Sedlacek, W. E. (1993). Employing noncognitive variables in admissions and retention in higher education. In Achieving diversity: Issues in the recruitment and retention of underrepresented racial/ethnic students in higher education. (Pp. 33-39). Alexandria VA: National Association of College Admission Counselors.

Employing Noncognitive Variables in the
Admission and Retention of Nontraditional Students

The evidence is in. Studies have demonstrated it; articles have documented it; and practical applications have confirmed it. If you want to successfully admit and retain nontraditional students you should employ noncongitive variables in your admission process.

This is a strong claim to make about any assessment procedure. Let's examine it in detail and see if it holds up. First, what is a nontraditional student?

NONTRADITIONALITY

Two of the biggest problems we have in admissions are deciding what to call "culturally-different" applicants, and determining how to identify them. Frequently, terminology can be a trap. We tend to feel we have solved a problem if we label it. As Westbrook and Sedlacek, (1991) have shown, in the last 40 years we have moved through terms such as "culturally deprived" in the 1950's through "disadvantaged" in the 1960's to "multicultural" and "diversity" in recent years to describe largely the same people; people who have had different experiences than white middle/upper middle class, mostly male, people in U.S. society tend to have prior to college application.

We're never sure what to call these people, but since language has an impact on feelings, I prefer to avoid negative terms such as "deprived" or "minority" and define an operational concept that is more neutral and can help in the admission process: the nontraditional applicant/student.

Traditionally, white middle/upper middle class males formed the bulk of the college and professional school applicant population. Admissions programs have decades of experience evaluating this type of applicant, but the further an applicant's background is from this paradigm, the less experience we have evaluating his/her potential for success in college. Operationally, then, nontraditional applicants are those for whom we have less experience interpreting the knowledge and skills in their backgrounds and relating that information to college success.

When we have less experience at a task, we tend to do it less expertly. Fortunately, tools have been developed which help us predict the potential college success of nontraditional applicants: noncognitive variables.

NONCOGNITIVE VARIABLES

Numerous studies provide strong evidence that noncognitive measures predict the success of nontraditional students better than traditional measures (standardized tests and previous grades). This is true if success is measured by college grades at the baccalaureate level (Sedlacek 1977, 1987, 1989, 1991; Tracey and Sedlacek 1984, 1985, 1987, 1988, 1989; White and Sedlacek, 1986) or in professional school (Bandalos and Sedlacek, 1989; Brown and Mareno, 1980; Sedlacek and Prieto, 1990). Or if success is measured as retention or graduation, noncognitive variables have more validity than other measures for both traditional and nontraditional students (Tracey and Sedlacek, 1987; 1989). But why should this be so?

Sternberg (1985, 1986) gives us a clue to why a relationship might exist between traditional experiences and noncognitive variables. He proposes that there are three basic ways a person may show ability. Componential intelligence is the ability to interpret information hierarchically and taxonomically in a relatively unchanging context. This is an ability associated with a traditional experience in our society. If your situation was not relatively stable, or if your context was other than traditional, you are likely to have more trouble showing your ability in this way. Traditional admission measures (standardized tests and prior grades) rely heavily on this type of intelligence.

Experiential intelligence involves the ability to interpret information in changing contexts; to be creative. A person with nontraditional experiences may have had to develop and demonstrate this kind of intelligence to be successful.

The third type of intelligence Sternberg calls contextual. This type has to do with the ability to understand and "work" the system to your advantage. For a person with a nontraditional background, it is critical to know how to interpret the system in terms that foster his/her development.

There is evidence that many different groups have nontraditional background experiences and can be more accurately assessed using noncognitive variables. These include, but are not limited to, Blacks (Sedlacek, 1987; Tracey and Sedlacek; 1984, 1985, 1987, 1988, 1989) Hispanics, (Arbona and Novy, 1990; Fuertes, Sedlacek and Westbrook in press; Patterson, Sedlacek and Perry, 1984), Asian-Americans (Wang, Sedlacek, and Westbrook (1992), women (Hirt, Hoffman and Sedlacek, 1983, Minatoya and Sedlacek, 1983), Native Americans (Trimble, 1988), international students (Boyer and Sedlacek (1988), athletes (Engstrom & Sedlacek, 1991, Sedlacek and Adams-Gaston, in press), students with physical disabilities (Patterson, Sedlacek, & Scales, 1988, Stovall and Sedlacek, 1985), commuting students (Foster, Sedlacek, and Hardwick, 1977, Wilkshire, 1989), older students (Peabody and Sedlacek, 1982; Schwalb and Sedlacek, 1990) and gays, lesbian, and bisexuals (D'Angelli, 1989)

The degree and form of nontraditional experiences vary considerably across groups, but with these and other groups, assessment using noncognitive variables can greatly facilitate admission decisions.

So what are these noncognitive variables, and how do we go about assessing students on them? The research discussed above is based on eight noncognitive variables defined as follows:

A short assessment instrument, the noncognitive questionnaire (NCQ), has been demonstrated to have validity in assessing the dimensions of these eight variables, and in predicting the grades and retention of nontraditional students, and the retention of traditional students, Sedlacek (1977, 1987, 1989, 1991), Sedlacek and Prieto (1990), Tracey and Sedlacek 1984, 1985, 1987, 1989).

It is also possible to use interviews in assessing the noncognitive variables (Prieto, Quinones, Elliott, Godldner, and Sedlacek, 1986), Westbrook and Sedlacek, 1988, Sedlacek, 1991). A discussion of the characteristics of high- and low-scoring students on each of the eight variables is presented in Exhibit 1.

Now we know that noncognitive variables exist; how to define them; who they work best for; and how students can be assessed with them. What we still need to know is how they can be fit into an existing assessment program. The next section discusses exactly that. Using the NCQ or the interviewing system, the evaluations based on the eight variables discussed above can be incorporated into any admission system. The examples discussed below employ both traditional and noncognitive variable assessment; and they demonstrate how noncognitive variables can be used in recruiting, admission and retention programs at different types of schools.

ADMISSION EXAMPLES EMPLOYING NONCOGNITIVE VARIABLES

There are three primary groups in admissions for whom the use of noncognitive variables is recommended. The first is the group of applicants that is in the "gray area" or "on the bubble". They have traditional admission assessments that are low but close to the usual minimum acceptance level at your institution. However, they are nontraditional applicants and you would like to admit some. Do you "lower your standards" on traditional measures to admit them? No; you use noncognitive variables to assess which have the best chance of succeeding at your institution without any special programs or services.

The second group consists of those nontraditional applicants with traditional scores considerably below your typical accepted applicants. You have some special programs for students, but how do you select participants? It should be no surprise that I suggest noncognitive variables as the answer.

The third group is often overlooked; nontraditional students with high scores on traditional measures. Here we can use nontraditional measures as the basis of a retention program coordinated with admission.

An example illustrating an admission procedure focused on each group follows.

A large state university had about 6,000 applications for 3,000 places in the first year class. The primary goal in the admission process was to select fairly but at the same time to maximize the number of nontraditional students, particularly African-Americans. The mean for all their applicants was 22 on the ACT composite. The mean ACTs for their 520 African-American general applicants was just about five points lower. If they went strictly by ACTs they would accept only about 150 African-American applicants and might get only half of those to actually enroll. These numbers were unacceptably low, but they did not want to lower the ACT scores required for admission, and did not want to require the students to go into special programs.

This institution decided to consider both traditional measures (ACTs and high school grades) and noncognitive measures (Noncognitive Questionnaire; NCQ)¹. They required the NCQ to be completed by all general applicants. They then developed two multiple regression equations based on some pilot studies and normative information in the NCQ Manual (Abler, Sedlacek, Tracey, in press). One regression equation was for traditional applicants and the other was for nontraditional applicants. The approximate weights given were about two-thirds weight to traditional measures and one-third to noncognitive variables in the traditional applicant equation. For nontraditional applicants, the NCQ was weighted about 60% compared to 40% for ACTs and grades.

Admission office personnel were trained to make a determination of traditionality of experience of each applicant to determine which equation to use. Any difficult decisions were assigned to a committee consisting of admissions staff, faculty, students and campus administrators. The assumption was that if the institution wanted nontraditional students they would have to consider this in their admission process and insure that they were as valid in their assessments of those nontraditional applicants as possible.

The institution did not change any other aspects of its admission, retention, or recruiting programs, or any of its special programs or curricular offerings. By employing the new procedures they admitted 310 of their African-American applicants and 208 actually enrolled. Their usual graduation rate after 6 years was about 60% for traditional students and about 30% for African-American students. After employing the new admission procedures the six year graduation rate for traditional students was 65% and 44% for African-Americans. While they did not employ a control group for comparison, officials at the institution were pleased with the new system and have begun to think of ways to provide recruiters and advisors with more admission information to perform those functions better.

Case 2 - Special Programs

A medium-sized private college had several special programs into which students who could not meet the fairly high admission standards were admitted. The means for their entering freshmen were 618 Verbal and 580 Math (SATs) with B+ high school grades.

Most of the funds for the programs were from public and private grants, but college administrators were not pleased with the high attrition rates (80% never graduated from the school) and isolation of the students in the program from the rest of the student body. The general student body, faculty, administrators and the students in the programs themselves felt the programs were for inferior students. Sixty percent of the students in the programs were Hispanic, 28% were Black and the rest were from a variety of other racial/cultural groups.

The admission office and the administrators and counselors in the special programs tried to work together smoothly with little success. Admission staff generally favored admitting students with the highest SATs and grades and staff in the programs tried to assess recommendations of teachers and an onsite interview, but were not able to quantify or articulate the impressions of the applicants in a way in which the admissions staff was comfortable.

The president of the school decided to implement a system based on noncognitive variables. Any applicant that did not meet the usual admission requirements could apply via a second procedure. They were sent the NCQ with a letter explaining that there many ways to do assessment and they (or their son or daughter) would be given every chance to demonstrate their qualifications for admission. Applicants were also requested to supply any further evidence of their accomplishments (artwork,writing, community work, etc.) that was not provided with their initial application.

Admission and program staff were trained in making assessments of noncognitive variables and a procedure which gave scores on each of the eight NCQ variables, as modified by their judgments of the additional materials supplied, was implemented. Evaluators began speaking in a common language focused on noncognitive concepts (e.g., long term goals) and more consensus on decisions was reached.

Additionally a program to educate the campus about the positive abilities and assessments of the students in the special programs was begun through workshops, written materials, videotapes, etc. The essence of the message was that their school does a thorough assessment of the abilities of all applicants and no one enters their school without the demonstrated positive attributes and experiences necessary to do well. They also discussed Sternberg's concepts of intelligence noted earlier as important for all to know. Their recruiters were also trained to emphasize these points in their work.

While it is too early to tell what will eventually occur at this institution, the special programs and their admission procedures seem to be moving in a positive direction. People are communicating better and feeling better about the programs and themselves. As one faculty member put it "I am a lot more comfortable knowing that there is a positive system based on some research behind this program. I should not be required to be a social worker, which is how I used to feel."

Is this really a problem? Any of us would feel fortunate to have that Native American student with high grades and test scores. Why ask for trouble? Because we too often lose those students due to problems with one or more noncognitive variables. Statistically, the reason we get lower correlations with standardized tests and college grades for nontraditional students is that not only do some students with low scores do better than we expect, but students with high scores do worse than expected.

A small selective institution with mean SATs in the mid 600's on each scale for their general student body was successful in attracting and enrolling Native American, Chicano, and African-American students with good grades and test scores, but they were not staying. While about 85% of the traditional students who matriculated eventually graduated, only 50% of the nontraditional students finished. Most left in good standing, somewhere in their first two years.

The school decided to implement a new advising system for all students, traditional and nontraditional, which relied heavily on noncognitive variables. All students were administered the NCQ at orientation, after admission, but prior to enrollment. A noncognitive profile was developed for each student and a specific prescriptive program was developed for each student. Faculty and staff were trained to be resource people specializing in one or more of the noncognitive variables. Thus, an individual or a student service might concentrate on self-concept development, another on understanding racism, and another group served as mentors (strong support persons). This proactive approach was embraced and seemed to be enjoyed by most on the campus. As one staff member put it "This kind of individualized attention is part of our tradition. We just needed to organize and focus it better."

Aside from the positive feelings generated, two years later retention was 90% for traditional students and up to 74% for nontraditional students. It may have been the increased energy and attention that caused the increases rather than noncognitive variables. However, noncognitive variables can provide a context for the activity and help provide a fresh look at what might be old issues.

CONCLUSIONS

Obviously, I would like you to try employing noncognitive variables in your admission or assessment programs. Whether you use the NCQ or not there is enough evidence of their utility to give them a try. I encourage you to apply Sternberg's experiential intelligence: be creative. No one method or application might work in all situations. Make an attempt, evaluate and share the results with colleagues. The more we can share experiences, the more likely we will all be able to do what we do better.

Positive Self-Concept or Confidence. High scorers feel confident of making it through graduation and make positive statements about themselves. They expect to do well in academic and nonacademic areas and assume they can handle new situations or challenges.

Low scorers express reasons why they might have to leave school and are not sure they have the ability to make it. They feel other students are more capable, and expect to get marginal grades. They feel they will have trouble balancing personal and academic life. They avoid new challenges or situations.

Realistic Self-Appraisal. High scorers appreciate and accept rewards as well as consequences of poor performance. They understand that reinforcement is imperfect and do not overreact to positive or negative feedback. They have developed a system of using feedback to alter behavior.

Low scorers are not sure how evaluations are done in school and overreact to the most recent reinforcement (positive and negative), rather than seeing it in a larger context. They do not know how they are doing in classes until grades are out. They do not have a good idea of how peers would rate their performance.

Understands and Deals with Racism. Racism is used here as a generic term covering all "isms". High scorers understand the role of the "system" in their life and how it treats nontraditional persons, often unintentionally. They have developed a method of assessing the cultural/racial/gender demands of the system and respond accordingly: assertively, if the gain is worth it, passively, if the gain is small or the situation ambiguous. They do not blame others for their problems or appear as a "Pollyanna" who does not see an "ism" that works, against them.

Low scorers are not sure how the "system" works and are preoccupied with racism or do not feel racism exists. They blame others for their problems and react with the same intensity to large and small issues concerned with racism. They do not have a successful method of handling racism that does not interfere with their personal and academic development.

Prefers Long-Range Goals to Short-Term or Immediate Needs.

High scorers can set goals and proceed for some time without reinforcement. They show patience and can see partial fulfillment of a longer-term goal. They are future- and past-oriented and do not see just immediate issues or problems. They show evidence of planning in academic and non-academic areas.

Low scorers show little ability to set and accomplish goals and are likely to proceed without clear direction. They rely on others to determine outcomes and live in the present. They do not have a plan for approaching a course, school in general, an activity, and so on. The goals they may have tend to be vague and unrealistic.

Availability of Strong Support Person. High scorers have identified and received help, support, and encouragement from one or more specific individuals. They do not rely solely on their own resources to solve problems. They are not "loners" and are willing to admit they need help when it is appropriate.

Low scorers show no-evidence of turning to others for help. They usually have no single support person, mentor, or close adviser. They do not talk about their problems and feel they can handle things on their own. Access to a previous support person may be reduced or eliminated as they enroll in new school, and they are not aware of the importance of a support person.

Successful Leadership Experience. High scorers have shown evidence of influencing others in academic or nonacademic areas. They are comfortable providing advice and direction to others and have served as mediators in disputes or disagreements among colleagues. They are comfortable taking action where it is called for.

Low scorers show no evidence that others turn to them for advice or direction. They are nonassertive and do not take the initiative. They are overly cautious and avoid controversy. They are not well-known by their peers.

Demonstrated Community Service. High scorers are identified with a group that is cultural, racial, gender-based, geographic etc. They have specific and long-term relationships in a community and have been active in community activities over a period of time. They have accomplished specific goals in a community setting.

Low scorers tend to have no involvement in a community as defined above. They have limited activities of any kind and are fringe members of any group to which they belong. They engage more in solitary rather than group activities (academic or nonacademic).

Knowledge Acquired in A field. High scorers know about a field or areas that they have not formally studied in school. They have a nontraditional, possibly culturally, racially or gender-based view of a field or profession. They have developed innovative ways to acquire information about a given subject or field.

Low scorers appear to know little about fields or areas that they have not studied in school. They show no evidence of learning from community or nonacademic activities and are traditional in their approach to learning. They have not received credit-by-examination for courses and may not be aware of credit-by-examination possibilities.

References