To qualify for the talent search, students must score at the 95th percentile or above on a nationally normed, standardized achievement test. (Sixth graders must score in the 97th percentile or higher.) However, CTD is aware that an increasing number of students are being home-schooled and that some school districts do not administer achievement tests on a regular basis. Students for whom qualifying scores are not available, but who are motivated and achieving at high levels, may apply for participation in MATS by means of a parent nomination form. In order to evaluate our procedures and practices, we compared the resulting SAT and ACT scores for students who qualified to participate in MATS by means of scores on school-given, standardized tests to students who qualified by means of a parent nomination form.
Subjects for the study consisted of students who participated in the Northwestern University's Midwest Academic Talent Search program sponsored by CTD from 1997-2001. The vast majority of the 124,401 MATS participants during this five-year period were 7th (40%) and 8th (50%) graders. Approximately 9% of the students were sixth graders at the time of participation in the talent search, and 1% were 9th graders. There was a slightly greater percentage of male participants (52%) than female (48%). And, the ethnic background of our sample population was primarily Caucasian/White (79 %). Approximately 5% of the participants reported their ethnicity as Asian or Pacific Islander, 3% as African-American, 1% as Multi-racial, and 1% as Hispanic.

Over the past five years 7,648 students qualified for participation in CTD’s talent search by means of a parent nomination. The number of parent-nominated students increased significantly from 1997 to 1999, but decreased in both 2000 and 2001. Overall, parent-nominated students accounted for just over 6% of talent search participants.

A comparison of the demographic characteristics of the parent-nominated students and test score qualifiers indicated a number of significant differences. There was a slightly greater ratio of females to males among parent-nominated students, as well as a slightly larger percentage of 6th graders, when compared to standard test-score qualifiers. And although only 397 students of all talent search participants during this five-year period reported that they were being home-schooled, the percentage of parent-nominated students who indicated that they were being home-schooled was slightly greater than that of students who qualified via test scores.Other significant differences in the background characteristics of parent-nominated talent search participants and students who qualified by means of test scores included geographic location and reported household income. While Michiganers accounted for just over a quarter (29%) of participants in the talent search as a whole, more than half (56%) of the parent-nominated students came from Michigan. A considerable percentage of parent nominated students also came from Wisconsin (19%) and Ohio (15%), but these percentages were comparable to the percentages of students, overall, coming from these states. The large number of Michiganers who avail themselves of the parent nomination process may be because Michigan uses a criterion-based performance assessment, the MEAP, which is not nationally normed, as a state assessment. MEAP scores are not used for entrance into the talent search and thus many Michigan schools use parent nomination to select students to participate in MATS.

With respect to household income, a slightly greater percentage of students who entered the talent search via test scores (21%) reported family incomes greater than or equal to $120,000, compared to parent-nominated students (14%). This may be due to the greater percentage of homeschooling families, who are likely one-income families, among parent-nominated students.

Finally, the majority of participants who qualified by means of standardized test scores opted to take the SAT (56%), compared to 37% who took the ACT and 7% who ended up not taking either test. By contrast the majority of parent nominated participants opted to take the ACT (51%); only 40% took the SAT and 8% took neither.

The ACT and SAT test scores of parent-nominated talent search participants and participants who qualified for the talent search via standardized test scores were compared using appropriate statistical procedures and tests. Standard test-score qualifiers tended to perform better in all subject areas on both the SAT and the ACT. On average standard test-score qualifiers scored 23 points better on the verbal portion of the SAT than parent-nominated students. (SAT scores range from 200 to 800.) The average verbal score for standard test-score qualifiers was 463, while the average verbal score for parent- nominated students was 440. This difference was considered statistically significant but is small in size. Similarly, on the math portion of the SAT, standard test-score qualifiers scored on average 19 points better than parent-nominated students. Again, the difference, while statistically significant, is small.

On the ACT (whose scores range from 2 to 36), standard test-score qualifiers performed on average 1.5 points higher on the reading and English subtests, eight-tenths of a point higher on the mathematics subtest, nine-tenths of a point higher on the science reasoning subtest, and their composite scores were on average 1.2 points higher. Though these differences were statistically significant, they are all small in size.

The differences in the average SAT and ACT scores of standard test-score qualifiers and parent-nominated students were analyzed independently for each year of the talent search. For the talent searches conducted in 2001, 2000 and 1999, the same pattern of results as described above were found. This was not true for 1997 and 1998. In these two years, average ACT reading and SAT verbal scores differences between these groups were also small but statistically significant. However, average ACT and SAT math score differences in 1998 and 1997 were not statistically significant.

Of particular interest to Talent Search program coordinators is the percentage of participants who receive scores that qualify them to participate in enrichment programs and special courses of study. On the SAT verbal and math tests, the proportion of parent-nominated students receiving scores within each of the top three SAT score categories (700-800, 600-690, 500-590) was somewhat lower than that of standard test-score qualifiers. Approximately 34% of standard test-score qualifiers received a verbal SAT score greater than or equal to 500, while 26% of parent-nominated students performed at this level. And on the math portion, 43% of standard test-score qualifiers received a score greater than or equal to 500, while 35% of parent-nominated students received scores in this range. The percentage of parent-nominated students scoring within each of the top three ACT score categories (30-36, 25-29, 20-24) was also slightly less than that of standard qualifiers. On the reading subtest, 55% of standard test-score qualifiers, compared to 42% of parent-nominated students, received a score in these ranges. On the English subtest, 50% of standard test-score qualifiers, compared to 37% of parent-nominated students, received scores in this range. Approximately 31% of standard test-score qualifiers, compared to 22% of parent-nominated students, performed at this level on the math subtest; and 55% of standard test-score qualifiers, compared to 45% of parent-nominated students, performed at this level on the science reasoning test. Finally, 51% of standard qualifiers received ACT composite scores greater than or equal to 25, while 38% of parent-nominated students received scores in this range. These findings held up when the data was analyzed separately for each year, with one exception. In 1997, the differences between the number of parent-nominated students and standard test-score qualifiers receiving scores in each category were not found to be statistically significant for the ACT math and the SAT math and science reasoning tests.

Our analysis revealed that students who enter the talent search on the basis of a parent nomination do, on average, earn scores on the SAT and ACT that are slightly lower than those of students who enter the talent search on the basis of standardized, in-grade achievement test scores. Parent-nominated students are therefore slightly less likely to qualify for accelerated programs. While the differences are statistically significant, they are small enough to not have much practical significance.

However, there are great benefits to including parent nomination as an additional or alternative means to identify students for the talent search. Due to parent nomination many more students, who would otherwise not have the opportunity to participate in the talent search, take off-level tests which better measure their abilities and obtain access to information and programs to develop their talents. The results of this study suggest that parent nomination is a viable alternative to use for identifying students to participate in talent searches when scores on standardized, in-grade achievement tests are not available.