Helping Families Understand Standardized Testing
Sunshine2016
CHAPTER3.1.docxCHAPTER 3 How Standardized Tests Are Used, Designed, and Selected
Benjamin LaFramboise/Pearson
Chapter Objectives
As a result of reading this chapter, you will be able to:
1. Discuss how standardized tests are used with infants and young children.
2. Describe the steps in standardized test design.
3. Explain the differences between test validity and test reliability.
4. Describe considerations for selecting and evaluating standardized tests.
Various methods and purposes for measuring and evaluating infants and young children were discussed in Chapter 02. In this chapter, wewill look at some ways that standardized tests are used with infants and young children. Specific examples of standardized tests and theirpurposes will be discussed.
3.1 How Standardized Tests Are Used with Infants and Young Children
Types of Standardized Tests
Many types of standardized tests are available for use with infants and young children. All are psychological tests, whether they measureabilities, achievements, aptitudes, interests, values, or personality characteristics. In the following sections, we will discuss each of these types oftests.
Ability refers to the current level of knowledge or skill in a particular area. Three types of psychological tests— intelligence tests , achievementtests , and aptitude tests —are categorized as ability tests because they measure facets of ability. Young children are often measured todetermine the progress of their development. This type of measure may assess ability in motor, language, social, or cognitive skills. Bayley’sScales of Infant Development, Third Edition (BSID-III) (Bayley, 2005), for example, is used to diagnose delays in development. More recently, theBSID-III was designed to learn about overall development. Children with intellectual disabilities might be assessed for adaptive functioning.Instruments such as the Vineland Adaptive Behavior Scales, Second Edition (Vineland™ II) (Sparrow, Cicchetti, & Balla, 2005) are administeredthrough parent and primary caregiver reports to determine communication, social, and daily living skills.
Achievement is the extent to which a person has acquired certain information or has mastered identified skills. An achievement test measuresability in that it evaluates the child’s achievement related to specific prior instruction. The Peabody Individual Achievement Test—Revised-Normative Update (PIAT-R/NU) (Markwardt, 1997) is a measure of achievement in mathematics, reading recognition, reading comprehension,spelling, and general information.
Aptitude is the potential to learn or develop proficiency in some area, provided that certain conditions exist or training is available. Anindividual may have a high aptitude for music or art, for example. Like achievement tests, aptitude tests also measure learned abilities. Anaptitude test measures the results of both general and incidental learning and predicts future learning.
Intelligence tests are ability tests in that they assess overall intellectual functioning. They are also aptitude tests because they assess aptitudefor learning and problem solving. The Stanford–Binet Intelligence Scales (SB5), Fifth Edition (Roid, 2003) is an example of an intelligence scalethat also measures individual aptitude.
Personality tests measure a person’s tendency to behave in a particular way. Such tests are used to diagnose children’s emotional strengthsand needs. Because an inventory is used to assess personality characteristics, the test is quite lengthy, usually containing several hundred itemsin a true–false format. Test items are answered by the parent or child or by both together and are analyzed to determine whether the child hascertain personality traits.
Teachers use standardized test results to design instruction for individual needs.
Suzanne Clouzeau/Pearson
Interest inventories are used to determine a person’s interest in a certain area or vocation and are not used with very young children. Aschool-age child may be given a reading interest inventory to provide the teacher with information that will serve as a guide when helping thechild select reading material.
Tests for Infants
Various psychological tests have been constructed for infants and young children. Examples that have been discussed previously are the Neonatal Behavioral Assessment Scale (NBAS), Third Edition (Brazelton & Nugent, 1995) and the Communication and Symbolic Behavior ScalesDevelopmental Profile (CSBS DP) (Prizant & Wetherby, 2002). They are examples of tests that have been normed. Such tests are challengingbecause of the child’s developmental limitations. Babies are particularly difficult to evaluate because of their short attention span. Their periodsof alertness are brief, and they have their own schedules of opportune moments for testing. In addition, developmental changes occur rapidly,making test results unreliable for more than a short time. Generally, because of these limitations, the validity and reliability of infant scales arequestionable. The tests are difficult to administer and interpret. Nevertheless, they are useful in evaluating the status of newborns and infants(Campbell et al., 1995; Hack et al., 2005; Wodrich, 1997). To better understand the types of infant and toddler measures, the following section isorganized into neonatal status, infant and toddler development, and diagnostic tests.
Neonatal Status
The status of a newborn can be determined using various measures. The Apgar Scale (Apgar, 1975), administered 1 minute and 5 minutes afterbirth, assesses the health of the newborn by evaluating the heart rate, respiratory effort, muscle tone, body color, and reflex irritability. Eachcharacteristic is scored on a scale of 0 to 2 for a maximum score of 10 points. A score of 7 to 10 indicates the infant is in good condition; ascore of 5 may indicate developmental difficulties. A score of 3 or below is very serious and indicates an emergency concerning the infant’ssurvival. The Brazelton Neonatal Behavioral Assessment Scale, another neonatal test (Als et al., 1979), measures temperamental differences,nervous system functions, and the capacity of the neonate to interact. Its purpose is to locate mild neurological dysfunctions and variations intemperament. A newer scale, the Neonatal Behavioral Assessment Scale, Fourth Edition (NBAS) (Brazelton & Nugent, 2011), is used withnewborns from the first day of life through the end of the first month. In this test, the infant’s competence is measured through behavioralitems. In addition to identifying the infant’s performance, if administered with the parents present, it can be used to help parents understandtheir infant’s signals and skills. This knowledge of child development generally and their baby’s competence specifically can facilitateimprovement in parenting skills (Widerstrom, Mowder, & Sandall, 1991). An adaptation of the NBAS to assess preterm infants came throughthe design of the Assessment of Preterm Infants’ Behavior (APIB) (Als, Butler, Kosta, & McAnulty, 2005). It includes many of the items in the NBAS,but refined them to be able to observe the preterm infant’s functioning. The Ounce Scale (Meisels et al., 2008) is another developmental scalesuitable for parents, child-care personnel, and Early Head Start teachers to administer. Used with children from birth to 3.6 years old, TheOunce Scale is organized around six developmental domains and helps parents observe developmental milestones.
Watch this video to learn about how the APGAR is administered and what it means. (www.youtube.com/watch?v=zY87wohJl9I)
Infant and toddler development
Infant development scales go beyond measuring neonatal status to focusing on development from 1 month to 2 years. The Gesell DevelopmentalSchedules (Ball, 1977) were the first scales devised to measure infant development. Gesell designed them to detect infants who were delayed indevelopment and might need special services. More recently, the Bayley Scales of Infant Development, Third Edition (BSID-III) (Bayley, 2005)were designed to learn about infants’ overall development, while the Communication and Symbolic Behavior Scales Developmental Profile (CSBSDP™) (Prizant & Wetherby, 2002) are used to assess communicative and symbolic development, including symbolic play and constructive play.The Mullen Scales of Early Learning (Mullen, 1995) measure cognitive functioning in infants, toddlers, and young children from birth to 68months. The assessment measures intellectual development through the child’s response to prepared activities. The Gesell and Bayleyinstruments are challenging to administer because of their length; however, because they are used to diagnose children with special needs, it isimportant to examine developmental milestones thoroughly.
The Devereux Early Childhood Assessment for Infants and Toddlers (DECA-I/T) (Powell, MacKrain, & LeBuffe, 2007) is an assessment designed tosupport the social and emotional development of infants and toddlers. There are two forms, the Infant Form for children 1–18 months old, andthe Toddler Form for children 18–36 months old.
The Denver II (Frankenburg et al., 1992) is a simple screening instrument designed to identify children who are likely to have significant delaysand need early identification and intervention, while the Adaptive Behavior Assessment System, Second Edition (ABAS-2) (Harrison & Oakland, 2003) assesses the strengths and weaknesses in adaptive skills. The Early Coping Inventory (ECI) (Zeitlin et al., 1988) assesses how well infantsand toddlers 4–36 months of age react and cope with different situations; in addition, the Infant/Toddler Symptom Checklist: A Screening Tool forParents (DeGangi et al., 1995) screens infants and toddlers who show disturbances in sleep, feeding, and self-calming. Used with children from7 to 30 months old, it can be administered by a parent or caregiver.
Diagnostic Tests
There are diagnostic tests for infants to identify developmental or physical disorders. As with developmental and screening tests for infants andtoddlers, it is very difficult to accurately acquire the needed information. The strategies for measuring lung function, for example, can beconsidered to be intrusive for infants (Panitch, 2004). Likewise, babies who have experienced a life-threatening event (ALTE) presentchallenges in what tests should be used, how to interpret the results, and how well the tests or assessment procedures will contribute to themany factors that can cause ALTE (Brand et al., 2005). Observational measures to assess children with spinal cord injury can result in lack ofagreement among the observers (Calhoun et al., 2009). Regardless, specialists in infant screening and diagnosis continue to research methodsthat provide the desired testing with minimal invasive methods and more dependable results. Figure 3-1 describes neonatal and infant tests.
FIGURE 3-1 Neonatal and infant tests
Tests for Preschool Children
Professionals have designed a variety of tests to evaluate development and to detect developmental problems during the preschool years. Justas the testing of infants and toddlers presents challenges to test administrators because of the children’s developmental circumstances, theevaluation of preschool children under age 6 must also be conducted with their developmental characteristics in mind. Instruments that assesscharacteristics used to identify developmental delays or to diagnose sources of disabilities that affect the child’s potential for learning areadministered to one child at a time. Test items are concrete tasks or activities that match the child’s ability to respond; nevertheless, validity andreliability are affected by such factors as the child’s limited attention span. As children enter the preschool years, more instruments areavailable for examining development and identifying potential developmental delays. To better understand the various types of measures,preschool tests are organized into screening, diagnostic, language, and achievement tests.
Screening Tests
Screening tests are administered to detect indicators that a child might have a developmental problem that needs to be further investigated.Screening tests can be contrasted with assessments that examine development in more depth and help determine strengths and possibledifficulties, as well as to determine what strategies need to be taken to address the child’s needs.
The Denver II (Frankenburg et al., 1992) was discussed earlier as a screening tool that can be used with infants and older children. It isadministered by a professional such as a pediatrician or educator. In contrast, the Ages and Stages Questionnaires, Third Edition (ASQ-3™)(Squires & Bricker, 2009) uses parental reporting. The parent can complete the questionnaire or participate in an interview with a professional.It is administered for children ages 1 month to 66 months.
The AGS Early Screening Profiles can be administered from ages 2 years to 6 years 11 months. They include parent–teacher questionnaires aswell as profiles in cognitive language, motor, and social development. The Developmental Indicators for the Assessment of Learning™, FourthEdition (DIAL™-4) (Mardell & Goldenberg, 2011) is also used to flag potential developmental delays. Administered to children ages 2 years 6months to 5 years 11 months, it includes direct observation and tasks presented to the child. The Early Screening Inventory—Revised (ESI-R),2008 edition (Meisels et al., 2008) has two forms: the ESI-P for ages 3 to 4.4 years, and the ESI-K for ages 4.5 to 6 years. It is used to screendevelopmental domains and uses cutoff scores to determine whether the child needs to be referred for further evaluation. A parentalquestionnaire is used to provide supplementary information. The Brigance Early Childhood Screens III (Brigance, 2013) include three screeningtools to examine development in the domains of physical development, language, academic/cognitive, self-help, and social-emotional. The threetools are: an Infant & Toddler Screen for children 0–35 months; an Early Preschool, Preschool Screen for children 3–5 years old; and a K & 1Screen for children in kindergarten and first grade. Finally, the FirstSTEp™: Screening Test for Evaluating Preschoolers (Miller, 1993) has 12subtests grouped into cognitive, communicative, and motor categories. There is also an optional social-emotional scale and adaptive behaviorchecklist. First Step has three levels: Level 1 is administered to children from ages 2 years 9 months to 3 years 8 months; Level II is for children3 years 9 months to 4 years 8 months; and Level III is administered to children 4 years 9 months to 6 years 2 months. A new Gesell Instituteinstrument, the Gesell Developmental Observation—Revised (2011), is used with 2½- to 6-year-old children. It measures child growth, academicachievement, and social and emotional development.
Baker School for Early Learning
Baker School is a community school that targets services for toddlers and preschool children from a nearby public housing development.The children in the housing development represent a variety of ethnic groups and languages. Some are from families that recentlyemigrated from another country. Teachers in the program need input from parents on their child’s current stage of development prior toentering the program. Parents can fill out the Ages and Stages Questionnaire with information about their child. The form includesquestions about behaviors, speaking abilities, and physical skills, as well as other indicators of development. Because the teachers aresensitive to possible language and literacy limitations, they are available if the parents need help filling in the information. In many casesthey read the questions to the parents and record their responses on the test form.
The screening tests just discussed cover various domains of development. The screening tools discussed next focus solely on social-emotionaldevelopment. These screening instruments look at social behaviors and require sensitive and careful collaboration between the home andschool because children’s behaviors are affected by environmental differences. While this type of screening is difficult to do accurately, social-emotional competence is very important and should be monitored (Meisels & Atkins-Burnett, 2005).
The Early Screening Project (ESP) (Walker, Severson, & Feil, 1995) is administered to children ages 3 to 6 years and is administered in threestages. Children are ranked in social interaction, adaptive behavior, maladaption behaviors, aggressive behaviors, and reactions to criticalevents. A parent questionnaire looks at how the child plays with other children, how the child interacts with caregivers, and social problemssuch as difficulties with self-esteem or social avoidance. An instrument that uses parent ratings is the PKBS–2: Preschool and KindergartenBehavior Scales, Second Edition (Merrell, 2003). Administered to children ages 3 to 6 years, it examines positive and problem behaviors. (Figure3-2 provides examples of items on screening tests.)
Diagnostic Tests
After a child has been screened and there are indicators that further evaluation is needed, tests for diagnostic assessment can be administered.Adaptive behavior instruments attempt to measure how well the young child has mastered everyday living tasks such as toileting and feeding. Vineland Adaptive Behavior Scales, 2nd Edition (Vineland™-II) (Sparrow, Cicchetti, & Balla, 2005) assesses the everyday behaviors of the childthat indicate level of development. The scale determines areas of strengths and needs in communication, daily living, socialization, and motorskills. Another instrument, the ABS-S:2™ Adaptive Behavior Scale–School 2nd Edition™ (Lambert, Nihira, & Leland, 2008), assesses adaptivebehavior for children 3–16 years old in 16 domains for social competence and independence. Figure 3-3 describes categories of adaptivebehaviors.
Preschool intelligence tests and adaptive behavior scales are used to diagnose children with intellectual disabilities. Although intelligencemeasures during the preschool years are generally unreliable because children’s IQs can change enormously between early childhood andadolescence, they are used with young children to measure learning potential.
FIGURE 3-2 Examples of items on screening tests
FIGURE 3-3 Some categories assessed in adaptive behaviors
The Stanford–Binet Intelligence Scales (SB5), Fifth Edition (Roid, 2003), the original IQ test, was designed to assess general thinking or problem-solving ability. It is valuable in answering questions about developmental delay and retardation. Another instrument, the Wechsler Preschool andPrimary Scale of Intelligence™, Third Edition (WPPSI™-III) (Wechsler, 2002), is useful in identifying signs of uneven development for childrenages 2 years 6 months to 7 years 3 months.
Other instruments address all domains of development. The Kaufman Assessment Battery for Children, Second Edition (KABC™-II) (Kaufman &Kaufman, 2004), Battelle Developmental Inventory™, Second Edition (BDI-2™) (Newborg, 2004), and Bracken Basic Concept Scale, Third Edition:Receptive (BBCS-3:R) (Bracken, 2006) have comprehensive assessments of development. Additionally, the Brigance Inventory of EarlyDevelopment III (IED III) (Brigance, 2013) is an assessment used with children from birth through developmental age 7 years. There are twoversions of the IED III, which is criterion-referenced, and the IED III-Standardized, which is norm-referenced.
The Devereux Early Childhood Assessment Preschool Program, Second Edition (DECA P-2) (Le Buffe & Naglieri, 2012) is a strengths-basedassessment system designed to promote resilience and positive social/emotional development for children ages 3–5 years. It can beadministered through classroom observations. It has items that examine positive and negative behaviors such as attention problems,aggression, depression, and emotional control.
Language Tests
The category of language tests for preschool children is very important because many children are at risk for school readiness because theyhave language deficits or their first language is not English. While some language tests for at-risk children are in English, others are available inboth English and Spanish, and occasionally other languages. The Preschool Language Scale, Fourth Edition (PLS-4) (Zimmerman, Steiner, & Pond, 2007) and Peabody Picture Vocabulary Test, Fourth Edition (PPVT-4) (Dunn & Dunn, 2007) provide information on a child’s language ability,which can help determine whether a child will benefit from a language enrichment program.
With the expanding numbers of English Language Learners (ELL) who are living in many states, language assessment tests are growing inimportance. Children who have limited English proficiency may be served in a bilingual program or ELL program. The Pre-LAS, Pre-IPT, and Woodcock-Muñoz Language Survey (discussed next) are available in English and Spanish editions. There are also forms of these tests for school-age children.
The preLAS Observational Assessment™ measures oral language proficiency for 3-year-olds and the preLAS (CTB/McGraw-Hill, 2000) measureslanguage proficiency and literacy skills for children in kindergarten through first grade. It is also used to make language placement decisions,monitor progress over time, and identify learner needs. The IDEA Proficiency Tests (Pre-IPT), Fourth Edition (Ballard & Tighe, 2006) aredesigned to evaluate language, reading, and writing skills in Spanish and English for children in pre-K to 12th grade. The Pre-IPT isadministered to 3- to 5-year-olds and can be used to determine when children are ready to be released from ELL programs. The Woodcock-Muñoz Language Survey®—Revised Normative Update (WMLS®-R NU) (Schrank et al., 2010) can be administered to children as young as age 2. Figure 3-4 lists different categories of preschool tests.
There are distinct phases of second language acquisition, as illustrated in this video. How might a child’s test results be affected during eachstage of second language acquisition? (www.youtube.com/watch?v=Eoca1Ou_6TE)
Tests for School-Age Children
For the child old enough to attend preschool and elementary school, many tests are available for use by teachers, school psychologists,program evaluators, and other personnel with responsibilities for students and the early childhood curriculum. In addition to preschoolprograms for children with disabilities, many states conduct programs for 4-year-old and kindergarten children as well. Descriptions of someof these assessments were included in the previous section on preschool tests. Likewise, some of the assessments in this section includeprekindergarten and kindergarten children. Although individual tests are available for some purposes in school-age programs, group testing isalso used. Group tests require the child to use paper and pencil; therefore, test results may be affected by the child’s ability to respond in thismanner. Test validity and reliability may be affected by the child’s ability both to respond in a group setting and to use a pencil to find and markresponses on the test. As students move into the primary grades, these factors become less important. The tests discussed in this section do notinclude the many tests designed by individual states to meet the grade testing requirements of NCLB. Instead, they address tests forunderstanding strengths and possible delays in language, cognitive, and motor development as children move into the primary grades.
FIGURE 3-4 Categories and characteristics of preschool tests
Many public school programs are designed for children at high risk for disabilities. A number of programs are available, including bilingual andEnglish language programs for children whose first language is not English, intervention programs for children with a physical or intellectualdisability, and preschool programs for children from low-income homes who lack the early childhood experiences that predict successfullearning. These programs may include a screening instrument to determine which children are eligible for special services. The WechslerIntelligence Scale for Children®, Fourth Edition (WISC®-IV) (Wechsler, 2003) and the Bender Visual-Motor Gestalt Test, Second Edition (Bender-Gestalt-II) (Bender, 2003) may be administered to a preschool or school-age child by a school psychologist or school diagnostician todetermine whether the child needs educational services for children with disabilities. Poor performance on the Bender-Gestalt-II by a school-age child indicates the need for further study of the child. The Beery-Buktenica Developmental Test of Visual-Motor Integration, Sixth Edition(BEERY™ VMI) (Beery, Buktenica, & Beery, 2010) is a similar test.
St. Pius Preschool
Areas of southwest Arkansas are experiencing an influx of people from Mexico and Central America who work at a large local paperfactory. Many of these families in one community attend St. Pius Catholic Church, and parishioners have seen the need to provide Englishclasses and other services for the parents as they adjust to a new country and language. As the parents found work, church members alsorecognized a need for child care. They decided to include a concentrated English language development program when they added a child-care center to their outreach activities.
As they began the program, the parishioners realized they needed to find a test that would indicate the children’s progress in learningEnglish as well as provide a language assessment to send to local Head Start, preschool, and kindergarten programs when the childrenwere transitioning out of the St. Pius school. They learned about the Pre-Language Assessment Survey (Pre-LAS) from public schoolcolleagues. After learning how to use the instrument, they were ready to start implementing the test to better help their very youngstudents learn English.
Achievement tests are useful when making decisions about instruction. If a child is exhibiting learning difficulties, a psychologist mightadminister the Peabody Individual Achievement Test—Revised Normative Update (PIAT-R/NU) (Markwardt, 1997) or the Wide Range AchievementTest 4 (WRAT 4) (Wilkinson & Robertson, 2006) to gain information about children’s math, reading, and spelling skills. The teacher mightadminister the Boehm Test of Basic Concepts, Third Edition (Boehm-3) (Boehm, 2000) to young children to determine their need for instructionin basic concepts or to assess successful learning of concepts previously taught.
Primary-grade teachers may also need specific information about a child having difficulties in the classroom. Diagnostic tests such as the Spache Diagnostic Reading Scales (Spache, 1981) can be administered by classroom teachers to pinpoint skills in which students need additionalinstruction. The Preschool Child Observation Record, Second Edition (COR) developed by the HighScope Educational Research Foundation(High/scope Educatonal Research Foundation, 2010) can be used in preschool through fifth grade in six developmental domains, includingsocial development. The Infant & Toddler Child Observation Record (High/Scope Educational Research Foundation, 2003) measuresdevelopmental domains and correlates with the Preschool COR. Figure 3-5 (on the next page) includes examples of initiative items relating toadults, other children, and social problem solving. The checklists can also be used in Head Start programs and child-care centers and withchildren who speak English as a second language.
FIGURE 3-5 Examples from the Preschool Child Observation Record
Source: Preschool Child Observation Record (COR). (2003). Ypsilanti, MI: High/Scope Educational Research Foundation, 2003, used with permission.
Group achievement tests are used to evaluate individual achievement, group achievement, and program effectiveness. All of the new testsdeveloped by individual states to provide accountability for student achievement are group achievement tests. A school district may administerachievement tests every year to determine each student’s progress, as well as to gain diagnostic information on the child’s need for futureinstruction. The same test results can be used at the district level to give information on student’s progress between and within schools and todetermine the effectiveness of the district’s instructional program.
Instructional effectiveness may also be evaluated at the state or national level. A state agency may administer statewide achievement tests towork toward establishing a standard of instructional effectiveness in all schools within the state. Test results can identify school districts thatboth exceed and fall below the set standard. Indicators of poor instructional areas in many school districts pinpoint weaknesses in the state’sinstructional program and facilitate specific types of improvement. As was discussed in Chapter 01, the No Child Left Behind Act, passed in2001, required all states to develop and administer tests to measure achievement in public schools. More recently, the Common Core of StateStandards was developed in an effort to measure achievement in all states (NAEYC, 2012). National assessments are made periodically topinpoint strengths and weaknesses in the educational progress of U.S. children in different subject areas. These findings are frequentlycompared with achievement results of students in other countries. Figure 3-6 lists tests for school-age children.
Though standardized achievement tests help educators and parents better understand the strengths and needs of school-age children, it isimportant to remember how they affect children’s self-esteem. In this video a student reflects on his experience with tests and calls attention tohis interests and characteristics that may not be measured on standardized tests. (https://www.youtube.com/watch?v=nA-y0Txmw8Y)
FIGURE 3-6 School-age tests
In this section, we discussed how standardized tests are used. Although the tests described include various types with different purposes, theprocess used for their development is essentially the same. The next part of the chapter will focus on how standardized tests are designed; thatis, the steps followed in the development of all standardized tests.
3.2 Steps in Standardized Test Design
Test designers follow a series of steps when constructing a new test. These steps ensure that the test achieves its goals and purposes. Inplanning a test, the developers first specify the purpose of the test. Next, they determine the test format. As actual test design begins, theyformulate objectives; write, try out, and analyze test items; and assemble the final test form. After the final test form is administered, thedevelopers establish norms and determine the validity and reliability of the test. As a final step, they develop a test manual containingprocedures for administering the test and statistical information on standardization results.
Specifying the Purpose of the Test
Every standardized test should have a clearly defined purpose. The description of the test’s purpose is the framework for the constructionof the test. It also allows evaluation of the instrument when design and construction steps are completed. The Standards for Educational andPsychological Testing (American Psychological Association [APA], 1999) has established guidelines for including the test’s purpose in thetest manual. In 2013 the standards were under revision. The 1999 standards are as follows:
B2. The test manual should state explicitly the purpose and applications for which the test is recommended.
B3. The test manual should describe clearly the psychological, educational and other reasoning underlying the test and the natureof the characteristic it is intended to measure. (p. 15)
Test designers should be able to explain what construct or characteristics the test will measure, how the test results will be used, and whowill take the test or to whom it will be administered.
The population for whom the test is intended is a major factor in test design. Tests constructed for infants and young children are verydifferent from tests designed for adults. As test developers consider the composition and characteristics of the children for whom they aredesigning the test, they must include variables such as age, intellectual or educational level, socioeconomic background, language andcultural background, and whether the young child can read.
Determining Test Format
Test format decisions are based on determinations made about the purpose of the test and the characteristics of the test takers. The testformat results from the developer’s decision on how test items will be presented and how the test taker will respond (Kaplan & Saccuzzo, 2013). One consideration is whether the test will be verbal or written. Although adults are most familiar with written tests, infants andyoung children are unable to read or write. Tests designed for very young children are usually presented orally by a test administrator. Analternative is to use a psychomotor response; the child is given an object to manipulate or is asked to perform a physical task. A morerecent approach is to conduct assessments during play or daily routines (Linder, 2008).
For older children, high school students, and adults, other test formats are possible. Test takers may respond to an alternative-choicewritten test such as one with true–false, multiple-choice , or matching items. The test may be given as a group test rather thanadministered as an individual test to one person at a time. Short-answer and essay items are also possibilities.
After the test designers have selected the format most appropriate for the test’s purpose and for the group to be tested, actual testconstruction begins. Experimental test forms are assembled after defining test objectives and writing test items for each objective.
Developing Experimental Forms
In preparing preliminary test forms, developers use the test purpose description as their guide. Test content is then delimited. If anachievement test for schoolchildren is to be written, for example, curriculum is analyzed to ensure that the test will reflect the instructionalprogram. If the achievement test is to be designed for national use, then textbook series, syllabi, and curricular materials are studied tocheck that test objectives accurately reflect curriculum trends. Teachers and curriculum experts are consulted to review the content outlinesand behavioral objectives that serve as reference points for test items.
The process of developing good test items involves writing, editing, trying out, and rewriting or revising test items. Before being tried out,each item for an achievement test may be reviewed and rewritten by test writers, teachers, and other experts in the field. Many more itemsthan will be used are written because many will be eliminated in the editing and rewriting stages (Mehrens & Lehman, 1991).
A preliminary test is assembled so that the selected test items can be tried out with a sample of students. The experimental test formsresemble the final form. Instructions are written for administering the test. The test may have more questions than will be used in the finalform because many questions will be revised or eliminated after the tryout. The sample of people selected to take the preliminary test issimilar to the population that will take the final form of the test.
The tryout of the preliminary test form is described as item tryout and analysis. Item analysis involves studying three characteristics of eachtest question: difficulty level, discrimination, and grade progression of difficulty (McMillan, 2013). The difficulty level of a question refers tohow many test takers in the tryout group answered the question correctly. Discrimination of each question involves the extent to which thequestion distinguishes between test takers who did well or poorly on the test. Test takers who did well should have been more successful inresponding to an item than test takers who did poorly. The item differentiates between people who have more or less knowledge or ability.The grade progression of difficulty refers to tests that are taken by students in different grades in school. If a test question has good gradeprogression of difficulty, a greater percentage of students should answer it correctly in each successively higher grade (Mehrens & Lehman, 1991; McMillan, 2013).
Assembling the Test
After item analysis is completed, the final form of the test is assembled. As a result of item analysis, test items have been reexamined,rewritten, or eliminated. Test questions or required behaviors to measure each test objective are selected for the test. If more than one testform is to be used, developers must ensure that alternative forms are equivalent in content and difficulty. Test directions are made finalwith instructions for both test takers and test administrators. In addition, information for test administrators includes details about thetesting environment and testing procedures.
Purpose of and Rationale for Selected Tests
The statement of purpose of a test describes the framework that will be used in designing the test. Two examples follow.
The Neonatal Behavioral Assessment Scale ((NBAS) represents a guide that helps parents, health care providers, and researchersunderstand the newborn’s language. The Scale looks at a wide range of behaviors and is suitable for examining newborns and infantsup to 2 months old. By the end of the assessment, the examiner has a behavioral “portrait” of the infant, describing the baby’sstrengths, adaptive responses, and possible vulnerabilities. The examiner shares this portrait with parents to develop appropriatecaregiving strategies aimed at enhancing the earliest relationship between babies and parents (Brazelton & Nugent, 2011, p. 1).
The Vineland Adaptive Behavior Scales, Second Edition (Pearson, 2008) is an individually administered, norm-referenced measure ofadaptive behavior or “personal and social self-sufficiency” for individuals ages birth–90 years. It is appropriate for educational, socialservices, health care, criminal justice, or military settings. The Vineland II helps measure adaptive behavior of individuals withintellectual disability, autism spectrum disorders (ASDs), attention deficit/hyperactivity disorder (ADHD), post-traumatic brain injury,hearing impairment, dementia, Alzheimer’s disease, and other conditions (Sparrow, Cicchettim, & Balla, 2006, p.1)
Standardizing the Test
Although test construction is complete when the final form is assembled and printed, the test has not yet been standardized. The final testform must be administered to another, larger sample of test takers to standardize each item. There are two types of standardized tests:criterion referenced and normed referenced. Norms provide the tool whereby children’s test performance can be compared with theperformance of a reference group. Criterion-referenced tests provide a description of mastery for each item.
A reference group that represents the children for whom the test has been designed is selected to take the test for the purpose ofestablishing norms or criterion mastery. The performance of the reference or sample group on the final test form during thestandardization process will be used to evaluate the test scores of individuals or groups who take the test in the future.
The norming group is chosen to reflect the makeup of the population for whom the test is designed. If a national school achievement test isbeing developed, the standardization sample consists of children from all sections of the country to include such variables as gender, age,community size, geographic area, socioeconomic status, and ethnic factors. For other types of tests, different characteristics may be used tomatch the norming sample with future populations to be tested.
Various kinds of norms can be established during the standardization process. Raw scores of sample test takers are converted into derivedscores or standard scores for purposes of comparison. Standard scores are achieved by calculating the raw score , or the number of itemsanswered correctly, into a score that can be used to establish a norm. Various types of standard scores can be used to compare the peopleselected to standardize the test with future populations that will be given the test. Each type of grade norm allows test users to interpret achild’s test scores in comparison with the scores of children used to norm the test (Miller, Linn, & Gronlund, 2012; Payne, 2002). Forexample, an age score is established by determining the norms for age groups when the test is given to the norming sample. The age normsdescribe the average performance of children of various ages. Likewise, grade norms or grade-equivalent norms are established bydetermining the average scores made by children at different grade levels in the norming group (Kaplan & Saccuzzo, 2013; McMillan, 2013).
Developing the Test Manual
The final step in test design is development of the test manual. The test developer describes the purpose of the test, the development of thetest, and the standardization procedures. Information on test validity and reliability is also included to give test users information on thedependability of the test. When explaining standardization information in the user’s manual, test developers describe the method used toselect the norming group. The number of individuals included in standardizing the test is reported, as well as the geographic areas, types ofcommunities, socioeconomic groups, and ethnic groups that they represent.
3.3 Differences Between Test Validity and Test Reliability
Norm information is important for establishing confidence in analyzing and interpreting the significance of test scores. Test users also needinformation demonstrating that the test will be valuable for the intended purposes. Therefore, the test manual must provide information onvalidity and reliability. Both types of dependability indicators are equally important in determining the quality of the test. Validity is thedegree to which the test serves the purpose for which it will be used; reliability is the extent to which a test is stable or consistent. Testvalidity can be determined through content validity, criterion-related validity, or construct validity.
When first designing a test, the developers describe its purpose. Test objectives or the test outlines provide the framework for the contentof the test. When a manual provides information on content validity , the test developers are defining the degree to which the test itemsmeasured the test objectives and fulfilled the purpose of the test. Thus, for example, on an achievement test, content validity is the extent towhich the content of the test represents an adequate sampling of the instructional program it is intended to cover. The content validity of areading test would be based on how well the test items measured the reading skills examined in the test. The content validity of amathematics test would look at the content of the objectives on the test and assess how well the test items measured that content.
Criterion-related validity is concerned with the validity of an aptitude test. Rather than analyzing course content, test items focus on skillsor tasks that predict future success in some area. The estimates of predictive validity are concerned with stability over time. For example, an intelligence quotient (IQ) test might be predictive of school achievement. Likewise, Scholastic Aptitude Test scores may predict whetherhigh school students will be successful in college. Validity is predictive because the criteria for success are the future grades the student willearn in college or the student’s future grade-point average.
Criterion-related validity may be concurrent validity , rather than predictive validity. Instead of using a future measure to determinevalidity, current measures are used. The outside criterion is assessed when the test is standardized. The developer of an intelligence testmay cite an existing intelligence test as the criterion to measure validity. The developer administers both intelligence tests to the samplegroup. If the new test scores correlate highly with scores on the existing test, they may be used to establish concurrent validity.
If a test measures an abstract psychological trait, the user’s manual will describe how the sample group was tested to establish constructvalidity. Construct validity is the extent to which a test measures a relatively abstract psychological trait such as personality, verbal ability,or mechanical aptitude (Miller, Linn, & Gronlund, 2012). Rather than examining test items developed from test objectives, one examinesconstruct validity by comparing test results with the variables that explain the behaviors. For example, suppose the construct is believed toinclude certain behavioral characteristics, such as sociability or honesty. An instrument’s construct validity can be checked by analyzing howthe trait is affected by changing conditions. Alternatively, an instrument may measure level of anxiety; its construct validity is determined bycreating experiments to find out what conditions affect anxiety (Miller, Linn, & Gronlund, 2012).
The validity of a test is the extent to which the test measures what it is designed to measure. Test users, however, are also interested in atest’s dependability or stability in measuring behaviors. Test developers, therefore, also establish and report on the reliability of theinstrument as part of the standardization process.
Test reliability is related to test item discrimination. When test items are analyzed after the initial item tryout, they are examined fordiscrimination power. After the final test form is administered to a norming sample, the items are analyzed again to ensure that theinstrument is fairly reliable. The whole test is analyzed, rather than individual test items. The test manual reports the test’s reliability asdetermined by using alternative-form, split-half, or test–retest reliability measures. A test’s reliability coefficient describes the degree towhich a test is free from error of measurement. If alternative-form reliability strategies are used, test developers construct two equivalentforms of the final test. Both forms are administered to the norming group within a short period. The correlation between the results on thetwo different forms measures the coefficient of reliability. For example, standardized achievement tests are published using severaldifferent forms of the test. To measure reliability, the norming group takes two forms of the test and then the results are compared to see ifthe performance on each of the tests was the same or very similar.
If a split-half reliability coefficient is used to establish reliability, the norming group is administered a single test, and scores on half of thetest are correlated with scores on the other half of the test. Split-half reliability is determined from the contents of a single test. A test withsplit-half reliability is also considered to have internal consistency ; that is, the items on each half of the test are positively correlated inmeasuring the same characteristics.
Test–retest reliability is also derived from the administration of a single test form. In this case, however, the test is administered to thenorming group and then is administered again after a short interval. The two sets of scores are compared to determine whether they wereconsistent in measuring the test objectives.
Factors that Affect Validity and Reliability
Despite the measures and procedures that are used to ensure validity and reliability in standardized tests, other factors can affect testoutcomes. Some common factors are reading ability, the physical condition of the testing room, memory, and the physical condition of theindividual taking the test. Thus, if the testing room is uncomfortably warm or a student had inadequate rest the night before the test, scoreswill be affected.
Lack of adherence to time limits and lack of consistency in test instructions affect test scores. Other factors are inconsistency in the rating ofessays from individual to individual and student guessing of test answers (Payne, 1997).
Validity is affected by such factors as unclear directions, difficulty of reading vocabulary on the test, and test items that are not appropriatefor the test objectives (Miller, Linn, & Gronlund, 2012). Reliability is affected by the number of test items or the length of the test, lack ofinter-rater reliability, and extraneous events that affect the testing situation (Miller, Linn, & Gronlund, 2012; McMillan, 2013).
These and other factors affect the possible errors on a test and the quality of the test. This variation in testing quality is accounted for in the standard error of measurement , discussed next.
Standard Error of Measurement
No matter how well designed, no test is completely free from error. Although there is a hypothetical true score , in reality it does not exist.The reliability of the test depends on how large the standard error of measurement is after analysis of the chosen method of determiningreliability. If the reliability correlations are poor, the standard error of measurement will be large. The larger the standard error ofmeasurement, the less reliable the test. Standard error of measurement is the estimate of the amount of variation that can be expected intest scores as a result of reliability correlations.
Several variables that are present during standardization affect test reliability, as discussed earlier. First is the size of the population sample.Generally, the larger the population sample, the more reliable the test will be. Second is the length of the test. Longer tests are usually morereliable than shorter tests. Longer tests have more test items, resulting in a better sample of behaviors. The more items that measure abehavior, the better the estimate of the true score and the greater the reliability. Strict adherence to test directions by test administratorscontributes to higher reliability, whereas variations in test instructions or the coaching of students can distort the reliability of test results.
The third variable that can affect standard error of measurement is the range of test scores obtained from the norming group. The widerthe spread of scores, the more reliably the test can distinguish among them. Thus, the range of scores demonstrates how well the testdiscriminates between good and poor students (Miller, Linn, & Gronlund, 2012). The spread of test scores can be related to the number ofstudents taking the test. The larger the testing sample, the more likely there will be a wider spread of test scores.
3.4 Considerations in Selecting and Evaluating Standardized Tests
Whenever a private school, public school district, preschool, or child-care center decides to use a test to evaluate children, educators mustdecide how to select the best test for that purpose. Those who select the test must determine the relevant questions to ask about the test.Brown (1983) identifies various factors that test users must consider: (1) the purpose of the testing, (2) the characteristics to be measured,(3) how the test results will be used, (4) the qualifications of the people who will interpret the scores and use the results, and (5) anypractical constraints. All these factors are important in selecting tests for young children. Because of the developmental limitations of youngtest takers, test formats must be compatible with their ability to respond. Developmental limitations include short attention span,undeveloped fine-motor skills, inability to use reading skills for test responses, and poor performance on group tests. Limitations in trainingand experience in those who administer the test are also factors in test selection.
Other relevant concerns, particularly in selecting tests for young children, are the costs involved, testing time, and ease of scoring and usingtest results (Kaplan & Saccuzzo, 2013). The test must be reasonable in cost, and the time needed to administer the test should be suitablefor young children.
A major issue is whether the test has quality. Is it a good test to use with the children? The person searching for an appropriate test willwant to examine the test manual for indications of how well the test was designed and normed. The test manual should include informationon the following:
1. Purpose of the test. The statement of purpose should include the rationale for the test, the characteristics the test is designed to measure,and the uses for the test.
2. Test design. The procedures and rationale for selecting test items and the development and trial of test forms should be explained.
3. Establishment of validity and reliability. The description should describe the procedures used to establish validity and reliability toinclude sufficient data on validity, reliability, and norms.
4. Test administration and scoring. Specific information should be given on how to administer and score the test and to interpret testresults. Information should be adequate for users to determine whether the test is practical and suitable for their purposes. Potentialproblems should be pointed out that can be encountered when administering and scoring the test (Kaplan & Saccuzzo, 2013). See Figure 3-7 for questions that should be answered in a test manual, including an acceptable coefficient of reliability.
FIGURE 3-7 Questions for test manuals about the quality of tests
Test users need extensive training in tests and measurements to interpret a test manual adequately. For many users, the explanations anddata reported in test manuals are complex and difficult to understand. A reader may have difficulty in deciding whether the reliabilitycoefficient is adequate, whether the size and demographic characteristics of the norming population are appropriate, or whether testcontent and format are suitable for the intended uses. To obtain additional help in understanding the suitability of the test, test users willwant to consult resources for test standards and reviews. The Standards for Educational and Psychological Testing (APA, 1999) includesstandards for tests, manuals, and reports. It also includes standards for reliability and validity, as well as information that should beincluded on the use of tests.
The Buros Institute of Mental Measurements is perhaps the most important source in identifying, describing, and evaluating published tests.The series of Tests in Print is a comprehensive bibliography of thousands of tests in five volumes. The most recent, Tests in Print VII(Murphy, Spies, & Plake, 2006), consists of two volumes. The tests are listed by type, and basic information is given about each test.
The Mental Measurements Yearbooks include descriptive information about tests, plus professional reviews. The content also includessources of information about test construction, validation, and use. Critical reviews of the tests are included. For example, the Stanford–Binet Intelligence Scale (Terman & Merrill, 1973) is the oldest and most highly regarded IQ test used in the United States. However, thefourth edition of the test (Thorndike, Hagen, & Sattler, 1986) was found to be significantly different from the earlier editions. Reviewerspointed out that users are given poor information on the accuracy of reliability scores, the test is less game-like and therefore likely to beless appealing to children, and it overrepresents parents from high occupational and educational levels in the sample of children used fornorming (Anastasi, 1989; Cronbach, 1989). The most recent edition is the Stanford-Binet Intelligence Scales (SB5), Fifth Edition (Roid, 2003).Educators choosing a test need to be informed of the quality of the test being considered for selection. The most recent yearbook is the Eighteenth Mental Measurements Yearbook (Spies, Carlson, & Geisinger, 2010).
A resource that is particularly helpful to people without a background in test design at a technical level is Test Critiques, Volumes I–X (Keyser& Sweetland, 1984–1994). It includes information about test design and use, as well as a critique of the tests. Other resources for testevaluation and selection are particularly suitable for users of early childhood tests.
Test reviews can be accessed online. The Buros Center for Testing provides search engines for Test Reviews Online (unl.edu/buros). Anotheronline source is Psychware Sourcebook (Krug, 1993). It identifies and describes computer products available for assessment in education,psychology, and business.
Brown (1983) summarized the steps in selecting and evaluating tests as follows:
1. Outline your general requirements: the purpose of testing, the characteristics to be measured, and the nature of the group to be tested.Consider also the qualifications of test users and practical considerations.
2. Identify what tests are available that appear to meet your needs. Here sources such as Tests in Print, the Mental Measurement Yearbooks,test publishers’ catalogs, and test compilations will be most helpful.
3. Obtain further information about these tests from texts, journals, reference books, and consultation with people who have used thistype of test.
4. Select the most promising tests. Obtain samples (specimen sets) of these tests.
5. Make a detailed evaluation of these tests, keeping in mind the unique requirements of your situation. On the basis of these evaluations,select the test(s) to be used.
6. If possible, conduct an experimental tryout of the test before putting it to use.
7. Use the test. Constantly monitor and evaluate its usefulness and effectiveness. (p. 463)
Summary
Psychological tests are administered to children of all ages beginning in infancy. The tests can be organized to measure abilities such as achievement, aptitudes, and intelligence. There are also measures to understand interests, attitudes, values, and personality characteristics. These types of psychological tests have been organized in this chapter according to age level. Thus, there are charts for tests for infants and toddlers, preschool children, and school-age children. Some of the measures, such as interest and attitude tests, are only appropriate for school-age children. The tests vary in quality depending on how they were designed and evaluated for validity and reliability. As a result, the choice of tests for a particular purpose should be examined for established quality indicators.
Despite their shortcomings, standardized tests are useful for test users. Because they have been carefully developed through a series of steps that ensure their dependability, educational institutions, in particular, use them to measure students’ characteristics. Good standardized tests are normed by using many individuals from various backgrounds who live in different parts of the United States. As a result, the tests also accurately measure the population to whom the tests are given.
Although the process of developing a standardized test may seem to be unnecessarily tedious, good test design requires careful planning and attention to each step. The ultimate validity and reliability of the test depend on the attention paid to design details, beginning with the definition of the test’s purpose and ending with the description of technical data about the test’s construction in the users’ manual.
Key Terms
achievement test 55
alternative-form reliability 72
aptitude test 55
concurrent validity 72
construct validity 72
content validity 71
criterion-referenced test 70
criterion-related validity 72
equivalent forms 70
grade norm 71
group test 69
individual test 69
intelligence quotient (IQ) 72
intelligence test 55
interest inventory 56
internal consistency 73
item analysis 69
multiple choice 69
norm 70
personality test 55
raw score 71
reliability 71
split-half reliability 72
standard error of measurement 73
test–retest reliability 73
true score 73
validity 71
Selected Organizations
Search for the following organizations online:
Child Care Exchange
Buros Institute of Mental Measurements
Test Reviews Online
References
Als, H., Butler, S., Kosta, S., & McAmulty, G. (2005). The assessment of preterm infants’ behavior (APIB): Furthering the understanding and measurement of neurodevelopmental competence in preterm and full-term infants. Mental Retardation and Developmental/Disabilities Research Reviews, 11, 94–102.
Als, H., Lester, B. M., Tronick, E., & Brazelton, T. B. (1982). Towards a research for the assessment of preterm infants’ behavior (APIB). In H. E. Fitzgerald, B. M. Lester, & M. W. Yogman (Eds.), Theory and research in behavioral pediatrics (Vol. 1, pp. 1–35). New York, NY: Plenum Press.
Als, H., Tronick, E., Lester, B. M., & Brazelton, T. B. (1979). Specific neonatal measures: The Brazelton Neonatal Behavioral Assessment Scale. In J. D. Osofsky (Ed.), Handbook of infant development (pp. 185–215). New York, NY: Wiley.
American Psychological Association. (1999). Standards for educational and psychological testing. Washington, DC: Author.
Anastasi, A. (1989). Review of the Stanford–Binet Intelligence Scale, Fourth Edition. In J. C. Conoley & J. J. Kramer (Eds.), The tenth Mental Measurements Yearbook (pp. 771–772). Lincoln, NE: University of Nebraska Press.
Apgar, V. (1975). A proposal for a new method of evaluation of a newborn infant. Anesthesia and Analgesia, 32, 260–267.
Ball, R. S. (1977). The Gesell Developmental Schedules. Journal of Abnormal Child Psychology, 5, 233–239.
Ballard, W., & Tighe, P. (2006). IDEA Proficiency Tests (Pre-IPT), Fourth Edition. Brea, CA: Author.
Bayley, N. (2005). Bayley Scales of Infant Development, Third Edition (BSID-III). San Antonio, TX: Pearson Assessments.
Beery, K. E., Buktenica, H., & Beery, N. (2010). Beery-Buktenica Developmental Test of Motor Integration, Sixth Edition. San Antonio, TX: Pearson Assessments.
Bender, L. (2003). Bender Visual Motor Gestalt Test for Children, Second Edition (Bender-Gestalt-II). San Antonio, TX: Pearson Assessments.
Boehm, A. E. (2000). Boehm Test of Basic Concepts, Third Edition. San Antonio, TX: Pearson Assessments.
Bracken, B. A. (1998). Bracken Basic Concept Scale—Revised (BBCS-R). San Antonio, TX: Pearson Assessments.
Brand, D. A., Altman, R. L., Puttill, K., & Edwards, K. S. (2005, April). Yield of diagnostic testing in infants who have had an apparent life-threatening event. Pediatrics, 115, 885–893.
Brazelton, T. B., Berry, T., & Nugent, J. K. (2011). Neonatal Behavioral Assessment Scale (NBAS), Fourth Edition. Cambridge, England: Cambridge University Press.
Brazelton, T. B. (1996). A window on the newborn’s world: More than two decades of experience with the Neonatal Behavioral Assessment Scale. In S. J. Meisels & E. Fenichel (Eds.), New visions for the developmental assessment of infants and young children (pp. 127–146). Washington, DC: Zero to Three: National Center for Infants, Toddlers, and Families.
Brigance, A. H. (2013). Brigance Early Childhood Screens, III. Woburn, MA: Curriculum Associates.
Brown, E. G. (1983). Principles of educational and psychological testing (3rd ed.). New York, NY: CBS College Publishing.
Buros, O. K. (1999). Tests in print V. Lincoln, NE: University of Nebraska Press.
Calhoun, C. L., Gaughan, J. P., Chafitz, R. S., & Mulcahey, M. J. (2009, Spring). A pilot study of observational motor assessment in infants and toddlers with spinal cord injury. Pediatric Physical Therapy, 21, 62–67.
Campbell, S. K., Kolobe, T. H.A., Osten, E. T., Lenke, M. L., & Girolami, G. L., (1995). Construct validity of the Test of Infant Motor Performance. Physical Therapy, 75, 585–596.
Cronbach, L. J. (1989). Review of the Stanford-Binet Intelligence Scale, Fourth Edition. In J. C. Conoley & J. J. Kramer (Eds.), The tenth Mental Measurements Yearbook (pp. 773–775). Lincoln, NE: University of Nebraska Press.
Cronbach, L. J. (1990). Essentials of psychological testing (5th ed.). New York, NY: Harper & Row.
CTB/McGraw-Hill. (2000). Pre-LAS 2000. Monterrey, CA: Author.
DeGangi, G., Poisson, S., Sickel, R., & Wiener, A. S. (1995). Infant/Toddler Symptom Checklist: A screening tool for parents (ITSC). San Antonio, TX: Pearson Assessments.
Dunn, L. M., & Dunn, L. (1997). Peabody Picture Vocabulary Test, Fourth Edition (PPVT-4). San Antonio, TX: Pearson Assessments
Frankenburg, W. K., Dodds, J., Archer, P., Shapiro, H., & Bresnick, B. (1992). Denver II. Denver, CO: Denver Developmental Materials.
Geisinger, K. F., Spies, R. A., Carlson, J. F., & Plake, B. S. (Eds.). (2007). The seventeenth Mental Measurements Yearbook. Lincoln, NE: University of Nebraska Press.
Gesell Institute. (2011). Gesell Developmental Observation—Revised. New Haven, CT: Author.
Gronlund, N. E. (1990). Measurement and evaluation in teaching (6th ed.). New York, NY: Macmillan.
Hack, M., Taylor, H. G., Drotar, D., Schluchter, M., Carter, L., Wilson-Costello, D., Klein, N., Friedman, H., Mercuri-Miinicih, N., & Morrow, M. (2005). Poor predictive validity of the Bayley Scales of Infant Development for cognitive function of extremely low birth weight children. Pediatrics, 116, 333–341.
Hammill, D. D., Pearson, N. A., & Voress, J. K. (1996). Test of Visual–Motor Integration. Austin, TX: PRO-ED.
Harrison, P., Kaufman, A. S., & Kaufman, N. L. (1990). AGS Early Screening Profiles (ESP). Bulverde, TX: Pearson Assessments.
Harrison, P., & Oakland, T. (2003). Adaptive Behavior Assessment System, Second Edition (APAS-II). Los Angeles, CA: Western Psychological Services.
High/Scope Educational Research Foundation. (2003). Preschool Child Observation Record. Ypsilanti, MI: Author.
Kaplan, R. M., & Saccuzzo, D. P. (2013). Psychological testing principles: Applications and issues (8th ed.). Belmont, CA: Wadsworth.
Kaufman, A., & Kaufman, N. (2005). Kaufman Assessment Battery for Children (K-ABC-II): Sampler manual (2nd ed.). San Antonio, TX: Pearson Assessments.
Keyser, D. J., & Sweetland, R. C. (1984–1994). Test critiques (Vols. I–X). Kansas City, MO: Test Corporation of America.
Krug, S. E. (1993). Psychware Sourcebook. Kansas City, MO: Test Corporation of America.
Lambert, N., Nihira, K., & Leland, H. (2008). Adaptive Behavior Scale-School, Second Edition (ABS-S:2 tm). North Tonawanda, NY: Multi-Health Systems.
Le Buffe, P. A., & Naglieri, J. A. (2007). Devereux Early Childhood Assessment, Infant and Toddler (DECA-I/T/). Villanova, PA: Devereux Center for Resilient Children.
LeBuffe, P. A., & Naglieri, J. A. (2012). Devereux Early Childhood Assessment Preschool Program, Second Edition (DECA P-2). Villanova, PA: Devereux Center for Resilient Children.
Linder, T. W. (2008). Transdisciplinary Play-Based Assessment-2. Baltimore, MD: Brookes.
Mardell, C. D., & Goldenberg, D. S. (2011). Developmental Indicators for the Assessment of Learning (Fourth Edition ed.). San Antonio, TX: Pearson Assessments.
Markwardt, F. (1997). Peabody Individual Achievement Test—Revised Normative Update.(PIAT-R/NU). San Antonio, TX: Pearson Assessments.
McMillan, J. H. (2007). Classroom assessment: Principles and practice for effective instruction (4th ed.). Boston, MA: Allyn & Bacon.
Mehrens, W. A., & Lehmann, I. J. (1991). Measurement and evaluation in education and psychology (4th ed.). New York, NY: Harcourt Brace.
Merrell, K. W. (2003). Preschool and Kindergarten Behavior Scales, Second Edition (PKBS-2). Austin, TX: PRO-ED.
Meisels, S. J., & Atkins-Burnett, S. (2004, January). The Head Start National Reporting System: A critique. Young Children, 59, 64–66.
Meisels, S. J., & Atkins-Burnett, S. A. (2005). Developmental screening in early childhood: A guide (5th ed.). Washington, DC: National Association for the Education of Young Children.
Meisels, S., Marsden, D. B., Dombro, A. L., Weston, D. R., & Jewkes, A. M. (2008). The Ounce Scale. San Antonio, TX: Pearson Assessments.
Meisels, S. J., Marsden, D. B., Wiske, M. S., & Henderson, L. W. (2008). Early Screening Inventory—Revised (ESI-R). New York, NY: Pearson Education.
Miller, L. J. (1993). First Step: Screening Test for Evaluating Preschoolers (First Step). San Antonio, TX: Pearson Assessments.
Miller, M. D., Linn, R. L., & Gronlund, N. M. (2012). Measurement and assessment in teaching (Eleventh ed.). Upper Saddle River, NJ: Pearson.
Mullen, E. M. (1995). Mullen Scales of Early Learning, AGS Edition. San Antonio, TX: Pearson Assessments.
Murphy, L. L., Spies, R. A., & Plake, B. S. (2006). Tests in print, VII. Lincoln, NE: University of Nebraska Press.
Newborg, J. (2004). Battelle Developmental Inventory—II (BDI-II). Chicago, IL: Riverside Publishing.
National Association for the Education of Young Children. (2012). Common Core Standards: Caution and opportunity for early childhood educators. Washington, DC: Author.
Nihira, K., & Lambert, N. (1993). AAMR Adaptive Behavior Scale—School (ABS-S:2). Washington, DC: American Association on Mental Retardation.
Panitch, H. B. (2004). The role of pulmonary testing in infants. NeoReviews, 5. Retrieved September 29, 2009, from http://www.neoreviews.aapublications.org
Payne, D. A. (1997). Applied educational assessment. Belmont, CA: Wadsworth.
Powell, G., MacKrain, M., & LeBuffe, P. (2007). Devereux Early Childhood Assessment Infant and Toddler (DECA-I/T) technical manual. Lewisville, NC: Kaplan Early Learning Company.
Prizant, B., & Wetherby, A. (2002). Communication and Symbolic Behavior Scales Developmental Profile (CSBS DP), First Normed Edition. Baltimore, MD: Brookes Publishing.
Roid, G. (2003). Stanford-Binet Intelligence Scales (SBS), Fifth Edition. Rolling Meadows, IL: Riverside Publishing.
Schrank, F. A., Wendling, B. J., Alvarado, C. G., & Woodcock, R. W. (2010). Woodcock-Munoz Language Survey—Revised Normative Update (WMLS-R NU). Rolling Meadows, IL: Riverside Publishing.
Spache, G. D. (1981). Diagnostic Reading Scales: Examiner’s manual. Monterey, CA: CTB/McGraw-Hill.
Sparrow, S. S., Cicchetti, D. V, & Balla, D. A. (2005). Vineland Adaptive Behavior Scale (2nd ed.). San Antonio, TX: Pearson Assessments.
Squires, J., & Bricker, D. (2009). Ages and Stages Questionnaire, Third Edition (ASQ:3). Baltimore, MD: Brookes.
Terman, L. M., & Merrill, M. A. (1973). Stanford–Binet Intelligence Scale: Manual for the third revision form L-M. Boston, MA: Houghton Mifflin.
Thorndike, R. L., Hagen, E. P., & Sattler, J. M. (1986). Stanford–Binet Intelligence Scale (Fourth ed.). Chicago, IL: Riverside.
U. S. Department of Health and Human Services Head Start Bureau. (2003). National Reporting System. Washington, DC: Author.
Walker, H. M., Severson, H. H., & Feil, E. G. (1995). Early Screening Project (ESP). Longmont, CO: Sopris West.
Wechsler, D. (2003). Wechsler Intelligence Scale for Children, Fourth Edition (WISC-IV). San Antonio, TX: Pearson Assessments.
Wechsler, D. (2002). Wechsler Preschool and Primary Scale of Intelligence, Third Edition (WPPSI-III). San Antonio, TX: Pearson Assessments.
Widerstrom, A. H., Mowder, B. A., & Sandall, S. R. (1991). At-risk and handicapped newborns and infants. Upper Saddle River, NJ: Prentice Hall.
Wilkinson, G. S., & Robertson, G. J. (2006). Wide Range Achievement Test 4 (WRAT4). Lutz, FL: Psychological Assessment Resources.
Wodrich, D. (1997). Children’s psychological testing: A guide for nonpsychologists. Baltimore, MD: Brookes.
Woodcock, R. W., & Muñoz-Sandoval, A. F. (2005). Woodcock-Muñoz Language Survey—Revised (WMLS-R). Chicago, IL: Riverside.
Zeitlin, S., Williamson, G. G., & Szczepanski, M. (1988). Early Coping Inventory (ECI): A measure of adaptive behavior. Bensenville, IL: Scholastic Testing Service.
Zimmerman, I. L., Steiner, V. G., & Pond, R. E. (2007). Preschool Language Scale, Fourth Edition (PLS-4). San Antonio, TX: Pearson Assessments.
