PSAT Component Scores as a Predictor of Success on AP Exam Performance for Diverse Students

Abstract

While studies have shown the positive effect of the Advanced Placement (AP) program on college readiness, there are still barriers preventing minority and low socioeconomic status (SES) students who possess high academic potential from participating in the opportunity that AP courses offer. One tool that could help identify students for participation in AP courses is their Preliminary Scholastic Aptitude Test (PSAT) results. Results from multiple regression analyses suggest that portions of the PSAT are significant predictors of AP test performance above and beyond student ethnicity and/or SES.

Keywords

college readiness AP exam PSAT minority socioeconomic status

Introduction

In an effort to increase college readiness and provide equitable opportunities for all students, the College Board sets a high goal. It seeks, by 2025, to increase the percentage of 25- to 34-year-olds who have attained an associate degree or higher to 55% (College Board, 2012b). As of 2009, the national college completion rate for 25- to 34-year-olds is 41.1% (Lee, Edwards, Menson, & Rawls, 2012). “Although 69.1 percent of Asians in that age range have earned an associate degree or higher, only 19.2 percent of Hispanics, 29.4 percent of African Americans and 48.7 percent of white students have done so” (Lee et al., 2012, p. 7). According to the College Board (2012b), in 2008 the College Completion Agenda was initiated in an effort to reestablish “the United States’ position as a global leader in education and ensuring its future prosperity” (p. 10). The goals were determined to be increasing the level of rigor, equitable collegiate academic exposure, and stressing science, technology, engineering, and math, also referred to as STEM education. The College Completion Agenda still grapples with the issue of equity in respect to preventing certain students who possess high academic potential from benefiting from the opportunity that Advanced Placement (AP) courses offer (College Board, 2012b). According to Mollison (2006), the average number of AP tests taken by high school students has changed nominally, increasing to 1.8 tests per examinee in 2005 from 1.7 tests per examinee in 1956. For a student, completing an AP course might serve as the affirmation necessary to continue with post secondary education. In 2001, almost 65% of 16- to 24-year-old White students enrolled in college in 2001, 55% of Blacks, and 50% of Hispanics decided to enroll (Bowen, Kurzwell, & Tobin, 2005). Relating college enrollment to income shows an even more significant disparity. Students from families with an income in the top quartile enroll in post secondary education by 82%, while only 54% of students from families with an income in the bottom quartile attend college (Bowen et al., 2005).

Sharing the same mission of increasing college readiness as well as providing equitable opportunities for all students, Public Law 107-110 of the 107th Congress of the United States, known as the No Child Left Behind Act (NCLB; 2002), was created. Among the issues addressed by NCLB are improving the academic achievement of the disadvantaged which is addressed through the Title I Program, and the AP programs in existence throughout the nation’s high schools (NCLB, 2003).

Regardless of the ongoing debate over the effectiveness of NCLB, its expectations and goals are clearly presented. One of the four principles of NCLB is to raise academic achievement to better prepare college-bound students (Goldrick-Rab & Mazzeo, 2005). High schools across the United States are faced with creating programs and courses that are rigorous and encompass critical thinking and processing skills so students are ready for the challenges of college coursework on completion from high school. Casement (2003) gives a brief history of the development of the AP program from a small elitist program to one accessible to almost all high school students. The AP program began almost 60 years ago so the brightest high school seniors could take college-level courses. By the 1980s, the AP program expanded so high school sophomores, juniors, and seniors were able to take courses in 19 different subject areas. In the last 10 to 12 years, the College Board has placed an emphasis on recruitment of underrepresented minority groups in an effort to ensure equity to all students (Casement, 2003).

Taking into account the structure and level of rigor of AP courses, past studies have shown the positive effect of the AP program on college readiness while providing equitable opportunities for all students. It is through the utilization of the AP framework that the researchers address methods to increase the number of students participating in AP courses and exams, paying special attention to the recruitment of minorities and economically disadvantaged students.

College Readiness

College readiness in upper level high school students can be something very subtle, mutable, and difficult to determine, because college readiness has numerous indicators and characteristics. Students may have a high level of cognitive abilities, but lack the self-regulation necessary to put those skills to most effective use. Students who are able to focus on detailed projects until completion but have limited time management strategies may miss deadlines even though they work hard to complete assignments. College readiness refers to the ability to successfully assimilate into a college culture and manage the academic rigor related to college coursework (Conley, 2007). Students who are successful in college need to be independent thinkers and self-reliant learners. Conley (2007) states that most college instructors expect students to make inferences, interpret data, analyze documents, support arguments with evidence, be able to solve complex problems, draw conclusions, conduct independent research, and ruminate on what they are being taught. Indicators of college readiness include problem-solving skills, critical thinking skills, research skills, critical reading skills, and expository writing skills (Conley, 2007). The obvious need is for the high school curriculum to become more aligned with what is expected of first year college students. Currently, according to Goldrick-Rab and Mazzeo (2005), NCLB is “more likely to have consequences for college preparation if its reforms are extended through the secondary level” (p. 121).

NCLB

The American federal government passed an elementary and secondary school reform in 2001 called the NCLB which required schools to close achievement gaps between majority and minority students (Goldrick-Rab & Mazzeo, 2005). Understanding that a well-prepared student will have a better chance at college success, Goldrick-Rab and Mazzeo (2005) believe “providing all students the opportunity to engage in upper-level coursework and providing support to perform well in those classes and remain in school will likely enhance college participation” (p. 108). This suggestion that school accountability under NCLB might close gaps in college readiness is dependent on how changes are administered by the states, and, if such measures as “tracking” and “ability grouping” are eliminated in an effort to raise expectations, because these measures sort students into various curriculum sequences (Goldrick-Rab & Mazzeo, 2005). Under the provisions of the NCLB, every public school throughout the United States is held to a higher level of accountability. Participation rates are now analyzed closer than ever. This is in place because, in prior years, some schools abused the exemption rule by excessively excluding students in certain subgroups from participating in testing. The debilitating effects are obvious when no assessment data on students in these subgroups is existent.

Preliminary Scholastic Aptitude Test (PSAT)/National Merit Scholarship Qualifying Test (NMSQT)

The PSAT/NMSQT provides 10th-grade students a chance to take a test similar to the Scholastic Aptitude Test (SAT) and offers 11th-grade students the opportunity to compete for National Merit Scholarships (Clement, De Rose, & Sutton, 2005). The PSAT is similar in design to the SAT, without an essay section, reducing the time of the PSAT by 1 hr. The PSAT consists of two verbal sections, two mathematics sections, and a writing skills section for a total time of 2 hr and 10 min (Camara & Millsap, 1998). Vaughn (2010) views the PSAT as an important counseling tool and precursor to higher education, noting that the PSAT measures skills which are developed over numerous years, both in and out of school. These skills are verbal reasoning, critical reading, math problem solving, and writing (Vaughn, 2010). In addition, Camara and Millsap (1998) have conducted research which suggests PSAT test results can be used to determine students who may enjoy a high rate of success in AP Program coursework. Their study showed PSAT data were moderately correlated with AP Exam scores. Ewing, Camara, and Millsap (2006) replicate this earlier study to validate use of PSAT to identify AP students. Clement et al. (2005) believe the PSAT is a good indicator of AP scores because the PSAT assesses the specific skills needed for success in college-level coursework. Conley (2007) believes exams must be aligned with successful college standards. Mayer (2010) also notes that the International Baccalaureate Diploma Program and the AP Programs have gained national recognition as schoolwide reform programs in recent years.

AP Coursework and Exams

The AP Program allows high school students to study advanced-level coursework and take AP Examinations, demonstrating proficiency and earning college credit for passing scores of 3, 4, or 5 (Ewing, 2006), although some colleges do not award credit for a score of 3. This program has grown since its 1956 inception (Mollison, 2006) as “a way of accelerating and fortifying the education of the nation’s future leaders in anticipation of cold war national security demands” (p. 34). Although the idea may have originally been to shorten college years in the anticipation of wartime concerns, the College Board AP program has grown to cover 35 subjects, with one fourth of 2004 high school graduates having taken an AP course (Mollison, 2006). Vaughn (2010) believes that the widespread increases in AP courses confirms a need to elevate academic rigor in high schools, adding that students who take AP courses and exams have a better understanding of college and university work. For students to be prepared to take AP Examinations, schools and districts need to not only focus on the quality and rigor of instruction in AP courses but also create prepared students prior to their enrollment in these courses (Dougherty, Mellor, & Jian, 2006).

Recruitment of Students in AP Courses

A challenge to the future success of the AP Program is recruitment of students who will thrive from the rigor of the program and successfully take AP Exams. The expansion and availability of PSAT test-taking help may recruit more students for AP Programs, according to Vaughn (2010). The College Board suggests all juniors take the PSAT, but sophomores taking the PSAT can see an increase in scores by their junior year. Vaughn sees this as an incentive for students to gauge mastery of skills needed to be successful in AP courses, although he agrees with Camara and Millsap (1998) that PSAT scores should not be used as the only or primary factor in determining success in AP courses. Vaughn’s study of an urban school district trying to expand PSAT test-taking to help identify students who may be successful in AP courses suggests PSAT scores can play a larger role in recruitment. The College Board has two tools to help identify prospective AP students: The AP Potential tool and the PSAT/NMSQT Summary of Answers and Skills (SOAS) Report (Ewing et al., 2006). Both tools can be used to assist in creating potential class rosters of students who have a high probability of success in AP courses. To be able to utilize the tools to maximum advantage, schools should make sure all students have access to PSAT tests. Palin (2001) also encourages schools to look at Grade Point Averages (GPAs) when considering recruitment measures.

Underrepresented Student Populations

When discussing the relevance of participation in AP programs in high schools by underrepresented student populations, consideration must be given to research done by Flowers in 2008. He used human capital theory to make the case for the positive correlation between participation in AP programs and labor market outcomes by race. His findings indicated there was evidence to support that AP participation in high school was associated with higher college entrance exam scores, GPAs in college, post graduate success, and income. While there were significant gains in scores among White and Asian populations, there was a noticeable improvement in gains in scores among Hispanics and African Americans as well, suggesting that AP programs are a valuable resource in the human capital potential of all students (Flowers, 2008).

Broward County Public Schools in Ft. Lauderdale, Florida, the nation’s sixth-largest school district with a diverse population, was the site Clement et al. (2005) chose to conduct a study using PSAT scores as a predictor for AP Programs. The PSAT was found to have a positive correlation with all AP Exam scores except Spanish language. Clement et al. also find that the PSAT was a more accurate predictor of AP Exam success than the Florida Comprehensive Assessment Test (FCAT). Although this study found the PSAT to be a predictor for AP Programs, it did not look at minority and socioeconomic status (SES) in particular and was located in the state of Florida, rather than the state of Texas.

In 1986, 85 of the 129 Garfield students who took the AP Calculus exam passed. Lichten and Wainer (2000) credit their teacher, Jaime Escalante, with remarkable recruitment, impressive teaching, and an educated base of student ability. These students were from the inner city and were primarily Latino. These students were recruited to take the AP Calculus exam even though some had PSAT scores below 60. Dougherty et al. (2006) state that in the last 10 years, AP incentive programs have been part of the budget for districts with predominately low-income and minority students and these districts have enjoyed greater participation by their students. These programs are instrumental in meeting the college readiness needs for underrepresented populations.

Despite the presence of highly qualified teachers and impressive AP programs, researchers need to recognize that some factors associated with enrollment in AP courses have more to do with social and cultural issues rather than academic ability. Walker and Pearsall (2012) describe some of the social, cultural, and economic factors involved that exist as barriers to Latino students when considering taking AP coursework. In a qualitative study focusing on a suburban high school in the Western United States, interviews from students and parents indicated several interesting findings (Walker & Pearsall, 2012). Significant factors inhibiting Latino enrollment in AP courses, gathered from student interviews, indicated peer relations and peer pressure, realization that college was not going to be financially possible and thus contributing to loss of motivation, fear of not doing well in AP courses, and feeling that they do not fit in to AP classrooms. Parents interviewed described the most significant factor limiting AP enrollment was cost. They were afraid high school program costs would be too high and did not have a good understanding of college financial aid (Walker & Pearsall, 2012). The Walker and Pearsall study also presented some sociocultural factors that impact achievement among Latino students. The biggest factor, as described by students and parents, was peer teasing. There was fear among the Latino students interviewed that they were not supposed to do better and be more successful than their family, friends, and neighbors. While most of the focus of this study was detailing the barriers inhibiting Latino students from taking AP courses, the researchers also found there were some factors that actually encouraged Latino participation in AP classes. Some of the factors listed by parents and students were the accessibility of parent support, and involvement as well as support from teachers and peers (Walker & Pearsall, 2012).

Purpose of the Study

There is a current push within school districts to increase the level of participation in AP classes to help students be more prepared for college on graduation. This push includes increasing the number of minority subpopulations and low-income students taking AP courses and ultimately, the AP Exams associated with those courses. If PSAT scores were found to significantly predict AP test performance, school districts may be encouraged to use PSAT scores as a less-biased method for identifying students who could benefit from AP class participation, which could result in more minority and low-SES students taking AP-level coursework.

Method

Participants

Purposive sampling was used by researchers to gather data for this study. Study participants consisted of high school students who had taken the PSAT and AP exams in Calculus, Biology, and/or U.S. History. All participants attend high school in the suburban setting surrounding a large metropolitan city in the Southeastern United States.

This study consisted of students from two fairly diverse school districts. District A has approximately 38,250 students. The ethnic breakdown of students in District A includes 17.9% Hispanic, 8.5% African American, and 25.4 % who were classified as Economically Disadvantaged. District B has approximately 21,097 students. The ethnic distribution of students in District B includes 54.7% Hispanic, 16.5% African American, and 65.4% who were classified as receiving free and reduced-price lunch. All demographic and ethnicity data were gathered from the Academic Excellence Indicator System (AEIS) located on the Texas Education Agency website. There were a total of 1,502 students sampled for this study. Of the 1,502 students, 197 students were Hispanic or African American, and 147 were classified as receiving free and reduced-price lunch.

Measure

The study undertaken by the researchers was quantitative and consisted of a non-experimental regression-based research design. There are two operational definitions that must be clarified for the purposes of this study. When the researchers discuss low SES, this means that the students qualify to receive free or reduced-priced lunch. Another definition to note is the term success when considering AP exam scores. For the purposes of this study, AP exam success will consist of the student scoring a three or higher on the designated exam.

Research Procedures/Data Collection

Prior to beginning research, Institutional Review Board (IRB) approval was secured. The specific procedures that followed were obtaining PSAT scores and AP test scores archival data from participating districts, collecting ethnicity data from participating districts, and collecting SES information (students on free and reduced-price lunch) from participating districts.

Instrumentation

The PSAT/NMSQT has been taken each year by more than 3.5 million students in approximately 23,000 schools since established in 1959 (College Board, 2012a). The instruments for this research study are the PSAT, also known as the National Merit Scholarship Qualifying Test (PSAT/NMSQT) and the following AP Program exams: AP Calculus, AP American History, and AP Biology.

The PSAT/NMSQT is divided into five subsections consisting of three verbal sections and two sections of mathematics. Critical reading (CR) sections alternate with math (M) sections, finishing with a writing skills (W) section. The timed test is 2 hr, 10 min. Internal reliability estimates for each section, according to the College Board, are 0.86 to 0.88 on critical reading, 0.84 to 0.87 on mathematics, and 0.84 to 0.89 on writing skills (College Board, 2012a). Content validity and construct validity are confirmed by experts in the field who have developed and reviewed the PSAT/NMSQT (College Board, 2012a).

The College Board currently has 34 AP Exams available to high school students. Three of these were the focus of our study. The AP Calculus, AP U.S. History, and AP Biology Exams are developed with content reviewed by the College Board Development Committee. This committee is made up of AP high school teachers and college and university professors. Each AP Exam, with the exception of AP Studio Art, consists of two sections. The multiple-choice sections are created by university faculty who teach courses which match the AP course (College Board, 2012a). Some multiple-choice questions are recycled each year to ensure “the statistical reliability of each AP Exam from year to year” (para. 2). The free response section consists of unique questions which are used only once. The College Board (2012a) reports that any question chosen by the Development Committee must be reviewed and revised numerous times, taking up to 2 years. Guidelines for difficulty and item discrimination are used to “ensure that each AP Exam will be of appropriate difficulty for the test-taking population and that each exam will distinguish among students with different levels of knowledge, skills, and abilities” (College Board, 2012a, para. 4). Each AP Exam must have “a coherence and internal consistency that reflects its purpose of assessing the students’ preparation in the subject” (College Board 2012a, para. 6). Each of the 34 AP exams must meet both predetermined content and statistical specifications, according to the College Board (2012a). Reliability numbers for the three AP Exams selected for this study are as follows, according to Bridgeman, Morgan, and Wang (1996) of the Educational Testing Service:

U.S. History AP Exam, composite score of 0.83

AP Biology, composite score of 0.93

AP Calculus, composite score of 0.92

Data Analysis

The data for this research were archival data from two different school districts in the southeastern United States. PSAT/NMSQT scores as well as demographic data and SES data were collected and arranged into a Microsoft Excel (2010) spreadsheet. Data were further divided between the five subsections of the PSAT/NMSQT. The researchers analyzed the data using the SPSS software program (SPSS Inc, 1999).

Multiple regression was used for statistical analysis of data, based on the fact that there were several predictors, or sub factors, of each independent variable as it related to the criterion variable. The dependent variable in this study was AP Exam scores. The predictor variables were the PSAT/NMSQT scores and demographic variables.

Results

Three multiple regression analyses were conducted to evaluate how well the three components of the PSAT score, ethnicity, and SES predicted AP exam performance. The predictors were the critical reading component (CR score), the math component (M score), the writing component of the PSAT test (W score), the student’s ethnicity (African American and Hispanic students vs. all others), and the student’s eligibility to receive free or reduced-price lunch, whereas the criterion variable was the student’s performance on the AP exams in Biology, U.S. History, and Calculus AB.

First, researchers considered the descriptive statistics of the three different multiple regression analyses. Table 1 identifies the descriptive statistics of all three regression analyses. There were 472 students who took the Biology AP exam (M = 2.89, SD = 1.51), 608 students who took the U.S. History AP exam (M = 3.20, SD = 1.19), and 923 students that took the Calculus AB AP exam (M = 3.32, SD = 1.45).

Table 1.

Descriptive Statistics for Students Who Took the Biology, U.S. History, and Calculus AB AP Exams.

Variables	M	SD	n
Biology score	2.89	1.51	472
CR score	57.77	10.72	472
M score	60.59	10.00	472
W score	57.20	23.35	472
Ethnicity	.20	.40	472
Free/reduced-price lunch	.10	.30	472
U.S. History score	3.20	1.19	608
CR score	56.21	10.00	608
M score	58.36	10.02	608
W score	54.99	9.74	608
Ethnicity	.29	.45	608
Free/reduced-price lunch	.15	.36	608
Calculus AB score	3.32	1.45	923
CR score	59.37	10.34	923
M score	64.94	8.40	923
W score	58.45	9.84	923
Ethnicity	.13	.33	923
Free/reduced-price lunch	.05	.21	923

Note. CR = critical reading; M = math; W = writing skills.

Correlation coefficients were computed among the five predictor variables and their relationship with the performance scores on the AP Biology, U.S. History, and Calculus AB exams. A p value of less than .05 was required for significance. Table 2 shows that all of the predictor variables were significant (p < .05) although some had a negative direction. The predictor variables Ethnicity and Free/Reduced-Price Lunch had a negative correlation indicating that as students moved from the all other ethnic classification to African American and/or Hispanic classification or from non–free and reduced-price lunch to free and reduced-price lunch classification, the mean scores decreased for all AP exams.

Table 2.

Correlations of Student AP Exam Scores Among the Five Predictor Variables.

	Biology score	U.S. History score	Calculus AB score
CR score
Pearson correlation	.66*	.61*	.45*
Significance (two-tailed)	.00	.00	.00
M score
Pearson correlation	.61*	.50*	.50*
Significance (two-tailed)	.00	.00	.00
W score
Pearson correlation	.21*	.53*	.44*
Significance (two-tailed)	.00	.00	.00
Ethnicity
Pearson correlation	−.23*	−.17*	−.18*
Significance (two-tailed)	.00	.00	.00
Free/reduced-price lunch
Pearson correlation	−.16*	−.08*	−.09*
Significance (two-tailed)	.00	.02	.00

Note. AP = Advanced Placement; CR = critical reading; M = math; W = writing skills.

p < .05.

Before looking at the linear combinations, the researchers reviewed multi-collinearity issues looking at both Tolerance and the variance inflation factor (VIF). None of the analyses were outside the acceptable limits for multi-collinearity.

The linear combination of the five predictors was significantly related to the Biology AP exam score, F(5, 466) = 84.46, p < .05. The sample multiple correlation coefficient was .69, indicating that 48% of the variance of the Biology AP exam score in the sample can be accounted for by the linear combination of the predictor variables. In Table 3, we present the coefficients of each of the predictors to see which are individually significant to AP Biology exam scores. The only predictors that had a significant effect (p < .05) on Biology AP scores were the CR score and the M score, while the other predictors were not significant.

Table 3.

Multiple Regression Analysis of PSAT Predictors on AP Biology Exam.

Model	Unstandardized coefficients (B)	Significance
Constant	−3.450	.000
CR score	.067	.000*
M score	.044	.000*
W score	−.003	.170
Ethnicity	−.097	.493
Free lunch	.257	.169

Note. PSAT = Preliminary Scholastic Aptitude Test; CR = critical reading; M = math; W = writing skills.

p < .05.

In the second multiple regression analysis, the linear combination of the 5 predictor variables also showed that they were significantly related to AP U.S. History scores, F(5, 602) = 79.86, p < .05. The sample multiple correlation coefficient for this analysis was .63 which indicates that 40% of the variance of the AP U.S. History exam in the sample can be accounted for by the linear combination of the predictor variables. Table 4 illustrates that the CR scores, the M scores, and the student’s classification of SES are individually significant (p < .05) whereas the W score and the student’s ethnicity had little weight to the overall significance in this analysis.

Table 4.

Multiple Regression Analysis of PSAT Predictors on AP U.S. History Exam.

Model	Unstandardized coefficients (B)	Significance
Constant	−1.667	.000
CR score	.055	.000*
M score	.019	.001*
W score	.012	.084
Ethnicity	.067	.475
Free lunch	.296	.010*

Note. PSAT = Preliminary Scholastic Aptitude Test; AP = Advanced Placement; CR = critical reading; M = math; W = writing skills.

p < .05.

In the final multiple regression analysis, the linear combination of the five predictor variables also indicated that they were significantly related to AP Calculus AB exam scores, F(5, 917) = 79.32, p < .05. The sample multiple correlation coefficient for this analysis was .55, indicating that 30% of the variance of the AP Calculus AB exam in the sample can be accounted for by the linear combination of the predictor variables. In Table 5, we present the coefficients of each of these predictors to see which were individually significant to the AP Calculus exam scores. This analysis suggests that all of the predictor variables are significant (p < .05) except for the variable indicating the student’s SES.

Table 5.

Multiple Regression Analysis of PSAT Predictors on AP Calculus Exam.

Model	Unstandardized coefficients (B)	Significance
Constant	−2.729	.000
CR score	.019	.005*
M score	.058	.000*
W score	.020	.004*
Ethnicity	−.355	.005*
Free lunch	.270	.175

Note. PSAT = Preliminary Scholastic Aptitude Test; AP = Advanced Placement; CR = critical reading; M = math; W = writing skills.

p < .05.

Discussion

Explanation of Results

This study examined the relationship between variables that could predict AP Exam outcomes. The five predictor variables were Critical Reading (CR) Score, Math (M) Score, Writing (W) Score, Ethnicity, and Free/reduced-price lunch status.

Correlations were analyzed among the five predictor variables and their relationship to performance on AP Exam scores in AP Biology, AP U.S. History, and AP Calculus AB. PSAT CR Score, M Score, and W score had a significant relationship with AP Exam scores for AP Biology, AP United States (U.S.) History and AP Calculus. Ethnicity and Free/reduced-price lunch status were also significantly correlated with all three AP results. Mean scores on the three AP Exams tended to decrease as ethnicity classification moved toward African American and Hispanic students and from non–free and reduced-price lunch to free and reduced-price lunch classification. All findings were in accordance with the literature reviewed.

The linear combination of the five predictor variables was significantly related to AP Biology exam scores. AP Biology results were impacted by both PSAT CR Scores and both M Scores, but there was no significant effect on AP Biology results by W Score, ethnicity, or free and reduced-price lunch status. According to the College Board (2012b), the AP Biology course is designed to develop advanced inquiry and reasoning skills, such as data collection design, data analysis, and application and connection of math concepts. The manner in which this course is presented does not appear to focus on writing skills, but rather laboratory-based skills. That ethnicity and free and reduced-price lunch status did not impact AP Biology results may indicate that the reasoning-based nature of AP Biology does not differentiate between ethnicity groups and SES groups (College Board, 2012b).

The linear combination of all five predictor variables also showed a significant relationship to AP U.S. History scores. CR Scores, M Scores, and student SES were significant predictors. W Score and ethnicity were not significant in the second regression analysis. Ethnicity factors may not have proved significant based on the sample of students as representative within school districts which may not have an equalized ethnic distribution.

The linear combination of the five predictor variables showed a significant relationship to AP Calculus AB Scores in the final regression analysis. All of the predictor variables are significant in this analysis, with exception of SES. That ethnicity was significant in AP Calculus scores may be understood more fully by examining what Lichten and Wainer (2000) call the Aptitude-Achievement Function, or the correlation between PSAT math scores and AP Calculus scores, which is a stronger indicator of success on AP Calculus exams than classroom grades are within all ethnic groups.

Limitations of the Study

There are some limitations for making generalizations with the results of this study. Because this study was conducted with data from only two school districts, the results can only be generalized to a pocketed region of southeast Texas. Also the two districts examined encouraged AP Calculus for all students with the necessary academic background, regardless of ethnicity or SES. Researchers could not assume they would find similar results when studying PSAT scores, ethnicity, and SES of students as predictors of AP exam performance in other regions/districts of Texas. Another limitation of this study was the stark contrasts of demographics between the two districts. With a more expansive study comprised of more Title I districts, researchers could possibly make more accurate generalizations with the results. This factor could be the focus of further research. While the participant group was diverse, it would have been useful to this study to have had a higher percentage of minority and low-SES students. SES effects may have been partially ameliorated because the participating districts both have cost reduction plans in place for the fees associated with the AP exams that are offered.

Implications for Further Research

That ethnicity and SES were significant factors in two of the three AP Exams studied merits concern. Three challenges present in education consist of closing the achievement gap, increasing college readiness, and increasing the number of high school students enrolling in colleges or universities, especially when it comes to underrepresented populations. A study by Flowers (2008) shows evidence that underrepresented student populations who participated in AP programs were linked to higher college GPAs and work income after college. While PSAT results can be a tool to help in predicting enrollment in AP courses, Dougherty et al. (2006) point out that the schools need to continue to encourage participation in AP coursework in low-income and minority students.

The College Completion Agenda seeks to increase the percentage of 25- to 34-year-olds who have an associate degree or higher to 55% by 2025 to reestablish the United States as a leader in education, and secure our position as a prosperous global superpower. In keeping with this goal, the education of everyone in the United States is imperative. To attain this goal, public, secondary education has the potential to significantly increase college readiness with a special focus on underrepresented populations. Promoting AP programs to low-income and minority students presents an opportunity for schools to address the three challenges previously mentioned while targeting the goals of the College Completion Agenda.

There is a significant disparity between college enrollment and family income (Bowen et al., 2005) and students from low-income groups may be unable to purchase or borrow costly AP study materials outside of what is offered at school (College Board, 2012b). The College Board (2012b) states that schools serving significant populations of minority and low-income students need support. Measures like the AP Potential website and grades should be utilized to target, encourage, and support underrepresented minorities and low-SES students. The approach should be deliberate, and counseling of students and their parents about the opportunities that participating in AP courses and AP exams can mean to the lives of the students. Having students and parents realize that a college education is feasible and attainable for them, and explaining measures like the PSAT scores, AP Potential, and grades can serve as proof that they do have the potential to be successful in AP courses, and in college.

Because the data show the PSAT W Score was only significant to the AP exam for Calculus AB, further insight into the improvement of writing skills and their impact on the Calculus AB exam should be explored. Our data show that schools wishing to increase participation and increase the number of students attaining a score of 3 or higher on the Calculus AB exam should focus on targeting the improvement of writing skills while keeping the Calculus curriculum and preparation for the AP exam unchanged to see if targeting the improvement of writing skills results in an increase in the number of students participating in AP Calculus and achieving a score of 3 or higher on the Calculus AB exam.

In addition, further research could explore the possible links to the processes required for writing, and the processes required for math. One possible link might prove a relationship between the need for a higher level of critical thinking and reasoning required in both, thereby explaining the results of the significance of the PSAT W score and the Calculus AB exam score, as well as what impact would occur with the other exams as well.

Based on the findings, researchers can expand the scope by replicating the study so that samples consist of minorities and low-SES students from non–Title I schools while looking at the same independent variables. The results can then be compared with minorities and low-SES participants from Title I schools looking at the same independent variables. The results might throw light on different challenges faced by non–Title I schools versus Title I schools in targeting underrepresented minorities and low-SES students so those challenges may be addressed.

Finally, the researchers believe that increased recruitment for AP course enrollment of diversity students based on PSAT scores is necessary. The College Board (2012b) suggests that schools can use the AP Potential tool to identify minority and low-income students who are likely to succeed in AP courses. Schools and school districts can use AP Potential results to invite students and parents from underserved backgrounds to sessions of schools’ AP night. The College Board (2012b) suggests having diversity students who have had success in AP programs speak to students and parents. Schools must effectively communicate that AP exam registration will be made cost-effective for low-income students with potential for success, removing the cost barrier. Once students and parents see possibilities as realities, diversity recruitment and enrollment in AP programs will escalate.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Author Biographies

Cristianne C. Richardson currently serves as a secondary educator in Clear Creek Independent School District. She has earned an MS in education from the University of Houston–Clear Lake and is currently completing an EdD in educational leadership. She has developed district curriculum and teacher training for independent study and advanced placement English Language Arts, having worked with both gifted and high-achieving students for 8 years.

Alejandro Gonzalez is an administrator in a suburban high school. He received his MS in education administration from Texas A&M University, Kingsville, and is currently working on an EdD in educational leadership. He has been in education for 15 years and has worked in Urban, Suburban, and Rural school districts as both a teacher and administrator.

Lonnie Leal currently serves as a campus administrator in Clear Creek ISD while working on an EdD in educational leadership at the University of Houston–Clear Lake. Throughout his career, he has served as a school teacher, assistant principal, and principal at the elementary (K-4), middle school (Grades 5-6), and intermediate levels (Grades 7-8). He has also served as a director of human resources.

Mary Z. Castillo currently serves as a family involvement coordinator. Her experience includes 9 years of working with English Language Learners and Title I students at different capacities. She has previously been a Bilingual/English as a Second Language (Bil/ESL) teacher, a district Bil/ESL curriculum specialist, and a district Bil/ESL coordinator.

Carol A. Carman completed her PhD in educational psychology with an emphasis in statistics, measurement, and evaluation from the University of Kansas in 2005. She is currently an associate professor of educational foundations at the University of Houston–Clear Lake. Her research interests include definitions of giftedness and identification methods.

References

Bowen

W. G.

Kurzwell

M. A.

Tobin

E. M.

(2005). Equity and excellence in American higher education. Charlottesville: University of Virginia Press.

Bridgeman

Morgan

Wang

(1996). Reliability of advanced placement examinations (U.S. Department of Education Research Report, ED400331, pp. 2-31). Princeton, NJ: Educational Testing Service.

Camara

Millsap

(1998). Using the PSAT/NMSQT and course grades in predicting the success in the Advanced Placement Program (Research Report No. 98-4). New York, NY: College Board.

Casement

(2003). Declining credibility for the AP program. Academic Questions, 16(4), 11-25.

Clement

R. W.

De Rose

D. S.

Sutton

C. O.

(2005). Do PSAT scores predict student success in Advanced Placement Programs? ERS Spectrum, 23(1), 31-41.

College Board. (2012a). Content reliability and validity. College Readiness Pathway. Retrieved from http://www.collegereadinesspathway.com/content_validity.html

College Board. (2012b). The 8th Annual AP Report to the Nation. Reston, VA: Author.

Conley

D. T.

(2007). Redefining college readiness. Eugene, OR: Educational Policy Improvement Center.

Dougherty

Mellor

Jian

(2006). The relationship between Advanced Placement and college graduation. National Center for Educational Accountability. Retrieved from http://www.nc4ea.org/resources/library/?startRow=26&nextNID=FBE98147-6F50-4C42-AACF0AD99D28DDE0

10.

Ewing

(2006). The AP program and student outcomes: A summary of research (College Board Research Note RN-29). New York, NY: The College Board.

11.

Ewing

Camara

W. J.

Millsap

R. E.

(2006). The relationship between PSAT/NMSQT scores and AP examination grades: A follow-up study (College Board Research Report No. 2006-1). New York, NY: The College Board.

12.

Flowers

L. A.

(2008). Racial differences in the impact of participating in advanced placement programs on educational and labor market outcomes. Educational Foundations, 22(1-2), 121-132.

13.

Goldrick-Rab

Mazzeo

(2005). What No Child Left Behind means for college access. Review of Research in Education, 29, 107-129.

14.

Lee

J. M.

Edwards

Menson

Rawls

(2012). The college completion agenda 2011 progress report executive summary. Reston, VA: College Board.

15.

Lichten

Wainer

(2000). The aptitude-achievement function: An aid for allocating educational resources, with an Advanced Placement example. Educational Psychology Review, 12, 201-228.

16.

Mayer

A. P.

(2010). Factors influencing the implementation of an international baccalaureate diploma program in a diverse urban high school. Journal of Advanced Academics, 22, 78-105.

17.

Mollison

(2006). Surviving a midlife crisis: Advancement Placement turns fifty. Education Next, 6(1), 34-39.

18.

No Child Left Behind Act of 2001, Pub. L. No. 107-110, § 115, Stat. 1425 (2002).

19.

No Child Left Behind Act of 2001, 20 U.S.C.A. § 6301 et seq. (West 2003).

20.

Palin

R. J.

(2001). PSAT and AP success. Organization of American Historians Magazine of History, 15(3), 55-56.

21.

SPSS Inc. (1999). SPSS Base 10.0 for Windows user’s guide. Chicago, IL: Author.

22.

Vaughn

(2010). Reform in an urban school district: The role of PSAT results in promoting Advanced Placement course-taking. Education and Urban Society, 42, 394-406.

23.

Walker

S. A.

Pearsall

L. D.

(2012). Barriers to advanced placement for Latino students at the high school level. Roeper Review, 34, 12-25. doi:10.1080/02783193.2012.627549