The following article has been excerpted from Identifying Gifted Students: A Practical Guide. This valuable resource offers up-to-date information for building an effective, defensible identification process. It acts as a hands-on, research-based guide. Designed for practicing professionals who must make decisions daily about identifying and serving gifted and talented students, this book acts as a handbook for establishing procedures that are effective in identifying gifted and talented students from diverse backgrounds.

When identifying gifted students, schools need to select qualitative and quantitative instruments that are technically adequate and that match gifted students’ characteristics and the school district’s program. Each state has specific rules that govern the types and kinds of assessments that should be used. In Texas, districts must identify students beginning in kindergarten and use a “minimum of three appropriate criteria that include both qualitative and quantitative measures . . . in languages they understand or with non-verbal based tests” (Texas Education Agency, 1996, Section 1, 1.5.2A, 1.5.4A, p. 4). These assessments must also include information from multiple sources for each area of giftedness served by the district (Texas Education Agency, Section 1, 1.5.1A, 1.1.5A, p. 4).

In meeting these guidelines, for example, a school district might select these five assessments for identification of students who are gifted in a specific academic area: teacher nomination, parent nomination, an intelligence test, an achievement test in the academic area, and a portfolio of work. These five assessments meet minimum standards: Quantitative and qualitative instruments are included and multiple sources—parent, teacher, and the student—are used. To be at an acceptable level, the school district will also want to make sure that the selected instruments are in a language that the students understand and that the standardized instruments have included bias studies in their technical manual.

Similar to Texas, most states rely on multiple criteria (Coleman, Gallagher, & Foster, 1994). This requirement is aligned with the Office for Civil Rights equal access concerns, which emphasize “multiple alternative referral sources” (Trice & Shannon, 2002; see Office for Civil Rights Checklist for Assessment of Gifted Programs in Appendix A at the end of this article.)

Multiple assessments are important for several reasons. First, no single test samples all behaviors (Salvia & Ysseldyke, 2001). Even intelligence tests vary according to theories and definitions that underlie the test design. Second, tests measuring the same trait may relate to one another, but produce different scores. For example, a student might score 130 on one intelligence test (in the very superior range) and 110 on another intelligence test (in the above-average range). Both tests may have good technical qualities, but simply sample different behaviors, be based on different definitions, be individually or group administered, or have different standard errors of measurement. Third, several sources of information (e.g., parent, teacher, student, and peers) will provide examples of behaviors across different settings and provide a broader picture of the gifted student. Gifted students may show more of their abilities at home or with friends. Coleman and Cross (2001) made the excellent point that certain behaviors are simply not exhibited in certain settings because the student doesn’t have the opportunity, or others, such as friends, teachers, or parents, might not understand or approve.

Richert (1991), on the other hand, has suggested that more is not always better—multiple measures can increase elitism in identification. For example, a district might choose to use grades, teacher nominations, and achievement tests, which might exclude gifted students who were underachievers, who didn’t have a strong academic background, or who were gifted in other areas besides academics. The National Report on Identification (Alvino, McDonnel, & Richert, 1981; Richert, 1985) offered these six principles in developing a comprehensive, fair identification system:

• Advocacy—is it in the best interests of students?
• Defensibility—is it based on best research and recommendations?
• Equity—does it provide equal opportunity for every child?
• Pluralism—does it use the broadest definition of giftedness?
• Comprehensiveness—does it serve many gifted students?
• Pragmatism—does it allow for modification and use accessible resources?

Along with the use of multiple assessments, these principles need to be considered when districts establish an identification process.

While all schools are required to identify students at least once a year (Texas Education Agency, 1996, Section 1.1.3A), some elect to identify students on an ongoing basis, referring students as they display characteristics in the classroom or in the community. Identification policies also need to address “furloughs, reassessment, exiting of students from program services, transfer students, and appeals of district decisions regarding program placement” (Texas Education Code, Chapter 89, 19 TAC Section 89.1[5]). Students who fit within these categories may be included in the overall identification process or may be treated on a case-by-case basis. For example, a student who transfers from another gifted program might be screened using the same identification process that is used for all students in the school district or may be placed within the program on a trial basis.

The identification process may vary. Some schools may choose to administer all of the assessments to all of the children at a particular grade level (e.g., kindergarten; Texas Education Agency, 1996, Section 1, 1.5.2R) and then decide which children will be referred to a final selection committee for placement into programs for gifted and talented children. Other districts may develop a three-phase process: nomination, screening, and selection (see Figure 5.1). At each of these phases, decisions must be made to determine which children progress to the next phase of assessment or placement. Some researchers suggest the addition of a validation phase in which identification procedures are evaluated by others outside of the school (Feldhusen, Asher, & Hoover, 1984; Feldhusen & Baska, 1985; Feldhusen, Hoover, & Sayler, 1990). This step is important in assuring that the process is valid, that is, that it identifies the gifted and talented students who need services.

A large group of students needs to be created during the nomination phase, even those who show only vague hints of giftedness. All students who exhibit any or some of the characteristics that indicate special gifts and talents should have an equal opportunity to be nominated (see Office for Civil Rights Checklist for Assessment of Gifted Programs in Appendix A at the end of this article). The placement of students in special education programs or with certain teachers who may or may not believe in gifted education should not preclude their inclusion in the nomination group. Every effort should be made to involve students from special populations, such as those with disabilities, from minority or lower income backgrounds, with limited English proficiency, and from rurally isolated areas. School districts may want to consider placing ads in local newspapers and sending flyers home in multiple languages that advertise the program and describe the identification process to parents. Nomination instruments may include teacher and parent checklists, group intelligence and achievement tests, portfolios of work, peer and self-nominations, teacher reports of students’ learning, and student background information.

If teachers are a part of the nomination process, they need to receive professional development training in the characteristics of gifted and talented students. With training, teachers identify more children (e.g., 85%) than untrained teachers (e.g., 40%; Gear, 1978). In addition, teachers at the high school level are better at identifying gifted students than at the middle school or elementary level (Cornish, 1968; Jacobs, 1971; Pegnato & Birch, 1959). In fact, parents are actually better at identifying very young children (e.g., 76%) than teachers (4.3%) when using an intelligence test as the criterion (Jacobs). Once trained, teachers should observe their students when they are involved in activities that are more open-ended and require more complex thinking and other behaviors. If the tasks are not challenging and require mostly single answers or low-level responses, gifted students do not have sufficient opportunities to demonstrate their higher level abilities. Some states ensure that opportunities are provided during the nomination process by using a “prereferral” process similar to special education. In this process, the teacher uses a variety of strategies in the classroom to determine if the student might be served within the general education program or needs services beyond the classroom (see Table 5.1). Strategies relate to motivation and research, rate or pacing, preference, and content and instruction.

View Detail

In all cases, the nomination instruments should be fair to the culture. Culturally appropriate measures usually (a) ensure that the student understands the purpose and the nature of the testing process, (b) minimize language, (c) include practice items, (d) minimize time constraints, and (e) present novel problems instead of narrow school-related information (Jensen, 1969).

Finally, multiple sources—parents, teachers, student, and peers—need to be used in the nomination phase to ensure equal access. Unless required by the program (e.g., writing, visual and performing arts), the formats of the assessments might also vary so that all students may perform using their strengths. For example, some activities might require verbal responses; others, manipulative responses; and still others, written responses.

Once all of the nomination information is collected, the identification committee may determine which students will proceed to the second phase: screening. While many districts choose 20–25% of their population for further screening, others may choose to administer all of the screening measures to the entire school population or to all students who are nominated for the program. No one instrument should be used as a single criterion. For example, movement to the screening phase should not be based on a cut-off score from a single measure, such as the 85th percentile on an achievement test or a single teacher nomination; rather, it should be based on successful performance on several assessments used during the nomination phase. These assessments might include parent, teacher, self, or peer checklists or observations; schoolwork that is a part of a portfolio; and achievement or aptitude tests. In this phase, the committee will want to include all students who appear to exhibit or have the potential to exhibit the desired qualities. A good rule of thumb might be that when in doubt, screen the student further.

During the screening phase, additional information is collected on the nominated students. Since the number of students in the screening phase will be smaller than in the nomination phase, a district might consider using individually administered measures or methods that allow for more clinical observations such as interviews, participation in a classroom for gifted students, or observations of the ways in which the student learns new information (e.g., dynamic assessment). Dynamic assessment focuses on the interaction between the gifted student and the task. The tasks should be problem-based and require complex strategies that discriminate among intelligent individuals (Geary & Brown, 1991; Kurtz & Weinert, 1989; Scruggs & Mastropieri, 1985). These novel tasks might provide opportunities for varying rates of learning, efficiency in retrieving information for solving problems, transfer to new tasks, and knowledge about a learner’s strategies (Johnsen, 1997). Again, all students should have opportunities to demonstrate their best performance levels. By the end of the nomination and screening phases, a school should have data from multiple sources and quantitative and qualitative assessments.

During the selection phase, the identification committee examines all of the data that have been collected on each child nominated and screened. The committee should be comprised of “at least three local district or campus educators who have received training in the nature and needs of gifted students” (Texas Education Agency, 1996, Section 1, 1.7A; 19 TAC Section 89.1[4]). All data from both the nomination and screening phases should be considered. To ensure objectivity, the committee may initially want to identify students by number only and add clinical or qualitative information later.

The identification committee determines which students are selected for which gifted program. While many districts may select 5–10% of their student population, this percentage will vary depending upon the number of programs provided for gifted and talented students and the number of students whose characteristics indicate the need for services that are not ordinarily provided by the school. Given the synthesis of the qualitative and quantitative information, the committee might also want to create a differentiation plan that includes specific programming based on the gifted student’s strengths and weaknesses, long- and short-term goals, classroom activities within the gifted and the general education program, and evaluation. These plans will set the stage for the next phase, which includes an annual evaluation of the identification procedure (Texas Education Agency, 1996, Section 5, 5.3A; TEC Section 11.251–11.253).

The following questions might be used to guide the school district in building a defensible procedure for identifying gifted and talented students:

1. Is the procedure based on best research and recommendations?
2. Does it match the district’s definition and program options?
3. Do all students have an equal opportunity to be nominated?
4. Are special populations considered in the nomination process?
5. Are all students able to demonstrate their strengths?
6. Are assessments fair to student cultures?
7. Are all students able to demonstrate their abilities in classroom activities?
8. Are multiple sources of information used?
9. Are all data considered during the selection phase?
10. Are the students’ data evaluated objectively?
11. Are all students who need a differentiated education being identified?
12. Do identified students perform well in the program that matches their gifts or talents?

The school district will want to continue collecting data to ensure that all gifted students are being served effectively. Data from the identification process may be correlated with future performance in the classroom, future performance on other assessments, future performance on assessments used for program evaluation purposes, and future performance in out-of-school settings.

Under the Fifth and Fourteenth Amendments of the U.S. Constitution or by state or federal statutes, due process procedures are imposed on school districts (Karnes & Marquardt, 1991). In 10 states, gifted students are afforded the same provisions as handicapped students. Other states, including Texas, have general due process procedures that are applicable to the gifted (Karnes & Marquardt).

To assure due process rights, a school district will want to identify clearly for parents and guardians time frames and steps in a locally developed appeals process (Texas Education Agency, 1996, Section 1, 1.2A, 1.2.6R; 19 TAC Section 89.1[5]). These steps may include meaningful parent meetings with (a) teachers, (b) the selection committee or building administrator, (c) a school district committee that would include the director or administrator responsible for the gifted program, and, finally, (d) the school board. If these meetings do not resolve the issues, the school district may want to bring in an impartial and professionally trained mediator. The mediator would discuss the important issues with the involved participants and try to resolve any remaining conflicts. If these conflicts are still not resolved, the parents or the school district may contact the state education agency and initiate a formal hearing. Hearing procedures generally allow parents to choose if the hearing is open or closed and if their child may attend. Both sides may choose to have counsel and present expert witnesses. If the formal hearing still does not resolve the conflicts, the parents or the school district may still choose to litigate in the federal or state court system. “Litigation should be the last resort . . . going to court is expensive, time consuming, adversarial, and emotionally draining” (Karnes & Marquardt, 1991, p. 37).

At each phase in the identification process, the committee needs to examine qualitative and quantitative information. These data may be organized in a variety of ways: case studies, profiles, matrices, or other forms. Whichever approach is used, the identification committee should follow these guidelines.

Guideline 1: Weighting of assessments. If each assessment has equal reliability and validity for identifying gifted and talented students, then each should have equal value in the decision-making process. The committee should not assign more weight or importance to one assessment over another. Following are five examples of “weighting,” each of which is inappropriate.

1. A single source (e.g., teacher nomination) or a cut-off score on a single test (e.g., 85th percentile on an achievement test) is used to nominate a student for the gifted program. This approach weights the instrument or the test in the overall identification process. Remember that untrained teachers may not refer economically disadvantaged children or may tend to nominate students who are like themselves (Peterson & Margolin, 1997).
2. Quantitative measures such as norm-referenced intelligence or achievement tests may be assigned more influence in the selection of students than qualitative measures such as product scores, parent nominations, or performance in informal lessons, because the former are judged to be more accurate or reliable than the latter.
3. Certain tests are assigned more points. Scores on an intelligence test might earn more points than a teacher nomination or teacher nominations might earn more points than parent nominations.
4. A single source (e.g., a teacher) provides the majority of the qualitative information such as grades, checklists, and product scores. One teacher’s ratings might be weighted three times as much as any other source of information.
5. Several subtests and the composite score that assess the same trait (e.g., achievement) are used from the same measure and counted each time. This means that a single measure receives a multiplied weight.

In summary, no one assessment or source of information should carry more weight than another. In selecting and designing a form, the committee will want to ensure that all assessments receive equal weight.

Guideline 2: Comparable scores. The committee may receive scores in various forms. These scores might include raw scores, percentiles, stanine scores, and standard scores. To interpret the different test scores, the committee needs to know (a) how they compare to one another and (b) what reference population or norm group is represented.

Raw scores that represent the total number of points a student earns on a checklist, a test, or a rating form are not interpretable until they have been converted to a standardized scoring system. Standard scores have an advantage over other types of scores because the measurement units are equal and can be averaged or manipulated (Feldhusen, Baska, & Womble, 1981).

Using the raw scores, the committee can determine standard scores by following the directions in Appendix B, Converting Raw Scores to Standard Scores (found at the end of this article). Once the raw scores have been converted to standard scores, comparisons may be made with other scores by using a conversion chart.

Test manuals and publishers provide conversion charts that compare various test scores with one another and to a normal distribution (e.g., the bell-shaped curve). For example, you will note in Table 5.2 that a performance at two standard deviations (SD) above the mean (M) represents a score that may be interpreted in these ways:

• The score is at the 98th percentile.
• The score is at the ninth stanine.
• The standard score is 130 (with a mean of 100 and standard deviation of 15).
• The student performed better than 98% of the students who took the test.
• The student is performing in the superior to very superior ranges.

View Detail

A score in the superior range or at the 95th percentile is comparable to a score at the eighth stanine and to a standard score of 124. (Note that age- and grade-equivalent scores are not represented since they are difficult to interpret and should not be used when comparing scores.)

The committee also needs to know what reference population or norm group is represented by the score. For example, when raw scores were converted to percentiles, were the students at all ability levels represented or only those nominated? If only the nomination pool scores were converted, the students would be compared with only a small percentage of the local population. On the other hand, if all students at a particular grade level were included, students would be compared with their entire local population. A student’s score will be higher when compared to all students his or her age and will be lower when compared only with students nominated for the gifted and talented program.

Local norms are also different from national norms. Unless the school system’s population is representative of the nation as a whole, local norms are likely to be different in some ways from the population on which a test was standardized. For example, more students may be from minority backgrounds or from middle or upper income levels than the national average. Therefore, it is important to consider the reference group when interpreting scores (Mills, Ablard, & Brody, 1993). As noted above, students whose scores are compared with a local nomination group may not appear to perform as well as when their scores are compared to an entire local population. In a similar fashion, students compared with a national gifted population may not appear to perform as well as when their scores are compared to an entire national population. Similarly, young kindergarten children who have summer birthdays may appear to do less well than older children from the same grade level. In summarizing data, the committee will want to ensure the scores are comparable and the reference groups are clearly understood.

Guideline 3: Error in measures. Every measure contains a certain amount of error. This error is estimated through the standard error of measurement. Depending on the measure’s reliability and standard deviation, the size of this error will vary across grade or age levels, across subtests, and between different tests. No single test score number should be construed as “the one true score.” A student’s true score will lie somewhere within a range of scores established by the standard error of measurement.

For example, suppose that David scores 120 on an intelligence test and the standard error of measurement (SEM) is 5 points. The interpreter of this score might say that 68% of the time David will score between 115 and 125 (plus 5 and minus 5 = one standard error of measurement); that 95% of the time David will score between 110 and 130 (plus 10 and minus 10 = two standard errors of measurement); and that 99% of the time David will score between 107 and 133 (plus 13 and minus 13 = 2.6 standard errors of measurement). (See Table 5.3 for additional standard errors of measurement.) Test manuals should report the standard error of measurement for each age or grade level or both. While more qualitative measures may not have a calculated standard error, the committee always should consider that some error is inherent in all methods and procedures that are used (see Appendix C for Calculating the Standard Error of Measurement).

Guideline 4. Best performance reported. Estimates of student potential come from their best performance. Attempts to compress all performance data into a single number to use as a cut-off score for entry into the gifted program can be misleading when student performance shows considerable variability. Student scores may actually range from the very superior level to the average level within one measure or across measures; a compressed single score is less likely to reveal these ranges. Committee members need to see the peaks and valleys in student performance. The committee should consider the student’s highest performance as indicative of his or her potential. The highest score is most often the truest (Tolan, 1992a, 1992b).

Guideline 5. Description of the student. While numbers are helpful in comparing certain kinds of data, not all information about the student can be described numerically. Therefore, space should be provided for anecdotal information or clinical observations (e.g., how he or she acquires new information and uses reasoning strategies). This qualitative information may be especially useful when attempting to match instructional strategies to student characteristics.

In summary, these five guidelines can be used in designing or selecting a form or process to organize multiple kinds of data. To evaluate a form or process under consideration, district staff could ask the following questions based on these guidelines:

1. Do all assessments receive equal weight or value?
2. Are the scores comparable?
3. Are errors in measures considered?
4. Does the form or process provide the opportunity for the identification committee to examine each student’s best performance?
5. Does the form or process allow the committee to consider anecdotal and other descriptive information?

Many possible forms and procedures meet the above guidelines. Each must (a) be based on best research and recommendations, (b) relate to the school’s definition and program, (c) use qualitative and quantitative assessments, (d) use multiple sources, and (e) be unbiased. Feldhusen and Baska (1985) cautioned against using forms that combine assessment data, particularly matrices. Borland (1989) suggested that matrices do more harm than good by adding disparate subscale scores from a variety of qualitative and quantitative instruments. The next section of this chapter therefore provides a few forms and procedures that a committee might use in organizing data to identify gifted and talented students.

Borland and Wright (1994) suggested that a case study approach is the best way to identify children from lower socioeconomic backgrounds. A case study provides more depth, shows growth of performance over time, and incorporates evidence from a variety of sources and settings. Clark (1997) suggested that a case study might include nomination forms, teacher reports of student functioning, family history and student background, peer identification, student inventory of interests, student work and achievements, student and parent interviews, and a variety of test protocols (intelligence, achievement, and creativity).

The cover page from a folder of evidence is included in Figure 5.2. At the top of the page is student demographic information. Note that the date of birth is included and is particularly important in the primary grades. Quantitative information is separated from qualitative information. In this case, the school district has set minimum stanine scores of two 8s and one 9 for the quantitative, but not for the qualitative information. The committee reviews each of these qualitative assessments using characteristics of gifted and talented students, product or performance rubrics, or both. Each piece of evidence is scored as meeting or not meeting the criteria. These criteria may be established for individual assessments or for the case study as a whole.

View Detail

In the example, Edward Ochoa scored 135 on the Reasoning subtest of the Screening Assessment for Gifted Elementary Students or the ninth stanine (see Figure 5.2). He also scored at the 95th percentile, the eighth stanine, on the math subtest of the Iowa Test of Basic Skills and is making grades in the 90s in math, science, and social studies. While the teacher’s appraisal was negative, he had a strong portfolio, which showed his mathematical reasoning and creative writing ability. He was rated outstanding in “depth of knowledge expressed,” “ability to see relationships and connections,” and “age and developmental appropriateness of product.” His peers and parent interviews also showed evidence of many characteristics of gifted and talented students. The counselor added, “Some have described Edward’s creative writing assignments as truly creative; his teacher sees them as strange. His peers see him as original, anxious to try new things, and impatient. At home, he likes to watch the Discovery Channel.” Overall, the committee agreed that Edward Ochoa had sufficient evidence for his placement in the school district’s program. Now, review the case study form using the criteria suggested earlier in this chapter:

1. Are the data from some measures more important than others? No, the committee considered quantitative and qualitative information as equally important. Information was acquired from quantitative (e.g., achievement and aptitude testing) and qualitative (e.g., observations, student performance, surveys, and interviews) sources.
2. Are the scores comparable? Broad bands of performance (e.g., stanines) were used to compare the objective indicators. Quantitative scores were not combined with one another or with the subjective indicators. After establishing interrater reliability, the trained committee discussed and rated each of the assessments for each of the nominated students.
3. Is the error in tests considered? Yes. Broad bands of performance were used instead of single test scores.
4. Does the form provide an opportunity for the identification committee to examine the student’s best performance? Definitely. Scores were listed separately for each quantitative instrument. The committee also reviewed each piece of qualitative evidence in Edward’s folder, including checklists, products, and anecdotal summaries.
5. Does the form provide a space for additional comments or anecdotal information? Definitely. A variety of individuals have described Edward’s performance—teachers, parents, peers, and himself.

The profile form in Figure 5.3 represents a way to display student data. In the upper left corner, student demographic characteristics (e.g., name, age, date of birth, gender, and school) are included. On the left side of the form, the district lists the measures used in the nomination and screening phases. These assessments match both local student and program characteristics within one district. The selection of these assessments most likely will vary depending on the district’s student population and program characteristics.

View Detail

In the upper right corner, the district lists scores that are comparable to one another. For example, line C relates all scores to a normal curve, line B to percentile scores, and line A to ranges of performance. By reading the scores from bottom to top, one can see that a mean score is comparable to the 50th percentile and comparable to a score in the average range. In like manner, a score at or above plus two standard deviations (+2 SD) is comparable to the 98th percentile and scores in the superior to very superior ranges. Additional lines of comparable scores may be added by your district.

In the example, to the right of the measures is the profile area where individual student data are recorded. Since the district wants to identify the top 5% of its population, a district line has been drawn at the 95th percentile. The district’s students who have at least three strengths are selected for its gifted program. This is indicated by scores to the right of the district line. Each district should decide where to place its line. Districts that raise the line above the 95th percentile may encounter problems since more test error is found above this range.

On the completed profile form, you will note in number 1, Product/Performance, out of a possible 8 points, Sarah’s six products from her academic portfolio earned an average of 6 points. Since only 8% of the local population achieved within this range, this score placed her in the superior range, with a standard score of 121 (M = 100, SD = 15; see Table 5.2). A standard error of measurement (1 SEM) of 4 points was calculated (see Appendix C). To achieve the 68% confidence level, 4 points (1 SEM) were added and subtracted to Sarah’s score (refer to Table 5.3). Sarah’s score now fell within the 117–125 range or within the above-average to superior ranges.

Sarah received a raw score of 20 points on the Renzulli teacher nomination checklist. When this raw score was converted to a standard score, Sarah’s score was .7 above the mean, 110, at the 75th percentile or in the average range. Again, a standard error of measurement (one SEM) of 5 points was calculated (see Appendix C). At the 68% confidence level, Sarah’s score fell within the 105–115 range or within the average to above-average range.

Sarah received a raw score of 30 points on a locally prepared parent nomination checklist. When this raw score was converted to a standard score, Sarah’s score was 1.8 above the mean, 127, at the 96th percentile or in the superior range. A standard error of measurement (one SEM) of 4 points was calculated. At the 68% confidence level, Sarah’s score fell within the 123–131 range or within the superior to very superior range.

She performed better than 90% of the students on the total battery of the California Achievement Test (CAT) with a 119 standard score. At the 68% confidence level, one SEM was added and subtracted to the standard score. Sarah’s score fell within the above-average to superior ranges, 114–124, or from the 83rd to the 95th percentile.

Finally, on the Wechsler Intelligence Scale for Children—Third Edition, Sarah obtained a full-scale intelligence quotient of 130 (i.e., mean of 100 with a standard deviation of 15). Given a 3-point standard error of measurement, 3 points were added and subtracted to Sarah’s score of 130 to achieve an 8% confidence level. Sarah’s score fell within the superior to very superior ranges or from 127 to 133.

Because Sarah obtained four scores at or to the right of the district line drawn at the 95th percentile, the committee recommended that she be included in the program. Now, review the completed profile form using the criteria suggested earlier in this chapter:

1. Are the data from some measures more important than others? No, all measures were considered to be equally important. Information was acquired from quantitative (e.g., the CAT and the WISC III) and qualitative (e.g., teacher and parent checklists and student product) sources. Student products were judged by the identification committee instead of the teacher to add an additional source of information and to avoid assigning a double weight to the teacher’s perceptions.
2. Are the scores comparable? Yes the percentiles are comparable to bands of performance that were established within the normal curve distribution (e.g., means and standard deviations). All district raw scores were converted to standard scores and then to percentiles, or to broad bands of performance. These scores were represented by “L” to indicate that local norms were used.
3. Is the error in tests considered? Yes. In an attempt to achieve the 68% confidence level, one standard error of measurement was added and subtracted to each student’s score and all scores were reported within broad bands of performance such as average, above-average, superior, and very superior.
4. Does the form provide an opportunity for the identification committee to examine Sarah’s best performance? Yes. Scores were not summed to obtain a cut-off score for entry into the program. The committee was able to examine Sarah’s strengths and weaknesses. In this example, Sarah’s relative strength was in her performance on the intelligence measure.
5. Does the form provide a space for additional comments or anecdotal information? Yes. The bottom of the form provided some space for these comments. Other anecdotal information could be attached to the profile form.

Another form for organizing data is the minimum scores approach. A minimum scores form with sample data is shown in Figure 5.4. At the top of the form, student demographic characteristics are included (e.g., name, age, gender, and school). On the left side of the form, the district again lists the measures used in the nomination and screening phases.

View Detail

To the right of the measure’s name is a column in which the SEM is written. The next column contains the minimum score. This minimum score corresponds to the district line on the profile. The major difference between the profile form and the minimum score form is that the standard error of measurement has been subtracted from the district-determined standard scores (i.e., the minimum score). On the profile form, standard errors of measurement are added and subtracted to each individual student’s score and the range of scores are then plotted on the graph. In the minimum scores approach, this calculation is figured for each individual measure before the student’s standard scores are recorded. In this way, the student is given an advantage of one or more standard errors of measurement. All measures must have the SEM subtracted.

For example, a district decides to select students for its gifted program who perform at or above the 95th percentile on three of five different measures. The 95th percentile corresponds to a standard score of 124. If a district wants to achieve a 68% confidence level, the committee will subtract approximately one standard error of measurement from the district determined minimum score (e.g., 124). If a measure has a 3-point standard error of measurement and the district wanted a 68% confidence level, the identification committee would subtract 3 points (i.e., one SEM) from 124 (i.e., 95th percentile or district line) and set the minimum entry score at 121, rather than 124.

The person who tallies the actual scores compares them to the minimum entry score. Students with scores at or beyond this minimum entry score receive a plus (+). If not at or beyond the minimum entry score, students receive a minus (-). Students who receive three plus marks are considered eligible for placement into the gifted program.

To use the minimum scores approach, consider Sarah’s scores again (see sample completed minimum scores approach). On the WISC III, 3 points (i.e., one SEM) were subtracted from 124 (i.e., the 95th percentile), making the minimum score 121. Since Sarah obtained a full-scale intelligence quotient of 130, she received another plus (+) in this category.

For the California Achievement Test, the district used the 95th percentile as a cut-off score or 124 standard score (M = 100, SD = 15). Considering the 68% confidence level and one SEM of 5 points, the minimum score is 119 (i.e., 124 - 5 = 119). Sarah scored exactly at this level and received a plus (+) in this category.

For the Renzulli Motivational Scale, the district converted the raw scores to standard scores and found that a raw score of 27 places a student in the superior range or at a standard score of 124. A standard error of measurement of 5 points was calculated. The 5 points were subtracted from 124 to achieve a 68% confidence level, and 119 was entered in the minimum entry score column. Since Sarah received 20 points, this translated into a standard score of 110. Thus, she did not meet the minimum entry score and received a minus (-) in this category.

For the locally developed parent nomination checklist, the district converted the raw scores to standard scores and found that a raw score of 27 places a student in the superior range or at a standard score of 124. A standard error of measurement of 4 points was calculated. The 4 points were subtracted from 124 to achieve a 68% confidence level, and 120 was entered in the minimum entry score column. Since Sarah received 30 points, this translated into a standard score of 127. Thus, she met the minimum entry score and received a plus (+) in this category.

Finally, the minimum entry score for products is 120. The score was derived by subtracting 4 points, 1 SEM from 124, the 95th percentile or the district cut-off line. Since Sarah’s score was 121, she would receive a plus (+) in this category.

In examining this minimum scores form, it is easy to see that Sarah received four plus marks and would be recommended for placement in the program for the gifted. Again, all guidelines were met: no weighting of measures occurred, the scores were comparable, error was calculated before the score was placed in the minimum entry score column, best performance could be noted, and space was provided for additional comments.

In summary, this chapter has recommended some forms that the district might use in the nomination screening and selection phases. A school will want to collect evaluation data on those who perform successfully in the program, those who don’t, and those who perform somewhere in-between. A district will then want to examine relationships between categories of youngsters and the measures it uses in selection processes. An ongoing evaluation process will be well-worth a district’s efforts in finding those students who truly need and can benefit from a differentiated curriculum to reach their full potential.

Alvino, J., McDonnel, R., & Richert, S. (1981). National survey of identification practices in gifted and talented education. Exceptional Children, 48, 124–132.

Borland, J. H. (1989). Planning and implementing programs for the gifted. New York: Teachers College Press.

Borland, J. H., & Wright, L. (1994). Identifying young, potentially gifted, economically disadvantaged students. Gifted Child Quarterly, 38, 164–171.

Clark, B. (1997). Growing up gifted (5th ed.). Upper Saddle River, NJ: Merrill.

Coleman, L. J., & Cross, T. L. (2001). Being gifted in school: An introduction to development, guidance, and teaching. Waco, TX: Prufrock Press.

Coleman, M. R., Gallagher, J. J., & Foster, A. (1994). Updated report on state policies related to the identification of gifted students. Chapel Hill: University of North Carolina.

Cornish, R. L. (1968). Parents’, pupils’, and teachers’ perceptions of a gifted child’s ability. Gifted Child Quarterly, 12, 14–17.

Feldhusen, J. F., Asher, J. W., & Hoover, S. M. (1984). Problems in the identification of giftedness, talent, or ability. Gifted Child Quarterly, 28, 149–151.

Feldhusen, J. F., & Baska, L. K. (1985). Identification and assessment of the gifted and talented. In J. F. Feldhusen (Ed.), Excellence in educating the gifted (pp. 87–88). Denver: Love.

Feldhusen, J. F., Baska, L. K., & Womble, S. R. (1981). Using standard scores to synthesize data in identifying the gifted. Journal for the Education of the Gifted, 4, 177–185.

Feldhusen, J. F., Hoover, S. M., & Sayler, M. (1990). Identifying and educating gifted students at the secondary level. Monroe, NY: Trillium Press.

Gear, G. (1978). Effects of training on teachers’ accuracy in identifying gifted children. Gifted Child Quarterly, 22, 90–97.

Geary, D. C., & Brown, S. C. (1991). Cognitive addition: Strategy choice and speed-of-processing differences in gifted, normal, and mathematically disabled children. Developmental Psychology, 27, 398–406.

Jacobs, J. (1971). Effectiveness of teacher and parent identification as a function of school level. Psychology in the Schools, 8, 140–142.

Jensen, A. (1969). How much can we boost IQ and scholastic achievement? Harvard Educational Review, 39(1), 1–24.

Johnsen, S. K. (1997). Assessment beyond definitions. Peabody Journal of Education, 72, 136–142.

Karnes, F. A., & Marquardt, R. G. (1991). Gifted children and the law. Dayton: Ohio Psychology Press.

Kurtz, B. E., & Weinert, F. E. (1989). Metacognition, memory performance, and causal attributions in gifted and average children. Journal of Experimental Child Psychology, 48, 45–61.

Mills, C., Ablard, K. E., & Brody, L. E. (1993). The Raven’s Progressive Matrices: Its usefulness for identifying gifted/talented students. Roeper Review, 15, 185–186.

Pegnato, C., & Birch, J. (1959). Locating gifted children in junior high schools: A comparison of methods. Exceptional Children, 25, 300–304.

Peterson, J. S., & Margolin, R. (1997). Naming gifted children: An example of unintended “production.” Journal for the Education of the Gifted, 21, 82–101.

Richert, S. (1985). Identification of gifted children in the United States: The need for pluralistic assessment. Roeper Review, 8, 68–72.

Richert, S. (1991). Rampant problems and promising practices in identification. In N. Colangelo & G. A. Davis (Eds.), Handbook of gifted education (pp. 81–96). Needham Heights, MA: Allyn and Bacon.

Salvia, J., & Ysseldyke, J. E. (2001). Assessment (8th ed.). Boston: Houghton-Mifflin.

Scruggs, T., & Mastropieri, M. (1985). Spontaneous verbal elaborations in gifted and nongifted youths. Journal for the Education of the Gifted, 9, 1–10.

Texas Education Agency, Division of Advanced Academic Services (1996). Texas state plan for the education of gifted/talented students. Austin, TX: Author.

Tolan, S. S. (1992a). Special problems of highly gifted children. Understanding Our Gifted, 4(3), 3, 5.

Tolan, S. S. (1992b). Parents vs. theorists: Dealing with the exceptionally gifted. Roeper Review, 15, 14–18.

Trice, B., & Shannon, B. (2002, April). Office for Civil Rights: Ensuring equal access to gifted education. Paper presented at the annual meeting of the Council for Exceptional Children, New York.

View Detail

View Detail

View Detail