test's
purpose (e.g., diagnosing a condition, identifying strengths or weaknesses).
4. Psychometric Evidence
 Provides data on the test’s reliability and validity, ensuring users are aware of the
strengths and limitations of the tool.
 Helps users decide if the test is suitable for their specific population or research.
5. Normative Comparisons
 Includes normative data to compare an individual's score against a reference
group.
  Assists in identifying outliers or diagnosing conditions based on statistical norms.
6. Training and Qualification
 Lists qualifications needed to administer, score, and interpret the test
appropriately.
 Ensures the test is used by trained professionals, preventing misuse.
7. Ensures Ethical Use
 Guides test users on maintaining confidentiality, obtaining informed consent, and
avoiding misuse of test results.
 Provides information on appropriate use cases and populations, discouraging
applications for which the test is not validated.
8. Adaptation and Flexibility
 Explains modifications for special populations (e.g., individuals with disabilities or
those from diverse cultural backgrounds).
 Ensures the test remains inclusive and adaptable to various contexts.
9. Supports Research and Development
 Acts as a resource for researchers studying the test or developing new measures
based on it.
 Provides detailed methodology for replicating or extending validation studies.
10. Legal and Professional Defense
 Serves as a reference in legal or professional contexts to justify test administration,
interpretation, and decision-making.
 Demonstrates adherence to professional guidelines and ethical standards.
11. Training and Education Tool
 Helps educators and supervisors train new professionals in using the test correctly.
 Acts as a teaching resource in psychology, counseling, or related fields.
12. Enhances Communication
 Provides templates and examples for creating client or stakeholder reports.
 Facilitates clear communication of test results and implications.

Information to be contained in the manual

Creating a manual for a psychological test involves providing a comprehensive guide
that ensures the test is used and interpreted consistently, reliably, and ethically
1. Introduction
 Purpose of the Test: Describe why the test was developed and what it aims to
measure (e.g., intelligence, personality, clinical symptoms).
  Target Population: Specify the demographic characteristics of the intended
population (e.g., age, education level, cultural background).
 Theoretical Background: Summarize the theoretical framework or psychological
constructs underpinning the test.
2. Test Development Process
 Conceptualization: Outline the rationale for the test and the constructs it
measures.
 Item Development:
o Steps in item creation (e.g., interviews, literature review).
o Type of items (e.g., multiple-choice, Likert scale, open-ended).
o Pilot testing of items.
 Validation:
o Types of validation used (e.g., content, criterion-related, construct validity).
o Results from validation studies.
 Reliability Testing:
o Methods used (e.g., test-retest, internal consistency).
o Reliability coefficients and their interpretation.
 Norm Development:
o Process of norming (e.g., sample characteristics, size, statistical methods).
o Presentation of norms (e.g., percentile ranks, standard scores).
3. Administration Instructions
 Testing Environment:
o Ideal conditions (e.g., quiet room, no interruptions).
o Necessary materials (e.g., test booklet, timer, pencils).
 Administration Guidelines:
o Instructions for the examiner.
o Timing details.
o Procedures for handling queries from test-takers.
 Special Populations:
o Modifications or accommodations for individuals with disabilities or other
special needs.
4. Scoring and Interpretation
  Scoring Procedures:
o Manual vs. computerized scoring.
o Scoring keys and error-checking methods.
 Derived Scores:
o Types of scores provided (e.g., raw scores, scaled scores, T-scores).
o Interpretation of scores in relation to norms.
 Interpretation Guidelines:
o Explanation of results for laypersons and professionals.
o Cautions against misuse or over-interpretation.
5. Psychometric Properties
 Validity:
o Evidence supporting the test’s validity.
o Studies comparing the test to established measures.
 Reliability:
o Internal consistency, test-retest reliability, and inter-rater reliability.
o Implications of reliability results.
 Standard Error of Measurement (SEM):
o Explanation of SEM and its application in interpretation.
 Cultural and Linguistic Adaptation:
o Details on cross-cultural adaptations or translations.
o Psychometric evidence for diverse groups.
6. Ethical Considerations
 Test User Qualifications:
o Required educational or professional background.
 Informed Consent:
o Guidelines for obtaining consent before test administration.
 Confidentiality:
o Handling and storing test results securely.
 Misuse Prevention:
o Potential misuse and how to prevent it.
7. Applications of the Test
  Clinical Use:
o Diagnostic purposes or therapy planning.
 Educational Use:
o Applications in schools (e.g., placement, learning disabilities).
 Occupational Use:
o Career counseling, employee selection, or organizational development.
 Research Use:
o Suitability for psychological or educational research.
8. Appendices
 Sample Test Items: Examples of test questions or tasks.
 Scoring Templates: Detailed scoring grids or templates.
 Norm Tables: Tables presenting norms across demographics.
 Sample Reports: Example reports for clients or stakeholders.
 References: Citations of studies, theoretical frameworks, and related literature.
9. Glossary
 Definitions of technical terms and constructs used in the manual.
10. Contact Information
 Details for obtaining test materials, technical support, or permissions.