Selection of Assessment Tools

1
Chapter 3

• Selection of Assessment Tools

2
Council of Exceptional Childrens Professional
Standards

• All special educators should possess a common
  core of knowledge and skills related to
  assessment
• Basic terminology used in assessment
• Legal provisions and ethical principles regarding
  assessment of individuals
• Screening, prereferral, referral and
  classification procedures
• Use and limitations of assessment instruments
• National, state, or provincial, and local
  accommodations and modifications

3
Legal Guidelines for Assessment

• Assessment is
• Nondiscriminatory
• Focuses on educational needs
• Comprehensive and multidisciplinary
• Assessment tools are technically adequate and are
  administered by trained professionals
• Rights of parents and students are protected

4
Professional Guidelines

• Standards for Educational and Psychological
  Testing (1985)
• Test publishers catalogs and manuals
• Mental Measurements Yearbooks series
• Tests in Print IV
• ERIC Clearinghouse on Assessment and Evaluation
  (www.ericae.net)

5
Evaluation Criteria

• The tool must fit the purpose of the assessment
• The assessment instrument must be appropriate for
  the student
• The assessment instrument should match the skills
  of the professional using it
• The assessment tool must be technically adequate
• The assessment instrument should be efficient

6
Measurement Terminology

• Measurement scales may be
• Nominal
• Ordinal
• Interval
• Ratio
• Descriptive statistics include measures of
• Central tendency
• Measures of variability
• Correlation

7
Evaluating Technical Quality of Assessment Tools

• Age, grade, and gender of norm group members
  should match characteristics of the students
  assessed
• Random selection is preferable
• Norm group should be representative of the
  population and should be of adequate size
• Test norms should be recent to reflect current
  standards

8
Reliability

• Test-retest reliability refers to consistency or
  measure from one administration to the next
• Split-half reliability is concerned with internal
  consistency
• Interrater or interobserver reliability is
  concerned with consistency among evaluators

9
Validity

• Content validity is the extent to which the
  instrument represents the content of interest
• Criterion-related validity relies on an outside
  criterion and may examine predictive validity or
  concurrent validity
• Construct validity examines the instrument and
  the theoretical construct it intends to measure

10
Measurement Error

• Related to variability and reliability of an
  assessment device
• Standard error of measurement can be quantified

11
Test Scores

• Informal Measures
• Frequency counts and percentages are interval
  data
• Criterion-referenced tests yield nominal data
• Rating scales provide ordinal data

12
Norm-Referenced Test Scores

• Age and grade equivalent are ordinal scores
• Percentile rank scores
• Standard scores use a set mean and standard
  deviation
• Stanines represent a range of performance

13
Normal Distribution
14
Relationship Between Types of Scores
15
Promoting Nonbiased Assessment

• For students whose primary language is not
  English, tests and test directions should be
  translated into the home language
• Interpreters may be used in the assessment of
  students who speak languages other than English
• Culture-free and culture-fair measures attempt to
  minimize bias

16
Promoting Nonbiased Assessment, continued

• Culture-specific measures relate directly to
  specific cultures
• Separate norms may be provided for students from
  diverse groups
• Pluralistic measures assess a broad range of
  skills
• Test administration procedures may be modified

17
Promoting Nonbiased Assessment, continued

• Dynamic assessment can be used to study the
  students learning ability
• Standardized tests can be replaced with informal
  procedures
• Standardized testing could be abolished

18
Selection of Nondiscriminatory Assessment Tools

• Norm groups should be representative of the race,
  culture, and gender of the student to be assessed
• Tests containing items that reflect cultural bias
  should be avoided
• The tools for assessment should minimize the
  effects of the disability